A multifunctional 3D co-culture system for studies of mammary tissue morphogenesis and stem cell biology

Abstract

Studies on the stem cell niche and the efficacy of cancer therapeutics require complex multicellular structures and interactions between different cell types and extracellular matrix (ECM) in three dimensional (3D) space. We have engineered a 3D in vitro model of mammary gland that encompasses a defined, porous collagen/hyaluronic acid (HA) scaffold forming a physiologically relevant foundation for epithelial and adipocyte co-culture. Polarized ductal and acinar structures form within this scaffold recapitulating normal tissue morphology in the absence of reconstituted basement membrane (rBM) hydrogel. Furthermore, organoid developmental outcome can be controlled by the ratio of collagen to HA, with a higher HA concentration favouring acinar morphological development. Importantly, this culture system recapitulates the stem cell niche as primary mammary stem cells form complex organoids, emphasising the utility of this approach for developmental and tumorigenic studies using genetically altered animals or human biopsy material, and for screening cancer therapeutics for personalised medicine.

Figure 1. An in vitro strategy for recapitulating mammary gland architecture in 3D.

(a,b) Whole-mount of murine mammary gland (10day gestation) showing branched epithelium invested within adipocyte-rich fat pad (bars 1mm and 200 µm respectively). Higher power immunohistochemical micrograph (c) of virginal gland showing distinctive architecture of luminal epithelial (cytokeratin 18+) and myoepithelial (cytokeratin 14+) bilayer (bar  = 200 µm). Phase contrast image (d) (bar  = 1mm) and S.E.M image (insert, bar  = 200 µm) of 7.5%HA/col scaffold showing interconnected pores that can support 3T3-L1 seeding and culture in 3D and differentiation to mature adipocytes. (e) S.E.M images of collagen and 7.5%HA/col scaffold following soaking in PBS for 24 hrs and 8 months showing collapse of porous structure in pure collagen scaffolds. (f) Stability of scaffolds by measurement of percentage weight retention during PBS incubation at 37°C. (g) 3T3-L1 seeded collagen scaffold following 8 days adipogenic culture. Haemotoylin/ Oil-red O staining (bar  = 200 µm). (h) Cross-section of thick (<2 mm) collagen scaffold following seeding with KIM-2 (GFP) for 48 hrs and 3T3-L1 (cell-tracker red) for 9 days (bar  = 500 µm). (i) Low-power observation of 3T3-L1/KIM-2 co-cultures (carmine stained) at 3 wks reveals branched epithelial structures with blunt ends (bar  = 200 µm).

Figure 2. 3T3-L1/KIM-2 co-culture in defined collagen/hyaluronic acid scaffolds generate bilayed, functional mammary epithelial organoids.

H&E (a,b) and immunohistochemical (c-o) comparison of epithelial organoid formations in vitro with murine gland (scale bars  = 50 µm, except j,o  = 30 µm). Examples of branched ducts in vivo (a) and in vitro (b,c). Like the native gland at 5days gestation, epithelial organoids display polarity with AQP5 (g) and ZO-1 (h) towards a luminal cavity and integrin β-1 (e) and basement membrane constituents colIV (l) and laminin (m) at the basal surface. SMA(+) myoepithelial cells surround luminal epithelial cells (j), characteristic of the epithelial bilayer in vivo (i). With lactogenic stiumuli, organoid acinar-like structures express β-casein in a luminal cavity (o) as in the native gland (n).

Figure 3. Mammary organoid fate is determined by scaffold composition.

(a) Scoring of acinar or ductal AQP5(+)/E-cad(+) epithelial organoids determined by long axis (y)/short axis (x) ratios. (b) Inclusion of HA within scaffold composition significantly elevates frequency of acinar-like to duct-like organoids and (c) the frequency of correct organoid polarity is significantly elevated under co-culture (CO) as opposed to mono-culture (MONO) conditions and in the presence of HA (ANOVA, * and ** represent p<0.05 and p<0.01 respectively by post-hoc students T-Test, n>3). (d) Mean number of E-cad(+) KIM-2 per organoid, measured at constant focal plane of immunohistochemically prepared sections. Western analysis of KIM-2 mono-culture (e) and KIM-2/3T3-L1 co-cultures (f) under KIM-2 maintenance media (control) or exposure for latter 2 wks to lactogenic media (lactogenic). (g) Inclusion of HA at 15% reduces cleaved caspase 3(+) cells and limits cell proliferation (h) at 7.5% and 15% compared to collagen control under lactogenic conditions by measurement of total Ki67(+) cells (* and ** represent p<0.05 and p<0.01 respectively by students T-Test). (i) PCR analysis of MMP2, MMP3 and MMP14 gene expression within 3D scaffolds of 3 wk KIM-2 mono-culture lactogenic specimens and 3T3-L1/KIM-2 co-culture lactogenic specimens. (j) Quantitative real-time PCR analysis of MMP2, MMP3 and MMP14 gene expression normalized to Cyclophillin A. Bars represent mean of at least 3 separate experiments performed in triplicate. Error bars represent SD. * represents statistical significance by students T-test (p<0.05)

Figure 4. 3T3-L1 seeded collagen/HA scaffolds support primary epithelial cell organoid development and provide a stem cell niche.

(a) 5 day gestation section of murine mammary gland expressing GFP-tagged histone H2B under the control of the Keratin 14 promoter. Cells of the myoepithelial/basal lineage are green while luminal cells do not express H2B-GFP. (b) Stem cell enriched (GFP+, CD24^med CD49f^high) fractions of these cells are capable of mammary gland regrowth following transplantation to a cleared fatpad. (c) Live cell confocal image representation (orthogonal view and oblique projection) of primary cell culture following 10days co-culture with 3T3-L1 showing basal localization of GFP+ cells around a branched organoid. Primary epithelial organoids co-cultured with 3T3-L1 in pure collagen (d,e) and 7.5%HA/col (f) scaffolds. Organoids form ductal and acinar-like structures enveloped in basement membrane (e) and exhibit CK14 basal cells arrayed around a clear lumen with correctly polarized E-cad (+) luminal epithelial cells expressing ZO-1 at the apical cell surface (f). (g,h) Sorted GFP+ label retaining cells form colonies that vary in their expression of cytoplasmic progesterone receptor (PR), and are beta catenin positive (i). (j,k) Colonies display positive and negative staining for p63 suggesting self-renewal or asymmetric cell division (indicated by arrows) (Bars  = 100 µm).