Introduction Telomeres, which cap and protect chromosomal DNA, shorten with each cell division reaching a critical point eventually when the cell is arrested in G1 phase and enters a state of cellular senescence. This process has been demonstrated in both normal ageing and chronic disease in diverse tissues. Murine studies demonstrated that telomere shortening within the liver predisposes to cirrhosis. Few studies have examined the effect of increasing age on telomere length within healthy human liver. Measuring telomere length in liver by Southern blotting assumes that results are representative for hepatocytes. However, liver comprises a diverse group of cells including hepatocytes, Kupffer cells, stellate cells, lymphocytes and cholangiocytes. A large volume four colour quantitative fluorescent in situ hybridisation (Q-FISH) technique was developed to measure telomere length within each cell type.

Aim To determine the effect of increasing age upon telomere length in healthy liver within hepatocytes, Kupffer cells, stellate cells, CD4 or CD8 lymphocytes and cholangiocytes.

Method Q-FISH was performed on paraffin-embedded tissue from 73 “time zero” liver biopsies obtained at liver transplantation from liver donors selected carefully for the absence of reperfusion injury and to include a wide age range (5–79 years). Hepatocytes, Kupffer cells, hepatic stellate cells, CD4 or CD8 lymphocytes and cholangiocytes were identified with monoclonal antibodies to Hepar-1, CD68, SMA, CD4, CD8 and CK-19, respectively; nuclei were counterstained with DAPI and telomeres identified with a specific Cy5-labelled PNA probe. Image acquisition and data analysis were performed with Olympus ScanR microscope and software. Mean telomere fluorescent intensity was measured and analysed using linear regression on GraphPad PRISM 5.0.

Results The Q-FISH methodology enabled large volume analysis of telomeres in various cell lineages within paraffin-embedded liver sections. A mean 40 000 hepatocytes, 30 000 Kupffer cells, 15 000 cholangiocytes and 4000 hepatic stellate cells, intrahepatic CD4 and CD8 lymphocytes were analysed per liver. Hepatic stellate cells had the longest telomeres at all ages whilst the others cell types had similar telomere lengths. Telomeres within Kupffer cells and stellate cells in healthy liver shortened, as might be expected, with increasing age (p=0.024 and p<0.0001, respectively). In contrast, telomeres in hepatocytes, cholangiocytes and lymphocytes in healthy livers did not shorten with increasing age.

Conclusion Cells within the healthy liver do not age equally. The disparity between age-related telomere shortening in Kupffer cells and stellate cells and maintained telomere length with age in cholangiocytes, hepatocytes and intrahepatic lymphocytes was unexpected but may reflect low turn over in hepatocytes and cholangiocytes. The absence of an effect of age on these cell lineages suggests that the potential for regeneration is maintained with age and has important implications for the choice of control tissues in studies of hepatic ageing.