Characterization and functionality of proliferative human Sertoli cells

Abstract

It has long been thought that mammalian Sertoli cells are terminally differentiated and nondividing postpuberty. For most previous in vitro studies immature rodent testes have been the source of Sertoli cells and these have shown little proliferative ability when cultured. We have isolated and characterized Sertoli cells from human cadaveric testes from seven donors ranging from 12 to 36 years of age. The cells proliferated readily in vitro under the optimized conditions used with a doubling time of approximately 4 days. Nuclear 5-ethynyl-2'-deoxyuridine (EdU) incorporation confirmed that dividing cells represented the majority of the population. Classical Sertoli cell ultrastructural features, lipid droplet accumulation, and immunoexpression of GATA-4, Sox9, and the FSH receptor (FSHr) were observed by electron and fluorescence microscopy, respectively. Flow cytometry revealed the expression of GATA-4 and Sox9 by more than 99% of the cells, and abundant expression of a number of markers indicative of multipotent mesenchymal cells. Low detection of endogenous alkaline phosphatase activity after passaging showed that few peritubular myoid cells were present. GATA-4 and SOX9 expression were confirmed by reverse transcription polymerase chain reaction (RT-PCR), along with expression of stem cell factor (SCF), glial cell line-derived neurotrophic factor (GDNF), and bone morphogenic protein 4 (BMP4). Tight junctions were formed by Sertoli cells plated on transwell inserts coated with fibronectin as revealed by increased transepithelial electrical resistance (TER) and polarized secretion of the immunoregulatory protein, galectin-1. These primary Sertoli cell populations could be expanded dramatically in vitro and could be cryopreserved. The results show that functional human Sertoli cells can be propagated in vitro from testicular cells isolated from adult testis. The proliferative human Sertoli cells should have important applications in studying infertility, reproductive toxicology, testicular cancer, and spermatogenesis, and due to their unique biological properties potentially could be useful in cell therapy.

Figure 1

Morphology consistent with Sertoli cells shown by cultured cells. Phase contrast microscopy of MM-HSE-3608 cells of passage 2 (A) and passage 4 (B) reveals features characteristic of Sertoli cells including branching cytoplasmic processes and irregularly shaped nuclei. Electron microscopy of MM-HSE-2305 cells of passage 2 (C–F) clearly shows the irregularity of the shape of the nucleus (NM; nuclear membrane) in addition to the prominent nucleolus (Nu). Lipid droplets (L) and abundant rough (R) and smooth (S) endoplasmic reticulum are evident within the cytoplasm (D, E). Dense chromatin (DC) is apparent along the nuclear envelope (F). Scale bars: 200 μm (A, B), 2.3 μm (C), 750 nm (D), 500 nm (E, F).

Figure 2

Percentage increase in number of Sertoli cells (MM-HSE-2305; P5) at 72 h postplating compared to the number of cells in control wells at 24 h postplating. There was a significant difference in the proliferation of the cells (indicated by bar, p < 0.0001) plated at 1.0 or 2.0 × 10³ cells per well and those that were plated at 4.0 × 10³ cells per well based upon one-way ANOVA. The number of cells initially plated (assessed at 24 h) at 1.0 and 2.0 × 10³ cells per well increased significantly by almost 50% over 48 h (**p = 0.004). The results are the mean of two independent experiments with quadruplicates, and are representative of three independent experiments.

Figure 3

Immunochemical staining reveals cell proliferation, expression of characteristic Sertoli cell markers, and few myoid cells. Fluorescent microscopy of MM-HSE-3608 (passage 6) cells after treatment with 2 μM EdU for 3 days (A) demonstrates nuclear uptake of EdU in almost all cells. Fluorescence microscopy of MM-HSE-2305 cells (passage 4) stained with AdipoRed reveals bright spots shown in grayscale that are indicative of lipid droplets (B). Immunostaining of HSE-MM-2305 cells (passage 3) for GATA-4 (C) demonstrates the expected nuclear expression pattern, whereas bisbenzamide (Hoescht) dye (blue) staining of nuclei and immunostaining for FSHr (D) is consistent with its expected cytoplasmic localization. Low detection of Vector Red from alkaline phosphatase activity (E) demonstrates that there are few myoid cells in the culture. Two red-stained myoid cells can be observed. Scale bars: 100 or 200 μm as indicated.

Figure 4

RT-PCR confirms cellular expression of genes that are characteristic of Sertoli cells. Lane 1 is MM-HSE-3608 cells; 2 is MM-HSE-1208.

Figure 5

Flow cytometry shows more than 90% of cells at passage 3–4 express the Sertoli cell markers GATA-4 and Sox9, and less than 7% express CD34, a marker for hematopoietic and endothelial progenitor cells. Gray histograms showing antibodies staining of MM-HSE-3608 (A) and MM-HSE-2305 (B) cells for CD13, CD29, CD34, CD44, Sox9, CD73, CD90, CD105, CD166, and GATA-4 compared to that of isotype control antibodies (white histograms). The percentage of cells positively stained for each antigen is shown.

Figure 6

Immunocytochemistry of human testes sections reveals expression of galectin-1. Photomicrographs of cryosections of normal adult testis stained with hematoxylin and eosin (A), and immunostained with galectin-1 antibody (B). The hematoxylin and eosin-stained cross section of the seminiferous tubule (A) shows the structural architecture of the tubule in which Sertoli and spermatogonial stem cells are surrounded by peritubular myoid cells, and germ cells fill the lumen. Leydig cells are located in the interstitial spaces between the tubules. High levels of galectin-1 (B) staining (green) can be observed on the perimeter of the tubule, and within the tubules associated with both Sertoli cells and differentiating germ cells. Scale bars: 50 μm.

Figure 7

A Ponceau S-stained blot (upper) of conditioned media from the apical (A lanes) and bottom (B lanes) transwell chambers containing a monolayer of Sertoli cells on cell culture inserts at 3 and 7 days postplating with MW markers (M lane) on the left side is presented. Greater apical (top chamber) secretion was detected of an unidentified protein running near the 260-kDa marker on the Ponceau S-stained loading control blot (top arrow). Growth medium only as a negative control is shown in lane C. In the immunoblot (lower) greater secretion of galectin-1 into conditioned media from apical (A lanes) compared to bottom chamber (B lanes) from transwells at 3 and 7 days postplating is revealed by the bands that are at approximately 14.5 kDa (bottom arrow).

Figure 8

Conditioned media from human Sertoli cells suppress lymphocyte proliferation. Proliferation of lymphocytes was significantly decreased by conditioned media from MM-HSE-2305 Sertoli cells (two-way ANOVA, p < 0.001) compared to HS-27 conditioned medium. Values represent the mean ± SD and are from three separate experiments with each concentration tested in triplicate.