Background Conventional dendritic cells (cDC) are innate immune cells specialized in antigen sampling and subsequent cross-presentation to immune cells, and are critical for an effective anti-tumor immune response. The relatively rare subset cDC type I (cDC1) has been associated with human cancer patient survival, and strategies aimed at increasing their abundance and activation in the tumor microenvironment (TME) are of interest. It is a challenge to use these rare cells in screening assays as they constitute less than 0.05% of cells in the blood and are difficult to isolate with high viability and purity.

Methods We compared two systems that generate a mixed culture containing cDC1 cells from human CD34+ hematopoietic stem cells (HSC) isolated from cord blood. In the first system, we developed a scalable in vitro differentiation method that expands and differentiates the CD34+ HSC to generate cDC1s over the course of 3.5 weeks. The second system utilized a humanized mouse model. Mice engineered to express human GM-CSF and IL-3 to support the myeloid compartment were engrafted with human CD34+ HSC and boosted twice with Flt3L-IgG. Femurs were taken at day 10 for bone marrow extraction and subsequent cDC1 culture.

Results The human CD34+ HSC and cytokine treatments in these methods yield an increase in number of cDC1 cells in a mixed cellular population. The cDC1 cells are phenotypically and functionally similar to those in the blood; they express high levels of CD141 and Clec9a, and respond to activating stimuli such as poly(I:C). The ex vivo cDC1 system using bone marrow from humanized mice is highly reproducible across laboratory operators and with multiple cord blood donors, and it is less laborious than the in vitro differentiation assay. A single operator can process 4 to 5 mouse femur pairs at one time to collect on average enough cells for 15 test compounds at a 5-pt dose curve with controls. Multicolor flow cytometry was used to further phenotype and assess functionality of other cell types present in the ex vivo culture.

Conclusions Both systems demonstrate that HSC-derived cDC1 cultures can be reliably generated with cDC1 phenotype and functionality, and may be used for screening assays in industry settings. The ex vivo system allows for a faster and higher throughput but requires supply of humanized mice.