Multipotent adult progenitor cells from bone marrow differentiate into functional hepatocyte-like cells

Robert E Schwartz, Morayma Reyes, Lisa Koodie, Yuehua Jiang, Mark Blackstad, Troy Lund, Todd Lenvik, Sandra Johnson, Wei-Shou Hu, Catherine M Verfaillie, Robert E Schwartz, Morayma Reyes, Lisa Koodie, Yuehua Jiang, Mark Blackstad, Troy Lund, Todd Lenvik, Sandra Johnson, Wei-Shou Hu, Catherine M Verfaillie

Abstract

We have derived from normal human, mouse, and rat postnatal bone marrow primitive, multipotent adult progenitor cells (MAPCs) that can differentiate into most mesodermal cells and neuroectodermal cells in vitro and into all embryonic lineages in vivo. Here, we show that MAPCs can also differentiate into hepatocyte-like cells in vitro. Human, mouse, and rat MAPCs, cultured on Matrigel with FGF-4 and HGF, differentiated into epithelioid cells that expressed hepatocyte nuclear factor-3beta (HNF-3beta), GATA4, cytokeratin 19 (CK19), transthyretin, and alpha-fetoprotein by day 7, and expressed CK18, HNF-4, and HNF-1alpha on days 14-28. Virtually all human, as well as a majority of rodent cells stained positive for albumin and CK18 on day 21; 5% (rodent) to 25% (human) cells were binucleated by day 21. These cells also acquired functional characteristics of hepatocytes: they secreted urea and albumin, had phenobarbital-inducible cytochrome p450, could take up LDL, and stored glycogen. MAPCs, which can be expanded in vitro and maintained in an undifferentiated state for more than 100 population doublings, can thus differentiate into cells with morphological, phenotypic, and functional characteristics of hepatocytes. MAPCs may therefore be an ideal cell for in vivo therapies for liver disorders or for use in bioartificial liver devices.

Figures

Figure 1
Figure 1
Phenotypic characterization of mMAPC differentiation to hepatocyte-like cells. The mMAPCs were cultured on Matrigel at 21.5 × 103 cells/cm2 with FGF-4 and HGF for 4–14 days. Immunohistochemical localization of (ac) CK18-FITC and HNF-3β-Cy3, (df) albumin-FITC and GATA4-Cy3, (gi) αFP-FITC and HNF-1α-Cy3, on day 4, 7, and 14, and (jl) CK18-FITC and albumin-Cy3 on day 14. Scale bar, 20 μm. Representative example of seven experiments.
Figure 2
Figure 2
Phenotypic characterization of hMAPC differentiation to hepatocyte-like cells. The hMAPCs were cultured on Matrigel at 3 × 104 cells/cm2 with FGF-4 for 21 days. Representative example of three experiments. Scale bar, 25 μm. (a and d) Day 0 and 14, stained with albumin-FITC and HNF-1α-Cy5 Ab’s. (b and e) Day 7 and 14, stained with CK18-FITC and GATA4-Cy5 Ab’s. (c and f) Day 7 and 14, stained with CK8-FITC and HNF-3β-Cy5 Ab’s. (gi) Day 21, stained with CK18-Cy3 and albumin-FITC Ab’s. (g) CK18-Cy3, (h) albumin-FITC, and (i) overlay. (jl) Day 21, stained with CK8-Cy3 and CK18-FITC Ab’s. (j) CK8-Cy3, (k) CK18-FITC, and (l) overlay.
Figure 3
Figure 3
Single hMAPC (clone A16) that differentiates to mesoderm and neuroectoderm, differentiates to hepatocyte-like cells when cultured with FGF-4 on Matrigel for 21 days. The hMAPCs were from clone A16, described elsewhere (M. Reyes, unpublished observations), generated from cultures in which five eGFP+ hMAPCs were added to 95 nontransduced hMAPCs and expanded to more than 5 × 106 cells. A16-derived hMAPCs were cultured with FGF-4 on Matrigel for 14 days and stained with Ab’s against CK18, albumin, and HNF-3β. Slides were evaluated by confocal microscopy for coexpression of eGFP and hepatocyte markers. Scale bar, 25 μm. (ac) FGF-4-induced MAPC evaluated for eGFP expression. (de) Same field as a and b stained with CK18-Cy3. eGFP+ cells express CK18. (f) Same field as c stained for albumin-Cy3 and HNF-3β–Cy5. eGFP+ cells express albumin and HNF-3β.
Figure 4
Figure 4
Quantitative RT-PCR and Western blot analyses confirm hepatocyte-like phenotype. (a) mMAPCs and (b) hMAPCs were cultured on Matrigel with FGF-4 and HGF or FGF-4 alone for 21 and 28 days, respectively. At multiple time points during culture, cells were harvested and underwent quantitative RT-PCR using the SYBR green method for mRNAs as indicated. The mRNA levels were normalized using β-actin as a housekeeping gene. Numbers under the blots for CK18, CK19, TTR, albumin, and CYP1B1 represent mRNA levels after treatment relative to undifferentiated MAPCs. For αFP, Cyp2b9, and Cyp2b13, numbers under the blots are relative to mRNA from the liver, because no transcripts were detected in undifferentiated MAPCs. Li, mouse or human liver mRNA; NT, no template. Representative example of five mouse and one human study. (c) hMAPCs (from b) were cultured on Matrigel with FGF-4 and HGF or FGF-4 alone for 21 days. HUH7 cells were cultured as described. Lysates of cells were separated by SDS-PAGE, transferred to Immuno-Blot PVDF membrane, and incubated for 1 minute with Ab’s against αFP, CK18, albumin, or β-actin. FN, FGF-4–induced hMAPCs on FN; M, FGF-4–induced hMAPCs on Matrigel; FH, FGF-4– and HGF-induced hMAPCs on Matrigel; Huh, Huh7 cell line used as control.
Figure 5
Figure 5
Urea and albumin production by mouse, rat, and human MAPCs cultured with FGF-4 and HGF on Matrigel. mMAPCs (n = 7), rMAPCs (n = 5), and hMAPCs (n = 2) were cultured on Matrigel with FGF-4 and HGF for 4–28 days. Gray area represents standard values for primary rat hepatocytes. (a) Urea production, expressed per cell number seeded and per hour. Studies were done in duplicate and repeated twice for each study. (b) Albumin production expressed per cell number seeded and per hour. Studies were done in duplicate and repeated twice for each study.
Figure 6
Figure 6
PROD assay on MAPC-derived hepatocytes from mice and humans. The mMAPCs and hMAPCs cultured on Matrigel with FGF-4 and HGF for 14 and 28 days were allowed to aggregate in the presence or absence of 1 mM phenobarbital for 4 days. PROD assay was done as described in Methods. (a and b) Undifferentiated mMAPC aggregates. (c and d) Primary rat hepatocytes. (e and f) mMAPC-derived hepatocyte-like spheroids. (g and h) hMAPC-derived hepatocyte-like spheroids. Representative example of four mouse studies including over 120 aggregates and one human study including over 50 aggregates.
Figure 7
Figure 7
LDL uptake by hMAPC-derived hepatocytes. Thee hMAPCs cultured with FGF-4 on Matrigel for 0–21 days were incubated with Dil-acil-LDL, fixed, and stained with anti–pan-CK, CK18, GATA4, and HNF-3β Ab’s. Representative example of three studies. Scale bar, 25 μm. (a) Day 21, incubated with Dil-acil-LDL and stained with HNF-3β-Cy5 Ab. (b) Same field as a stained with pan-CK-FITC and HNF-3β-Cy5 Ab. Inset, higher-magnification three-color view. Note the presence of binucleated cells. (c) Day 21, incubated with Dil-acil-LDL and stained with GATA4-Cy5 Ab. (d) Same field as b stained with GATA4-Cy5 and CK18-FITC Ab. Inset, higher magnification showing three-color view. Note the presence of binucleated cells.
Figure 8
Figure 8
FGF-4–induced MAPCs produce glycogen. The rMAPCs and hMAPCs were cultured on Matrigel with FGF-4 and HGF and FGF-4 alone, respectively. Cells were fixed with methanol and stained by PAS. Glycogen storage is seen as accumulation of magenta staining. Representative example of three studies. Scale bar, 25 μm. (a) Undifferentiated hMAPCs, (b) day 21 hMAPCs, (c) rMAPCs, (d) rat primary hepatocytes.
Figure 9
Figure 9
FGF-4–induced MAPCs exhibit polarization. The hMAPCs cultured with FGF-4 on Matrigel for 28 days. Representative example of two studies. Arrowhead, polarized cell; arrow, nonpolarized cell; star, binucleated cell. Scale bar, 25 μm. Staining for c-met-Cy3 (a), albumin-FITC (b), and overlap (c). Staining for c-met-Cy3 (d), CD26-FITC (e), and overlap (f).

Source: PubMed

3
Se inscrever