Impairment of 7F2 osteoblast function by simulated partial gravity in a Random Positioning Machine.

The multifaceted adverse effects of reduced gravity pose a significant challenge to human spaceflight. Previous studies have shown that bone formation by osteoblasts decreases under microgravity conditions, both real and simulated. However, the effects of partial gravity on osteoblasts' function are less well understood. Utilizing the software-driven newer version of the Random Positioning Machine (RPMSW), we simulated levels of partial gravity relevant to future manned space missions: Mars (0.38 G), Moon (0.16 G), and microgravity (Micro, ~10-3 G). Short-term (6 days) culture yielded a dose-dependent reduction in proliferation and the enzymatic activity of alkaline phosphatase (ALP), while long-term studies (21 days) showed a distinct dose-dependent inhibition of mineralization. By contrast, expression levels of key osteogenic genes (Alkaline phosphatase, Runt-related Transcription Factor 2, Sparc/osteonectin) exhibited a threshold behavior: gene expression was significantly inhibited when the cells were exposed to Mars-simulating partial gravity, and this was not reduced further when the cells were cultured under simulated Moon or microgravity conditions. Our data suggest that impairment of cell function with decreasing simulated gravity levels is graded and that the threshold profile observed for reduced gene expression is distinct from the dose dependence observed for cell proliferation, ALP activity, and mineral deposition. Our study is of relevance, given the dearth of research into the effects of Lunar and Martian gravity for forthcoming space exploration.

Introduction

Loss of structural skeletal mineral is a health complication facing any human presence in space or, to a yet unknown degree, on other terrestrial bodies with reduced gravity, like the Moon (~0.16 G) or Mars (~0.38 G). The detrimental effects of microgravity on the musculoskeletal system have been known, albeit not fully understood, since the onset of manned spaceflight. The effect was confirmed as astronauts and cosmonauts began to spend increasingly more time in space[1]. In studies conducted on the International Space Station (ISS), significant trabecular volumetric bone mineral density (vBMD) losses in the spine and hip were noted with femoral vBMD showing an average loss of 2.7%/month and losses at the trabecular hip of 2.3%/month[2]. If not mitigated by proper countermeasures, loss of bone mass poses an increased risk of fracture in the workplace (space) and negatively impacts quality of life on return to Earth.

The primary cause of bone loss in microgravity (and presumably also under the partial-gravity conditions found on Mars or Moon) is believed to be the inhibition of osteoblast and osteocyte activity, resulting in decreased bone mineralization[3], concomitant with elevated osteoclast resorption. The upset equilibrium results in a substantial loss of bone mass over time, as mentioned above. The effects of microgravity on osteoblasts have been previously explored: in experiments carried out in orbital microgravity (at ~10−4–10−6 G)[4] osteoblast proliferation was inhibited, osteogenic differentiation was delayed, and the expression of genes controlling bone differentiation was reduced[5-7], while the bone resorption by osteoclasts increased aggressively[8,9]. The result is an overall loss of bone mineral density, wherein load-bearing bones are prone to atrophy[10], although one study reports a small increase in skull bone mineral density[11]. These effects have been replicated when osteoblasts and preosteoblasts were cultured on earth in microgravity analogs (rotating wall vessel (RWV) bioreactors and clinostats) that recapitulate certain aspects of microgravity. These experiments demonstrated that simulated microgravity (Micro) conditions (equivalent to ~10−3 G[12,13]), suppress osteoblast differentiation[12,14] and alter osteoclast function and survival[15], similar to the effects observed in orbital microgravity, making these venues cost-effective substitutes for gravity modeling and experimentation.

While alternatives like parabolic flights exist for short-term studies, the experimental gold standards for simulating microgravity on Earth are rotating clinostats[16] or microgravity-simulating bioreactors[12,17]. There is ample evidence in the literature suggesting that the results of ground-based modeled microgravity studies mimic real space conditions[18-20]. In comparative studies, similar degrees of inhibition of proliferation and differentiation have been observed both in osteoblast-like cells cultured in true-microgravity in space as well as in experiments carried out in simulated microgravity using clinostats[21]. The Random Positioning Machine (RPM), a 3D clinostat, randomizes orientation with respect to Earth’s gravitation field so that a sample’s gravity vector averages out to close to zero, providing a state of simulated microgravity[22]. Using outer and inner frames, fixed to two separate axes, the RPM can rotate independently in three dimensions and simulate partial gravities beyond microgravity, such as Moon and Mars. This is an advantage the RPM has over the two-dimensional rotation of the rotating wall vessel (RWV) or 2D clinostat. In vitro studies carried out in the RPM have yielded results similar to those performed in actual (orbital) microgravity in bone-marrow-derived mesenchymal stem cells (MSCs)[23] and 2T3 preosteoblasts[24]. An alternative approach to the operation of an RPM (RPMSW) is to control the range of clinostat motion through the use of specific path files that move the arms of the device through pre-determined paths. These paths are not random, rather they are set for every iteration of the experiment, allowing for the simulation of distinct partial gravities: in our case simulating the reduced gravity levels encountered for Mars, Moon, and Microgravity[25-27].

For the sake of scientific rigor and within the context of this paper, the term Simulated Partial Gravity (SPG) or Simulated Microgravity (SMG) refers specifically to the respective reduction in the net gravitational vector as generated by the motion-averaged movement of the 3D clinostat or RPM. The use of the terms “Mars” and “Moon” does not indicate that we actually recreated the respective extraterrestrial partial-gravity conditions. Rather, operating the RPM in the specific path files generates earth-based analog environments that simulate many of the environmental conditions expected at 0.38 G (Mars) and 0.16 G (Moon).

In this study, we cultured murine preosteoblasts (7F2 cells) in the RPM under conditions that simulate microgravity (10−3 G), and the partial gravities on Moon (0.16 G), and Mars (0.38 G). Specifically, we mostly focused on the effects of altered simulated gravity conditions on several distinct stages of initial osteoblastic cell functions, such as proliferation and osteogenic differentiation[28]. In addition, we also evaluated the effects of these simulated partial-gravity conditions on later-stage matrix mineralization. We hypothesized that the inhibition of these stage-specific osteoblast functions would depend on the simulated partial-gravity levels. In testing this hypothesis, we observed that the inhibition of two of these osteoblast functions, proliferation, and maturation (ALP-enzymatic activity and mineral deposition) was dose-dependent. By contrast, we found a distinct threshold profile for the inhibition of osteogenic marker gene expression.

Materials and methods

Materials

Alpha-minimum essential medium (a-MEM) and Fetal Bovine Serum (FBS) were purchased from Gibco Life Technologies (Carlsbad, CA, USA). L-ascorbic acid (AA), β–glycerophosphate (β-GP), para-Nitrophenylphosphate (pNPP), Alizarin Red, and Tri Reagent® for processing tissues were purchased from Sigma-Aldrich (St. Louis, MO, USA). Quant-iT™ PicoGreen™ dsDNA Assay Kit was purchased from Invitrogen Molecular Probes (Eugene, OR, USA) via Thermo Fisher Scientific. TaqMan Fast Universal PCR Master Mix (2X) and Taqman primers were purchased through Applied Biosystems (Foster City, CA, USA). RNeasy Protect Mini Kits were purchased from Qiagen (Hilden, GER).

Cell culture techniques

7F2 murine preosteoblasts (American Type Culture Collection, Manassas, VA, USA, CRL-12557) were cultured in α-MEM media supplemented with 10 mM HEPES, 10% FBS, 1% streptomycin and penicillin, and maintained in a humidified, 37 °C, 5% CO2/air incubator (maintenance medium). For osteogenic induction, the cells were cultured in osteogenic media containing the above complete alpha-MEM culture medium supplemented with 10 mM β-glycerophosphate and 10 µg/ml ascorbic acid (differentiation medium). The media was changed every 3 days. In addition to frequent equilibrated media changes, oxygenation was maintained by using modified culture flasks utilizing silicon caps in place of ventilated caps, and shear stress was minimized by having the flask completely filled with culture medium and devoid of air bubbles, which are notoriously detrimental to maintaining the slow-shear, simulated microgravity conditions[29], thus maintaining the concept of near-solid body (“zero headspace”). Cells were cultured as 2D monolayers in T-12.5 Falcon™ Tissue Culture Treated Flasks (Fisher Scientific, Waltham, MA, USA) retrofitted with Fischer-brand Silicone Recessed Septum Stoppers (internal diameter: 14.5–15.5 mm) for enhanced gas exchange. The accessories used are shown in Supplementary Fig. 1, including the 3D printed mounting cage for loading and unloading flasks and the perfusion system used to eliminate bubbles and change media.

Random Positioning Machine

To simulate reduced gravity, all experiments were carried out in the second generation of the software-driven Random Positioning Machine (RPMSW 2.0) (originally DutchSpace Airbus, Leiden, Netherlands, now Yuri GmbH, Meckenbeuren, Germany). The Mode of operation of this RPM differs from that of the earlier generation, which was developed for generating completely random paths, thus creating a simulated microgravity condition, and used for years in research[18-20]. The “mode of operation” of the RPMSW for simulating micro and partial gravity (see Supplementary Fig. 3) was described by Benavides Damm et al.[30] and was previously used to test the effects of modeled Mars gravity in plants[25,31]. To simulate partial gravity the RPM utilizes pre-determined path files, developed, and validated by the manufacturer (Dutch Airbus), in which the software directs the motion of the RPM arms in a non-random fashion. These paths will have a degree of preference along the Earth gravity vector, and the result is a net positive gravity that is greater than net-zero but <1 G normal. The motion of the RPM in random mode over time can be visualized as a sphere while the motion over time of a path file can be visualized as a prolate spheroid. In the center of this spheroid, a sample is not weightless, as with a net-zero sphere, but instead experiences partial gravity, where the larger the eccentricity of the spheroid the higher the level of gravity. Additional validation of the simulation of partial gravity via vector averaging was achieved through maintenance software of the RPM, which converts feedback from the frames into a real-time vector average (Supplementary Fig. 4).

Alkaline phosphatase (ALP) activity assay

Alkaline phosphatase (ALP) enzymatic activity was used as a marker for osteoblastic differentiation and quantitated spectrophotometrically. Following a room-temperature PBS wash, cell monolayers were scraped in 250 µl PBS and transferred into 1 ml microcentrifuge tubes. The cells were then lysed with 250 µl 0.2% Triton in PBS, to a final concentration of 0.1% (v/v) Triton X-100 (500 µl), followed by one freeze-thawing cycle (−80 °C/RT). After thawing and centrifugation (2000 × g, 1 min), the supernatants were used to determine ALP activity according to the protocol of Lin et al.[32] with some modifications: the buffer used was 10 mM MgCl2, 0.5 M AMP (2-Amino-2-Methyl-1-Propanol), supplemented with 9 mM of the ALP substrate, p-nitrophenylphosphate (pNPP). The lysate was diluted 10×. Color development was read in situ in an Infinite 200 PRO multimode plate reader (Tecan Group Ltd., Switzerland) every 2 min for 14 min at 405 nm. Readings were converted to concentration (nM) with a standard curve based on 4-Nitrophenol. ALP results are presented as enzyme activity over time (the rate of p-nitrophenol production from the p-nitrophenylphosphate substrate) and normalized to cell number as calculated from DNA content, using PICO green (as below). The normalized results are expressed as the amount of substrate converted (ng) over time per number of cells (ng/min/10k cells).

PICO-green assay for cell proliferation

PicoGreen dsDNA Quantitation Reagent (Invitrogen, Eugene OR, USA) was supplied as a 1-ml concentrated dye solution in anhydrous dimethylsulfoxide (DMSO) and used following the manufacturer’s protocol, with precedent in osteoblast studies[33]. 100 µl of the 0.1% Triton monolayer lysate supernatant (see above) was removed, diluted 400×, and added to a 96-well plate. 100 µl of the combined PicoGreen Reagent (1:200 PicoGreen diluted in the TE buffer supplied with the kit) was added to each sample. After mixing and incubation for 5 min at room temperature, protected from light, fluorescence was measured on an Infinite 200 PRO multimode plate reader (Tecan Group Ltd., Switzerland) at 485 nm excitation, 535 nm emission. Standard curves were constructed from known cell numbers (counted in triplicate).

Alizarin Red and dissolved TECO assay for mineralization

Mineralization of the cultures was assessed qualitatively by Alizarin Red staining, essentially as previously described[34]. In brief, following washing with PBS and fixation with 10% neutral buffered formalin for 15 min, the cultures were stained using 0.5% Alizarin Red S (pH 4.2) Digital images of the stained calcified nodules were evaluated using ImageJ software (National Institutes of Health, NIH). Mineralization was also quantified using a commercially available, colorimetric calcium quantification kit (Teco Diagnostics, Anaheim CA), following destructive decalcification of the cultures in 0.6 N HCl and analyzing the supernatant according to the manufacturer’s instructions. Separate standard curves were established for each experiment.

RNA extraction and real-time PCR for osteogenic marker gene expression (ALPL, RUN, ON)

The expression of select osteogenic marker genes was determined by quantitative PCR (qPCR) essentially as previously described, with some minor modifications[34]. In brief, total RNA was extracted from 7F2 cells using a modified version of a hybrid Tri Reagent (Sigma-Aldrich, USA)/RNEasy® protocol. RNA was quantified using a NanoDrop Spectrophotometer (Thermo Scientific, Waltham, MA), and concentrations were brought to a uniform level using additional RNase-free water. RNA was reverse transcribed to cDNA using the high-capacity cDNA reverse transcription kit (Applied Biosystems, Foster City, CA) according to the manufacturer’s instructions. The cDNA was amplified in TaqMan Fast universal PCR master mix with TaqMan assay primers and probes according to the manufacturer’s instruction. Genes of interest were: ALPL (Mm00475834_m1), RUNx2 (Mm00501584_m1) and Sparc/osteonectin/BM40 (Mm00486332_m1). Quantitative PCR (qPCR) was performed in a RealPlex Real-Time PCR System (Eppendorf, Enfield, CT) with fast thermal cycling as described by Taqman (Applied Biosystems). The level of expression of each gene was normalized to the level of expression of a common standard housekeeping gene, glyceraldehyde 3-phosphate dehydrogenase (GAPDH) to determine the fold change in up/downregulation of the genes of interest using the comparative CT method (2-ΔΔCT)[35].

Statistical analysis

Throughout the text, unless otherwise specified, statistical differences between the samples were assessed by ANOVA and post hoc analysis using Tukey’s HSD (honestly significant difference). Mean absolute error (MAE) was used to determine the difference between modeled projections. Results were plotted using Excel or JMP Pro. Data are presented as means ± standard deviation. P < 0.05 was considered significant and noted as “*”, P < 0.01 and P < 0.001 were noted as “**” and “***”, respectively.