Three-Dimensional Microfluidic Tri-Culture Model of the Bone Marrow Microenvironment for Study of Acute Lymphoblastic Leukemia

Abstract

Acute lymphoblastic leukemia (ALL) initiates and progresses in the bone marrow, and as such, the marrow microenvironment is a critical regulatory component in development of this cancer. However, ALL studies were conducted mainly on flat plastic substrates, which do not recapitulate the characteristics of marrow microenvironments. To study ALL in a model of in vivo relevance, we have engineered a 3-D microfluidic cell culture platform. Biologically relevant populations of primary human bone marrow stromal cells, osteoblasts and human leukemic cells representative of an aggressive phenotype were encapsulated in 3-D collagen matrix as the minimal constituents and cultured in a microfluidic platform. The matrix stiffness and fluidic shear stress were controlled in a physiological range. The 3-D microfluidic as well as 3-D static models demonstrated coordinated cell-cell interactions between these cell types compared to the compaction of the 2-D static model. Tumor cell viability in response to an antimetabolite chemotherapeutic agent, cytarabine in tumor cells alone and tri-culture models for 2-D static, 3-D static and 3-D microfluidic models were compared. The present study showed decreased chemotherapeutic drug sensitivity of leukemic cells in 3-D tri-culture models from the 2-D models. The results indicate that the bone marrow microenvironment plays a protective role in tumor cell survival during drug treatment. The engineered 3-D microfluidic tri-culture model enables systematic investigation of effects of cell-cell and cell-matrix interactions on cancer progression and therapeutic intervention in a controllable manner, thus improving our limited comprehension of the role of microenvironmental signals in cancer biology.

Fig 1. Design considerations of a tumor model.

Fig 2. A typical 3-D microfluidic tri-culture model.

(A) The microfluidic platform consists of 4 microchannels, each 500 μm in width, 75 μm in height and 2 cm in length. The divergent angles are 30°. Near the inlets and outlet there are perfusion channels. The enlarged boxed area illustrates multiple cell types embedded in a 3-D matrix. (B) Optical image of a microfluidic platform. (C) SEM images of top view (upper panel) and side view (lower panel) of the perfusion channels.

Fig 3. Cell-cell interactions in the tumor models.

(A) Confocal and Imaris reconstruction images were presented of the tri-culture of SUP-B15 tumor cells (green), BMSC (red) and HOB (blue) across the tumor models at low magnification (20X), high magnification (40X) and high magnification cross sections (CS) of 40X images. The cell-cell interactions in z-direction in the boxed region of the CS images were enlarged in Row 4, where the heights for 2-D and 3-D CS images were 10 μm and 30 μm, respectively. (B) 3-D reconstruction images rotated via x- and y-axis revealing cell-cell interactions.

Fig 4. Confocal images of SUP-B15 and BMSC co-cultured in 2-D static, 3-D static and 3-D dynamic models.

The yellow arrowheads and white arrows point to SUP-B15 cells and BMSC, respectively. The actin filaments were stained with phalloidin in red, the nuclei were stained with DAPI in blue, and the nuclei of proliferating cells were stained with Ki67 in green. Scale bars: 20 μm.

Fig 5. Viability analysis using Annexin V.

(A) Representative flow cytometry analysis for untreated (left) and Ara-C treated (right) 3-D static tri-culture models. (B) Representative immunofluorescence image analysis of untreated (left) and treated (right) 3-D static tri-culture models. The immunofluorescence images are the projection of z-stack confocal images. Confocal images were reconstructed using Imaris and also presented from a cross-sectional (CS) view. Scale bars: 20 μm.

Fig 6. Comparison in chemoresistance of tumor cells to Ara-C among a variety of tumor models.

(A) The effects of dimensionality, tri-culture and interstitial flow on the viability of tumor cells. (B) The effects of 3-D culture on tumor cell viability under the tri-culture condition. Significant difference between two groups is indicated by + where p < 0.1, * where p < 0.05.