Isolation and time lapse microscopy of highly pure hepatic stellate cells

Abstract

Hepatic stellate cells (HSC) are the main effector cells for liver fibrosis. We aimed at optimizing HSC isolation by an additional step of fluorescence-activated cell sorting (FACS) via a UV laser. HSC were isolated from livers of healthy mice and animals subjected to experimental fibrosis. HSC isolation by iohexol- (Nycodenz) based density centrifugation was compared to a method with subsequent FACS-based sorting. We assessed cellular purity, viability, morphology, and functional properties like proliferation, migration, activation marker, and collagen expression. FACS-augmented isolation resulted in a significantly increased purity of stellate cells (>99%) compared to iohexol-based density centrifugation (60-95%), primarily by excluding doublets of HSC and Kupffer cells (KC). Importantly, this method is also applicable to young animals and mice with liver fibrosis. Viability, migratory properties, and HSC transdifferentiation in vitro were preserved upon FACS-based isolation, as assessed using time lapse microscopy. During maturation of HSC in culture, we did not observe HSC cell division using time lapse microscopy. Strikingly, FACS-isolated, differentiated HSC showed very limited molecular and functional responses to LPS stimulation. In conclusion, isolating HSC from mouse liver by additional FACS significantly increases cell purity by removing contaminations from other cell populations especially KC, without affecting HSC viability, migration, or differentiation.

Figure 1

Optimization of the isolation of primary hepatic stellate cells (HSC) based on iohexol density gradient centrifugation and fluorescence-activated cell sorting (FACS) (schematic depiction). In both strategies for cell purification, the mouse is anaesthetized before surgery, and the liver is then perfused via the Vena portae and drained through the Vena cava inferior using a two-step perfusion of the enzymes pronase and collagenase (a). Liver cells are harvested by gently tearing the liver into bits (b), followed by a postdigestion using a combination of both enzymes (c). The liver cells are subjected to iohexol density gradient centrifugation, after which HSC and Kupffer cells are located in the interphase between iohexol and buffer (d). The enriched HSC layer containing HSC, HSC-Kupffer cell doublets, and cellular debris can be used directly for cell culture of HSC (e) or can be cleared from HSC-Kupffer cell doublets and cellular debris using FACS based on the autofluorescence of retinol, using the UV laser of the cell sorter, resulting in highly pure HSC (f).

Figure 2

Gating strategy for the purification of hepatic stellate cells using fluorescence-activated cell sorting. Cells are first gated based on their forward and sideward scattering (a), doublets are excluded from sideward (b) and forward scattering (c), and hepatic stellate cells (HSC) are selected based on the UV light excitation of retinol (vitamin A) (d). A detailed cell-type specific staining of the Kupffer cell marker F4/80 demonstrated that the large (here: FSC-A > 100 in the plot, dotted gate) retinol⁺ cells are Kupffer cell- (KC-) HSC doublets that stain positive for F4/80 (right hand side), whereas the smaller retinol⁺ cells are F4/80⁻ HSC (left hand side). By placing a sorting gate as depicted in (d) (middle plot, black line), selected HSC (left plot) are pure and do not contain contaminating KC.

Figure 3

Comparison of the purity of hepatic stellate cells isolated via iohexol gradient without or with subsequent cell sorting. Purity of hepatic stellate cells (HSC) before (left top) and after fluorescence-activated cell sorting (FACS, right top) (a) based on their retinol autofluorescence only (dashed line) or additional exclusion of HSC-Kupffer cell doublets (highly pure real HSC, straight line). The statistical summary of retinol⁺ cells compared to HSC (with HSC-KC doublets excluded) is depicted (b). Analysis of purity using cell type-specific markers for major cell populations. Decorin is considered as a marker for hepatic stellate cells, C-type lectin domain family 4f (Clec4f) is a gene expressed by Kupffer cells, the platelet/endothelial cell adhesion molecule 1 (Pecam-1, CD31) is mainly expressed by endothelial cells, and albumin is a marker for hepatocytes (c). Data are given relative to the expression of β-actin of the cells derived from perfusion and digestion. Mean data ± SD of n = 4 independent experiments.

Figure 4

Cultures of hepatic stellate cells isolated via iohexol density gradient centrifugation without or with subsequent cell sorting. Hepatic stellate cells (HSC) were isolated from 40-week-old C57BL/6J mice using enzymatic digestion of the liver based on pronase and collagenase, followed by density gradient centrifugation in 8% iohexol. Cells were cultured directly after the gradient for one day (a) and four days (b), where Kupffer cells (indicated by arrows) can be found in the HSC culture. Highly pure HSC after additional fluorescence-activated cell sorting (FACS) after one (c) and four days of culture are shown (d). Expression analysis of desmin (e) and phalloidin (f) of HSC after four days of culture, indicating proper maturation of HSC.

Figure 5

Hepatic stellate cell activation in vivo and isolation of hepatic stellate cells from livers of injured mice. Chronic toxic liver injury was induced in 8-week-old C57BL/6 mice by 6 weeks of carbon tetrachloride (CCl₄) treatment, and control treatment was done using corn oil. Mice were sacrificed 48 hours after the last injection of (CCl₄) and liver sections were stained for hematoxylin eosin (a), Sirius red (stains collagen fibres, a hallmark of fibrosis) (b), and α smooth muscle action (αSMA) which targets activated hepatic stellate cells, mediators of fibrosis. Morphometric quantification of Sirius red confirms fibrosis progression (d). Quantitative real-time PCR indicates upregulation of collagen 1 (Col1A1) and αSMA mRNA in liver sections (e). Application of the gating strategy of fluorescence-activated cell sorting (FACS) to isolate hepatic stellate cells from livers of mice that underwent six weeks of repetitive CCl₄-based liver injury (f) compared to vehicle corn oil-treated control mice (g). Flow cytometric analysis of HSC purity before and after sorting demonstrated that the purification was successful (h). Mean ± SD of three independent experiments, n = 12 for control and 16 for fibrotic mice; ^∗∗∗ P < 0.001, ^∗∗ P < 0.005, and ^∗ P < 0.05 (unpaired Student's t-test).

Figure 6

Hepatic stellate cell functionality in vitro. Hepatic stellate cells (HSC) were isolated from 40-week-old C57BL/6J mice using enzymatic digestion of the liver based on pronase and collagenase, followed by density gradient centrifugation in 8% iohexol and fluorescence-activated cell sorting (FACS). The HSC were cultured in DMEM with 10% fetal calf serum, and some plates were stimulated with lipopolysaccharides (LPS) at 100 ng/mL (after five days of culture) for another 48 hours. Changes in the cell number during culture were determined from time lapse microscopy (a) and statistical summary (b). HSC were cultured for five days on tissue culture-treated polystyrene in DMEM with 10% fetal calf serum including culture inserts for self-insertion (“scratch assay”). To start horizontal migration, the plastic inserts were removed and HSC migrated (c) and were quantified using software (d). HSC were cultured for designated periods and quantitative real-time PCR was performed to study the expression of α smooth muscle actin (αSMA), collagen 1 (Col1A1), or the transforming growth factor β (TGFβ) as markers of HSC activation. Gene expression was normalized to β-actin expression of cells that were lysed directly after isolation at day zero (e). Mean ± SD of three independent experiments.