An efficient method to isolate Kupffer cells eliminating endothelial cell contamination and selective bias

Abstract

Multicolor flow cytometry and cell sorting are powerful immunologic tools for the study of hepatic mϕ, yet there is no consensus on the optimal method to prepare liver homogenates for these analyses. Using a combination of mϕ and endothelial cell reporter mice, flow cytometry, and confocal imaging, we have shown that conventional flow-cytometric strategies for identification of Kupffer cells (KCs) leads to inclusion of a significant proportion of CD31^hi endothelial cells. These cells were present regardless of the method used to prepare cells for flow cytometry and represented endothelium tightly adhered to remnants of KC membrane. Antibodies to endothelial markers, such as CD31, were vital for their exclusion. This result brings into focus recently published microarray datasets that identify high expression of endothelial cell-associated genes by KCs compared with other tissue-resident mϕ. Our studies also revealed significant and specific loss of KCs among leukocytes with commonly used isolation methods that led to enrichment of proliferating and monocyte-derived mϕ. Hence, we present an optimal method to generate high yields of liver myeloid cells without bias for cell type or contamination with endothelial cells.

Keywords: Kupffer cell; endothelial cell; flow cytometry.

Figure 1

(A) Identification of F4/80^hiCD11b^lo KCs in murine liver. Lineage = CD3, CD19, Ly6G, and Siglec F. (B) Representative GFP and CD31 expression by F4/80^hiCD11b^lo liver cells from Cdh5‐Cre‐ERT2;mTmG mice. Proportion of GFP⁺ and CD31^hi cells shown (mean, n = 7, 3 experiments). (C) Representative GFP and CD31 expression by F4/80^hiCD11b^lo liver cells from Cdh5‐Cre‐ERT2;mTmG mice ± tamoxifen. (D) The frequency of CD31^hiGFP⁺ cells amongst KCs, lung interstitial and alveolar mϕ. Liver (n = 7, 3 experiments) and lung (n = 4, 2 experiments). (E and F) Geometric mean fluorescence intensity (GeoMFI) of dTomato expression by CD31^loGFP⁻ KCs (E) and CD31^hiGFP⁺ cells (F) (n = 3/group, representative of 3 experiments)

Figure 2

(A) Representative expression of CD45 and F4/80 by total CD31^hi liver cells (n = 6, 2 experiments). (B) Representative histograms (n = 6, 2 experiments) and delta MFI (n = 3/group, representative of 3 experiments) of ICAM‐2 and LYVE‐1 expression by F4/80^hiCD31^hi cells and F4/80^hiCD31^lo KCs. Significance determined by t‐test. (C) Confocal image of liver from Cdh5‐Cre‐ERT2;mTmG mice (dTomato = nonrecombined cells; GFP = recombined cells) stained with F4/80 (blue). Scale bar, 20 μm. (D) Proportion of F4/80⁺ Cdh5 ⁺ cells from (C) (10 FoV at 40× magnification of 4 livers). (E) Representative maximal projections of confocal z‐stacks of FACS‐purified F4/80^hiCD31^hi. CD31 (green), F4/80 (blue), CD45 (red), and merge (purple). White arrows indicate areas of punctate surface F4/80 and CD45 staining by CD31⁺ cells (n = 5 from 2 separate experiments). Scale bar, 10 μm. (F) Representative Tim4, Clec4f, and CD31 expression by F4/80^hiCD11b^lo cells. (G) Representative mApple and CD31 expression by F4/80^hiCD11b^lo cells from Csf1r‐mApple transgenic mice or their negative littermate controls (n = 9, 3 experiments). (H) Proportion of BrdU⁺Ki67⁺ cells amongst CD31^hi and CD31^lo F4/80^hiCD11b^lo cells after administration of CSF1‐Fc or PBS (n = 4, 1 experiment). Significance determined by 1‐way ANOVA. (I) Overlay of all CD45⁺CD31^hi cells (red) onto CD45⁺CD31^lo cells. (J) Representative maximal projections of confocal z‐stacks of FACS sorted CD45⁺F4/80⁻CD31^hi cells, demonstrating a CD45⁺ cell (red) enveloped in an endothelial cell (green)

Figure 3

(A) The frequency of CD31^hi cells of all live, single cells. (B and C) Representative flow plots (B) and replicate data (C) of the proportion of CD31^hi cells amongst F4/80^hiCD11b^lo cells obtained with different protocols (300 g, Percoll, 50 g) (n = 4–8/group, 2 experiments). (D) Improved gating strategy incorporating CD31 to identify endothelial contaminants from the F4/80^hiCD11b^lo KCs population. Significance determined by 1‐way ANOVA

Figure 4

(A and B) Representative flow plots of all isolated (A) and discarded (B) live CD45⁺CD31^lo cells showing proportions of F4/80^hiCD11b^lo and F4/80^loCD11b^hi populations from livers prepared for flow cytometry by 1 of 3 different protocols. (C–E) The relative frequency of F4/80^hiCD11b^lo KCs, F4/80^lo cells granulocytes, B cells and T cells as a proportion of all CD45⁺CD31^lo leukocytes in the cell isolate linked to the frequency in the discard for the various protocols (n = 4–8/group, 2 experiments). Significance determined by multiple t‐test (Holm‐Sidak correction). (F and G) Number of CD45⁺ cells or CD45⁺F4/80^hiCD31^lo KCs isolated by either 300 g or Percoll method (n = 8/group, 2 experiments). Significance determined by t‐test and 1‐way ANOVA respectively. (H) Nonhost chimerism amongst KCs and F4/80^lo cells from liver of WT > WT tissue‐protected BM chimeric mice isolated by 300 g or Percoll protocol (n = 8/group, 2 experiments). (I) Proportion of Ki67⁺ KCs isolated from liver by 300 g or Percoll protocol (n = 4/group, representative of 2 experiments). Significance determined by t‐test