Optimizing CD34+ cell doses for large-scale translational, single donor humanized mouse experiments

Optimizing CD34+ cell doses for large-scale translational, single donor humanized mouse experiments | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Article Optimizing CD34+ cell doses for large-scale translational, single donor humanized mouse experiments Dirk Busch, Julius Schuetz, Sarah Dötsch, Georg Schmidt, Fabian Mohr, and 8 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-6382310/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted You are reading this latest preprint version Abstract Humanized mice are a powerful tool in preclinical assessment of novel therapeutics. However, generation of large homogenous cohorts is believed to be hampered by the limited number of human stem cells retrievable from individual donors, like from umbilical cord blood specimens. Here, we present a streamlined CD34+ cell isolation method that improves yield and purity compared to standard isolation protocols, enabling more efficient use of umbilical cord blood donations. Enriched CD34+ stem cell preparations yield robust humanization of NSG-SGM3 mice, even with very low input doses, down to a few thousand cells. Using our findings to effectively humanize large cohorts of mice (in this study up to 30 mice from a single stem cell donor), we resolve differences in cytokine release syndrome for two different CAR-T cell products within single-donor settings. These optimizations facilitate larger or smaller humanization experiments, with reduced economical and logistical burden for critical translational research. Biological sciences/Biological techniques/Experimental organisms/Model vertebrates/Mouse Health sciences/Medical research/Experimental models of disease Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Introduction Humanized mice are an important tool for preclinical studies of human diseases where the immune system plays a central role, such as cancer. Through the introduction of human cells, they harbor a human immune system within a small rodent model. Thus, this model allows humanized readouts that are not hindered by the many constraints and differences between the murine and human immune systems, such as tolerance and elimination. Despite the strength in evaluating different research topics, for example, anaphylaxis, virus infections, or immunotherapies, humanized mice unfortunately remain a niche tool, as they are restricted primarily by their complexity in generation and readout, as well as small cohort sizes [ 1 ]. Humanization models are based on the injection of human cells into highly immunocompromised and xeno-tolerant mice strains (e.g. NSG, NOG). While injecting peripheral mononuclear cells (PBMCs) allows to establish a human immune system within the recipient mice, it leads to induction of graft–versus-host disease (GvHD) demise in 2–3 weeks, allowing only limited and strongly biased analysis. A more elegant animal model is generated by only injecting CD34 + stem cells. After engraftment, these stem cells allow the reconstitution of human, xeno-tolerant, immune cells. Early work showed a B-cell biased reconstitution following CD34 + cell transfer in NSG mice, leading to the development of NSG mice transgenic for human cytokines. The NSG-SGM3 mouse – transgenic for human stem cell factor (SCF), granulocyte-macrophage colony-stimulating factor (GM-CSF) and interleukin 3 (IL3) – is one such model and has significantly improved myeloid cell reconstitution [ 2 ]. Especially when examining immune responses in toto , myeloid cells play an important role, as seen by findings from Norelli et al. , which show that the deadly cytokine release syndrome (CRS) in CAR T cell transfer is predominantly driven by “bystander” monocytes, only seen in humanized NSG-SGM3 mice [ 3 ]. Especially CD4 + CAR T cells seem to be the driver of CRS in humanized NSG-SGM3 mice[ 4 ]. CD34 + cells can be isolated from adult bone marrow, mobilized peripheral blood, fetal liver cells or umbilical cord blood (UCB) and contain human stem and progenitor cells. Adult sources often only result in transient, partial and low humanization levels [ 5 , 6 ] and thus, are a difficult tool for achieving stable and comprehensive humanized mouse models. In comparison, fetal liver and UCB cells allow full lineage differentiation and long-term reconstitution. Fetal liver cells are problematic to obtain, thereby also giving rise to only very limited options. In contrast, UCB is considered a by-product of birth, thus it is more accessible compared to other stem cells sources and still contains potent CD34 + stem cells. Therefore, UCB is the preferred choice for most researchers. The most common CD34 + cell isolation method from UCB is magnetic bead-based isolation, which relies on magnetic beads conjugated to anti-CD34 antibodies to positively isolate hematopoietic stem cells (HSC). However, inconsistencies in yield and variability in purity are observed with this approach. We have previously established fragment antigen binding (Fab)-based isolation protocols, [ 7 ] which were subsequently adapted into a column-based isolation system named traceless affinity cell sorting (TACS) [ 8 , 9 ]. This technology uses Fab multimers (Fab monomers associating to a backbone) that bind surface markers on target cells, with the added advantage of reversibility of the catching reagents (Fab fragments). After the addition of D-biotin, the multimer complex disassembles. Specifically biotin disrupts the binding between the multimer-backbone and the Fab molecules, leaving monomeric Fab-epitope interactions, from which Fab dissociated, yielding Fab-free cells. This reversibility is especially important when isolating cells, such as T cells, where surface markers such as CD3 are critical for cell function. Residual antibodies from magnetic bead-based methods can interfere with cell functionality, but Fab-based isolation mitigates this issue. CD34 is known to be involved in cell adhesion and hence may be affected by residual antibodies from magnetic bead-based isolation protocols, potentially imparting the CD34 + cells functionality [ 10 ]. Additionally, the TACS method offers the advantage of isolating cells directly from whole blood, bypassing density gradient separation and reducing cell loss. This approach is expected to enable higher CD34 + yields than magnetic bead isolation and thus, permit larger humanized mice cohorts. Despite extensive development and improvement in mouse models, a key challenge remains expanding cohort sizes due to the limited availability of CD34 + cells. Previously, researchers pooled CD34 + cells from multiple donors or only used small cohorts per donor, which are both sub-optimal solutions, particularly in comparative experiments. Pooling CD34 + cells from different donors can have unforeseen immunological consequences, and typically, only one infused donor achieves long-term engraftment [ 11 , 12 ]. The issue of limited cohort sizes is aggravated when replicating diseases in humanized mice with specific leucocyte traits, such as human-leucocyte antigen (HLA) or Killer cell Ig-like receptor (KIR) types. Modelling diseases within specific immunological constraints still requires suitable cohort sizes while working with limited HLA- or KIR-type specific CD34 + cells. Under these constraints, researchers require sufficient humanized mice for downstream applications without compromising humanization levels, as low humanization levels can confound results. Here we describe optimal CD34 + cell doses, titrated in the low range, for both excellent humanization levels and large single donor huNSG-SGM3 cohorts. Methods Cell source material UCB was collected from full-term deliveries after maternal informed written consent and kindly provided by the “Klinik und Poliklinik für Frauenheilkunde “of the Technical University Munich (reference no. 617/19S). UCB was collected in UCB blood bags (Macopharma, Germany, Cat.# MSC1200PU) or EDTA tubes (Saarsted, Germany, Cat.# 01.1605). Magnetic bead isolation for CD34 cells All UCB was processed within 6 hours of collection. CD34 + cells were isolated as per the manufacturer’s instruction using the CD34 MACS system (Miltenyi Biotec, Germany, Cat.# 130-046-702). In brief, UCB was diluted 1:1 in phosphate buffered saline (pH = 7.40) (PBS) (Merck, Germany, Cat.# 56064C), layered on 1,077 g/mL Biocoll (Merck, Germany, Cat.# L6715-BC) and centrifuged for 20 min at 800g with no brake. Mononuclear cells were transferred into FACS (PBS 1x (Merck, Germany) containing 0.5% (w/v) bovine serum albumin (BSA) (Sigma-Aldrich, Germany, Cat.# 05470), pH = 7.40) buffer and washed before adding Fcγ blocker and anti-CD34 magnetic beads. After 30 min, CD34 + cells were eluted using the MidiMACS separator magnet and LS columns (both Miltenyi Biotec, Germany, Cat.# 130-042-301 & 130-042-401). 5% of isolated cells were transferred for purity and exact CD34 + cell count for flow cytometry analysis. The remaining 95% of isolated cells were cryopreserved in a 250 µL mix of 10% dimethyl sulfoxide (Merck, Germany, Cat.# 67-68-5) and 90% fetal calf serum (FCS) (GE Healthcare, United Kingdom). Cells were stored in -80°C and/or liquid nitrogen. Traceless Affinity Cell Selection (TACS) for CD34 + cells CD34 + cells were isolated from whole UCB by adapting our previously established protocols [ 9 ]. Following aseptic technique, Poly-Prep chromatography columns (Bio-Rad, USA) were cleaned using 80% (v/v) Ethanol and loaded with 1 mL Streptactin agarose beads (IBA Lifesciences, Germany, Cat.# 6-6350-002) together with 6 mL PBS (Merck, Germany). The agarose beads were left to settle, and columns were stored at 4° C in sterile 50 mL Falcon tubes (Greiner Bio-one, Germany) until use. Immediately before isolation, 45 µg CD34-targeting Fab-fragments (IBA Lifesciences, Germany) dissolved in 1 mL sterile PBS was loaded onto the column to coat the agarose bead bed. After washing with sterile PBS, UCB diluted 1:20 (v/v) with citrate phosphate dextrose solution (Sigma-Aldrich, Germany Cat.# C7165) was loaded directly onto the column. The CD34 + cells were retained via binding to the Fab molecules on the agarose bead surface. After washing with four runs of 10 mL PBS, cells were eluted using 1 mM D-biotin (Merck, Germany, Cat.# 2031) in FACS buffer and collected for further analysis. Cryopreservation was performed in section “Magnetic Bead isolation for CD34 + cells”. Thawing of cryopreserved CD34 cells CD34 + cells were thawed in warm water and immediately transferred into warm RPMI 1640 media (ThermoFisher Scientific, Germany, Cat.#11875093) containing 10% (v/v) FCS (GE Healthcare, United Kingdom). Cells were spun down and resuspended in RPMI containing 10% FCS medium. 5% of the volume was used for flow cytometry analysis, and the remaining cells were rested at 37° C, 5% CO 2 until flow cytometry analysis was done (approximately 1-1.5 hours). After flow cytometry quality control and cell count analysis, the cells were centrifuged down, and resuspended in the desired volume of FCS for the desired cell dose and used for humanization. Generation of HuNSG-SGM3 NOD. Cg-Prkdc scid Il2rg tm1Wjl Tg(CMV-IL3, CSF2, KITLG)1Eav/MloySzJ (NSG-SGM3) mice were purchased from Jackson Laboratory (USA) and housed under specific-pathogen-free (SPF) conditions. We only used female mice housed in groups of 3–6 littermates per cage. Mice within a cage were given the same treatment. At 4 weeks of age, mice were irradiated with 1 Gy from a Caesium-137 source. 16–24 hours later, NSG-SGM3 mice were intravenously injected with 150 µL FCS containing varying doses of CD34 + cells. Cell dose titration using very low, low and high CD34 + cell doses was performed to assess humanization rates as outlined in Supplementary Table 1. The CD34 + cell numbers represent the total living cells per mouse as assessed by flow cytometry. Each mouse only received cells from one donor. After injection, mice were monitored regularly for clinical signs of GvHD (e.g. skin & fur changes, hunched back, excessive weight loss). Assessment of humanization levels Mice were bled via the tail vein every 2 weeks for 12 weeks. Peripheral mouse blood was erythrocyte lysed using Tris-buffered (10% v/v) Ammonium Chloride (90% v/v) pH 7.50) (ACT) (Roth, Germany, Cat.#77-86-1 and 12125-02-9) and stained for various human leucocyte surface markers before flow cytometry analysis described below (see flow cytometry). 12 weeks after CD34 + cell injection, mice were sacrificed by cervical dislocation. Spleen and BM (bone marrow from tibia and femur from each side) were harvested. Spleens were mashed through a 70 µm Nylon-mesh to create a single-cell suspension. BM was extracted by inserting a 25 G syringe and flushing out the BM cells with Dulbecco's Modified Eagle medium (ThermoFisher Scientific, Germany, Cat.# A4192101). A BM single-cell suspension was generated by rigorously pipetting the extracted cells. Single cell suspensions were centrifuged down, and residual erythrocytes in the spleen and BM samples lysed using ACT. Cells were washed and a portion of the single-cell suspensions were used for downstream analysis. Humanization levels were defined as \(\:hum\left(\%\right)=\frac{human\:CD45+}{all\:CD45+}\times\:100\%\) . Flow cytometry Cells were transferred into a V-bottom 96-well plate and incubated at 4° C with various surface antibodies for 20 minutes. 10 µL of whole blood was used for CD34 + analysis in whole UCB and after washing with FACS buffer stained with human anti-huCD34 (4H11; eBiosciences, Germany; Cat.# 11-0349-42) and anti-CD45 (HI30; ExBio, Czech Republic; Cat.# PO-684-T100). The CD34 + cell isolation product was assessed by staining with anti-huCD34, anti-huCD45, anti-huCD3 (UCHT1; eBiosciences, Germany; Cat.# 17-0038-42), anti-huCD19 (J3-119; Beckman Coulter, USA; Cat.# A07770) and anti-huCD33 (WM53; BioLegend, USA; Cat.# 303416). The humanization and reconstitution levels were analyzed through anti-muCD45.1 (A20; Biolegend, USA; Cat.# 110708), anti-huCD45, anti-huCD3, anti-huCD19, anti-huCD33 and anti-huCD56 (TULY56; eBiosciences, Germany; Cat.# 11-0566-42). After surface staining, cells were washed, and 1:1000 (v/v) propidium iodide (Merck, Germany) was added for live-dead discrimination. Additionally, 123 eCounting Beads (Thermofisher, Germany, Cat.# 01-1234-42) were added to each sample for accurate cell counts. The gating strategy is outlined in Supplementary Fig. 1A and B. Cells were acquired on a Cytoflex S flow cytometer (Beckman Coulter, USA), and whole blood flow cytometry was performed on a Cyan ADP flow cytometer (Beckman Coulter, USA). CFC assay A human HSC colony-forming assay, which only allows for myeloid and erythroid development, was used to analyze the composition and clonogenicity of the CD34 + CD45 + cells. A small fresh aliquot of the final isolated CD34 + cells containing approximately 200 cells was resuspended in the CFC media (R&D Systems, USA; Cat.# HSC005) and plated for 14 days. The aliquots were collected from both the magnetic bead and TACS protocols. After 14 days the plates were blinded, and the number and type of colonies were visually scored. Bone histology Femur and tibia of one side were harvested. Excess soft tissues were removed and bones were washed in PBS before being fixed through immersion in 4% PFA. After 48 hours the samples were washed and transferred into 70% (v/v) ethanol until further processing. Tissues were decalcified in 10% EDTA, pH 7.4 at 37°C for 1 week, before being processed in a Leica ASP300S tissue processor, embedded in paraffin wax, and sectioned at 5 µm using a Leica rotary microtome. Slides were stained for H&E using a Leica XL Autostainer. Immunohistochemistry for human-specific anti-huCD45 (Agilent, USA; Cat.# M070101) was performed following antigen retrieval in pH 6.0 10 mM tri-sodium citrate buffer, 0.05% Tween-20, as previously described [ 13 , 14 ]. Briefly, endogenous peroxidase activity was quenched with 3% hydrogen peroxide (Sigma-Aldrich, Australia, Cat.#7722-84-1) for 15 minutes. Non-specific binding sites were blocked with 2% BSA (Sigma-Aldrich, Australia, Cat.# 9048-46-8) for 30 min before overnight incubation with primary antibodies diluted in 2% BSA at 4°C. Normal mouse IgG was used as a negative control. Positive immunoreactivity was detected using the Dako Envision + Dual Link System HRP and the Liquid DAB + 2-component immunohistochemistry visualization system (both Agilent, USA). Cytokine release syndrome (CRS) model Mice were humanized using CD34 + cells from a single donor. 8 weeks after humanization, mice were intravenously injected with 0.5x10 6 CD19 + Raji leukemia tumor (ATCC, UK). One week later, mice were weighed and intravenously injected. 0.8x10 6 CD8 + CAR T-cells or mock transduced CD8 + T cells from the same donor. Mice were examined and weighed daily for signs of CRS. Blood was collected from the tail vein mice at days 3 and 5 after T cell transfer, and serum levels of IL-6 and IL-10 were measured using LEGENDplex immunoassay kit (Multi-Analyte Flow Assay Kit, Biolegend, Germany, Cat.# 740267) per the manufacturer’s instructions. Data analysis Flow cytometry data were analyzed using FlowJo 10 (Becton, Dickinson & Company, USA). Data was stored using Microsoft Excel (Microsoft, USA). Graphs were created, and statistical analysis was performed using GraphPad Prism 9 (GraphPad Software, USA). The manuscript was written in Microsoft Word (Microsoft, USA). Results Donor-to-donor variations exclude many donations on grounds of low CD34 + cell count Isolation of CD34 + stem cells using established protocols leads to extensive cell loss. Current humanization protocols recommend 10 5 CD34 + cells per mouse, and together with a poor yield, this hampers humanized mice cohort sizes (Fig. 1 A)[ 15 , 16 ]. First, we set out to establish how many CD34 + stem cells could reliably be collected using established magnetic bead isolation protocols (Fig. 1 A). The cord blood collections were performed immediately following birth. UCB cells were not pooled as the preference was to work only with single-donor material. To ensure generation of robust data and comprehensive assessment of CD34 + populations within a large cohort, we analyzed cord blood from 56 donors. Within our analyzed cord blood (n = 56), we observed substantial donor-to-donor variation in the CD34 + cell counts with a mean of 2.68 × 10 4 cells/ml and a standard deviation of ± 1.55 × 10 4 CD34 + CD45 + cells/mL (Fig. 1 B). This translates to an average of 7.5 × 10 5 CD34 + cells per collection from standard UBC donation volumes of 20–60 ml (Fig. 1 C), which would suffice for a handful number of humanized mice. Next, we wanted to analyze if the processing following collection impacted the average number of available stem cells. It has been described that cell processing, especially gradient centrifugation, is associated with significant cell loss [ 17 ]. Thus, to determine the cell loss during CD34 + cell isolation and storage (cryopreservation) we analyzed CD34 + cell count using flow cytometry at each step of the isolation process. In line with data from other groups, there is a sizeable percentual cell loss during cell isolation and cryopreservation, resulting in on average just 36.1% cells available for in vivo application (Fig. 1 D). The thawed injection product was pure for CD34 + cells with only few CD45 + cells (Fig. 1 E) and virtually absent CD3 + T cells (Fig. 1 F). TACS leads to higher CD34 + cell purity and yield compared to magnetic bead isolation TACS offers an opportunity to isolate CD34 + cells using more streamlined processes, which we optimized to isolate cells directly from whole UCB (Supplementary Fig. 2). Thus, we sought to compare the efficacy of this system to standard magnetic bead-based isolation protocols (mag. bead). TACS led to a higher yield and purity of CD34 + cells and a more consistent final product, with reduced variability compared to mag. bead-isolated cells (CV yield 33% versus 43.1%; CV purity 26.9% versus 43.7%) (Fig. 2 A-B). While TACS resulted in a greater yield of CD34 + CD45 + cells, this increase was not statistically significant. However, TACS consistently produced a higher purity of isolated CD34 + cells with less variability across 26 unique UBC donations (Fig. 2 C). Notably, both isolation protocols produced CD34 + cell products with comparable phenotypical of progenitor cells (HSC, Myeloid-primed progenitors (MPP), Lympho-myeloid primed progenitors (LMPP)) (Fig. 2 D). We further analyzed the composition and clonogenicity of the CD34 + CD45 + cells through a 14-day CFC assay (Fig. 2 E). As expected, the CFC assay, which only supports myeloid and erythroid lineage development, did not produce lymphoid colonies. Most colonies were erythroid (CFU-E/BFU-E), followed by pure granulocytic colonies (CFU-G) (Fig. 2 F-G). Less than 5% of colonies consisted of granulocytic, monocytic, macrophagic and erythroid cells (CFU-GEMM), which were derived from multipotent progenitors (MPPs) or HSCs (Fig. 2 F). The colony distribution of TACS-isolated cells mirrored that of mag. bead-isolated cells, with no observable differences in the proportional composition of colony types between the two methods (Fig. 2 F). Additionally, by controlling the initial input dose to 200 CD34 + CD45 + cells, we determined that both methods produced approximately 100 colonies, further indicating equivalent clonogenic potential (Fig. 2 G). Thus, we conclude that TACS provides superior CD34 + cell purity and yield while maintaining the same progenitor cell phenotypic and functional characteristics as mag. bead isolation. Humanized levels are equivalent between TACS and mag. bead-isolated CD34 + cells using an in vivo engraftment efficiency model To assess the functional engraftment capacity of CD34 + cells isolated by TACS, we compared the in vivo performance to that of mag. bead-isolated cells using a humanized NSG-SGM3 mouse model. To remove donor variability, we strictly used TACS and mag. bead-isolated cells from the same donor for all comparisons. Peripheral humanization was tracked over 12 weeks in peripheral blood, while bone marrow (BM) and spleen humanization levels were analyzed at the endpoint. In a lead-up experiment with mag. bead-isolated cells, we observed a wider humanization spread and thus, increase in CV at doses below 15,000 (Supplementary Fig. 3A-B). Thus, we performed a dilution series of cell inputs in doses between 15,000 – hoping to stress the system and exacerbate differences between the isolated cells. For donor 1, we compared humanization outcomes between mag. bead- and TACS-isolated cells at very low doses of 3,750, 7,500 and 15,000 CD34 + CD45 + cells per mouse. Across all doses and timepoints, there were no significant differences in the peripheral humanization levels between the two isolation methods (Fig. 3 A). However, due to the limited number of mice (1–2 per group), we sought to increase the statistical power by replicating this experiment with a larger cohort of mice. For donor 2, we tested the very low input doses of 1,200 and 12,000 CD34 + CD45 + cells with 5–6 mice per group. Consistent with the first experiment, humanization levels were comparable between mag. bead- and TACS-isolated cells across all doses tested. Only at 12 weeks did 12,000 TACS-isolated cells have slightly higher humanization levels (P = 0.0230) (Fig. 3 B). Together, these results establish that TACS-isolated cells perform similarly to the CD34 + cells isolated using mag. beads. Central humanization, particularly in the BM and spleen, is an integral part of functional humanization in these mice models. Thus, we assessed whether TACS-isolated cells functionally humanized organs similarly to mag. bead-isolated cells. For donor 1, we observed similar humanization levels in both BM and spleen at all doses (Fig. 3 C). Notably, at the 3,750 cell dose, there was increased variability in humanization levels, likely due to suboptimal cell input approaching the threshold for maximum achievable engraftment levels in NSG-SGM3 mice (Fig. 3 C). To further challenge the system and amplify any functional difference between the two isolation methods, donor 2 cells were tested at an even lower input dose of 1,200 cells. As expected, the 1,200 cell dose resulted in lower and more variable engraftment levels in both BM and spleen than the 12,000 cell dose. However, no significant differences were detected between the isolation methods (Fig. 3 D). We conclude that the TACS system produces CD34 + CD45 + cells that are functionally equivalent to those isolated by standard mag. beads for in vivo humanization, even at low input doses. Low CD34 + input doses result in robust and predictable humanization Although our TACS system performed equivalent to mag. beads in terms of yield and humanization, the current gold standard of the field is mag. bead-based isolation. Thus, using our established humanization model, we investigated the influence of varying mag. bead-isolated CD34 + cell doses on immune system humanization. To first establish a baseline humanization trajectory in NSG-SGM3 mice, we injected 4 week old mice with a high dose (60,000) of living CD34 + cells, based on previously published doses [ 16 ]. Peripheral blood was analyzed every 2 weeks for 12 weeks using flow cytometry (Supplementary Fig. 4A). Mice did not show signs of GvHD or illness (Supplementary Fig. 4B), even with high humanization percentages greater than 45% of human CD45 + cells from total CD45% cells present in peripheral blood (Supplementary Fig. 4C-D), Moreover, engrafted CD34 + cells possessed capacity to differentiate into T-, B- and myeloid cell lineages (Supplementary Fig. 4E-F). At the endpoint 12 weeks after CD34 + cell engraftment, bone marrow and spleen humanization was robust and maintained lineage differentiation (Supplementary Fig. 4G-H). We next investigated the influence that very low CD34 + cell doses have on humanization levels in NSG-SGM3 mice (Fig. 4 A). Therefore, we titrated the number of transplanted cells down to 3,750 viable CD34 + cells per mouse, showing that at very low cell doses reconstitution is slower but still robust, achieving > 45% human CD45 + cells in peripheral blood (Fig. 4 B). Our results also showed that the humanization CV was relatively stable across all but the lowest doses, suggesting that the lowest dose of CD34 + cells led to less predictability and uniformity of engraftment compared to the moderate and high cell doses (Fig. 4 C). The robust BM humanization observed with the highest cell dose of 60,000 was reproducible with much lower cell doses. Mice receiving a very low dose of 7,500 CD34 + cells or more all achieved similar spleen humanization levels (Fig. 4 D). In the spleen, humanization levels were high, exceeding 80%, and while absolute cell numbers showed no statistically significant differences across all doses, humanization levels were different; however, the biological relevance remains uncertain (Fig. 4 D and E). Immunohistochemistry confirmed the presence of human CD45 + cells in the bone marrow at endpoint in the very low, low, and high cell dose groups (Fig. 4 F). Important for downstream application is also a homogeneity of engraftment of human CD45 + cell populations across all humanized mice. We assessed this by looking at the relative deviation across blood, BM, and spleen (Supplementary Fig. 3A-B). Especially in blood and BM, very low doses of 3,750 and 7,500 cells show larger relative spread of human CD45 + cells measured between different mice than very low 15,000, low 30,000 and high 60,000 input doses. The spleen deviation was very small across the board, with particularly less deviation in cell doses of 15,000 or greater. The plateauing CV% seems to indicate that the stem cell niche is saturated with far fewer CD34 + cells than previously thought (Supplementary Fig. 3B). Development of the human leucocyte compartment is not affected by lower CD34 + cell doses Historically the goal of humanized mice was to engraft human CD34 + stem cells within the mouse model system. Next generation humanized mouse model strains, such as the NSG-SGM3, have focused on not only engraftment but robust lineage differentiation of the stem cells. The human cytokines produced by the NSG-SGM3 mice strain support robust myeloid differentiation and previous publications have extensively characterized and compared this strain to the NSG model [ 18 , 19 ]. We assessed the leucocyte compartment for the presence of T cells, B cells and CD33 + myeloid cells, demonstrating that while the early reconstitution is dominated by CD33 + myeloid cells, CD19 + B cells and CD3 + T cells reconstitute later after week 6 and week 10 (Fig. 5 A-B). We previously observed that absolute hCD45 + and CD33 + cell counts remain constant after 6 weeks (Supplementary Fig. 4D-E) and that the shift from a B cell dominated to a T cell dominated compartment results in an absolute decrease in B cells, simultaneous to an absolute increase in T cells in the analyzed organs (Supplementary Fig. 5E-F). This trend is found across all cell doses and with remarkable stability (Fig. 5 B). Only the T cell emergence is slowed in the lowest dose. It is likely that these reconstitution dynamics reflect the differentiation of CD34 + cells in the huNSG-SGM3 model and deviations from these dynamics result from engraftment of CD34 − cells in the injection product; something researchers should be particularly aware of. Across different organs we also saw substantially different leucocyte compartments, with CD33 + myeloid cells dominating the BM and spleen favoring CD19 + B cells. This trend is consistent across all doses, although in lower CD34 + cell doses T cell reconstitution is delayed (Fig. 5 C). Further the reconstitution of a humanized immune system using lower stem cell doses allows to study large cohorts of mice. This is important, as large cohorts of single-donor humanized mice are required in a variety of applications, for example in studying allergies, infectious diseases, or cancer. Large cohorts of humanized NSG-SGM3 mice can be generated using 15,000 CD34 + cells per mouse and serve as a potent model for CAR-T cell-induced cytokine release syndrome One of the applications of huNSG-SGM3 mice is modelling cytokine-release syndrome (CRS) following CAR T-cell therapy [ 3 ]. CRS can be lethal and is a hinderance in the adoption of CAR-T-cell therapy for human treatment. It also underlies significant inter-donor severity, thus using pooled donor cells is not an appropriate approach. As a proof of concept, we used the previously established very low 15,000 CD34 + cell dose to robustly humanize 30 NSG-SGM3 mice from a single UCB donor. After reconstitution for 8 weeks, we injected CD19 + tumor cells followed by anti-CD19 CAR T cells one week later (Fig. 6 A). We transferred CAR T cells after 8 weeks of humanization, when the leucocyte compartment contains CD33 cells but is not yet dominated by CD3 cells. Reconstitution dynamics can vary by lab and donor, so this timing should also ensure a more stable system and robust results for researchers. Across all 30–mice, we saw high humanization levels, with an expected B cell dominated hCD45 + compartment (Fig. 6 B-C). With such a cohort size, we were able to compare three conditions with 6–8 mice per group, which would have not been possible using older protocols with high cell numbers (Fig. 6 A). Following CAR-T cell injection, we monitored mice for clinical features of CRS. Mock transduced T cells did not elicit CRS, while all clones of CAR-T cells caused substantial weigh loss in line with CRS (Fig. 6 D). We further confirmed CRS by measuring inflammatory cytokines IL-6 and IL-10 at 3 and 5 days after CAR-T cell transfer. Each CAR-T clone elicited significantly elevated IL-6 and IL-10 in the serum of the mice. (Fig. 6 E). Both weight loss and IL-6 levels are well in line with previously published results by Norrelli et al. , who reported that CRS is marked by high systemic IL-6 levels and profound weight loss, and that CRS severity correlates with tumor burden [ 3 ]. Discussion Humanized mice are a powerful tool for assessing human-like immune responses in small animals. Previous work has focused, in part, on optimizing the host through transgenic cytokine expressions [ 20 , 21 ]. This has led to an assortment of potential host strains, each appropriate to slightly different experiments. Nonetheless, the described workflows for humanizing mouse immune systems are largely ubiquitous but also tedious. Additionally, the shortage of donor material typically leads to the mixing of hematopoietic stem cell donors to gain sufficient numbers to perform experiments. This mixing may potentially lead to unnatural diverse and complex HLA compositions reducing many benefits of using human immune cells to a minimum as the models are reduced artificially. To circumvent this artificial obstruction, we here describe a workflow for the humanization of mouse immune cells and highlight constraints due to CD34 + cell doses. Total CD34 + cell counts in a single UCB donation are limited and primarily restricts its clinical use to pediatric stem cell transfer [ 22 , 23 ]. To develop a more stable transfer system we first defined the cell product we are working with. The CD34 + cell count per mL and total cell collection amount is variable. (Fig. 1 B & C) Furthermore, cell isolation significantly reduces available cells. Overall, the purity of stem cells was high, but the total amount was so limited that it hinders larger cohort humanization experiments. In particular, large cell losses are frequently observed during density gradient centrifugation, a non-affinity-based method that is a prerequisite for magnetic bead-based cell isolation. This challenge is not unique to CD34 + cells, but is common across all rare cell types and is observed for example also in isolating circulating tumor cells, highlighting an inherent limitation of density gradient centrifugation [ 24 ]. Despite this limitation, we still achieve satisfactory results when using density gradient centrifugation. However, alternative methods, such as whole blood isolation protocols, present their own challenges, particularly when processing larger volumes of blood [ 25 , 26 ]. Hence, a further optimization could be using an alternative wholeblood-based isolation protocol, such as traceless affinity cell selection (TACS) developed in our lab [ 27 ]. Here, we demonstrated that CD34 + cell isolation using the TACS method was a simpler and more efficient process compared to mag. based isolation. Moreover, the TACS approach more frequently produced CD34 + cells with higher yield and purity while maintaining functional equivalent to those isolated via magnetic beads. Residual contaminating cell populations, particularly T cells, can lead to GvHD reaction in humanized mice, highlighting the importance of CD34 + cell purity for experimental procedures [ 28 ]. Nonetheless, we purposely abstained from this more “modern” and innovative isolation method in the cell dosing study, as mag. bead isolation still remains the most widespread isolation method and thus, appeals to the broadest audience. As a baseline comparison we used high 60,000 CD34 + cells, a cell count known to elicit robust humanization levels across all compartments [ 16 ]. We then methodically reduced the input dose to evaluate the robustness of the system, hereby showing that a reduction of transferred human stem cells down to a very low dose of 3,750 living CD34 + cells still led to robust albeit lower humanization of the immune system (Fig. 1 B). This is in line with other publications showing that from cell types with a high plasticity and inherent stemness, even a singular cell is able to reconstitute a whole organism [ 29 , 30 ]. Our reported reconstitution kinetics with an early myeloid dominance and a later emergence of T cells is in-line with various previous studies [ 31 , 32 ]. Earlier work from Lang et al. , report that T cell emergence coincides with B cell maturation and lymph node occupation in humanized mice, suggesting engraftment kinetics may be influenced by tissue microenvironment interactions. Similarly, early myeloid dominance and later T cell emergence is observed in immune system reconstitution in humans receiving UCB transplantation and may reflect the natural timing of the lineage differentiation pathways within biological systems [ 33 ]. For downstream applications, humanization levels between mice must be consistent and also sufficient, else risking falsifying results. Furthermore, a careful balance between available cells, mice numbers and humanization output is crucial for a successful workflow. Using our workflow, we determined that a very low dose of 15,000 CD34 + cells provided an optimal balance between utilization and output. In fact our observed reconstitution levels where higher and comparable to previous humanization levels shown for NSG-SGM3 mice by Maser et al. and Coughlan et al. , who both used very high doses of 100,000 CD34 + cells per mouse [ 19 , 34 ]. At our reduced and very low dose, reconstitution levels in the human CD45 + compartment was on par with our saturating and previous reported dose, while also allowing four- and six-fold more humanized mice respectively. Conversely, the dose reduction also allows similar cohort sizes from fewer donor cells. This drastically increases the available (suboptimal) CD34 + collections for small cohort humanization projects. From our data, we emphasize that current protocols recommending upwards of 100,000 CD34 + cells per mouse exceed the necessary minimum effective threshold for successful humanization [ 35 ]. Alternatively, the current unnecessarily very high recommended dosing is an entry barrier to humanized mice and our proposed cell dose reduction is an opportunity to enhance resource utilization and experimental efficiency. We demonstrate that the reduced cell dose allows in vivo analysis of CRS elicited by three different CAR T-cell constructs in the setting of a single donor CD34 humanization (Fig. 4 ). Robust group sizes (6–8 mice) were achieved and all CAR constructs elicited weight loss and elevated inflammatory cytokine levels consistent with CRS. This model is also excellent in screening for better compositions and designs of CAR T-cells, as Arcangeli et al. recently showed by primarily using naïve T-cells for CAR T-cell generation [ 36 ]. While our work focuses on a single donor proof of concept, future research should assess inter-donor efficacy and potency. For example, using humanized mice with HLA matched tumor and CD34 + cells, different off-the-shelf CAR T cell products could be assessed easing the selection of the most suitable product for the patient (or HLA composition). Inter donor variability is becoming a larger and larger factor as our understanding of the immune system and immunotherapies increases. This study emphasizes CD34 + cell dose reduction in order to significantly increase humanization cohorts and/or increasing incorporation of suboptimal CD34 + donations. Ultimately, we hope to lower the entry barrier of humanized mice, making this technology a more widely available research tool. Conclusion Recapitulating, humanized mice are powerful tool in preclinical experiments, but current protocols hamper their usage due to very high cell dosages. Here we optimize the NSG-SGM3 mouse model by decreasing the cell dose by six-fold to previously published protocols, without a reduction in humanization levels or kinetics. In our proof-of-concept experiment this allowed us to simultaneously evaluate 4 groups (6–8 mice) of humanized mice with markedly lower total CD34 + cell usage. We further present a simple method for isolating functionally equivalent CD34 + cells from whole UCB, achieving higher yield and purity compared to standard magnetic bead-based isolation protocols. We hope that our findings will increase uptake of humanized mice projects and more importantly make single donor humanized mice the gold standard for immunotherapies. Declarations IRB The cord blood biobank was approved by the local ethics committee (reference number 617/19S). IACUC Regarding animal experimentation approval and oversight by an animal care committee, I can state that “all experiments with humanized mice were approved by the local administration (Regierung von Oberbayern) and overlooked by the local animal care committee”. If you need the exact animal experimentation approval number, please let me know (I am currently traveling). Acknowledgements This work was supported by the Deutsche Forschungsgemeinschaft (DFG) SFB-TRR 338/1–452881907 [projects A01], and SFB 1371–395357507 [project P04]. References Chuprin, J., et al., Humanized mouse models for immuno-oncology research. Nature Reviews Clinical Oncology, 2023. 20 (3): p. 192-206. Wunderlich, M., et al., Improved multilineage human hematopoietic reconstitution and function in NSGS mice. PLoS One, 2018. 13 (12): p. e0209034. Norelli, M., et al., Monocyte-derived IL-1 and IL-6 are differentially required for cytokine-release syndrome and neurotoxicity due to CAR T cells. Nat Med, 2018. 24 (6): p. 739-748. Bove, C., et al., CD4 CAR-T cells targeting CD19 play a key role in exacerbating cytokine release syndrome, while maintaining long-term responses. Journal for Immunotherapy of Cancer, 2023. 11 (1): p. e005878. Noort, W., et al., Similar myeloid recovery despite superior overall engraftment in NOD/SCID mice after transplantation of human CD34+ cells from umbilical cord blood as compared to adult sources. Bone marrow transplantation, 2001. 28 (2): p. 163-171. Wiekmeijer, A.-S., et al., Sustained Engraftment of Cryopreserved Human Bone Marrow CD34+ Cells in Young Adult NSG Mice. BioResearch Open Access, 2014. 3 (3): p. 110-116. Stemberger, C., et al., Novel serial positive enrichment technology enables clinical multiparameter cell sorting. PLoS One, 2012. 7 (4): p. e35798. Mohr, F., et al., Minimally manipulated murine regulatory T cells purified by reversible Fab Multimers are potent suppressors for adoptive T-cell therapy. Eur J Immunol, 2017. 47 (12): p. 2153-2162. Mohr, F., et al., Efficient immunoaffinity chromatography of lymphocytes directly from whole blood. Sci Rep, 2018. 8 (1): p. 16731. Ohnishi, H., et al., Regulation of cell shape and adhesion by CD34. Cell Adh Migr, 2013. 7 (5): p. 426-33. Haspel, R.L. and K.K. Ballen, Double cord blood transplants. Stem Cell Reviews, 2006. 2 (2): p. 81-86. Barker, J.N., et al., Transplantation of 2 partially HLA-matched umbilical cord blood units to enhance engraftment in adults with hematologic malignancy. Blood, 2005. 105 (3): p. 1343-1347. Martine, L.C., et al., Engineering a humanized bone organ model in mice to study bone metastases. Nat Protoc, 2017. 12 (4): p. 639-663. Shafiee, A., et al., Immune system augmentation via humanization using stem/progenitor cells and bioengineering in a breast cancer model study. Int J Cancer, 2018. 143 (6): p. 1470-1482. Kang, Y.K., et al., Humanizing NOD/SCID/IL-2Rγnull (NSG) mice using busulfan and retro-orbital injection of umbilical cord blood-derived CD34(+) cells. Blood Res, 2016. 51 (1): p. 31-6. Wunderlich, M., et al., Improved multilineage human hematopoietic reconstitution and function in NSGS mice. PLOS ONE, 2018. 13 (12): p. e0209034. Mata, M.F., et al., A modified CD34+ hematopoietic stem and progenitor cell isolation strategy from cryopreserved human umbilical cord blood. Transfusion, 2019. 59 (12): p. 3560-3569. Wunderlich, M., et al., A New Immunodeficient Mouse Strain, NOD/SCID IL2Rγ−/− SGM3, Promotes Enhanced Human Hematopoietic Cell Xenografts with a Robust T Cell Component. Blood, 2009. 114 (22): p. 3524. Maser, I.-P., et al., The Tumor Milieu Promotes Functional Human Tumor-Resident Plasmacytoid Dendritic Cells in Humanized Mouse Models. Frontiers in Immunology, 2020. 11 . Walsh, N.C., et al., Humanized mouse models of clinical disease. Annual Review of Pathology: Mechanisms of Disease, 2017. 12 : p. 187-215. Martinov, T., et al., Building the Next Generation of Humanized Hemato-Lymphoid System Mice. Frontiers in Immunology, 2021. 12 . Eapen, M., et al., Effect of graft source on unrelated donor haemopoietic stem-cell transplantation in adults with acute leukaemia: a retrospective analysis. Lancet Oncol, 2010. 11 (7): p. 653-60. Ballen, K.K., E. Gluckman, and H.E. Broxmeyer, Umbilical cord blood transplantation: the first 25 years and beyond. Blood, 2013. 122 (4): p. 491-498. Bacon, K., et al., Past, Present, and Future of Affinity-based Cell Separation Technologies. Acta Biomaterialia, 2020. 112 : p. 29-51. Sajay, B.N.G., et al., Microfluidic platform for negative enrichment of circulating tumor cells. Biomedical Microdevices, 2014. 16 (4): p. 537-548. Bankó, P., et al., Technologies for circulating tumor cell separation from whole blood. Journal of Hematology & Oncology, 2019. 12 (1): p. 48. Mohr, F., et al., Efficient immunoaffinity chromatography of lymphocytes directly from whole blood. Scientific Reports, 2018. 8 (1): p. 16731. Huey, D.D. and S. Niewiesk, Production of Humanized Mice through Stem Cell Transfer. Curr Protoc Mouse Biol, 2018. 8 (1): p. 17-27. Graef, P., et al., Serial Transfer of Single-Cell-Derived Immunocompetence Reveals Stemness of CD8+ Central Memory T Cells. Immunity, 2014. 41 (1): p. 116-126. Notta, F., et al., Isolation of Single Human Hematopoietic Stem Cells Capable of Long-Term Multilineage Engraftment. Science, 2011. 333 (6039): p. 218-221. Scherer, S.D., et al., An immune-humanized patient-derived xenograft model of estrogen-independent, hormone receptor positive metastatic breast cancer. Breast Cancer Research, 2021. 23 (1): p. 100. Lang, J., et al., Studies of lymphocyte reconstitution in a humanized mouse model reveal a requirement of T cells for human B cell maturation. J Immunol, 2013. 190 (5): p. 2090-101. Chiesa, R., et al., Omission of in vivo T-cell depletion promotes rapid expansion of naive CD4+ cord blood lymphocytes and restores adaptive immunity within 2 months after unrelated cord blood transplant. Br J Haematol, 2012. 156 (5): p. 656-66. Coughlan, A.M., et al., Myeloid Engraftment in Humanized Mice: Impact of Granulocyte-Colony Stimulating Factor Treatment and Transgenic Mouse Strain. Stem Cells and Development, 2016. 25 (7): p. 530-541. Verma, B. and A. Wesa, Establishment of Humanized Mice from Peripheral Blood Mononuclear Cells or Cord Blood CD34+ Hematopoietic Stem Cells for Immune-Oncology Studies Evaluating New Therapeutic Agents. Current Protocols in Pharmacology, 2020. 89 (1): p. e77. Arcangeli, S., et al., CAR T cell manufacturing from naive/stem memory T lymphocytes enhances antitumor responses while curtailing cytokine release syndrome. The Journal of clinical investigation, 2022. 132 (12). Additional Declarations There is NO Competing Interest. Supplementary Files 250405SchuXXtzetal7.png Supplementary Figure 1 – Flow cytometry gating strategy A: exemplary gating strategy. Events were gated for all leucocytes, single-events, living and on all CD45+ cells B: human CD45 + cells were further sub-quantified into lineages according to CD3, CD19 and CD33 positive cells 250405SchuXXtzetal8.png Supplementary Figure 2 – Optimization of the CD34 TACS protocol A: CD34+CD45+ yields from different TACS protocols showing the optimization from the starting to the optimized TACS protocol. Starting protocol was using the original system using erythrocyte-depleted starting material ([9]), the semi-optimized system with whole UCB starting material run through a modified streptactin-agarose bed, and the optimized system using a different CD34 Fab clone. B: Yield and C: Purity for the different protocols (one point represents one isolation. Data represents mean ± SD). 250405SchuXXtzetal9.png Supplementary Figure 3 – Humanization consistency A: relative deviation between different doses at week 12 in blood, BM and spleen B: Data from (A) expressed in %CV for each compartment 250405SchuXXtzetal10.png Supplementary Figure 4 – Current humanization workflow A: humanization workflow for 4 week old NSG-SGM3 mice receiving 60,000 CD34 + cells per mouse (n = X). B: mice showed a steady increase in absolute and relative weight following radiation and humanization C - D: humanization levels and hCD45 + cells in peripheral blood for 12 weeks after CD34 + cell injection. E - F: percentual and absolute number of CD3 + T cells, CD19+ B cells and CD33+ myeloid cells. G: Organ humanization and hCD45+ compartment composition week 12 after transfer of 60,000 CD34+ cells, H: Absolute hCD45+ counts in BM and spleen at 12 weeks after transfer of 60,000 CD34+ cells. SupplementaryTable1.docx Cite Share Download PDF Status: Under Review Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-6382310","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":447592829,"identity":"c3bf9629-7371-4e49-890b-c3fa30c3f418","order_by":0,"name":"Dirk Busch","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAAwklEQVRIiWNgGAWjYDCCwwzMDIwNQIYE8wEQH8wmVgtbYgNxWg7AtfAYEqeF7zjvYwPGHYft+Wf3fH/Mm8Mg209Ii+RhduMExjOHE2fcObuxmXcbg/FMQtYYHGZjPsDYdjjBQCIXrCVxwwEitdgbSOQ8BGvZT4yWBKAWxg0SOYwQWwj7hY3ZIPFMeuKMG2mGM+dukzCeQcgWvvPHmCU+7rC255+R/ODD2202sv0NhKwBgQQEU4IY9aNgFIyCUTAKCAEASFJBybzp1dIAAAAASUVORK5CYII=","orcid":"https://orcid.org/0000-0001-8713-093X","institution":"Technichal University of Munich","correspondingAuthor":true,"prefix":"","firstName":"Dirk","middleName":"","lastName":"Busch","suffix":""},{"id":447592830,"identity":"1a2855ef-b2a2-45e3-b54f-68f73d8feee2","order_by":1,"name":"Julius Schuetz","email":"","orcid":"","institution":"Technichal University of Munich","correspondingAuthor":false,"prefix":"","firstName":"Julius","middleName":"","lastName":"Schuetz","suffix":""},{"id":447592831,"identity":"c3cb5edf-cdba-4064-ae5d-7ef48cb302d9","order_by":2,"name":"Sarah Dötsch","email":"","orcid":"","institution":"Technichal University of Munich","correspondingAuthor":false,"prefix":"","firstName":"Sarah","middleName":"","lastName":"Dötsch","suffix":""},{"id":447592832,"identity":"f90ffa62-ddd4-4e59-9fa0-96606000252a","order_by":3,"name":"Georg Schmidt","email":"","orcid":"","institution":"Rechts der Isar Hospital","correspondingAuthor":false,"prefix":"","firstName":"Georg","middleName":"","lastName":"Schmidt","suffix":""},{"id":447592833,"identity":"f9a1e99c-336a-412e-ae75-e51e52331f3f","order_by":4,"name":"Fabian Mohr","email":"","orcid":"","institution":"Technical University of Munich","correspondingAuthor":false,"prefix":"","firstName":"Fabian","middleName":"","lastName":"Mohr","suffix":""},{"id":447592834,"identity":"3ec8bdb5-8ea2-45ef-a6d7-61cd5d20f030","order_by":5,"name":"Jack Barton","email":"","orcid":"","institution":"Technical University of Munich","correspondingAuthor":false,"prefix":"","firstName":"Jack","middleName":"","lastName":"Barton","suffix":""},{"id":447592835,"identity":"c48cbe26-1d94-40f2-9a92-890d06846c26","order_by":6,"name":"Linda Warmuth","email":"","orcid":"https://orcid.org/0000-0003-3324-7731","institution":"Technical University of Munich","correspondingAuthor":false,"prefix":"","firstName":"Linda","middleName":"","lastName":"Warmuth","suffix":""},{"id":447592836,"identity":"da2eb6cf-3ca0-42ef-ba06-90ff73bb6f9e","order_by":7,"name":"Pascal Winterhalter","email":"","orcid":"https://orcid.org/0009-0001-2672-5734","institution":"Technical University of Munich","correspondingAuthor":false,"prefix":"","firstName":"Pascal","middleName":"","lastName":"Winterhalter","suffix":""},{"id":447592837,"identity":"282fa062-732e-46de-a072-2461c146693d","order_by":8,"name":"Anna Hichholzer","email":"","orcid":"","institution":"Technical University of Munich","correspondingAuthor":false,"prefix":"","firstName":"Anna","middleName":"","lastName":"Hichholzer","suffix":""},{"id":447592838,"identity":"32c3ba1c-04dc-4d74-8c77-55290f2b60f7","order_by":9,"name":"Monica Casucci","email":"","orcid":"https://orcid.org/0000-0003-4500-9060","institution":"Innovative Immunotherapies Unit, Division of Immunology, Transplantation and Infectious Diseases, IRCCS San Raffaele Scientific Institute","correspondingAuthor":false,"prefix":"","firstName":"Monica","middleName":"","lastName":"Casucci","suffix":""},{"id":447592839,"identity":"aaab3a08-e0e5-42a7-b1c0-bf5f416c62be","order_by":10,"name":"Elvira D'Ippolito","email":"","orcid":"https://orcid.org/0000-0003-0113-6797","institution":"Technical University of Munich and German Center for Infection Research (DZIF), Partner Site Munich","correspondingAuthor":false,"prefix":"","firstName":"Elvira","middleName":"","lastName":"D'Ippolito","suffix":""},{"id":447592840,"identity":"97dd345f-0bc4-491c-b15c-192403569f95","order_by":11,"name":"Dietmar Hutmacher","email":"","orcid":"https://orcid.org/0000-0001-5678-2134","institution":"Queensland University of Technology","correspondingAuthor":false,"prefix":"","firstName":"Dietmar","middleName":"","lastName":"Hutmacher","suffix":""},{"id":447592841,"identity":"0eceef72-00d7-4f06-9720-7bf6a3ddf385","order_by":12,"name":"Jacqui A McGovern","email":"","orcid":"https://orcid.org/0000-0002-4993-6745","institution":"Faculty of Engineering, School of Mechanical, Medical and Process Engineering, Queensland University of Technology, Brisbane","correspondingAuthor":false,"prefix":"","firstName":"Jacqui","middleName":"A","lastName":"McGovern","suffix":""}],"badges":[],"createdAt":"2025-04-05 13:36:11","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-6382310/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-6382310/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":81665219,"identity":"8366c18a-eca9-44cd-b171-b8230e6aa73f","added_by":"auto","created_at":"2025-04-30 02:47:11","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":101279,"visible":true,"origin":"","legend":"\u003cp\u003eIsolation workflow and cell loss\u003c/p\u003e\n\u003cp\u003eA: workflows for mice humanization results in inevitable CD34\u003csup\u003e+\u003c/sup\u003e cell loss during cell isolation and cryopreservation. This results in unpredictable mouse cohort size.\u003c/p\u003e\n\u003cp\u003eB: CD34\u003csup\u003e+\u003c/sup\u003e cells per millilitre UCB in full term newborns detected through flow cytometry (n=52) (line represents mean)\u003c/p\u003e\n\u003cp\u003eC: significant variations in the CD34\u003csup\u003e+\u003c/sup\u003e cell count and collection volumes leads to limitations in the available CD34\u003csup\u003e+\u003c/sup\u003e cells (n=23) (line represents mean)\u003c/p\u003e\n\u003cp\u003eD: percentual cell loss during magnetic bead isolation (n=30) and cryopreservation (n=6) means only 36% of collected CD34+ cells are available for humanization (line represents mean)\u003c/p\u003e\n\u003cp\u003eE: representative flow cytometry plot of thawed CD34\u003csup\u003e+\u003c/sup\u003e cell product\u003c/p\u003e\n\u003cp\u003eF: Quantification of CD34+ injection product analyzed through flow cytometry (n=7)\u003c/p\u003e","description":"","filename":"250405SchuXXtzetal1.png","url":"https://assets-eu.researchsquare.com/files/rs-6382310/v1/27cddbe17edd879d14da72ae.png"},{"id":81665218,"identity":"629093b1-7802-437b-b98a-80b8797e85c7","added_by":"auto","created_at":"2025-04-30 02:47:10","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":78965,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cu\u003eMag. bead versus TACS isolation\u003c/u\u003e\u003c/p\u003e\n\u003cp\u003eA: CD34\u003csup\u003e+\u003c/sup\u003eCD45\u003csup\u003e+\u003c/sup\u003e yield from whole blood using magnetic bead (n=26) or TACS (n=15), each symbol represents one donor (numbers represent mean, bars represent mean ± SD).\u003c/p\u003e\n\u003cp\u003eB: CD34\u003csup\u003e+\u003c/sup\u003eCD45\u003csup\u003e+\u003c/sup\u003e purity for the same donors in A (numbers represent mean, bars represent mean ± SD).\u003c/p\u003e\n\u003cp\u003eC: X-Y scatter plot of purity versus yield for either magnetic beads (red) or TACS (blue) isolation. Statistical significance was calculated using a two-tailed Welsh’s t-test.\u003c/p\u003e\n\u003cp\u003eD: Flow cytometric analysis of CD34\u003csup\u003e+\u003c/sup\u003eCD45\u003csup\u003e+\u003c/sup\u003e using CD90 and CD45RA to analyze frequency of HSC, MPP and LMPP within the CD34\u003csup\u003e+\u003c/sup\u003eCD45\u003csup\u003e+\u003c/sup\u003e cells (n=3 and from the same donors for both groups)\u003c/p\u003e\n\u003cp\u003eE: A simplified HSC differentiation tree. Not all lineages differentiate during the CFC assay, which are denoted by the red X through the lineage branches.\u003c/p\u003e\n\u003cp\u003eF: Relative distribution of the different colony types after 14 days of incubation\u003c/p\u003e\n\u003cp\u003eG: Total number of colonies after plating 200 CD34+CD45+ directly after isolation with either magnetic beads or TACS.\u003c/p\u003e\n\u003cp\u003eIn (A) and (B), statistical significance was assessed through paired multiple t-tests\u003c/p\u003e","description":"","filename":"250405SchuXXtzetal2.png","url":"https://assets-eu.researchsquare.com/files/rs-6382310/v1/c1a3c86bcd88d8a6bcb376ce.png"},{"id":81665610,"identity":"adfc8bdf-d4ae-4658-a433-e40264609ce7","added_by":"auto","created_at":"2025-04-30 02:55:11","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":58450,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cu\u003ePeripheral blood humanization in NSG-SGM3 mice after injection of donor-matched CD34+CD45+ cells isolated using magnetic beads or TACS\u003c/u\u003e\u003c/p\u003e\n\u003cp\u003eA: Comparison of magnetic bead and TACS isolated CD34+ cells from donor 1 (n=1-2 mice per group).\u003c/p\u003e\n\u003cp\u003eB: Comparison of magnetic bead and TACS isolated CD34+ cells from donor 2 (n= 5-6 mice per group).\u003c/p\u003e\n\u003cp\u003eC: Comparison of magnetic bead and TACS isolated CD34\u003csup\u003e+\u003c/sup\u003e cells from donor 1 (n=1-2 mice per group) for central humanization in BM and spleen.\u003c/p\u003e\n\u003cp\u003eD: Comparison of magnetic bead and TACS isolated CD34\u003csup\u003e+\u003c/sup\u003e cells from donor 2 (n= 5-6 mice per group) for central humanization in BM and spleen.\u003c/p\u003e\n\u003cp\u003eIn (A) – (D), statistical significance was assessed through 2way ANOVA.\u003c/p\u003e","description":"","filename":"250405SchuXXtzetal3.png","url":"https://assets-eu.researchsquare.com/files/rs-6382310/v1/dbb5c30003850f5f89b2c5b7.png"},{"id":81665221,"identity":"a965fbf8-c6f8-49f8-a17a-3f073dff9daa","added_by":"auto","created_at":"2025-04-30 02:47:11","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":668691,"visible":true,"origin":"","legend":"\u003cp\u003eCD34\u003csup\u003e+\u003c/sup\u003e cell dose titration\u003c/p\u003e\n\u003cp\u003eA: \u003cem\u003ein vivo\u003c/em\u003e workflow for establishing humanized NSG-SGM3 mice. Shown are representative flow cytometry plots for each week of a single mouse from the 3,750, 5,000 and 60,000 CD34\u003csup\u003e+\u003c/sup\u003e cell group\u003c/p\u003e\n\u003cp\u003eB: Humanization (hCD45\u003csup\u003e+\u003c/sup\u003e/all CD45\u003csup\u003e+\u003c/sup\u003e) in peripheral blood after CD34+ injection over 12 weeks. (2-3 donors per group, n= 7-13 mice per group)\u003c/p\u003e\n\u003cp\u003eC: humanization CV across different weeks for different doses\u003c/p\u003e\n\u003cp\u003eD: organ humanization 12 weeks after CD34\u003csup\u003e+\u003c/sup\u003e injection shows consistent and robust humanization levels.\u003c/p\u003e\n\u003cp\u003eE: absolute hCD45\u003csup\u003e+\u003c/sup\u003e cell counts in each organ\u003c/p\u003e\n\u003cp\u003eF: Representative immunohistochemistry of bone marrow to assess hCD45+ in very low (7,500), low (30,000) and high (60,000) cell doses\u003c/p\u003e\n\u003cp\u003eIn (D) and (E), statistical significance was assessed through 1way ANOVA with 60,000 dose acting as the control.\u003c/p\u003e","description":"","filename":"250405SchuXXtzetal4.png","url":"https://assets-eu.researchsquare.com/files/rs-6382310/v1/fe838696d39548469d584c30.png"},{"id":81665066,"identity":"24030583-481e-4071-ad4f-def7ac4cb987","added_by":"auto","created_at":"2025-04-30 02:39:11","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":177435,"visible":true,"origin":"","legend":"\u003cp\u003eCD34\u003csup\u003e+\u003c/sup\u003e cell dose titration – hCD45 subsets\u003c/p\u003e\n\u003cp\u003eA: representative flow cytometry staining plots showing the hCD45\u003csup\u003e+\u003c/sup\u003e compartment in a mouse from either the 3,750, 15,000 and 60,000 cell group for weeks 6, 8, 10 and 12.\u003c/p\u003e\n\u003cp\u003eB: Composition of the hCD45\u003csup\u003e+\u003c/sup\u003e compartment after CD34\u003csup\u003e+\u003c/sup\u003e injection over the 12 weeks after injection for different CD34\u003csup\u003e+\u003c/sup\u003e cell doses. (2-3 donors per group, n= 7-13 mice per group)\u003c/p\u003e\n\u003cp\u003eC: composition of the hCD45\u003csup\u003e+\u003c/sup\u003e compartment in BM and spleen at 12 weeks\u003c/p\u003e","description":"","filename":"250405SchuXXtzetal5.png","url":"https://assets-eu.researchsquare.com/files/rs-6382310/v1/ee8e7955e60e859994146e92.png"},{"id":81665223,"identity":"3efcce10-ac30-4e35-ad97-a673bd567c5c","added_by":"auto","created_at":"2025-04-30 02:47:11","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":42632,"visible":true,"origin":"","legend":"\u003cp\u003eCRS model\u003c/p\u003e\n\u003cp\u003eA: workflow using a reduced 15,000 CD34+ cell dose per mouse from a single cord blood donation to robustly humanize 30 NSG-SGM3 mice. CD19+ tumor cells were injected 8 weeks after CD34\u003csup\u003e+\u003c/sup\u003e injection and one-week later CAR T cells transferred. Mice were monitored for signs of CRS for 7 days after T cell transfer.\u003c/p\u003e\n\u003cp\u003eB \u0026amp; C: all 30 mice were robustly humanized and showed an expected B cell dominated hCD45\u003csup\u003e+\u003c/sup\u003e compartment 8 weeks after humanization.\u003c/p\u003e\n\u003cp\u003eD: following CAR T cell transfer mice receiving different CAR T cells showed weight loss consistent with CRS\u003c/p\u003e\n\u003cp\u003eE: mice serum was analyzed for pro-inflammatory cytokines IL-6 and IL-10 at 3 and 5 days after CAR T cell transfer.\u003c/p\u003e","description":"","filename":"250405SchuXXtzetal6.png","url":"https://assets-eu.researchsquare.com/files/rs-6382310/v1/d9cd83dfe59f74ca97cf3815.png"},{"id":81665716,"identity":"79885fae-d47d-4413-b7b7-9c6cff9647ce","added_by":"auto","created_at":"2025-04-30 03:03:13","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":2083378,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-6382310/v1/3742f688-f1d0-4208-b065-eed960cf7c9e.pdf"},{"id":81665057,"identity":"cabbd938-2e57-4c3b-bd66-f165c172f9a7","added_by":"auto","created_at":"2025-04-30 02:39:11","extension":"png","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":188493,"visible":true,"origin":"","legend":"\u003cp\u003eSupplementary Figure 1 – Flow cytometry gating strategy\u003c/p\u003e\n\u003cp\u003eA: exemplary gating strategy. Events were gated for all leucocytes, single-events, living and on all CD45+ cells\u003c/p\u003e\n\u003cp\u003eB: human CD45\u003csup\u003e+\u003c/sup\u003e cells were further sub-quantified into lineages according to CD3, CD19 and CD33 positive cells\u003c/p\u003e","description":"","filename":"250405SchuXXtzetal7.png","url":"https://assets-eu.researchsquare.com/files/rs-6382310/v1/4447054a6f174ee8d4804513.png"},{"id":81665052,"identity":"3eef2968-d1e0-4e84-8c58-57f153d007b2","added_by":"auto","created_at":"2025-04-30 02:39:10","extension":"png","order_by":2,"title":"","display":"","copyAsset":false,"role":"supplement","size":36007,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cu\u003eSupplementary Figure 2 – Optimization of the CD34 TACS protocol\u003c/u\u003e\u003c/p\u003e\n\u003cp\u003eA: CD34+CD45+ yields from different TACS protocols showing the optimization from the starting to the optimized TACS protocol. Starting protocol was using the original system using erythrocyte-depleted starting material ([9]), the semi-optimized system with whole UCB starting material run through a modified streptactin-agarose bed, and the optimized system using a different CD34 Fab clone.\u003c/p\u003e\n\u003cp\u003eB: Yield and C: Purity for the different protocols (one point represents one isolation. Data represents mean ± SD).\u003c/p\u003e","description":"","filename":"250405SchuXXtzetal8.png","url":"https://assets-eu.researchsquare.com/files/rs-6382310/v1/235b6dc4df5a3d219cbd7e11.png"},{"id":81665054,"identity":"c1385c03-3bfb-437a-a2d9-51106941f905","added_by":"auto","created_at":"2025-04-30 02:39:11","extension":"png","order_by":3,"title":"","display":"","copyAsset":false,"role":"supplement","size":28492,"visible":true,"origin":"","legend":"\u003cp\u003eSupplementary Figure 3 – Humanization consistency\u003c/p\u003e\n\u003cp\u003eA: relative deviation between different doses at week 12 in blood, BM and spleen\u003c/p\u003e\n\u003cp\u003eB: Data from (A) expressed in %CV for each compartment\u003c/p\u003e","description":"","filename":"250405SchuXXtzetal9.png","url":"https://assets-eu.researchsquare.com/files/rs-6382310/v1/e69907b0b34cca97e2644400.png"},{"id":81665058,"identity":"4c7e4428-4b29-4950-8b3c-d1a602acadc1","added_by":"auto","created_at":"2025-04-30 02:39:11","extension":"png","order_by":4,"title":"","display":"","copyAsset":false,"role":"supplement","size":101311,"visible":true,"origin":"","legend":"\u003cp\u003eSupplementary Figure 4 – Current humanization workflow\u003c/p\u003e\n\u003cp\u003eA: humanization workflow for 4 week old NSG-SGM3 mice receiving 60,000 CD34\u003csup\u003e+\u003c/sup\u003e cells per mouse (n = X).\u003c/p\u003e\n\u003cp\u003eB: mice showed a steady increase in absolute and relative weight following radiation and humanization\u003c/p\u003e\n\u003cp\u003eC - D: humanization levels and hCD45\u003csup\u003e+\u003c/sup\u003e cells in peripheral blood for 12 weeks after CD34\u003csup\u003e+\u003c/sup\u003e cell injection.\u003c/p\u003e\n\u003cp\u003eE - F: percentual and absolute number of CD3\u003csup\u003e+\u003c/sup\u003e T cells, CD19+ B cells and CD33+ myeloid cells.\u003c/p\u003e\n\u003cp\u003eG: Organ humanization and hCD45+ compartment composition week 12 after transfer of 60,000 CD34+ cells,\u003c/p\u003e\n\u003cp\u003eH: Absolute hCD45+ counts in BM and spleen at 12 weeks after transfer of 60,000 CD34+ cells.\u003c/p\u003e","description":"","filename":"250405SchuXXtzetal10.png","url":"https://assets-eu.researchsquare.com/files/rs-6382310/v1/f091a9b049d9cd4adeef84fe.png"},{"id":81665220,"identity":"08f509bc-3194-494c-bef9-989c98d930ef","added_by":"auto","created_at":"2025-04-30 02:47:11","extension":"docx","order_by":5,"title":"","display":"","copyAsset":false,"role":"supplement","size":14320,"visible":true,"origin":"","legend":"","description":"","filename":"SupplementaryTable1.docx","url":"https://assets-eu.researchsquare.com/files/rs-6382310/v1/011d2107c19c9487b36be152.docx"}],"financialInterests":"There is \u003cb\u003eNO\u003c/b\u003e Competing Interest.","formattedTitle":"Optimizing CD34+ cell doses for large-scale translational, single donor humanized mouse experiments","fulltext":[{"header":"Introduction","content":"\u003cp\u003eHumanized mice are an important tool for preclinical studies of human diseases where the immune system plays a central role, such as cancer. Through the introduction of human cells, they harbor a human immune system within a small rodent model. Thus, this model allows humanized readouts that are not hindered by the many constraints and differences between the murine and human immune systems, such as tolerance and elimination. Despite the strength in evaluating different research topics, for example, anaphylaxis, virus infections, or immunotherapies, humanized mice unfortunately remain a niche tool, as they are restricted primarily by their complexity in generation and readout, as well as small cohort sizes [\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eHumanization models are based on the injection of human cells into highly immunocompromised and xeno-tolerant mice strains (e.g. NSG, NOG). While injecting peripheral mononuclear cells (PBMCs) allows to establish a human immune system within the recipient mice, it leads to induction of graft\u0026ndash;versus-host disease (GvHD) demise in 2\u0026ndash;3 weeks, allowing only limited and strongly biased analysis. A more elegant animal model is generated by only injecting CD34\u003csup\u003e+\u003c/sup\u003e stem cells. After engraftment, these stem cells allow the reconstitution of human, xeno-tolerant, immune cells. Early work showed a B-cell biased reconstitution following CD34\u003csup\u003e+\u003c/sup\u003e cell transfer in NSG mice, leading to the development of NSG mice transgenic for human cytokines. The NSG-SGM3 mouse \u0026ndash; transgenic for human stem cell factor (SCF), granulocyte-macrophage colony-stimulating factor (GM-CSF) and interleukin 3 (IL3) \u0026ndash; is one such model and has significantly improved myeloid cell reconstitution [\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e]. Especially when examining immune responses \u003cem\u003ein toto\u003c/em\u003e, myeloid cells play an important role, as seen by findings from Norelli \u003cem\u003eet al.\u003c/em\u003e, which show that the deadly cytokine release syndrome (CRS) in CAR T cell transfer is predominantly driven by \u0026ldquo;bystander\u0026rdquo; monocytes, only seen in humanized NSG-SGM3 mice [\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e]. Especially CD4\u0026thinsp;+\u0026thinsp;CAR T cells seem to be the driver of CRS in humanized NSG-SGM3 mice[\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eCD34\u003csup\u003e+\u003c/sup\u003e cells can be isolated from adult bone marrow, mobilized peripheral blood, fetal liver cells or umbilical cord blood (UCB) and contain human stem and progenitor cells. Adult sources often only result in transient, partial and low humanization levels [\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e, \u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e] and thus, are a difficult tool for achieving stable and comprehensive humanized mouse models. In comparison, fetal liver and UCB cells allow full lineage differentiation and long-term reconstitution. Fetal liver cells are problematic to obtain, thereby also giving rise to only very limited options. In contrast, UCB is considered a by-product of birth, thus it is more accessible compared to other stem cells sources and still contains potent CD34\u003csup\u003e+\u003c/sup\u003e stem cells. Therefore, UCB is the preferred choice for most researchers.\u003c/p\u003e \u003cp\u003eThe most common CD34\u003csup\u003e+\u003c/sup\u003e cell isolation method from UCB is magnetic bead-based isolation, which relies on magnetic beads conjugated to anti-CD34 antibodies to positively isolate hematopoietic stem cells (HSC). However, inconsistencies in yield and variability in purity are observed with this approach. We have previously established fragment antigen binding (Fab)-based isolation protocols, [\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e] which were subsequently adapted into a column-based isolation system named traceless affinity cell sorting (TACS) [\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e, \u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e]. This technology uses Fab multimers (Fab monomers associating to a backbone) that bind surface markers on target cells, with the added advantage of reversibility of the catching reagents (Fab fragments). After the addition of D-biotin, the multimer complex disassembles. Specifically biotin disrupts the binding between the multimer-backbone and the Fab molecules, leaving monomeric Fab-epitope interactions, from which Fab dissociated, yielding Fab-free cells. This reversibility is especially important when isolating cells, such as T cells, where surface markers such as CD3 are critical for cell function. Residual antibodies from magnetic bead-based methods can interfere with cell functionality, but Fab-based isolation mitigates this issue. CD34 is known to be involved in cell adhesion and hence may be affected by residual antibodies from magnetic bead-based isolation protocols, potentially imparting the CD34\u003csup\u003e+\u003c/sup\u003e cells functionality [\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e]. Additionally, the TACS method offers the advantage of isolating cells directly from whole blood, bypassing density gradient separation and reducing cell loss. This approach is expected to enable higher CD34\u003csup\u003e+\u003c/sup\u003e yields than magnetic bead isolation and thus, permit larger humanized mice cohorts.\u003c/p\u003e \u003cp\u003eDespite extensive development and improvement in mouse models, a key challenge remains expanding cohort sizes due to the limited availability of CD34\u003csup\u003e+\u003c/sup\u003e cells. Previously, researchers pooled CD34\u003csup\u003e+\u003c/sup\u003e cells from multiple donors or only used small cohorts per donor, which are both sub-optimal solutions, particularly in comparative experiments. Pooling CD34\u003csup\u003e+\u003c/sup\u003e cells from different donors can have unforeseen immunological consequences, and typically, only one infused donor achieves long-term engraftment [\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e, \u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e]. The issue of limited cohort sizes is aggravated when replicating diseases in humanized mice with specific leucocyte traits, such as human-leucocyte antigen (HLA) or Killer cell Ig-like receptor (KIR) types. Modelling diseases within specific immunological constraints still requires suitable cohort sizes while working with limited HLA- or KIR-type specific CD34\u003csup\u003e+\u003c/sup\u003e cells. Under these constraints, researchers require sufficient humanized mice for downstream applications without compromising humanization levels, as low humanization levels can confound results. Here we describe optimal CD34\u003csup\u003e+\u003c/sup\u003e cell doses, titrated in the low range, for both excellent humanization levels and large single donor huNSG-SGM3 cohorts.\u003c/p\u003e"},{"header":"Methods","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003eCell source material\u003c/h2\u003e \u003cp\u003eUCB was collected from full-term deliveries after maternal informed written consent and kindly provided by the \u0026ldquo;Klinik und Poliklinik f\u0026uuml;r Frauenheilkunde \u0026ldquo;of the Technical University Munich (reference no. 617/19S). UCB was collected in UCB blood bags (Macopharma, Germany, Cat.# MSC1200PU) or EDTA tubes (Saarsted, Germany, Cat.# 01.1605).\u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003eMagnetic bead isolation for CD34 cells\u003c/h3\u003e\n\u003cp\u003eAll UCB was processed within 6 hours of collection. CD34\u003csup\u003e+\u003c/sup\u003e cells were isolated as per the manufacturer\u0026rsquo;s instruction using the CD34 MACS system (Miltenyi Biotec, Germany, Cat.# 130-046-702). In brief, UCB was diluted 1:1 in phosphate buffered saline (pH\u0026thinsp;=\u0026thinsp;7.40) (PBS) (Merck, Germany, Cat.# 56064C), layered on 1,077 g/mL Biocoll (Merck, Germany, Cat.# L6715-BC) and centrifuged for 20 min at 800g with no brake. Mononuclear cells were transferred into FACS (PBS 1x (Merck, Germany) containing 0.5% (w/v) bovine serum albumin (BSA) (Sigma-Aldrich, Germany, Cat.# 05470), pH\u0026thinsp;=\u0026thinsp;7.40) buffer and washed before adding Fcγ blocker and anti-CD34 magnetic beads. After 30 min, CD34\u003csup\u003e+\u003c/sup\u003e cells were eluted using the MidiMACS separator magnet and LS columns (both Miltenyi Biotec, Germany, Cat.# 130-042-301 \u0026amp; 130-042-401). 5% of isolated cells were transferred for purity and exact CD34\u003csup\u003e+\u003c/sup\u003e cell count for flow cytometry analysis. The remaining 95% of isolated cells were cryopreserved in a 250 \u0026micro;L mix of 10% dimethyl sulfoxide (Merck, Germany, Cat.# 67-68-5) and 90% fetal calf serum (FCS) (GE Healthcare, United Kingdom). Cells were stored in -80\u0026deg;C and/or liquid nitrogen.\u003c/p\u003e\n\u003ch3\u003eTraceless Affinity Cell Selection (TACS) for CD34 + cells\u003c/h3\u003e\n\u003cp\u003eCD34\u003csup\u003e+\u003c/sup\u003e cells were isolated from whole UCB by adapting our previously established protocols [\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e]. Following aseptic technique, Poly-Prep chromatography columns (Bio-Rad, USA) were cleaned using 80% (v/v) Ethanol and loaded with 1 mL Streptactin agarose beads (IBA Lifesciences, Germany, Cat.# 6-6350-002) together with 6 mL PBS (Merck, Germany). The agarose beads were left to settle, and columns were stored at 4\u0026deg; C in sterile 50 mL Falcon tubes (Greiner Bio-one, Germany) until use.\u003c/p\u003e \u003cp\u003eImmediately before isolation, 45 \u0026micro;g CD34-targeting Fab-fragments (IBA Lifesciences, Germany) dissolved in 1 mL sterile PBS was loaded onto the column to coat the agarose bead bed. After washing with sterile PBS, UCB diluted 1:20 (v/v) with citrate phosphate dextrose solution (Sigma-Aldrich, Germany Cat.# C7165) was loaded directly onto the column. The CD34\u003csup\u003e+\u003c/sup\u003e cells were retained via binding to the Fab molecules on the agarose bead surface. After washing with four runs of 10 mL PBS, cells were eluted using 1 mM D-biotin (Merck, Germany, Cat.# 2031) in FACS buffer and collected for further analysis. Cryopreservation was performed in section \u0026ldquo;Magnetic Bead isolation for CD34\u003csup\u003e+\u003c/sup\u003e cells\u0026rdquo;.\u003c/p\u003e\n\u003ch3\u003eThawing of cryopreserved CD34 cells\u003c/h3\u003e\n\u003cp\u003eCD34\u003csup\u003e+\u003c/sup\u003e cells were thawed in warm water and immediately transferred into warm RPMI 1640 media (ThermoFisher Scientific, Germany, Cat.#11875093) containing 10% (v/v) FCS (GE Healthcare, United Kingdom). Cells were spun down and resuspended in RPMI containing 10% FCS medium. 5% of the volume was used for flow cytometry analysis, and the remaining cells were rested at 37\u0026deg; C, 5% CO\u003csub\u003e2\u003c/sub\u003e until flow cytometry analysis was done (approximately 1-1.5 hours). After flow cytometry quality control and cell count analysis, the cells were centrifuged down, and resuspended in the desired volume of FCS for the desired cell dose and used for humanization.\u003c/p\u003e\n\u003ch3\u003eGeneration of HuNSG-SGM3\u003c/h3\u003e\n\u003cp\u003eNOD.\u003cem\u003eCg-Prkdc\u003c/em\u003e\u003csup\u003e\u003cem\u003escid\u003c/em\u003e\u003c/sup\u003e\u003cem\u003eIl2rg\u003c/em\u003e\u003csup\u003e\u003cem\u003etm1Wjl\u003c/em\u003e\u003c/sup\u003e Tg(CMV-IL3, CSF2, KITLG)1Eav/MloySzJ (NSG-SGM3) mice were purchased from Jackson Laboratory (USA) and housed under specific-pathogen-free (SPF) conditions. We only used female mice housed in groups of 3\u0026ndash;6 littermates per cage. Mice within a cage were given the same treatment. At 4 weeks of age, mice were irradiated with 1 Gy from a Caesium-137 source. 16\u0026ndash;24 hours later, NSG-SGM3 mice were intravenously injected with 150 \u0026micro;L FCS containing varying doses of CD34\u003csup\u003e+\u003c/sup\u003e cells. Cell dose titration using very low, low and high CD34\u0026thinsp;+\u0026thinsp;cell doses was performed to assess humanization rates as outlined in Supplementary Table\u0026nbsp;1. The CD34\u003csup\u003e+\u003c/sup\u003e cell numbers represent the total living cells per mouse as assessed by flow cytometry. Each mouse only received cells from one donor. After injection, mice were monitored regularly for clinical signs of GvHD (e.g. skin \u0026amp; fur changes, hunched back, excessive weight loss).\u003c/p\u003e \u003cdiv id=\"Sec8\" class=\"Section2\"\u003e \u003ch2\u003eAssessment of humanization levels\u003c/h2\u003e \u003cp\u003eMice were bled via the tail vein every 2 weeks for 12 weeks. Peripheral mouse blood was erythrocyte lysed using Tris-buffered (10% v/v) Ammonium Chloride (90% v/v) pH 7.50) (ACT) (Roth, Germany, Cat.#77-86-1 and 12125-02-9) and stained for various human leucocyte surface markers before flow cytometry analysis described below (see flow cytometry).\u003c/p\u003e \u003cp\u003e12 weeks after CD34\u003csup\u003e+\u003c/sup\u003e cell injection, mice were sacrificed by cervical dislocation. Spleen and BM (bone marrow from tibia and femur from each side) were harvested. Spleens were mashed through a 70 \u0026micro;m Nylon-mesh to create a single-cell suspension. BM was extracted by inserting a 25 G syringe and flushing out the BM cells with Dulbecco's Modified Eagle medium (ThermoFisher Scientific, Germany, Cat.# A4192101). A BM single-cell suspension was generated by rigorously pipetting the extracted cells. Single cell suspensions were centrifuged down, and residual erythrocytes in the spleen and BM samples lysed using ACT. Cells were washed and a portion of the single-cell suspensions were used for downstream analysis.\u003c/p\u003e \u003cp\u003eHumanization levels were defined as \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:hum\\left(\\%\\right)=\\frac{human\\:CD45+}{all\\:CD45+}\\times\\:100\\%\\)\u003c/span\u003e\u003c/span\u003e.\u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003eFlow cytometry\u003c/h3\u003e\n\u003cp\u003eCells were transferred into a V-bottom 96-well plate and incubated at 4\u0026deg; C with various surface antibodies for 20 minutes. 10 \u0026micro;L of whole blood was used for CD34\u003csup\u003e+\u003c/sup\u003e analysis in whole UCB and after washing with FACS buffer stained with human anti-huCD34 (4H11; eBiosciences, Germany; Cat.# 11-0349-42) and anti-CD45 (HI30; ExBio, Czech Republic; Cat.# PO-684-T100). The CD34\u003csup\u003e+\u003c/sup\u003e cell isolation product was assessed by staining with anti-huCD34, anti-huCD45, anti-huCD3 (UCHT1; eBiosciences, Germany; Cat.# 17-0038-42), anti-huCD19 (J3-119; Beckman Coulter, USA; Cat.# A07770) and anti-huCD33 (WM53; BioLegend, USA; Cat.# 303416). The humanization and reconstitution levels were analyzed through anti-muCD45.1 (A20; Biolegend, USA; Cat.# 110708), anti-huCD45, anti-huCD3, anti-huCD19, anti-huCD33 and anti-huCD56 (TULY56; eBiosciences, Germany; Cat.# 11-0566-42). After surface staining, cells were washed, and 1:1000 (v/v) propidium iodide (Merck, Germany) was added for live-dead discrimination. Additionally, 123 eCounting Beads (Thermofisher, Germany, Cat.# 01-1234-42) were added to each sample for accurate cell counts. The gating strategy is outlined in Supplementary Fig.\u0026nbsp;1A and B.\u003c/p\u003e \u003cp\u003eCells were acquired on a Cytoflex S flow cytometer (Beckman Coulter, USA), and whole blood flow cytometry was performed on a Cyan ADP flow cytometer (Beckman Coulter, USA).\u003c/p\u003e\n\u003ch3\u003eCFC assay\u003c/h3\u003e\n\u003cp\u003eA human HSC colony-forming assay, which only allows for myeloid and erythroid development, was used to analyze the composition and clonogenicity of the CD34\u003csup\u003e+\u003c/sup\u003eCD45\u003csup\u003e+\u003c/sup\u003e cells. A small fresh aliquot of the final isolated CD34\u003csup\u003e+\u003c/sup\u003e cells containing approximately 200 cells was resuspended in the CFC media (R\u0026amp;D Systems, USA; Cat.# HSC005) and plated for 14 days. The aliquots were collected from both the magnetic bead and TACS protocols. After 14 days the plates were blinded, and the number and type of colonies were visually scored.\u003c/p\u003e \u003cdiv id=\"Sec11\" class=\"Section2\"\u003e \u003ch2\u003eBone histology\u003c/h2\u003e \u003cp\u003eFemur and tibia of one side were harvested. Excess soft tissues were removed and bones were washed in PBS before being fixed through immersion in 4% PFA. After 48 hours the samples were washed and transferred into 70% (v/v) ethanol until further processing. Tissues were decalcified in 10% EDTA, pH 7.4 at 37\u0026deg;C for 1 week, before being processed in a Leica ASP300S tissue processor, embedded in paraffin wax, and sectioned at 5 \u0026micro;m using a Leica rotary microtome. Slides were stained for H\u0026amp;E using a Leica XL Autostainer. Immunohistochemistry for human-specific anti-huCD45 (Agilent, USA; Cat.# M070101) was performed following antigen retrieval in pH 6.0 10 mM tri-sodium citrate buffer, 0.05% Tween-20, as previously described [\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e, \u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e]. Briefly, endogenous peroxidase activity was quenched with 3% hydrogen peroxide (Sigma-Aldrich, Australia, Cat.#7722-84-1) for 15 minutes. Non-specific binding sites were blocked with 2% BSA (Sigma-Aldrich, Australia, Cat.# 9048-46-8) for 30 min before overnight incubation with primary antibodies diluted in 2% BSA at 4\u0026deg;C. Normal mouse IgG was used as a negative control. Positive immunoreactivity was detected using the Dako Envision\u0026thinsp;+\u0026thinsp;Dual Link System HRP and the Liquid DAB\u0026thinsp;+\u0026thinsp;2-component immunohistochemistry visualization system (both Agilent, USA).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec12\" class=\"Section2\"\u003e \u003ch2\u003eCytokine release syndrome (CRS) model\u003c/h2\u003e \u003cp\u003eMice were humanized using CD34\u003csup\u003e+\u003c/sup\u003e cells from a single donor. 8 weeks after humanization, mice were intravenously injected with 0.5x10\u003csup\u003e6\u003c/sup\u003e CD19\u003csup\u003e+\u003c/sup\u003e Raji leukemia tumor (ATCC, UK). One week later, mice were weighed and intravenously injected. 0.8x10\u003csup\u003e6\u003c/sup\u003e CD8\u003csup\u003e+\u003c/sup\u003e CAR T-cells or mock transduced CD8\u003csup\u003e+\u003c/sup\u003e T cells from the same donor. Mice were examined and weighed daily for signs of CRS. Blood was collected from the tail vein mice at days 3 and 5 after T cell transfer, and serum levels of IL-6 and IL-10 were measured using LEGENDplex immunoassay kit (Multi-Analyte Flow Assay Kit, Biolegend, Germany, Cat.# 740267) per the manufacturer\u0026rsquo;s instructions.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec13\" class=\"Section2\"\u003e \u003ch2\u003eData analysis\u003c/h2\u003e \u003cp\u003eFlow cytometry data were analyzed using FlowJo 10 (Becton, Dickinson \u0026amp; Company, USA). Data was stored using Microsoft Excel (Microsoft, USA). Graphs were created, and statistical analysis was performed using GraphPad Prism 9 (GraphPad Software, USA). The manuscript was written in Microsoft Word (Microsoft, USA).\u003c/p\u003e \u003c/div\u003e"},{"header":"Results","content":"\u003cdiv id=\"Sec15\" class=\"Section2\"\u003e\n \u003ch2\u003eDonor-to-donor variations exclude many donations on grounds of low CD34\u003csup\u003e+\u003c/sup\u003e cell count\u003c/h2\u003e\n \u003cp\u003eIsolation of CD34\u003csup\u003e+\u003c/sup\u003e stem cells using established protocols leads to extensive cell loss. Current humanization protocols recommend 10\u003csup\u003e5\u003c/sup\u003e CD34\u003csup\u003e+\u003c/sup\u003e cells per mouse, and together with a poor yield, this hampers humanized mice cohort sizes (Fig. \u003cspan class=\"InternalRef\"\u003e1\u003c/span\u003eA)[\u003cspan class=\"CitationRef\"\u003e15\u003c/span\u003e, \u003cspan class=\"CitationRef\"\u003e16\u003c/span\u003e]. First, we set out to establish how many CD34\u003csup\u003e+\u003c/sup\u003e stem cells could reliably be collected using established magnetic bead isolation protocols (Fig. \u003cspan class=\"InternalRef\"\u003e1\u003c/span\u003eA). The cord blood collections were performed immediately following birth. UCB cells were not pooled as the preference was to work only with single-donor material. To ensure generation of robust data and comprehensive assessment of CD34\u003csup\u003e+\u003c/sup\u003e populations within a large cohort, we analyzed cord blood from 56 donors. Within our analyzed cord blood (n\u0026thinsp;=\u0026thinsp;56), we observed substantial donor-to-donor variation in the CD34\u003csup\u003e+\u003c/sup\u003e cell counts with a mean of 2.68 \u0026times; 10\u003csup\u003e4\u003c/sup\u003e cells/ml and a standard deviation of \u0026plusmn;\u0026thinsp;1.55 \u0026times; 10\u003csup\u003e4\u003c/sup\u003e CD34\u003csup\u003e+\u003c/sup\u003eCD45\u003csup\u003e+\u003c/sup\u003e cells/mL (Fig. \u003cspan class=\"InternalRef\"\u003e1\u003c/span\u003eB). This translates to an average of 7.5 \u0026times; 10\u003csup\u003e5\u003c/sup\u003e CD34\u003csup\u003e+\u003c/sup\u003e cells per collection from standard UBC donation volumes of 20\u0026ndash;60 ml (Fig. \u003cspan class=\"InternalRef\"\u003e1\u003c/span\u003eC), which would suffice for a handful number of humanized mice. Next, we wanted to analyze if the processing following collection impacted the average number of available stem cells. It has been described that cell processing, especially gradient centrifugation, is associated with significant cell loss [\u003cspan class=\"CitationRef\"\u003e17\u003c/span\u003e]. Thus, to determine the cell loss during CD34\u003csup\u003e+\u003c/sup\u003e cell isolation and storage (cryopreservation) we analyzed CD34\u003csup\u003e+\u003c/sup\u003e cell count using flow cytometry at each step of the isolation process. In line with data from other groups, there is a sizeable percentual cell loss during cell isolation and cryopreservation, resulting in on average just 36.1% cells available for \u003cem\u003ein vivo\u003c/em\u003e application (Fig. \u003cspan class=\"InternalRef\"\u003e1\u003c/span\u003eD). The thawed injection product was pure for CD34\u003csup\u003e+\u003c/sup\u003e cells with only few CD45\u003csup\u003e+\u003c/sup\u003e cells (Fig. \u003cspan class=\"InternalRef\"\u003e1\u003c/span\u003eE) and virtually absent CD3\u003csup\u003e+\u003c/sup\u003e T cells (Fig. \u003cspan class=\"InternalRef\"\u003e1\u003c/span\u003eF).\u003c/p\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec16\" class=\"Section2\"\u003e\n \u003ch2\u003eTACS leads to higher CD34\u0026thinsp;+\u0026thinsp;cell purity and yield compared to magnetic bead isolation\u003c/h2\u003e\n \u003cp\u003eTACS offers an opportunity to isolate CD34\u003csup\u003e+\u003c/sup\u003e cells using more streamlined processes, which we optimized to isolate cells directly from whole UCB (Supplementary Fig. 2). Thus, we sought to compare the efficacy of this system to standard magnetic bead-based isolation protocols (mag. bead). TACS led to a higher yield and purity of CD34\u003csup\u003e+\u003c/sup\u003e cells and a more consistent final product, with reduced variability compared to mag. bead-isolated cells (CV\u003csup\u003eyield\u003c/sup\u003e 33% versus 43.1%; CV\u003csup\u003epurity\u003c/sup\u003e 26.9% versus 43.7%) (Fig. \u003cspan class=\"InternalRef\"\u003e2\u003c/span\u003eA-B). While TACS resulted in a greater yield of CD34\u003csup\u003e+\u003c/sup\u003eCD45\u003csup\u003e+\u003c/sup\u003e cells, this increase was not statistically significant. However, TACS consistently produced a higher purity of isolated CD34\u003csup\u003e+\u003c/sup\u003e cells with less variability across 26 unique UBC donations (Fig. \u003cspan class=\"InternalRef\"\u003e2\u003c/span\u003eC).\u003c/p\u003e\n \u003cp\u003eNotably, both isolation protocols produced CD34\u003csup\u003e+\u003c/sup\u003e cell products with comparable phenotypical of progenitor cells (HSC, Myeloid-primed progenitors (MPP), Lympho-myeloid primed progenitors (LMPP)) (Fig. \u003cspan class=\"InternalRef\"\u003e2\u003c/span\u003eD). We further analyzed the composition and clonogenicity of the CD34\u003csup\u003e+\u003c/sup\u003eCD45\u003csup\u003e+\u003c/sup\u003e cells through a 14-day CFC assay (Fig. \u003cspan class=\"InternalRef\"\u003e2\u003c/span\u003eE). As expected, the CFC assay, which only supports myeloid and erythroid lineage development, did not produce lymphoid colonies. Most colonies were erythroid (CFU-E/BFU-E), followed by pure granulocytic colonies (CFU-G) (Fig. \u003cspan class=\"InternalRef\"\u003e2\u003c/span\u003eF-G). Less than 5% of colonies consisted of granulocytic, monocytic, macrophagic and erythroid cells (CFU-GEMM), which were derived from multipotent progenitors (MPPs) or HSCs (Fig. \u003cspan class=\"InternalRef\"\u003e2\u003c/span\u003eF). The colony distribution of TACS-isolated cells mirrored that of mag. bead-isolated cells, with no observable differences in the proportional composition of colony types between the two methods (Fig. \u003cspan class=\"InternalRef\"\u003e2\u003c/span\u003eF). Additionally, by controlling the initial input dose to 200 CD34\u003csup\u003e+\u003c/sup\u003eCD45\u003csup\u003e+\u003c/sup\u003e cells, we determined that both methods produced approximately 100 colonies, further indicating equivalent clonogenic potential (Fig. \u003cspan class=\"InternalRef\"\u003e2\u003c/span\u003eG). Thus, we conclude that TACS provides superior CD34\u003csup\u003e+\u003c/sup\u003e cell purity and yield while maintaining the same progenitor cell phenotypic and functional characteristics as mag. bead isolation.\u003c/p\u003e\n \u003ch3\u003e\u003cspan type=\"Underline\" class=\"Underline\" name=\"Emphasis\"\u003eHumanized levels are equivalent between TACS and mag. bead-isolated CD34\u0026thinsp;+\u0026thinsp;cells using an\u003c/span\u003e \u003cspan type=\"ItalicUnderline\" class=\"ItalicUnderline\" name=\"Emphasis\"\u003ein vivo\u003c/span\u003e \u003cspan type=\"Underline\" class=\"Underline\" name=\"Emphasis\"\u003eengraftment efficiency model\u003c/span\u003e\u003c/h3\u003e\n \u003cp\u003eTo assess the functional engraftment capacity of CD34\u003csup\u003e+\u003c/sup\u003e cells isolated by TACS, we compared the \u003cem\u003ein vivo\u003c/em\u003e performance to that of mag. bead-isolated cells using a humanized NSG-SGM3 mouse model. To remove donor variability, we strictly used TACS and mag. bead-isolated cells from the same donor for all comparisons. Peripheral humanization was tracked over 12 weeks in peripheral blood, while bone marrow (BM) and spleen humanization levels were analyzed at the endpoint. In a lead-up experiment with mag. bead-isolated cells, we observed a wider humanization spread and thus, increase in CV at doses below 15,000 (Supplementary Fig. 3A-B). Thus, we performed a dilution series of cell inputs in doses between 15,000 \u0026ndash; hoping to stress the system and exacerbate differences between the isolated cells. For donor 1, we compared humanization outcomes between mag. bead- and TACS-isolated cells at very low doses of 3,750, 7,500 and 15,000 CD34\u003csup\u003e+\u003c/sup\u003eCD45\u003csup\u003e+\u003c/sup\u003e cells per mouse. Across all doses and timepoints, there were no significant differences in the peripheral humanization levels between the two isolation methods (Fig. \u003cspan class=\"InternalRef\"\u003e3\u003c/span\u003eA). However, due to the limited number of mice (1\u0026ndash;2 per group), we sought to increase the statistical power by replicating this experiment with a larger cohort of mice. For donor 2, we tested the very low input doses of 1,200 and 12,000 CD34\u003csup\u003e+\u003c/sup\u003eCD45\u003csup\u003e+\u003c/sup\u003e cells with 5\u0026ndash;6 mice per group. Consistent with the first experiment, humanization levels were comparable between mag. bead- and TACS-isolated cells across all doses tested. Only at 12 weeks did 12,000 TACS-isolated cells have slightly higher humanization levels (P\u0026thinsp;=\u0026thinsp;0.0230) (Fig. \u003cspan class=\"InternalRef\"\u003e3\u003c/span\u003eB). Together, these results establish that TACS-isolated cells perform similarly to the CD34\u003csup\u003e+\u003c/sup\u003e cells isolated using mag. beads.\u003c/p\u003e\n \u003cp\u003eCentral humanization, particularly in the BM and spleen, is an integral part of functional humanization in these mice models. Thus, we assessed whether TACS-isolated cells functionally humanized organs similarly to mag. bead-isolated cells. For donor 1, we observed similar humanization levels in both BM and spleen at all doses (Fig. \u003cspan class=\"InternalRef\"\u003e3\u003c/span\u003eC). Notably, at the 3,750 cell dose, there was increased variability in humanization levels, likely due to suboptimal cell input approaching the threshold for maximum achievable engraftment levels in NSG-SGM3 mice (Fig. \u003cspan class=\"InternalRef\"\u003e3\u003c/span\u003eC). To further challenge the system and amplify any functional difference between the two isolation methods, donor 2 cells were tested at an even lower input dose of 1,200 cells. As expected, the 1,200 cell dose resulted in lower and more variable engraftment levels in both BM and spleen than the 12,000 cell dose. However, no significant differences were detected between the isolation methods (Fig. \u003cspan class=\"InternalRef\"\u003e3\u003c/span\u003eD). We conclude that the TACS system produces CD34\u003csup\u003e+\u003c/sup\u003eCD45\u003csup\u003e+\u003c/sup\u003e cells that are functionally equivalent to those isolated by standard mag. beads for in vivo humanization, even at low input doses.\u003c/p\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec17\" class=\"Section2\"\u003e\n \u003ch2\u003eLow CD34\u003csup\u003e+\u003c/sup\u003e input doses result in robust and predictable humanization\u003c/h2\u003e\n \u003cp\u003eAlthough our TACS system performed equivalent to mag. beads in terms of yield and humanization, the current gold standard of the field is mag. bead-based isolation. Thus, using our established humanization model, we investigated the influence of varying mag. bead-isolated CD34\u003csup\u003e+\u003c/sup\u003e cell doses on immune system humanization. To first establish a baseline humanization trajectory in NSG-SGM3 mice, we injected 4 week old mice with a high dose (60,000) of living CD34\u003csup\u003e+\u003c/sup\u003e cells, based on previously published doses [\u003cspan class=\"CitationRef\"\u003e16\u003c/span\u003e]. Peripheral blood was analyzed every 2 weeks for 12 weeks using flow cytometry (Supplementary Fig.\u0026nbsp;4A). Mice did not show signs of GvHD or illness (Supplementary Fig.\u0026nbsp;4B), even with high humanization percentages greater than 45% of human CD45\u003csup\u003e+\u003c/sup\u003e cells from total CD45% cells present in peripheral blood (Supplementary Fig. 4C-D), Moreover, engrafted CD34\u003csup\u003e+\u003c/sup\u003e cells possessed capacity to differentiate into T-, B- and myeloid cell lineages (Supplementary Fig. 4E-F). At the endpoint 12 weeks after CD34\u003csup\u003e+\u003c/sup\u003e cell engraftment, bone marrow and spleen humanization was robust and maintained lineage differentiation (Supplementary Fig. 4G-H).\u003c/p\u003e\n \u003cp\u003eWe next investigated the influence that very low CD34\u003csup\u003e+\u003c/sup\u003e cell doses have on humanization levels in NSG-SGM3 mice (Fig. \u003cspan class=\"InternalRef\"\u003e4\u003c/span\u003eA). Therefore, we titrated the number of transplanted cells down to 3,750 viable CD34\u003csup\u003e+\u003c/sup\u003e cells per mouse, showing that at very low cell doses reconstitution is slower but still robust, achieving\u0026thinsp;\u0026gt;\u0026thinsp;45% human CD45\u003csup\u003e+\u003c/sup\u003e cells in peripheral blood (Fig. \u003cspan class=\"InternalRef\"\u003e4\u003c/span\u003eB). Our results also showed that the humanization CV was relatively stable across all but the lowest doses, suggesting that the lowest dose of CD34\u003csup\u003e+\u003c/sup\u003e cells led to less predictability and uniformity of engraftment compared to the moderate and high cell doses (Fig. \u003cspan class=\"InternalRef\"\u003e4\u003c/span\u003eC). The robust BM humanization observed with the highest cell dose of 60,000 was reproducible with much lower cell doses. Mice receiving a very low dose of 7,500 CD34\u003csup\u003e+\u003c/sup\u003e cells or more all achieved similar spleen humanization levels (Fig. \u003cspan class=\"InternalRef\"\u003e4\u003c/span\u003eD). In the spleen, humanization levels were high, exceeding 80%, and while absolute cell numbers showed no statistically significant differences across all doses, humanization levels were different; however, the biological relevance remains uncertain (Fig. \u003cspan class=\"InternalRef\"\u003e4\u003c/span\u003eD and E). Immunohistochemistry confirmed the presence of human CD45\u003csup\u003e+\u003c/sup\u003e cells in the bone marrow at endpoint in the very low, low, and high cell dose groups (Fig. \u003cspan class=\"InternalRef\"\u003e4\u003c/span\u003eF).\u003c/p\u003e\n \u003cp\u003eImportant for downstream application is also a homogeneity of engraftment of human CD45\u003csup\u003e+\u003c/sup\u003e cell populations across all humanized mice. We assessed this by looking at the relative deviation across blood, BM, and spleen (Supplementary Fig. 3A-B). Especially in blood and BM, very low doses of 3,750 and 7,500 cells show larger relative spread of human CD45\u003csup\u003e+\u003c/sup\u003e cells measured between different mice than very low 15,000, low 30,000 and high 60,000 input doses. The spleen deviation was very small across the board, with particularly less deviation in cell doses of 15,000 or greater. The plateauing CV% seems to indicate that the stem cell niche is saturated with far fewer CD34\u003csup\u003e+\u003c/sup\u003e cells than previously thought (Supplementary Fig. 3B).\u003c/p\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec18\" class=\"Section2\"\u003e\n \u003ch2\u003eDevelopment of the human leucocyte compartment is not affected by lower CD34\u003csup\u003e+\u003c/sup\u003e cell doses\u003c/h2\u003e\n \u003cp\u003eHistorically the goal of humanized mice was to engraft human CD34\u003csup\u003e+\u003c/sup\u003e stem cells within the mouse model system. Next generation humanized mouse model strains, such as the NSG-SGM3, have focused on not only engraftment but robust lineage differentiation of the stem cells. The human cytokines produced by the NSG-SGM3 mice strain support robust myeloid differentiation and previous publications have extensively characterized and compared this strain to the NSG model [\u003cspan class=\"CitationRef\"\u003e18\u003c/span\u003e, \u003cspan class=\"CitationRef\"\u003e19\u003c/span\u003e]. We assessed the leucocyte compartment for the presence of T cells, B cells and CD33\u003csup\u003e+\u003c/sup\u003e myeloid cells, demonstrating that while the early reconstitution is dominated by CD33\u003csup\u003e+\u003c/sup\u003e myeloid cells, CD19\u003csup\u003e+\u003c/sup\u003e B cells and CD3\u003csup\u003e+\u003c/sup\u003e T cells reconstitute later after week 6 and week 10 (Fig. \u003cspan class=\"InternalRef\"\u003e5\u003c/span\u003eA-B). We previously observed that absolute hCD45\u003csup\u003e+\u003c/sup\u003e and CD33\u003csup\u003e+\u003c/sup\u003e cell counts remain constant after 6 weeks (Supplementary Fig. 4D-E) and that the shift from a B cell dominated to a T cell dominated compartment results in an absolute decrease in B cells, simultaneous to an absolute increase in T cells in the analyzed organs (Supplementary Fig. 5E-F). This trend is found across all cell doses and with remarkable stability (Fig. \u003cspan class=\"InternalRef\"\u003e5\u003c/span\u003eB). Only the T cell emergence is slowed in the lowest dose. It is likely that these reconstitution dynamics reflect the differentiation of CD34\u003csup\u003e+\u003c/sup\u003e cells in the huNSG-SGM3 model and deviations from these dynamics result from engraftment of CD34\u003csup\u003e\u0026minus;\u003c/sup\u003e cells in the injection product; something researchers should be particularly aware of. Across different organs we also saw substantially different leucocyte compartments, with CD33\u003csup\u003e+\u003c/sup\u003e myeloid cells dominating the BM and spleen favoring CD19\u003csup\u003e+\u003c/sup\u003e B cells. This trend is consistent across all doses, although in lower CD34\u003csup\u003e+\u003c/sup\u003e cell doses T cell reconstitution is delayed (Fig. \u003cspan class=\"InternalRef\"\u003e5\u003c/span\u003eC). Further the reconstitution of a humanized immune system using lower stem cell doses allows to study large cohorts of mice. This is important, as large cohorts of single-donor humanized mice are required in a variety of applications, for example in studying allergies, infectious diseases, or cancer.\u003c/p\u003e\n \u003ch3\u003e\u003cspan type=\"Underline\" class=\"Underline\" name=\"Emphasis\"\u003eLarge cohorts of humanized NSG-SGM3 mice can be generated using 15,000 CD34\u003c/span\u003e \u003csup\u003e\u0026nbsp;\u003cspan type=\"Underline\" class=\"Underline\" name=\"Emphasis\"\u003e+\u003c/span\u003e\u0026nbsp;\u003c/sup\u003e \u003cspan type=\"Underline\" class=\"Underline\" name=\"Emphasis\"\u003ecells per mouse and serve as a potent model for CAR-T cell-induced cytokine release syndrome\u003c/span\u003e\u003c/h3\u003e\n \u003cp\u003eOne of the applications of huNSG-SGM3 mice is modelling cytokine-release syndrome (CRS) following CAR T-cell therapy [\u003cspan class=\"CitationRef\"\u003e3\u003c/span\u003e]. CRS can be lethal and is a hinderance in the adoption of CAR-T-cell therapy for human treatment. It also underlies significant inter-donor severity, thus using pooled donor cells is not an appropriate approach. As a proof of concept, we used the previously established very low 15,000 CD34\u003csup\u003e+\u003c/sup\u003e cell dose to robustly humanize 30 NSG-SGM3 mice from a single UCB donor. After reconstitution for 8 weeks, we injected CD19\u003csup\u003e+\u003c/sup\u003e tumor cells followed by anti-CD19 CAR T cells one week later (Fig. \u003cspan class=\"InternalRef\"\u003e6\u003c/span\u003eA). We transferred CAR T cells after 8 weeks of humanization, when the leucocyte compartment contains CD33 cells but is not yet dominated by CD3 cells. Reconstitution dynamics can vary by lab and donor, so this timing should also ensure a more stable system and robust results for researchers. Across all 30\u0026ndash;mice, we saw high humanization levels, with an expected B cell dominated hCD45\u003csup\u003e+\u003c/sup\u003e compartment (Fig. \u003cspan class=\"InternalRef\"\u003e6\u003c/span\u003eB-C). With such a cohort size, we were able to compare three conditions with 6\u0026ndash;8 mice per group, which would have not been possible using older protocols with high cell numbers (Fig. \u003cspan class=\"InternalRef\"\u003e6\u003c/span\u003eA). Following CAR-T cell injection, we monitored mice for clinical features of CRS. Mock transduced T cells did not elicit CRS, while all clones of CAR-T cells caused substantial weigh loss in line with CRS (Fig. \u003cspan class=\"InternalRef\"\u003e6\u003c/span\u003eD). We further confirmed CRS by measuring inflammatory cytokines IL-6 and IL-10 at 3 and 5 days after CAR-T cell transfer. Each CAR-T clone elicited significantly elevated IL-6 and IL-10 in the serum of the mice. (Fig. \u003cspan class=\"InternalRef\"\u003e6\u003c/span\u003eE). Both weight loss and IL-6 levels are well in line with previously published results by \u003cem\u003eNorrelli et al.\u003c/em\u003e, who reported that CRS is marked by high systemic IL-6 levels and profound weight loss, and that CRS severity correlates with tumor burden [\u003cspan class=\"CitationRef\"\u003e3\u003c/span\u003e].\u003c/p\u003e\n\u003c/div\u003e"},{"header":"Discussion","content":"\u003cp\u003eHumanized mice are a powerful tool for assessing human-like immune responses in small animals. Previous work has focused, in part, on optimizing the host through transgenic cytokine expressions [\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e, \u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e]. This has led to an assortment of potential host strains, each appropriate to slightly different experiments. Nonetheless, the described workflows for humanizing mouse immune systems are largely ubiquitous but also tedious. Additionally, the shortage of donor material typically leads to the mixing of hematopoietic stem cell donors to gain sufficient numbers to perform experiments. This mixing may potentially lead to unnatural diverse and complex HLA compositions reducing many benefits of using human immune cells to a minimum as the models are reduced artificially.\u003c/p\u003e \u003cp\u003eTo circumvent this artificial obstruction, we here describe a workflow for the humanization of mouse immune cells and highlight constraints due to CD34\u003csup\u003e+\u003c/sup\u003e cell doses. Total CD34\u003csup\u003e+\u003c/sup\u003e cell counts in a single UCB donation are limited and primarily restricts its clinical use to pediatric stem cell transfer [\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e, \u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e]. To develop a more stable transfer system we first defined the cell product we are working with. The CD34\u003csup\u003e+\u003c/sup\u003e cell count per mL and total cell collection amount is variable. (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eB \u0026amp; C) Furthermore, cell isolation significantly reduces available cells. Overall, the purity of stem cells was high, but the total amount was so limited that it hinders larger cohort humanization experiments.\u003c/p\u003e \u003cp\u003eIn particular, large cell losses are frequently observed during density gradient centrifugation, a non-affinity-based method that is a prerequisite for magnetic bead-based cell isolation. This challenge is not unique to CD34\u003csup\u003e+\u003c/sup\u003e cells, but is common across all rare cell types and is observed for example also in isolating circulating tumor cells, highlighting an inherent limitation of density gradient centrifugation [\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e]. Despite this limitation, we still achieve satisfactory results when using density gradient centrifugation. However, alternative methods, such as whole blood isolation protocols, present their own challenges, particularly when processing larger volumes of blood [\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e, \u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e]. Hence, a further optimization could be using an alternative wholeblood-based isolation protocol, such as traceless affinity cell selection (TACS) developed in our lab [\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e]. Here, we demonstrated that CD34\u003csup\u003e+\u003c/sup\u003e cell isolation using the TACS method was a simpler and more efficient process compared to mag. based isolation. Moreover, the TACS approach more frequently produced CD34\u003csup\u003e+\u003c/sup\u003e cells with higher yield and purity while maintaining functional equivalent to those isolated via magnetic beads. Residual contaminating cell populations, particularly T cells, can lead to GvHD reaction in humanized mice, highlighting the importance of CD34\u003csup\u003e+\u003c/sup\u003e cell purity for experimental procedures [\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e]. Nonetheless, we purposely abstained from this more \u0026ldquo;modern\u0026rdquo; and innovative isolation method in the cell dosing study, as mag. bead isolation still remains the most widespread isolation method and thus, appeals to the broadest audience.\u003c/p\u003e \u003cp\u003eAs a baseline comparison we used high 60,000 CD34\u003csup\u003e+\u003c/sup\u003e cells, a cell count known to elicit robust humanization levels across all compartments [\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e]. We then methodically reduced the input dose to evaluate the robustness of the system, hereby showing that a reduction of transferred human stem cells down to a very low dose of 3,750 living CD34\u003csup\u003e+\u003c/sup\u003e cells still led to robust albeit lower humanization of the immune system (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eB). This is in line with other publications showing that from cell types with a high plasticity and inherent stemness, even a singular cell is able to reconstitute a whole organism [\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e, \u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eOur reported reconstitution kinetics with an early myeloid dominance and a later emergence of T cells is in-line with various previous studies [\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e, \u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e]. Earlier work from Lang \u003cem\u003eet al.\u003c/em\u003e, report that T cell emergence coincides with B cell maturation and lymph node occupation in humanized mice, suggesting engraftment kinetics may be influenced by tissue microenvironment interactions. Similarly, early myeloid dominance and later T cell emergence is observed in immune system reconstitution in humans receiving UCB transplantation and may reflect the natural timing of the lineage differentiation pathways within biological systems [\u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eFor downstream applications, humanization levels between mice must be consistent and also sufficient, else risking falsifying results. Furthermore, a careful balance between available cells, mice numbers and humanization output is crucial for a successful workflow. Using our workflow, we determined that a very low dose of 15,000 CD34\u003csup\u003e+\u003c/sup\u003e cells provided an optimal balance between utilization and output. In fact our observed reconstitution levels where higher and comparable to previous humanization levels shown for NSG-SGM3 mice by \u003cem\u003eMaser et al.\u003c/em\u003e and \u003cem\u003eCoughlan et al.\u003c/em\u003e, who both used very high doses of 100,000 CD34\u003csup\u003e+\u003c/sup\u003e cells per mouse [\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e, \u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e]. At our reduced and very low dose, reconstitution levels in the human CD45\u003csup\u003e+\u003c/sup\u003e compartment was on par with our saturating and previous reported dose, while also allowing four- and six-fold more humanized mice respectively. Conversely, the dose reduction also allows similar cohort sizes from fewer donor cells. This drastically increases the available (suboptimal) CD34\u003csup\u003e+\u003c/sup\u003e collections for small cohort humanization projects. From our data, we emphasize that current protocols recommending upwards of 100,000 CD34\u003csup\u003e+\u003c/sup\u003e cells per mouse exceed the necessary minimum effective threshold for successful humanization [\u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e]. Alternatively, the current unnecessarily very high recommended dosing is an entry barrier to humanized mice and our proposed cell dose reduction is an opportunity to enhance resource utilization and experimental efficiency.\u003c/p\u003e \u003cp\u003eWe demonstrate that the reduced cell dose allows \u003cem\u003ein vivo\u003c/em\u003e analysis of CRS elicited by three different CAR T-cell constructs in the setting of a single donor CD34 humanization (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003e). Robust group sizes (6\u0026ndash;8 mice) were achieved and all CAR constructs elicited weight loss and elevated inflammatory cytokine levels consistent with CRS. This model is also excellent in screening for better compositions and designs of CAR T-cells, as \u003cem\u003eArcangeli et al.\u003c/em\u003e recently showed by primarily using na\u0026iuml;ve T-cells for CAR T-cell generation [\u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e36\u003c/span\u003e]. While our work focuses on a single donor proof of concept, future research should assess inter-donor efficacy and potency. For example, using humanized mice with HLA matched tumor and CD34\u003csup\u003e+\u003c/sup\u003e cells, different off-the-shelf CAR T cell products could be assessed easing the selection of the most suitable product for the patient (or HLA composition). Inter donor variability is becoming a larger and larger factor as our understanding of the immune system and immunotherapies increases. This study emphasizes CD34\u003csup\u003e+\u003c/sup\u003e cell dose reduction in order to significantly increase humanization cohorts and/or increasing incorporation of suboptimal CD34\u003csup\u003e+\u003c/sup\u003e donations. Ultimately, we hope to lower the entry barrier of humanized mice, making this technology a more widely available research tool.\u003c/p\u003e"},{"header":"Conclusion","content":"\u003cp\u003eRecapitulating, humanized mice are powerful tool in preclinical experiments, but current protocols hamper their usage due to very high cell dosages. Here we optimize the NSG-SGM3 mouse model by decreasing the cell dose by six-fold to previously published protocols, without a reduction in humanization levels or kinetics. In our proof-of-concept experiment this allowed us to simultaneously evaluate 4 groups (6\u0026ndash;8 mice) of humanized mice with markedly lower total CD34\u003csup\u003e+\u003c/sup\u003e cell usage. We further present a simple method for isolating functionally equivalent CD34\u003csup\u003e+\u003c/sup\u003e cells from whole UCB, achieving higher yield and purity compared to standard magnetic bead-based isolation protocols. We hope that our findings will increase uptake of humanized mice projects and more importantly make single donor humanized mice the gold standard for immunotherapies.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003eIRB The cord blood biobank was approved by the local ethics committee (reference number 617/19S).\u003c/p\u003e\n \u003cp\u003eIACUC Regarding animal experimentation approval and oversight by an animal care committee, I can state that \u0026ldquo;all experiments with humanized mice were approved by the local administration (Regierung von Oberbayern) and overlooked by the local animal care committee\u0026rdquo;. If you need the exact animal experimentation approval number, please let me know (I am currently traveling).\u003c/p\u003e\n\u003ch2\u003eAcknowledgements\u003c/h2\u003e\n\u003cp\u003eThis work was supported by the Deutsche Forschungsgemeinschaft (DFG) SFB-TRR 338/1\u0026ndash;452881907 [projects A01], and SFB 1371\u0026ndash;395357507 [project P04].\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eChuprin, J., et al., \u003cem\u003eHumanized mouse models for immuno-oncology research.\u003c/em\u003e Nature Reviews Clinical Oncology, 2023. \u003cstrong\u003e20\u003c/strong\u003e(3): p. 192-206.\u003c/li\u003e\n\u003cli\u003eWunderlich, M., et al., \u003cem\u003eImproved multilineage human hematopoietic reconstitution and function in NSGS mice.\u003c/em\u003e PLoS One, 2018. \u003cstrong\u003e13\u003c/strong\u003e(12): p. e0209034.\u003c/li\u003e\n\u003cli\u003eNorelli, M., et al., \u003cem\u003eMonocyte-derived IL-1 and IL-6 are differentially required for cytokine-release syndrome and neurotoxicity due to CAR T cells.\u003c/em\u003e Nat Med, 2018. \u003cstrong\u003e24\u003c/strong\u003e(6): p. 739-748.\u003c/li\u003e\n\u003cli\u003eBove, C., et al., \u003cem\u003eCD4 CAR-T cells targeting CD19 play a key role in exacerbating cytokine release syndrome, while maintaining long-term responses.\u003c/em\u003e Journal for Immunotherapy of Cancer, 2023. \u003cstrong\u003e11\u003c/strong\u003e(1): p. e005878.\u003c/li\u003e\n\u003cli\u003eNoort, W., et al., \u003cem\u003eSimilar myeloid recovery despite superior overall engraftment in NOD/SCID mice after transplantation of human CD34+ cells from umbilical cord blood as compared to adult sources.\u003c/em\u003e Bone marrow transplantation, 2001. \u003cstrong\u003e28\u003c/strong\u003e(2): p. 163-171.\u003c/li\u003e\n\u003cli\u003eWiekmeijer, A.-S., et al., \u003cem\u003eSustained Engraftment of Cryopreserved Human Bone Marrow CD34+ Cells in Young Adult NSG Mice.\u003c/em\u003e BioResearch Open Access, 2014. \u003cstrong\u003e3\u003c/strong\u003e(3): p. 110-116.\u003c/li\u003e\n\u003cli\u003eStemberger, C., et al., \u003cem\u003eNovel serial positive enrichment technology enables clinical multiparameter cell sorting.\u003c/em\u003e PLoS One, 2012. \u003cstrong\u003e7\u003c/strong\u003e(4): p. e35798.\u003c/li\u003e\n\u003cli\u003eMohr, F., et al., \u003cem\u003eMinimally manipulated murine regulatory T cells purified by reversible Fab Multimers are potent suppressors for adoptive T-cell therapy.\u003c/em\u003e Eur J Immunol, 2017. \u003cstrong\u003e47\u003c/strong\u003e(12): p. 2153-2162.\u003c/li\u003e\n\u003cli\u003eMohr, F., et al., \u003cem\u003eEfficient immunoaffinity chromatography of lymphocytes directly from whole blood.\u003c/em\u003e Sci Rep, 2018. \u003cstrong\u003e8\u003c/strong\u003e(1): p. 16731.\u003c/li\u003e\n\u003cli\u003eOhnishi, H., et al., \u003cem\u003eRegulation of cell shape and adhesion by CD34.\u003c/em\u003e Cell Adh Migr, 2013. \u003cstrong\u003e7\u003c/strong\u003e(5): p. 426-33.\u003c/li\u003e\n\u003cli\u003eHaspel, R.L. and K.K. Ballen, \u003cem\u003eDouble cord blood transplants.\u003c/em\u003e Stem Cell Reviews, 2006. \u003cstrong\u003e2\u003c/strong\u003e(2): p. 81-86.\u003c/li\u003e\n\u003cli\u003eBarker, J.N., et al., \u003cem\u003eTransplantation of 2 partially HLA-matched umbilical cord blood units to enhance engraftment in adults with hematologic malignancy.\u003c/em\u003e Blood, 2005. \u003cstrong\u003e105\u003c/strong\u003e(3): p. 1343-1347.\u003c/li\u003e\n\u003cli\u003eMartine, L.C., et al., \u003cem\u003eEngineering a humanized bone organ model in mice to study bone metastases.\u003c/em\u003e Nat Protoc, 2017. \u003cstrong\u003e12\u003c/strong\u003e(4): p. 639-663.\u003c/li\u003e\n\u003cli\u003eShafiee, A., et al., \u003cem\u003eImmune system augmentation via humanization using stem/progenitor cells and bioengineering in a breast cancer model study.\u003c/em\u003e Int J Cancer, 2018. \u003cstrong\u003e143\u003c/strong\u003e(6): p. 1470-1482.\u003c/li\u003e\n\u003cli\u003eKang, Y.K., et al., \u003cem\u003eHumanizing NOD/SCID/IL-2R\u0026gamma;null (NSG) mice using busulfan and retro-orbital injection of umbilical cord blood-derived CD34(+) cells.\u003c/em\u003e Blood Res, 2016. \u003cstrong\u003e51\u003c/strong\u003e(1): p. 31-6.\u003c/li\u003e\n\u003cli\u003eWunderlich, M., et al., \u003cem\u003eImproved multilineage human hematopoietic reconstitution and function in NSGS mice.\u003c/em\u003e PLOS ONE, 2018. \u003cstrong\u003e13\u003c/strong\u003e(12): p. e0209034.\u003c/li\u003e\n\u003cli\u003eMata, M.F., et al., \u003cem\u003eA modified CD34+ hematopoietic stem and progenitor cell isolation strategy from cryopreserved human umbilical cord blood.\u003c/em\u003e Transfusion, 2019. \u003cstrong\u003e59\u003c/strong\u003e(12): p. 3560-3569.\u003c/li\u003e\n\u003cli\u003eWunderlich, M., et al., \u003cem\u003eA New Immunodeficient Mouse Strain, NOD/SCID IL2R\u0026gamma;\u0026minus;/\u0026minus; SGM3, Promotes Enhanced Human Hematopoietic Cell Xenografts with a Robust T Cell Component.\u003c/em\u003e Blood, 2009. \u003cstrong\u003e114\u003c/strong\u003e(22): p. 3524.\u003c/li\u003e\n\u003cli\u003eMaser, I.-P., et al., \u003cem\u003eThe Tumor Milieu Promotes Functional Human Tumor-Resident Plasmacytoid Dendritic Cells in Humanized Mouse Models.\u003c/em\u003e Frontiers in Immunology, 2020. \u003cstrong\u003e11\u003c/strong\u003e.\u003c/li\u003e\n\u003cli\u003eWalsh, N.C., et al., \u003cem\u003eHumanized mouse models of clinical disease.\u003c/em\u003e Annual Review of Pathology: Mechanisms of Disease, 2017. \u003cstrong\u003e12\u003c/strong\u003e: p. 187-215.\u003c/li\u003e\n\u003cli\u003eMartinov, T., et al., \u003cem\u003eBuilding the Next Generation of Humanized Hemato-Lymphoid System Mice.\u003c/em\u003e Frontiers in Immunology, 2021. \u003cstrong\u003e12\u003c/strong\u003e.\u003c/li\u003e\n\u003cli\u003eEapen, M., et al., \u003cem\u003eEffect of graft source on unrelated donor haemopoietic stem-cell transplantation in adults with acute leukaemia: a retrospective analysis.\u003c/em\u003e Lancet Oncol, 2010. \u003cstrong\u003e11\u003c/strong\u003e(7): p. 653-60.\u003c/li\u003e\n\u003cli\u003eBallen, K.K., E. Gluckman, and H.E. Broxmeyer, \u003cem\u003eUmbilical cord blood transplantation: the first 25 years and beyond.\u003c/em\u003e Blood, 2013. \u003cstrong\u003e122\u003c/strong\u003e(4): p. 491-498.\u003c/li\u003e\n\u003cli\u003eBacon, K., et al., \u003cem\u003ePast, Present, and Future of Affinity-based Cell Separation Technologies.\u003c/em\u003e Acta Biomaterialia, 2020. \u003cstrong\u003e112\u003c/strong\u003e: p. 29-51.\u003c/li\u003e\n\u003cli\u003eSajay, B.N.G., et al., \u003cem\u003eMicrofluidic platform for negative enrichment of circulating tumor cells.\u003c/em\u003e Biomedical Microdevices, 2014. \u003cstrong\u003e16\u003c/strong\u003e(4): p. 537-548.\u003c/li\u003e\n\u003cli\u003eBank\u0026oacute;, P., et al., \u003cem\u003eTechnologies for circulating tumor cell separation from whole blood.\u003c/em\u003e Journal of Hematology \u0026amp; Oncology, 2019. \u003cstrong\u003e12\u003c/strong\u003e(1): p. 48.\u003c/li\u003e\n\u003cli\u003eMohr, F., et al., \u003cem\u003eEfficient immunoaffinity chromatography of lymphocytes directly from whole blood.\u003c/em\u003e Scientific Reports, 2018. \u003cstrong\u003e8\u003c/strong\u003e(1): p. 16731.\u003c/li\u003e\n\u003cli\u003eHuey, D.D. and S. Niewiesk, \u003cem\u003eProduction of Humanized Mice through Stem Cell Transfer.\u003c/em\u003e Curr Protoc Mouse Biol, 2018. \u003cstrong\u003e8\u003c/strong\u003e(1): p. 17-27.\u003c/li\u003e\n\u003cli\u003eGraef, P., et al., \u003cem\u003eSerial Transfer of Single-Cell-Derived Immunocompetence Reveals Stemness of CD8+ Central Memory T Cells.\u003c/em\u003e Immunity, 2014. \u003cstrong\u003e41\u003c/strong\u003e(1): p. 116-126.\u003c/li\u003e\n\u003cli\u003eNotta, F., et al., \u003cem\u003eIsolation of Single Human Hematopoietic Stem Cells Capable of Long-Term Multilineage Engraftment.\u003c/em\u003e Science, 2011. \u003cstrong\u003e333\u003c/strong\u003e(6039): p. 218-221.\u003c/li\u003e\n\u003cli\u003eScherer, S.D., et al., \u003cem\u003eAn immune-humanized patient-derived xenograft model of estrogen-independent, hormone receptor positive metastatic breast cancer.\u003c/em\u003e Breast Cancer Research, 2021. \u003cstrong\u003e23\u003c/strong\u003e(1): p. 100.\u003c/li\u003e\n\u003cli\u003eLang, J., et al., \u003cem\u003eStudies of lymphocyte reconstitution in a humanized mouse model reveal a requirement of T cells for human B cell maturation.\u003c/em\u003e J Immunol, 2013. \u003cstrong\u003e190\u003c/strong\u003e(5): p. 2090-101.\u003c/li\u003e\n\u003cli\u003eChiesa, R., et al., \u003cem\u003eOmission of in vivo T-cell depletion promotes rapid expansion of naive CD4+ cord blood lymphocytes and restores adaptive immunity within 2 months after unrelated cord blood transplant.\u003c/em\u003e Br J Haematol, 2012. \u003cstrong\u003e156\u003c/strong\u003e(5): p. 656-66.\u003c/li\u003e\n\u003cli\u003eCoughlan, A.M., et al., \u003cem\u003eMyeloid Engraftment in Humanized Mice: Impact of Granulocyte-Colony Stimulating Factor Treatment and Transgenic Mouse Strain.\u003c/em\u003e Stem Cells and Development, 2016. \u003cstrong\u003e25\u003c/strong\u003e(7): p. 530-541.\u003c/li\u003e\n\u003cli\u003eVerma, B. and A. Wesa, \u003cem\u003eEstablishment of Humanized Mice from Peripheral Blood Mononuclear Cells or Cord Blood CD34+ Hematopoietic Stem Cells for Immune-Oncology Studies Evaluating New Therapeutic Agents.\u003c/em\u003e Current Protocols in Pharmacology, 2020. \u003cstrong\u003e89\u003c/strong\u003e(1): p. e77.\u003c/li\u003e\n\u003cli\u003eArcangeli, S., et al., \u003cem\u003eCAR T cell manufacturing from naive/stem memory T lymphocytes enhances antitumor responses while curtailing cytokine release syndrome.\u003c/em\u003e The Journal of clinical investigation, 2022. \u003cstrong\u003e132\u003c/strong\u003e(12).\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"nature-portfolio","isNatureJournal":true,"hasQc":false,"allowDirectSubmit":false,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"","title":"Nature Portfolio","twitterHandle":"","acdcEnabled":false,"dfaEnabled":false,"editorialSystem":"ejp","reportingPortfolio":"","inReviewEnabled":true,"inReviewRevisionsEnabled":false},"keywords":"","lastPublishedDoi":"10.21203/rs.3.rs-6382310/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-6382310/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"Humanized mice are a powerful tool in preclinical assessment of novel therapeutics. However, generation of large homogenous cohorts is believed to be hampered by the limited number of human stem cells retrievable from individual donors, like from umbilical cord blood specimens. Here, we present a streamlined CD34+ cell isolation method that improves yield and purity compared to standard isolation protocols, enabling more efficient use of umbilical cord blood donations. Enriched CD34+ stem cell preparations yield robust humanization of NSG-SGM3 mice, even with very low input doses, down to a few thousand cells. Using our findings to effectively humanize large cohorts of mice (in this study up to 30 mice from a single stem cell donor), we resolve differences in cytokine release syndrome for two different CAR-T cell products within single-donor settings. These optimizations facilitate larger or smaller humanization experiments, with reduced economical and logistical burden for critical translational research.","manuscriptTitle":"Optimizing CD34+ cell doses for large-scale translational, single donor humanized mouse experiments","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-04-30 02:39:06","doi":"10.21203/rs.3.rs-6382310/v1","editorialEvents":[],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"communications-biology","isNatureJournal":true,"hasQc":false,"allowDirectSubmit":false,"externalIdentity":"commsbio","sideBox":"Learn more about [Communications Biology](http://www.nature.com/commsbio/)","snPcode":"","submissionUrl":"","title":"Communications Biology","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"ejp","reportingPortfolio":"Communications Series","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"b30af6f3-cb9b-4bfb-a962-7d40197eb8c5","owner":[],"postedDate":"April 30th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[{"id":47643417,"name":"Biological sciences/Biological techniques/Experimental organisms/Model vertebrates/Mouse"},{"id":47643418,"name":"Health sciences/Medical research/Experimental models of disease"}],"tags":[],"updatedAt":"2025-05-13T07:20:39+00:00","versionOfRecord":[],"versionCreatedAt":"2025-04-30 02:39:06","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-6382310","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-6382310","identity":"rs-6382310","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}