Human iPSC-derived neural progenitor stem cells were obtained from Axol Bioscience (Little Chesterford, UK) and were differentiated to cerebral cortical neurons in approximately 35–40 days following the recommended protocol. Phenotyping of fully differentiated cortical neurons was performed by flow cytometry using anti-CUTL1 and anti-Ctip2 (both from Abcam, Cambridge, UK) and isotype-matched control antibodies (BD Biosciences, Franklin Lakes, NJ) in accordance with the recent literature (Saito et al., 2011; Sakakibara et al., 2012).

Fluorescence intensities were measured by FACS Calibur (BD Biosciences) and data were analyzed by the FlowJo software (Tree Star, Ashland, OR). Further information about the characterization of cerebral cortical neurons (phenotyping and transcriptome analysis) is available at the company website (http://www.axolbio.com/page/cortical-neurons).

Leukocyte-enriched buffy coats were obtained from healthy blood donors drawn at the Regional Blood Center of the Hungarian National Blood Transfusion Service (Debrecen, Hungary) in accordance with the written approval of the Director of the National Blood Transfusion Service and the Regional and Institutional Ethics Committee of the University of Debrecen, Faculty of Medicine (Debrecen, Hungary). Written informed consent was obtained from the donors prior blood donation and their data were processed and stored according to the directives of the European Union. Peripheral blood mononuclear cells (PBMCs) were separated by a standard density gradient centrifugation with Ficoll-Paque Plus (Amersham Biosciences, Uppsala, Sweden). Monocytes were purified from PBMCs by positive selection using immunomagnetic cell separation with anti-CD14 microbeads according to the manufacturer's instruction (Miltenyi Biotech, Bergisch Gladbach, Germany). After separation on a VarioMACS magnet, 96–99% of the cells were CD14⁺ monocytes as measured by flow cytometry. For moDC generation, monocytes were cultured in 12-well tissue culture plates at a density of 2 × 10⁶ cells/ml in AIMV medium (Invitrogen, Carlsbad, CA) supplemented with 80 ng/ml GM-CSF (Gentaur Molecular Products, Brussels, Belgium) and 100 ng/ml IL-4 (Peprotech EC, London, UK). On day 2, the same amounts of GM-CSF and IL-4 were added to the cell cultures, and moDCs were harvested on day 5. For moMAC differentiation, 50 ng/ml M-CSF (Gentaur) was added to the monocyte cultures on days 0 and 2, and fully differentiated macrophages were collected on day 4. Phenotyping of moDCs and moMACs was carried out by flow cytometry using anti-CD209-PE, anti-CD14-PE (Beckman Coulter, Hialeah, FL), anti-CD68-PE, anti-HLA-DR-FITC, and isotype-matched control antibodies (BD Biosciences).

Induction of hypoxia was performed following the modified version of a previously described protocol (Wu and Yotnda, 2011). To allow culture media to de-gas, 12-well plates containing serum-free AIMV were pre-incubated in low-oxygen atmosphere (94.5% N₂, 5% CO₂, 0.5% O₂) for 4 h using a hypoxia chamber (Billups-Rothenberg, San Diego, CA). Cells were then seeded on the plates, placed in the chamber and were incubated at 37°C for up to 6 h under similar hypoxia conditions. An automated regulator (with built-in flow meter and oxygen-sensor) was used to ensure and maintain the proper composition of gas mixture within the chamber. After hypoxia treatment cells were removed from the chamber and were either immediately lysed (for Western blot or QPCR), or placed on ice (for Annexin V-FITC staining and flow cytometry analysis).

N,N-dimethyltryptamine (R&D Systems, Abingdon, UK) and BD1063 dihydrochloride (Tocris, Bristol, UK) were used at working concentrations of 1–200 and 1–100 μM, respectively. BD1063 treatments always preceded the addition of DMT by 30 min to allow successful antagonism. DMT, as a controlled substance (Schedule I drug), was used with the approval and monitoring of the Hungarian Institute for Forensic Sciences and the Hungarian National Police Department.

To prepare cell lysates for Western blotting and QPCR measurements, in vitro cell cultures were sampled after 6 h of treatment (if not stated otherwise).

Real-time quantitative polymerase chain reaction (QPCR) was performed as described previously (Szabo et al., 2016). Briefly, total RNA was isolated by TRIzol reagent (Invitrogen, Carlsbad, CA). 1.5–2 μg of total RNA were reverse transcribed using SuperScript II RNase H reverse transcriptase (Invitrogen) and Oligo(dT)15 primers (Promega, Madison, WI). Gene-specific TaqMan assays (Applied Biosystems, Foster City, CA) were used to perform QPCR in a final volume of 12 μl in triplicates using AmpliTaq Gold DNA polymerase and ABI StepOnePlus real-time PCR instrument (Applied Biosystems). Amplification of 36B4 was used as a normalizing control. Cycle threshold values (Ct) were determined by using the StepOne 2.1 software. Constant threshold values were set for each gene throughout the study. The sequence of the primers and probes are available upon request.

Cells were lysed in Laemmli buffer and the protein extracts were tested by Ab specific for Sig-1R/OPRS1, HIF-1α, ATF6, p65, phospho-p65 (S536) (all from Abcam), and β-actin (Sigma, Schnelldorf, Germany) diluted at 1:500 and 1:000, respectively. Anti-rabbit Ab conjugated to horseradish peroxidase (GE Healthcare, Little Chalfont Buckinghamshire, UK) was used as the secondary Ab at a dilution of 1:5000. The SuperSignal enhanced chemiluminescence system was used for probing target proteins (Thermo Scientific, Rockford, IL). After the membranes had been probed for Sig-1R/OPRS1 or HIF-1α, they were stripped and re-probed for β-actin.

The percentage of apoptotic cells was assessed by using an Annexin V apoptosis kit (BioVision, CA, USA) following the manufacturer's recommendations. The rate of necrotic cell death was also monitored simultaneously by measuring membrane integrity. Necrotic cell death was quantified based on the loss of membrane integrity and the uptake of propidium iodide (PI). Upon stimulation cells were harvested and stained with PI (10 μg/ml) and analyzed immediately by flow cytometry.

Gene-specific siRNA knockdown was performed by Silencer Select siRNA (Applied Biosystems) transfection using Gene Pulser Xcell instrument (Bio-Rad, Hercules, CA). Pulse conditions were square-wave pulse, 500 V, 0.5 ms. Immediately after electroporation, moMACs/moDCs were transferred to pre-warmed, fresh medium supplemented with penicillin, streptomycin, and L-glutamine. Gene knockdown in neurons was performed with the LyoVec transfection system (InvivoGen, San Diego, CA) according to the manufacturer's recommendations. Silencing of Sig-1R gene expression was performed by using a mix of three of the available Sig-1R siRNAs. Silencer negative control non-targeting siRNA (Applied Biosystems) was used as a negative control. The efficacy of siRNA treatments was tested 2 days post-transfection by Western blotting.

Data are presented as mean ± SEM. A t-test was used for comparison of two groups followed by Bonferroni correction. Two-way ANOVA was used for multiple comparisons. Differences were considered to be statistically significant at p < 0.05 (^*).

In this work, we used three experimental models—moMACs, moDCs, and iPSC-derived neurons—to investigate the effects of DMT-mediated activation of the Sig-1R in hypoxia. In a previous study, we have already showed that human moMACs and moDCs express high levels of the Sig-1R gene and protein (Szabo et al., 2014). Though the expression of Sig-1R has also been demonstrated in human neural cell types (reviewed in Frecska et al., 2013), it has not been investigated in NSCs and in iPSC-derived cortical neurons. Therefore, we first sought for the Sig-1R expression profile of iPSC-derived neurons during the differentiation process. We found that NSCs have a low baseline expression of Sig-1R at both the mRNA and protein levels and its expression increases during the differentiation of cells into cortical neurons (Figure 1). The expression of Sig-1R becomes prominent after the 14th (mRNA) and 21st (protein) days of differentiation where significant changes were detected as compared to the baseline Sig-1R expression of NSCs (Figures 1A,B). The highest level of Sig-1R was detected at the end of the differentiation process (Figure 1).

Based on our previous findings about the detectable levels of Sig-1R in human iPSC-derived neurons (Figure 1), moMACs, and moDCs (Szabo et al., 2014), we next investigated whether the in vitro treatment of these cell types with DMT, as a natural endogenous ligand of Sig-1R (Fontanilla et al., 2009), influences their survival in hypoxia. Human tissues experience a wide range of diverse oxygen tensions that profoundly differ from that of the inhaled ambient oxygen (21%, 160 mm Hg). By the time inhaled oxygen reaches tissues and organs its tension drops to 2–9% (14–65 mm Hg; Brahimi-Horn and Pouysségur, 2007). Thus, in various experimental human tissue cultures, 2–9% in vitro oxygen level is considered as physiologic normoxia (Simon and Keith, 2008; Mohyeldin et al., 2010). Certain low vascular density tissues may experience even lower oxygen tensions, however, 1% or lower level of oxygen is often regarded as a hypoxic environment in the literature (Semenza, 1999; Liu and Simon, 2004). In our experiments we used severe hypoxic treatment of cells where, in accordance with other studies (Lee et al., 2013; Harrison et al., 2015), the level of oxygen was set to 0.5% in a hypoxia chamber.

Incubation in hypoxic environment rapidly induced cell death in all cultures as assessed by Annexin V-FITC staining and subsequent flow cytometry analysis (Figure 2). The ratio of necrotic cells, as measured with PI-staining, never exceeded 6% median values (±1%, n = 3 in neuron, and ±3%, n = 6 in moMAC/DC cultures; data not included in Figure 2). By using different concentrations (1–200 μM) we found that in vitro administration of 10–200 μM DMT increases the survival of iPSC-derived cortical neurons (Figure 2A), moMACs (Figure 2B), and moDCs (Figure 2C) in hypoxia. This survival-boosting effect was statistically significant at DMT concentrations of as low as 10 μM (neurons) or 50 μM (moMAC/DCs) as compared to the non-DMT-treated hypoxia controls (Figure 2). Non-DMT-treated control cultures (hypoxia controls) of iPSC-derived neurons showed significant increase in the ratio of apoptotic cells even after 1 h of hypoxia treatment as compared to normoxic controls (Figure 2A). Six hours of hypoxia resulted in a median of 19% (±2%, n = 3) survival rate in case of non-treated iPSC-derived cortical neuron control cultures, while the administration of 10 and 50 μMs DMT elevated this value to a median of 31% (±6%, p = 0.037) and 64% (±5%, p = 0.006), respectively (Figure 2A). Interestingly, in cases of moMACs and moDCs significant amounts of apoptotic cells could be detected only after 6 h of hypoxia (Figures 2B,C). Furthermore, no difference was seen between the modulating effects of 50 and 200 μM concentrations of DMT (Figure 2). Normoxic cultures exhibited no sign of change in cellular viability neither in controls (Figures 2A–C) nor in DMT-treated (normoxia+1–200 μM DMT) cases (data not shown). Since 50 μM DMT, an achievable serum concentration reported by previous in vivo human (Strassman and Qualls, 1994) and animal studies (Shen et al., 2011), was found to be optimal for modulating the survival capacity of all cell types this concentration was used in further experiments.

Our results demonstrated that severe hypoxia (0.5% O₂) induced apoptotic death of human primary cells and DMT treatment could prevent this phenomenon (Figure 2). The expression of the transcription factor HIF-1α is strongly induced by hypoxia in many cell types, however, it is subject to ubiquitination and rapid degradation under normoxia (Huang et al., 1998; Kallio et al., 1999). The molecular basis of this process is an O₂-dependent hydroxylation of proline residues. Under hypoxia, HIF-1α is promptly assembled and carries out the downstream control of the expression of many genes related to hypoxic stress including VEGF. Thus, increased expressions of both HIF-1α and its target gene VEGF often signify hypoxic stress or state (Semenza, 2002). We next aimed to investigate whether HIF-1α was involved or affected in this process.

We found that 6 h of hypoxia treatment greatly induced the protein-level expression of HIF-1α in human iPSC-derived neurons, moMACs, and moDCs, and the administration of 50 μM DMT prevented this increase in all cell types (Figures 3A,B). In normoxia control experiments, DMT alone did not influence HIF-1α expressions (Figure 3A). The DMT-mediated inhibition of HIF-1α protein expression under hypoxia was found to be statistically significant as compared to hypoxia controls (Figure 3B). These results were consistent with our subsequent findings showing significantly decreased mRNA expressions of VEGF upon DMT treatment in all cell types under hypoxia (Figure 3C). Furthermore, we also monitored other stress-related pathways including NF-κB and the ER-stress sensor Activating transcription factor 6 (ATF6; Elbarghati et al., 2008). We found no alterations in the protein level expression of native p65 NF-κB subunit neither could detect its phosphorylated form in our cell types in hypoxic vs. normoxic conditions within the time-range of observation (data not shown). Interestingly, the level of the 50 kDa transcriptionally active form of ATF6 showed some degree of decrease when cells were treated with DMT under hypoxia as compared to non-treated controls. However, this change did not appear to be statistically significant (Figure S1).

In previous experiments we showed that Sig-1R is expressed in human iPSC-derived neurons (Figure 1), moMACs, and moDCs (Szabo et al., 2014). We also demonstrated that DMT treatment of these cell types can significantly increase their survival under severe hypoxia (Figure 2) and decrease their expression of HIF-1α and VEGF (Figure 3). As DMT has previously been described as a natural endogenous agonist of Sig-1R (Fontanilla et al., 2009) we next tested whether Sig-1R was involved in the observed cell biological effects of DMT.

To clarify the DMT-dependent modulatory role of Sig-1R in human primary cells and to check its contribution to the observed phenomena we performed Sig-1R gene knockdown experiments. Specific silencing of the Sig-1R gene resulted in a >93% (±5%, n = 3) of downregulation of the Sig-1R protein in human iPSC-derived neurons as compared to non-transfected and scrambled siRNA controls, and >96% (±3%, n = 4) and >91% (±5%, n = 4) in human moMACs and moDCs, respectively (Figure 4A). Using the same treatment protocols as in Figures 2, 3, we found that specific silencing of Sig-1R abrogated the modulatory potential of DMT on the survival (Figure 4B) and HIF-1α protein expression of human iPSC-derived neurons, moMACs, and moDCs (Figure 4C). In DMT-treated Sig-1R knockdown cultures, the survival of all cell types was found to be significantly lower as compared to controls (Figure 4B). Similarly, Sig-1R gene silencing resulted in the ablation of the modulatory capacity of DMT on HIF-1α protein expression (Figure 4C). These findings were further validated by an additional set of experiments in which we used the selective Sig-1R antagonist BD1063 to block Sig-1R-mediated signals (Figure S2). The effects of DMT on cellular survival and HIF-1α expression is thus dependent on the Sig-1R in human primary iPSC-derived neurons and monocyte-derived microglia-like cells.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.