Revealing the inner workings of human HSC adhesion

In this issue of Blood, Rak et al identify cytohesin 1 (CYTH1) as an important intracellular mediator of human hematopoietic stem cell (HSC) adhesion that plays a role in HSC homing and lodging in the bone marrow and subsequent engraftment.¹  This finding contributes to our understanding of the mechanisms involved in regulating HSC-niche interaction, which we know is critical in regulating HSC function.² 

Much effort over the years has revealed many cell surface players responsible for the ability of HSCs to interact with the bone marrow microenvironment.³  Integrins in particular have emerged as critical players in the adhesion of HSCs to the extracellular matrix, with integrin β1 (ITGβ1) knockout leading to severe homing defects.⁴  However, very little is known about the intracellular players that link integrin activation to HSC migration and localization.

CYTH1 is a guanine-nucleotide exchange factor for multiple guanosine triphosphate (GTP)–binding proteins. Known to form a complex with ITGβ1 and integrin αL to mediate adhesion to ICAM1,⁵  it was recently shown to cooperate with ITGβ2 in neutrophils,⁶  lead to Rho activation in dendritic cells,⁷  and regulate migration of natural killer cells.⁸  Still highly understudied in the hematopoietic system, CYTH1 was selected by the authors as the top hit of an original short hairpin RNA (shRNA) library in vitro screening. Rak and colleagues reasoned that the traditional approach of separating hematopoietic stem and progenitor cells (HSPCs) adhering to stroma by means of washing the medium is far from physiological because it introduces high shear stress, and decided to take advantage of gravity, which naturally acts on all cells in any organisms, to identify genes that precisely regulate the adhesion properties of human HSPCs. Using the analogy of a car, which moves thanks to the interaction of the wheels with the road, while the steering wheel determines the direction of movement, this screening allowed the identification of wheel components, and the ensuing study focused on a factor akin to the wheels’ axis. All top hits identified, including CD90, ITGα5, MMRN1, NEDD9, PPFIA1, and ROBO1 had previously been described as components of cellular molecular motors. Importantly, known genes involved in HSPC localization but acting through different mechanisms (eg, chemoattraction in the case of CXCR4), were not highlighted by this screening.

In this study, CYTH1 is knocked down via shRNA in human HSPCs, resulting in their reduced adhesion to both retronectin (mediated by ITGβ1) and ICAM1 (mediated by ITGβ2) and reduced integrin activation in in vitro assays. When human HSCs deficient in CYTH1 are challenged to engraft in immunodeficient NOD.Cg-Prkdc^scidIl2rg^tm1Wjl/SzJ (NSG) mice, they show reduced ability to home to the bone marrow, lodge within the parenchyma, and drive efficient long-term engraftment. A member of another family of guanine exchange factors and linked to the regulation of Rho GTPases, Vav1, was previously shown to contribute to murine HSPC localization in the bone marrow and subsequent engraftment,⁹  with Vav1 knockout HSPCs showing a phenotype similar to that of the human cytohesin-deficient HSPCs in the Rak et al study. Rak et al take advantage of more recently developed time-lapse intravital microscopy of engrafting HSPCs¹⁰  to unravel the mechanism behind the observed aberrant HSPC localization. Fluorescently labeled CYTH1-deficient HSPCs were injected in NSG recipient mice, and intravital microscopy revealed that most cells found 4 days later were unable to settle in the marrow parenchyma in the same way as control cells.

The homing and lodgment defects observed are not as severe as the engraftment reduction, and the low levels of engrafted hematopoiesis show balanced differentiation. The cell cycle profile of cytohesin-deficient cells is unaffected, suggesting that a fraction of the observed HSPCs are able to engraft. This raises the questions of whether the nonengrafting HSPCs prematurely differentiate or die, whether the few engrafting HSPCs are those showing localization and migration more similar to control cells, and what is the differentiation and functional state of both control and knockdown cells observed. Even though CYTH1 is efficiently knocked down in all HSPCs assessed through functional studies, some engraftment is obtained. This fact points to the complexity and robustness of the phenomena studied, whereby deficiency in one component does not completely obliterate a cellular function—in this case adhesion and migration—in an entire cell population, and other mechanisms can still rescue CYTH1 deficiency. Future studies will indicate whether this is achieved through highly defined and uniform compensatory mechanisms or is the result of stochastic and variable activation of other components of the HSPCs’ molecular motors.