Single-cell gene expression analysis reveals diversity among human spermatogonia

STUDY QUESTION

Is the molecular profile of human spermatogonia homogeneous or heterogeneous when analysed at the single-cell level?

SUMMARY ANSWER

Heterogeneous expression profiles may be a key characteristic of human spermatogonia, supporting the existence of a heterogeneous stem cell population.

WHAT IS KNOWN ALREADY

Despite the fact that many studies have sought to identify specific markers for human spermatogonia, the molecular fingerprint of these cells remains hitherto unknown.

STUDY DESIGN, SIZE, DURATION

Testicular tissues from patients with spermatogonial arrest (arrest, n = 1) and with qualitatively normal spermatogenesis (normal, n = 7) were selected from a pool of 179 consecutively obtained biopsies. Gene expression analyses of cell populations and single-cells (n = 105) were performed. Two OCT4-positive individual cells were selected for global transcriptional capture using shallow RNA-seq. Finally, expression of four candidate markers was assessed by immunohistochemistry.

PARTICIPANTS/MATERIALS, SETTING, METHODS

Histological analysis and blood hormone measurements for LH, FSH and testosterone were performed prior to testicular sample selection. Following enzymatic digestion of testicular tissues, differential plating and subsequent micromanipulation of individual cells was employed to enrich and isolate human spermatogonia, respectively. Endpoint analyses were qPCR analysis of cell populations and individual cells, shallow RNA-seq and immunohistochemical analyses.

MAIN RESULTS AND THE ROLE OF CHANCE

Unexpectedly, single-cell expression data from the arrest patient (20 cells) showed heterogeneous expression profiles. Also, from patients with normal spermatogenesis, heterogeneous expression patterns of undifferentiated (OCT4, UTF1 and MAGE A4) and differentiated marker genes (BOLL and PRM2) were obtained within each spermatogonia cluster (13 clusters with 85 cells). Shallow RNA-seq analysis of individual human spermatogonia was validated, and a spermatogonia-specific heterogeneous protein expression of selected candidate markers (DDX5, TSPY1, EEF1A1 and NGN3) was demonstrated.

LIMITATIONS, REASONS FOR CAUTION

The heterogeneity of human spermatogonia at the RNA and protein levels is a snapshot. To further assess the functional meaning of this heterogeneity and the dynamics of stem cell populations, approaches need to be developed to facilitate the repeated analysis of individual cells.

WIDER IMPLICATIONS OF THE FINDINGS

Our data suggest that heterogeneous expression profiles may be a key characteristic of human spermatogonia, supporting the model of a heterogeneous stem cell population. Future studies will assess the dynamics of spermatogonial populations in fertile and infertile patients.

LARGE SCALE DATA

RNA-seq data is published in the GEO database: GSE91063.

STUDY FUNDING/COMPETING INTEREST(S)

This work was supported by the Max Planck Society and the Deutsche Forschungsgemeinschaft DFG-Research Unit FOR 1041 Germ Cell Potential (grant numbers SCHO 340/7-1, SCHL394/11-2). The authors declare that there is no conflict of interest.

Rapamycin-insensitive companion of mTOR (RICTOR) amplification defines a subset of advanced gastric cancer and is sensitive to AZD2014-mediated mTORC1/2 inhibition

Background

Targeting oncogenic genomic aberrations is an established therapeutic strategy in multiple tumor types. Molecular classification has uncovered a number of novel targets, and rapamycin-insensitive companion of mTOR (RICTOR) amplification has been identified in lung cancer. Further investigation assessing the therapeutic potential of RICTOR amplification as a novel target across advanced cancers is needed.

Patients and methods

Tumor samples from 640 patients with metastatic solid tumors, primarily gastrointestinal and lung cancers were prospectively subjected to a next-generation sequencing (NGS) assay to identify molecular targets. Samples with NGS-detected RICTOR amplification were confirmed with FISH. A RICTOR-amplified patient-derived cell (PDC) line was generated and used to investigate the effectiveness of selective AKT, mTORC1, and mTORC1/2 inhibition.

Results

NGS identified 13 (2%) of 640 patients with RICTOR-amplified tumors (6 gastric, 3 NSCLC, 1 SCLC, 1 CRC, 1 sarcoma, 1 MUO). Of the 13 patients, seven patients had RICTOR protein overexpression by IHC. The prevalence of RICTOR amplification in gastric cancer by NGS was 3.8% (6/160). FISH testing confirmed amplification (RICTOR/control >2) in 5/13 (38%) of samples, including four gastric cancers and one lung cancer. Treatment of a RICTOR amplified PDC with a selective AKT (AZD5363), selective mTORC1 (everolimus), dual mTORC1/2 (AZD2014), and the multi-target kinase inhibitor pazopanib demonstrated preferential sensitivity to the mTORC1/2 inhibitor (AZD2014). Knockdown of RICTOR reversed PDC sensitivity to AZD2014, validating the importance of RICTOR amplification to the PDC line.

Conclusions

RICTOR amplification is a rare but therapeutically relevant genomic alteration across solid tumors. Our results support further pre-clinical and clinical investigation with AZD2014 in RICTOR amplified gastric cancer and highlights the importance of genomic profiling.

EphrinB1 promotes cancer cell migration and invasion through the interaction with RhoGDI1

Eph receptors and their corresponding ephrin ligands have been associated with regulating cell–cell adhesion and motility, and thus have a critical role in various biological processes including tissue morphogenesis and homeostasis, as well as pathogenesis of several diseases. Aberrant regulation of Eph/ephrin signaling pathways is implicated in tumor progression of various human cancers. Here, we show that a Rho family GTPase regulator, Rho guanine nucleotide dissociation inhibitor 1 (RhoGDI1), can interact with ephrinB1, and this interaction is enhanced upon binding the extracellular domain of the cognate EphB2 receptor. Deletion mutagenesis revealed that amino acids 327–334 of the ephrinB1 intracellular domain are critical for the interaction with RhoGDI1. Stimulation with an EphB2 extracellular domain-Fc fusion protein (EphB2-Fc) induces RhoA activation and enhances the motility as well as invasiveness of wild-type ephrinB1-expressing cells. These Eph-Fc-induced effects were markedly diminished in cells expressing the mutant ephrinB1 construct (Δ327–334) that is ineffective at interacting with RhoGDI1. Furthermore, ephrinB1 depletion by siRNA suppresses EphB2-Fc-induced RhoA activation, and reduces motility and invasiveness of the SW480 and Hs578T human cancer cell lines. Our study connects the interaction between RhoGDI1 and ephrinB1 to the promotion of cancer cell behavior associated with tumor progression. This interaction may represent a therapeutic target in cancers that express ephrinB1.

Thermophilic Ferritin 24mer Assembly and Nanoparticle Encapsulation Modulated by Interdimer Electrostatic Repulsion

Protein cage self-assembly enables encapsulation and sequestration of small molecules, macromolecules, and nanomaterials for many applications in bionanotechnology. Notably, wild-type thermophilic ferritin from Archaeoglobus fulgidus (AfFtn) exists as a stable dimer of four-helix bundle proteins at a low ionic strength, and the protein forms a hollow assembly of 24 protomers at a high ionic strength (∼800 mM NaCl). This assembly process can also be initiated by highly charged gold nanoparticles (AuNPs) in solution, leading to encapsulation. These data suggest that salt solutions or charged AuNPs can shield unfavorable electrostatic interactions at AfFtn dimer-dimer interfaces, but specific "hot-spot" residues controlling assembly have not been identified. To investigate this further, we computationally designed three AfFtn mutants (E65R, D138K, and A127R) that introduce a single positive charge at sites along the dimer-dimer interface. These proteins exhibited different assembly kinetics and thermodynamics, which were ranked in order of increasing 24mer propensity: A127R < wild type < D138K ≪ E65R. E65R assembled into the 24mer across a wide range of ionic strengths (0-800 mM NaCl), and the dissociation temperature for the 24mer was 98 °C. X-ray crystal structure analysis of the E65R mutant identified a more compact, closed-pore cage geometry. A127R and D138K mutants exhibited wild-type ability to encapsulate and stabilize 5 nm AuNPs, whereas E65R did not encapsulate AuNPs at the same high yields. This work illustrates designed protein cages with distinct assembly and encapsulation properties.

Rapamycin Interferes With Postdepletion Regulatory T Cell Homeostasis and Enhances DSA Formation Corrected by CTLA4-Ig

Previously, we demonstrated that alemtuzumab induction with rapamycin as sole maintenance therapy is associated with an increased incidence of humoral rejection in human kidney transplant patients. To investigate the role of rapamycin in posttransplant humoral responses after T cell depletion, fully MHC mismatched hearts were transplanted into hCD52Tg mice, followed by alemtuzumab treatment with or without a short course of rapamycin. While untreated hCD52Tg recipients acutely rejected B6 hearts (n = 12), hCD52Tg recipients treated with alemtuzumab alone or in conjunction with rapamycin showed a lack of acute rejection (MST > 100). However, additional rapamycin showed a reduced beating quality over time and increased incidence of vasculopathy. Furthermore, rapamycin supplementation showed an increased serum donor-specific antibodies (DSA) level compared to alemtuzumab alone at postoperation days 50 and 100. Surprisingly, additional rapamycin treatment significantly reduced CD4(+) CD25(+) FoxP3(+) T reg cell numbers during treatment. On the contrary, ICOS(+) PD-1(+) CD4 follicular helper T cells in the lymph nodes were significantly increased. Interestingly, CTLA4-Ig supplementation in conjunction with rapamycin corrected rapamycin-induced accelerated posttransplant humoral response by directly modulating Tfh cells but not Treg cells. This suggests that rapamycin after T cell depletion could affect Treg cells leading to an increase of Tfh cells and DSA production that can be reversed by CTLA4-Ig.

Evaluation of interface trap densities and quantum capacitance in carbon nanotube network thin-film transistors

The interface trap density in single-walled carbon nanotube (SWNT) network thin-film transistors (TFTs) is a fundamental and important parameter for assessing the electronic performance of TFTs. However, the number of studies on the extraction of interface trap densities, particularly in SWNT TFTs, has been insufficient. In this work, we propose an efficient technique for extracting the energy-dependent interface traps in SWNT TFTs. From the measured dispersive, frequency-dependent capacitance-voltage (C-V) characteristics, the dispersive-free, frequency-independent C-V curve was obtained, thus enabling the extraction and analysis of the interface trap density, which was found to be approximately 8.2 × 10(11) eV(-1) cm(-2) at the valence band edge. The frequency-independent C-V curve also allows further extraction of the quantum capacitance in the SWNT network without introducing any additional fitting process or parameters. We found that the extracted value of the quantum capacitance in SWNT networks is lower than the theoretical value in aligned SWNTs due to the cross point of SWNTs on the SWNT network. Therefore, the method proposed in this work indicates that the C-V measurement is a powerful tool for obtaining deep physical insights regarding the electrical performance of SWNT TFTs.