Use of postmortem human dura mater and scalp for deriving human fibroblast cultures

Abstract

Fibroblasts can be collected from deceased individuals, grown in culture, reprogrammed into induced pluripotent stem cells (iPSCs), and then differentiated into a multitude of cell types, including neurons. Past studies have generated iPSCs from somatic cell biopsies from either animal or human subjects. Previously, fibroblasts have only been successfully cultured from postmortem human skin in two studies. Here we present data on fibroblast cell cultures generated from 146 scalp and/or 53 dura mater samples from 146 postmortem human brain donors. In our overall sample, the odds of successful dural culture was almost two-fold compared with scalp (OR = 1.95, 95% CI: [1.01, 3.9], p = 0.047). Using a paired design within subjects for whom both tissues were available for culture (n = 53), the odds of success for culture in dura was 16-fold as compared to scalp (OR = 16.0, 95% CI: [2.1-120.6], p = 0.0007). Unattended death, tissue donation source, longer postmortem interval (PMI), and higher body mass index (BMI) were associated with unsuccessful culture in scalp (all p<0.05), but not in dura. While scalp cells proliferated more and grew more rapidly than dura cells [F (1, 46) = 12.94, p<0.008], both tissues could be generated and maintained as fibroblast cell lines. Using a random sample of four cases, we found that both postmortem scalp and dura could be successfully reprogrammed into iPSC lines. Our study demonstrates that postmortem dura mater, and to a lesser extent, scalp, are viable sources of living fibroblasts for culture that can be used to generate iPSCs. These tissues may be accessible through existing brain tissue collections, which is critical for studying disorders such as neuropsychiatric diseases.

Figure 1. Fibroblast characterization: FSP-1 protein expression by immunofluorescence staining and cell proliferation assay in dura and scalp. A.

The morphology of postmortem fibroblast cells generated from (a) dura and (b) scalp. Cultured cells from both sources macroscopically looked similar to what is seen in living skin fibroblast cells, with more enriched cytoplasm and spindle-shaped nuclei under phase-contrast microscopy. Cells from (c) dura and (d) scalp express cytoplasmic Fibroblast Specific Protein-1 (FSP-1) (green). Original scale bars = 35 µm. B. Results from cell proliferation assay in 8 fibroblast cell lines (dura and scalp from 4 individuals) in five different densities. Cell viability was determined in 24 hrs and 48 hrs by WST-8 assay. Values are the mean of results from six wells. Bars ± SE. Scalp fibroblast cell lines grew 1.27-fold faster in the same period than dura fibroblast cells. C. Differences in cell proliferation between scalp and dura by one-way ANOVA; scalp cell growth was significantly more rapid than dura cell growth at 24 hr and 48 hr intervals [F (1, 46) = 12.94, p<0.008].

Figure 2. Percentage of successful growth in dura (n = 53) vs. scalp (n = 146) samples.

Sample of 146 scalp and 53 dural specimens. The odds of successful culture in dura were nearly 2-fold compared to scalp.

Figure 3. iPSCs generated from one dura fibroblast line express pluripotency markers and differentiate to neuronal fates.

Undifferentiated iPSCs express pluripotency markers NANOG (A) and SOX2 (B). Upon neural differentiation, these cells express neuroectoderm marker SOX1 (C). Temporal gene expression analysis also shows a decrease in pluripotency marker OCT4 (POU5F1) and an increase in PAX6 expression by quantitative RT-PCR (D). iPSC-derived neurons express βIII-tubulin and MAP2 (E–G).