Survival Pathways Are Differently Affected by Microgravity in Normal and Cancerous Breast Cells

International Journal of Molecular Sciences

Article Survival Pathways Are Differently Affected by Microgravity in Normal and Cancerous Breast Cells

Noemi Monti 1,2, Maria Grazia Masiello 3, Sara Proietti 3, Angela Catizone 4 , Giulia Ricci 5 , Abdel Halim Harrath 6 , Saleh H. Alwasel 6, Alessandra Cucina 3,7 and Mariano Bizzarri 1,2,*

1 Department of Experimental Medicine, Sapienza University of Rome, 00161 Rome, Italy; [email protected] 2 Systems Biology Group Lab, Sapienza University of Rome, 00161 Rome, Italy 3 Department of Surgery “Pietro Valdoni”, Sapienza University of Rome, 00161 Rome, Italy; [email protected] (M.G.M.); [email protected] (S.P.); [email protected] (A.C.) 4 Department of Anatomy, Histology, Forensic-Medicine and Orthopedics, Section of Histology and Embryology, Sapienza University of Rome, 00161 Rome, Italy; [email protected] 5 Department of Experimental Medicine, Università degli Studi della Campania “Luigi Vanvitelli”, 80138 Naples, Italy; [email protected] 6 Department of Zoology, College of Science, King Saud University, Riyadh 11451, Saudi Arabia; [email protected] (A.H.H.); [email protected] (S.H.A.) 7 Azienda Policlinico Umberto I, 00161 Rome, Italy * Correspondence: [email protected]; Tel.: +39-4976-6606

Abstract: Metazoan living cells exposed to microgravity undergo dramatic changes in morphological and biological properties, which ultimately lead to apoptosis and phenotype reprogramming. How- ever, apoptosis can occur at very different rates depending on the experimental model, and in some cases, cells seem to be paradoxically protected from programmed cell death during weightlessness. Citation: Monti, N.; Masiello, M.G.; These controversial results can be explained by considering the notion that the behavior of adherent Proietti, S.; Catizone, A.; Ricci, G.; cells dramatically diverges in respect to that of detached cells, organized into organoids-like, floating Harrath, A.H.; Alwasel, S.H.; Cucina, structures. We investigated both normal (MCF10A) and cancerous (MCF-7) breast cells and found A.; Bizzarri, M. Survival Pathways that appreciable apoptosis occurs only after 72 h in MCF-7 cells growing in organoid-like structures, Are Differently Affected by Microgravity in Normal and in which major modifications of cytoskeleton components were observed. Indeed, preserving cell Cancerous Breast Cells. Int. J. Mol. attachment to the substrate allows cells to upregulate distinct Akt- and ERK-dependent pathways Sci. 2021, 22, 862. https://doi.org/ in MCF-7 and MCF-10A cells, respectively. These findings show that survival strategies may differ 10.3390/ijms22020862 between cell types but cannot provide sufficient protection against weightlessness-induced apoptosis alone if adhesion to the substrate is perturbed. Received: 22 December 2020 Accepted: 13 January 2021 Keywords: microgravity; apoptosis; cytoskeleton; Akt; ERK; vinculin; survival pathways; Published: 16 January 2021 space biomedicine

Publisher’s Note: MDPI stays neu- tral with regard to jurisdictional clai- ms in published maps and institutio- 1. Introduction nal affiliations. Microgravity is one of the major challenges facing astronauts during spaceflight. Exposure to microgravity has been associated with various significant changes in the organization and function of several tissues [1]. Namely, numerous reports have highlighted Copyright: © 2021 by the authors. Li- that weightlessness—experienced in real (i.e., on board the International Space Station) or censee MDPI, Basel, Switzerland. in simulated microgravity (i.e., performed with specific tools, like the random position- This article is an open access article ing machine, RPM)—may foster programmed cell death in living cells [2–4]. Cell death distributed under the terms and con- may be promoted by activating different pathways, including apoptosis, autophagy, and ditions of the Creative Commons At- anoikis, and controversial results and discrepancies have been described [5]. Divergences tribution (CC BY) license (https:// among findings gathered to date may depend on differences in experimental design, tech- creativecommons.org/licenses/by/ nical tools (especially when simulated weightlessness is considered), and cell types [6]. 4.0/).

Int. J. Mol. Sci. 2021, 22, 862. https://doi.org/10.3390/ijms22020862 https://www.mdpi.com/journal/ijms Int. J. Mol. Sci. 2021, 22, 862 2 of 26

Furthermore, cells growing in microgravity are frequently partitioned into two different clusters—cells adhering to the substrate or, alternatively, organized into floating, organoid- like structures [7]—and it is likely that these two different phenotypes may respond in a very different way to weightlessness. Indeed, microgravity-dependent loss of adherence and cytoskeleton remodeling in floating cells can increase apoptosis when compared to cells that are growing adherent to the substrate [8]. Given that the rate of apoptosis is usually estimated by mixing cells displaying different adhesion capabilities when exposed to weightlessness, it is conceivable that the measured incidence of apoptosis is in fact an “average” obtained from different cell samples, each one displaying a specific cell death program. Additionally, as cells seem to efficiently cope with the stressful condition of microgravity by self-limiting the overall incidence of apoptosis when exposed for long periods [9], it can be hypothesized that living organism can activate pro-survival strategies. Undoubtedly, these are very relevant questions for properly planning scientific inquiries related to space biomedicine studies. Moreover, different responses triggered by normal or cancerous cells in the presence of biophysical stressful conditions could, in principle, suggest useful insights for a better appraisal of apoptotic-related processes occurring in different pathological conditions, including cancer. To address such issues, we planned to investigate apoptosis and survival pathways in both normal (MCF10A) and neoplastic (MCF-7) breast cells cultured in RPM versus cells growing in normal gravity condition (1 g). These populations display different adhesive responses to the substrate, as MCF-7 cells are partitioned into adherent and nonadherent clusters when cultured in simulated microgravity [6]. Our findings show that the early apoptotic response is actively counteracted by the concomitant activation of two different survival pathways—ERK and Akt—occurring in MCF10A and MCF-7 cells, respectively. However, at later times, a significant increase in apoptosis can be recorded in non-adherent MCF-7 cells only, indicating that loss of adherence and the concomitant disruption of cytoskeleton promoted by microgravity can ultimately overcome the pro-survival strategy enacted by cancerous MCF-7 cells.

2. Results 2.1. Microgravity Induces the Emergence of Two Stable Populations in Both MCF10A and MCF7 Microgravity induces astonishing modifications in cell morphology. Shape changes undergo early after microgravity exposition and are clearly recognizable after six hours (data not shown). After 24 h of microgravity conditioning, two different morphologic phenotypes spontaneously emerge, and cells are partitioned almost equally (49% versus 51%) into two phenotypically different populations. MCF10A cells split into two distinct morphological phenotypes, recognizable for size and shape features (large and small cells, RPM-L and RPM-S, respectively), growing adherently to the substrate (Figure1). In MCF-7 cells, one population is represented by flat, spindle cells adhering to the substrate (RPM-AD), while clumps or rounded cells (RPM-CLUMP) constitute the second subset, floating in the culture medium and associated with discrete cluster-resembling organoid- like structures (Figure2A,B). A cell’s phenotypes emerging after exposure to weightlessness have been demonstrated to be stable and viable over long periods (>72 h, data not shown), and the relative proportion of the two morphological clusters remains invariant for the full period of observation, suggesting that the intrinsic dynamics between the two populations are operating close to the equilibrium. This behavior has been previously observed in other types of breast cancer (MDA-MB-231) [10] as well as in cells obtained from different tissues (bones, lymphocytes, and thyroids) [11,12]. Int. J. Mol. Sci. 2021, 22, x FOR PEER REVIEW 3 of 26 Int. J. Mol. Sci. 2021, 22, 862 3 of 26 Int. J. Mol. Sci. 2021, 22, x FOR PEER REVIEW 3 of 26

Figure 1. MicrophotographsFigure 1. Microphotographs of MCF-10A ofby MCF-10A optical microscopy. by optical microscopy.MCF-10A cell MCF-10A line in the cell on-ground line in the control on-ground condition Figure 1. Microphotographs(A,C) and exposedcontrol of MCF-10A conditionto microgravity by ( Aoptical,C) ( andB ,Dmicroscopy.) exposed for 24 h. to Magnification microgravityMCF-10A cell ×320. (B line,D )The forin theimages 24 h.on-ground Magniﬁcation have been control cut× out320. condition (C The,D) images to highlight (A,C) and exposeddifferences to microgravity havein both been area (B cut,D and) outfor cell (24C , Dprofile.h.) Magnification to highlight Cropped differences images×320. The show in images both that areain have microgravity, and been cell cut proﬁle. out tw o( C Croppeddistinct,D) to highlightpopulations images show emerge differences in both area and cell profile. Cropped images show that in microgravity, two distinct populations emerge (evidenced thatin blue in microgravity,and light blue), two displaying distinct populationsdifferent fractal emerge dimension (evidenced and roundness in blue and values, light blue),as reported displaying in Table 1. (evidenced in blueScale and bar light 10 µm. blue), displaying different fractal dimension and roundness values, as reported in Table 1. different fractal dimension and roundness values, as reported in Table1. Scale bar 10 µm. Scale bar 10 µm.

Figure 2. Microphotographs of MCF-7 by optical microscopy. MCF-7 cell line in the on-ground control condition (A) and exposed to microgravity (B) for 24 h. MCF7 cells in microgravity were distributed into two populations. The first, repre- FigureFigure 2. MicrophotographsMicrophotographssented by floating-clump of of MCF-7 MCF-7 by by(random optical optical microscopy.positioning microscopy. machine MC MCF-7F-7 (RPMcell cell line)-CLUMP) line in the in theon-ground cells, on-ground are indicated control control conditionby black condition arrows, (A) and (A and) the sec- exposed to microgravityond are composed (B) for 24 adherent h. MCF7 cells cells (RPM-A in microgravityD.) Magnification were dist ×100.ributed Scale into bar two100 µm. populations. The first, repre- and exposed to microgravity (B) for 24 h. MCF7 cells in microgravity were distributed into two populations. The first, sented by floating-clump (random positioning machine (RPM)-CLUMP) cells, are indicated by black arrows, and the sec- represented by floating-clump (random positioning machine (RPM)-CLUMP) cells, are indicated by black arrows, and the ond are composed adherent cells (RPM-AD.) Magnification ×100. Scale bar 100 µm. second are composed adherent cells (RPM-AD.) Magnification ×100. Scale bar 100 µm.

Int. J. Mol. Sci. 2021, 22, 862 4 of 26

Table 1. Roundness, solidity, and fractal dimension (FD) in MCF-7 and MCF-10A at 24 h and 72 h. Mean values ± SD in on-ground control cells, RPM adherent cells, and RPM cell clumps. ** p < 0.01 versus on-ground control cells; *** p < 0.001 vs. on-ground control and RPM adherent cells by ANOVA followed by Bonferroni post-test. (Sample size = 50 cells for each group).

MCF-7 (breast cancer cells, poorly invasive) Fractal Roundness ± SD Solidity ± SD Demension ±SD (FD) 24 h on ground 0.646 ±0.0151 0.802 ±0.101 1.1560 ±0.0356 RPM (adherent cells) 0.667 ±0.139 0.844 ±0.130 *** 1.1381 ±0.0230 RPM (cell clumps) 0.783 ±0.111 *** 0.921 ±0.034 *** 1.1275 ±0.1017 *** 72 h on ground 0.619 ±0.149 0.759 ±0.100 1.1478 ±0.1296 RPM (adherent cells) 0.654 ±0.143 *** 0.773 ±0.093 1.1633 ±0.0333 RPM (cell clumps) 0.743 ±0.130 *** 0.897 ±0.063 *** 1.1385 ±0.0248 MCF-10A (non tumorigenic mammary) Fractal Roundness ±SD Solidity ±SD Demension ±SD (FD) 24 h on ground 0.709 0.154 0.788 0.259 1.344 0.088 RPM 0.713 0.135 0.821 0.215 1.371 0.096 ** 72 h on ground 0.726 0.131 0.856 0.132 1.391 0.092 RPM 0.711 0.125 0.860 0.090 1.443 0.081 ***

2.2. Morphological Parameters To assess how these changes are mirrored by simultaneous modifications in quantitative shape parameters, we investigated roundness (RO) and fractal dimension (FD) in both cell types after 24 h of weightlessness exposition. In both MCF10A and MCF-7, the two cell clusters enacted by exposure to simulated microgravity show significant differences in roundness and fractal dimension (Table1). Differences can still be observed when these parameters are compared with those measured in cells growing in normal gravity. Namely, by cutting out MCF10A growing in normal gravity or in simulated microgravity, respectively, one can easily appreciate the reported morphological differences (Figure1). Overall, such findings confirm that weightlessness promotes the partitioning of cells into distinct phenotypes, thus influencing the subsequent fate and behavior of each one of these clusters. Six studies carried out on cells cultured in simulated weightlessness have reported small, albeit significant, changes in growth and cell cycle distribution. Therefore, we as- sessed those parameters in both cell populations under scrutiny. In MCF10A, no significant differences were found in cell cycle distribution among cells cultured in normal or reduced gravity (data not shown), while in MCF-7 cells, a striking increase in both S and G2M phase was observed for RPM-CLUMP cells, as previously reported [6]. Proliferation, at 24 and 72 h after exposition to weightlessness was almost unchanged in MCF10A (Figure3A). As expected, a significant decrease in growth rate was observed in MCF-7, especially in the RPM-CLUMP cluster at 72 h (Figure3B). Overall, these findings suggest that floating cells experience appreciable stress, which is subsequently followed by an increase in the G2M fraction that favors cell growth arrest and consequent apoptosis enhancement, as previously observed in other experimental models of cellular stress [13]. Int. J. Mol. Sci. 2021, 22, 862 5 of 26 Int. J. Mol. Sci. 2021, 22, x FOR PEER REVIEW 5 of 26

Figure 3. EffectEffect of of microgravity microgravity on on proliferation proliferation of of MCF-10A MCF-10A (A (A) )and and MCF-7 MCF-7 (B ()B cells) cells at at24 24 and and 72 72 h after h after exposition exposition to 4 toweightlessness. weightlessness. Results Results showing showing the number the number of cells of cells×10 ×represents104 represents the mean the meanvalue value± SD of± threeSD of independent three independent experiments. ** p < 0.01 vs. on ground (OG) by ANOVA followed by Bonferroni post-test. experiments. ** p < 0.01 vs. on ground (OG) by ANOVA followed by Bonferroni post-test.

2.3. Cytoskeleton Changes 2.3. Cytoskeleton Changes Changes in cytoskeleton (CSK) architecture have been observed during simulated Changes in cytoskeleton (CSK) architecture have been observed during simulated and real microgravity, as early as after a few minutes [14]. Indeed, CSK self-organized and real microgravity, as early as after a few minutes [14]. Indeed, CSK self-organized processes are highly sensitive to the gravity field [15], and CSK proteins are deregulated processes are highly sensitive to the gravity field [15], and CSK proteins are deregulated even before any change in gene expression pattern can be recorded [16]. CSK changes are even before any change in gene expression pattern can be recorded [16]. CSK changes instrumental in driving progressive modifications in cell shape, and some of these archi- are instrumental in driving progressive modifications in cell shape, and some of these architecturaltectural alterations alterations play playa pivotal a pivotal role rolein tr iniggering triggering a wide a wide range range of death of death processes, processes, in- includingcluding apoptosis apoptosis and and anoikis anoikis [17]. [17]. Accordin Accordingly,gly, in in recent recent decades, the drug-mediateddrug-mediated interference with CSK remodeling has been activelyactively pursued to findfind a successful strategy of cancer treatment [[18].18]. Therefore, wewe exploredexplored CSKCSK changes changes in in both both MCF10A MCF10A and and MCF-7 MCF- cells7 cells in in microgravity. microgravity. Overall, Overall, both both cell cell types types underwentunderwent appreciableappreciable CSK modifications,modifications, although the most profound were recorded—asrecorded—as expected—inexpected—in floatingfloating MCF-7, in whichwhich cell detachment plays a pivotal role in promoting a number of different programmed cell death pathways, including anoikis.

2.4. F-actin In MCF10A MCF10A cells cells exposed exposed to to microgravity, microgravity, F-actin F-actin shows shows a different a different distribution distribution pat- patterntern when when compared compared to tocells cells cultured cultured on on ground. ground. Cortical Cortical F-actin F-actin can can be be recognized as parallel bundles to plasma membrane in 1 g cultured cells (Figure 44A),A), whereaswhereas inin RPM-RPM- cultured MCF-10 cells, actin bundles are orthogonalorthogonal to the periphery of plasma membrane (Figure4 4B,B, arrows). arrows). In In MCF-7 MCF-7 cells, cells, after after 24 24 h of h simulatedof simulated microgravity, microgravity, both both RPM-AD RPM-AD and RPM-CLUMPand RPM-CLUMP cells displaycells display a large a rearrangementlarge rearrangement of F-actin of F-actin when comparedwhen compared to 1g control to 1g cellscontrol (on cells ground, (on ground, OG). In controlOG). In MCF-7 control cells, MCF-7 cytosolic cells, F-actincytosolic fibers F-actin appear fibers well appear organized well inorganized bundles in associated bundles associated with the cell with plasma the cell membrane, plasma membrane, with recognizable with recognizable stress fibers stress and afibers well-shaped and a well-shaped network of network stress fibers of stress in the fibers cytosol in (Figurethe cytosol4C). (Figure In RPM-AD 4C). In cells, RPM-AD stress fiberscells, stress are slightly fibers reduced,are slightly while reduced, F-actin while is mostly F-actin localized is mostly at thelocalized cell border at the (Figure cell border4D). On(Figure the contrary,4D). On the in RPM-CLUMPScontrary, in RPM-CLUMPS cells, the actin cells, meshwork the actin losesmeshwork its organization, loses its organ- and actinization, filaments and actin appear filaments short, appear fragmented, short, and fragmented, significantly and reduced, significantly mostly reduced, recognizable mostly at therecognizable perinuclear at the level perinuclear of floating level cells of that floating form thecells organoid-like that form the structure, organoid-like whereas structure, stress fiberswhereas are stress dramatically fibers are reduced dramatically (Figure reduced4E). (Figure 4E).

Int. J. Mol. Sci. 2021, 22, 862 6 of 26 Int. J. Mol. Sci. 2021, 22, x FOR PEER REVIEW 6 of 26

FigureFigure 4. 4. ConfocalConfocal images images of of F-actin F-actin distribution distribution pattern pattern in in MCF-10A MCF-10A ( (AA,,BB)) and and MCF-7 MCF-7 ( (CC––EE)) cultured cultured in in normal normal gravity gravity (OG)(OG) and and microgravity microgravity (RPM) (RPM) conditions conditions for for 24 24 h. MCF-10A MCF-10A cells cells cultured cultured in in RPM RPM conditions conditions ( (BB)) show show actin actin bundles bundles orthogonal to the periphery of plasma membrane (arrows; scale bar 37.50 µm), while F-actin fibers appear well-organized orthogonal to the periphery of plasma membrane (arrows; scale bar 37.50 µm), while F-actin fibers appear well-organized in in MCF-7 cultured in the on-ground condition (C, arrows, scale bar 20 µm). In RPM-AD cells, stress fibers are slightly MCF-7 cultured in the on-ground condition (C, arrows, scale bar 20 µm). In RPM-AD cells, stress fibers are slightly reduced, reduced, while F-actin was mostly localized at the cell border (D). On the contrary, in RPM-CLUMPS cells the actin fila- mentswhile F-actinare fragmented was mostly and localizedsignificantly at the reduced cell border (E), mostly (D). On positioned the contrary, at the in perinuclear RPM-CLUMPS level cells of floating the actin cells filaments €. are fragmented and significantly reduced (E), mostly positioned at the perinuclear level of floating cells €. 2.5. Microtubules 2.5. MicrotubulesIn both MCF-7 and MC10A cells exposed to weightlessness, we recorded the disap- pearanceIn both of polarized MCF-7 and fluorescence MC10A cells associat exposeded with to weightlessness, the microtubule-organizing we recorded thecenter. dis- However,appearance in ofadherent polarized cells—in fluorescence MCF-7 associatedas well as in with MCF10A—microtubules the microtubule-organizing are still center.iden- tifiableHowever, as filaments in adherent that cells—in unusually MCF-7 proceed as well from as in the MCF10A—microtubules perinuclear space toward are still the iden- cell membrane.tifiable as filaments Instead, in that RPM-CLUMP unusually proceed cells, th frome tubulin the perinuclearnetwork shows space a dramatic toward the disor- cell ganization,membrane. and Instead, tubulin in RPM-CLUMPloses its polarization cells, the and tubulin messily network aggregates shows around a dramatic the nucleus disor- (Figureganization, 5A–E). and Similar tubulin findings loses its have polarization already been and messilyreported aggregates for thyroid around cells [19]. the nucleus (Figure5A–E). Similar findings have already been reported for thyroid cells [19].

Int. J. Mol. Sci. 2021, 22, 862 7 of 26 Int. J. Mol. Sci. 2021, 22, x FOR PEER REVIEW 7 of 26

Figure 5. ImmunofluorescenceImmunoﬂuorescence microscopy microscopy images images of of alpha-tubulin alpha-tubulin (in (in green) green) distribution distribution pattern pattern in MCF-10A ( (A,,B)) and and MCF-7 ( C––E)) cultured in normal gravitational (OG) and microgravity (RPM) conditions for for 24 h. Nuclei Nuclei (in (in blue) were stainedstained with TO-PRO-3 iodide. Scale bar 20 µm.µm.

2.6. Intermediate Filaments: Cytokeratins 2.6. IntermediateCytokeratins Filaments: (CKs)—composed Cytokeratins of intermediate filaments (IF)—are the third most relevantCytokeratins constituent (CKs)—composed of the cytoskeleton. of intermediateUsually, the filamentscytokeratin (IF)—are network the surrounding third most therelevant nucleus constituent displays ofa “net the cytoskeleton. architecture”, Usually, resembling the cytokeratin a crate in which network meshes surrounding are regular the innucleus size and displays depicting a “net an architecture”, “alveolar-like” resembling structure a in crate both in whichMCF10A meshes and areMCF-7 regular cells in [20]. size Instead,and depicting MCF-7 an cells “alveolar-like” exposed to structureweightlessn in bothess exhibit MCF10A a loosely and MCF-7 organized cells[ 20perinuclear]. Instead, cytokeratinMCF-7 cells network, exposed toas weightlessnesspreviously reported exhibit [21]. a loosely Yet, while organized changes perinuclear were minimally cytokeratin un- detectablenetwork, asin previouslyMCF10A (data reported not shown), [21]. Yet, modifications while changes in were the intermediate minimally undetectable filament dis- in tributionMCF10A were (data remarkably not shown), evident modifications in MCF-7 in cells, the intermediate especially in filament the RPM-CLUMP distribution cluster were (Figureremarkably 6A,B). evident Given inthat MCF-7 the intermediate cells, especially filament in the organization RPM-CLUMP is instrumental cluster (Figure in6 mod-A,B). ulatingGiven that the theeffects intermediate of external filament physical organization stresses [22]—because is instrumental the incell modulating responds to the change effects inof the external mechanical physical environment stresses [22 ]—becauseby fine tuning the the cell aggregation responds to of change IFs from in the concentration mechanical fieldsenvironment of the soluble by fine pool—it tuning theis tempting aggregation to speculate of IFs from that concentration microgravity-dependent fields of the solublereduc- tionpool—it of tensile is tempting forces exerted to speculate by the that milieu microgravity-dependent upon cells could have reductiona significant of tensileinfluence forces on differentexerted bybiological the milieu functions,upon cells including could have apop a significanttosis. Indeed, influence loosening on differentof the cytokeratin biological networkfunctions, and including degradation apoptosis. of cytokeratin Indeed, loosening8/18 have been of the associated cytokeratin with network the early and steps degra- of apoptosisdation of cytokeratinin both normal 8/18 [23] have and been cancer associated cells [24]. with It is the conceivable early steps that of apoptosis disorganization in both normal [23] and cancer cells [24]. It is conceivable that disorganization of the cytokeratin of the cytokeratin meshwork as well as the reduced cytokeratin content—as suggested by meshwork as well as the reduced cytokeratin content—as suggested by the reduced CK the reduced CK density observed in confocal analysis—may contribute to apoptosis in density observed in confocal analysis—may contribute to apoptosis in microgravity. microgravity.

Int. J. Mol. Sci. 2021, 22, 862 8 of 26 Int. J. Mol. Sci. 2021, 22, x FOR PEER REVIEW 8 of 26 Int. J. Mol. Sci. 2021, 22, x FOR PEER REVIEW 8 of 26

FigureFigure 6. 6. ConfocalConfocal imagesimages ofof cytokeratin cytokeratin distributiondistribution patternpattern inin MCF-7MCF-7 culturedcultured inin normal normal gravitational gravitational (OG)(OG) ( (AA)) and and microgravitymicrogravity (RPM-CLUMPS)(RPM-CLUMPS) (( BB))) conditionsconditions forfor for 2424 h.h. Scale Scale barbar 1010 µm.µµm.m. ●●• DesmosomesDesmosomes plaques plaques at at the the cell–cell cell–cell junction junction areas; ▀ perinuclear regions surrounding cell nuclei; meshes of the cytokeratin network in the perinuclear regions are areas; ▀ perinuclearperinuclear regions regions surroundi surroundingng cell cell nuclei; nuclei; meshes meshes ofof thethe cytokeratincytokeratin network in the perinuclear regions are looser in microgravity than in 1 g control; ▲ plaques reduced both in number and in network density. Within the yellow looser in microgravity than than in in 1 1 g g control; control; ▲ plaquesplaques reducedreduced both in number and in network density. Within the yellow squares, the distribution of cytokeratins hasN been magnified. Red arrows indicate the relevant structures under scrutiny. squares, the distribution of cytokeratins hashas beenbeen magniﬁed.magnified. RedRed arrowsarrows indicateindicate thethe relevantrelevant structuresstructures underunder scrutiny.scrutiny.

2.7.2.7. VinculinVinculin 2.7. Vinculin VinculinVinculin isis aa cytoskeletalcytoskeletal proteinprotein associatedassociated withwith cell-matrixcell-matrix andand cell–cellcell–cell junctionsjunctions requiredrequiredVinculin forfor anchoringanchoring is a cytoskeletal F-actinF-actin protein toto thethe membrane.membrane. associated InwithIn thisthis cell-matrix way,way, vinculinvinculin and cell–cell isis indispensableindispensable junctions forfor re- thethequired mechanicalmechanical for anchoring stabilizationstabilization F-actin oftoof thecellcell membrane. morphologymorphology In [25].[25]. this way,VinculinVinculin vinculin distributiondistribution is indispensable isis severelyseverely for af-theaf- fectedfectedmechanical byby simulatedsimulated stabilization microgravimicrogravi of cell morphologyty,ty, especiallyespecially [ 25 inin]. thethe Vinculin MCF-MCF-77 distribution RPM-CLUMPRPM-CLUMP is severely clustercluster affected(Figure(Figure 7A–F).7A–F).by simulated InIn MCF-7MCF-7 microgravity, RPMRPM adherentadherent especially cells,cells, invinculinvinculin the MCF-7 waswas RPM-CLUMPchieflychiefly diminisheddiminished cluster inin(Figure proximityproximity7A–F). ofof fo-fo-In calcalMCF-7 adhesionadhesion RPM (FA),(FA), adherent whilewhile cells, beingbeing vinculin stillstill recognizablerecognizable was chiefly withindiminishedwithin thethe cytosolcytosol in proximity (Figure(Figure of 7B).7B). focal Instead,Instead, adhesion inin floatingfloating(FA), while MCF-7MCF-7 being cells,cells, still vinculinvinculin recognizable waswas dramaticallydramatically within the cytosol reducedreduced (Figure andand 7disappearsdisappearsB). Instead, almostalmost in floating com-com- pletelypletelyMCF-7 from from cells, the the vinculin cytosol cytosol was (Figure (Figure dramatically 7C). 7C). Western Western reduced blot blot assay andassay disappears showed showed that that almost vinculin vinculin completely was was reduced reduced from ininthe bothboth cytosol MCF10AMCF10A (Figure andand7C). MCF-7MCF-7 Western cellscells blot (Figure(Figure assay 7G,H)7G,H) showed andand that thisthis vinculin reductionreduction was isis associatedassociated reduced in withwith both aa remarkableremarkableMCF10A and decreasedecrease MCF-7 in in cells F-actin F-actin (Figure andand7 G,H) stressstress and fibefibe thisrs,rs, togethertogether reduction withwith is associated thethe disappearancedisappearance with a remarkable of of lamel- lamel- lipodialipodiadecrease andand in pseudopodia F-actinpseudopodia and stress formation.formation. fibers, togetherTheseThese changeschanges with the areare disappearance likelylikely instrumentalinstrumental of lamellipodia inin hamperinghampering and cells’cells’pseudopodia adhesionadhesion formation. toto thethe substrate.substrate. These In changesIn turn,turn, reductionreduction are likely in instrumentalin cellcell adhesivenessadhesiveness in hampering impairsimpairs cells’ bothboth adhe-FAKFAK sion to the substrate. In turn, reduction in cell adhesiveness impairs both FAK activation activationactivation andand actinactin polymerization,polymerization, therebythereby hinderinghindering migration.migration. Concurrently,Concurrently, thosethose and actin polymerization, thereby hindering migration. Concurrently, those factors favor factorsfactors favorfavor cellcell detachmentdetachment andand thethe subsequentsubsequent formationformation ofof floatingfloating organoid-likeorganoid-like cell detachment and the subsequent formation of floating organoid-like structures [26]. structuresstructures [26].[26].

Int. J. Mol. Sci. 2021, 22, 862 9 of 26 Int. J. Mol. Sci. 2021, 22, x FOR PEER REVIEW 9 of 26

FigureFigure 7. Confocal 7. Confocal images images of vinculinof vinculin (green) (green) and and F-actinF-actin (red) distribution distribution in inMCF-7 MCF-7 cultured cultured in normal in normal gravitational gravitational conditions (A,D) and microgravity conditions (B,E,C,F) for 24 h. Vinculin distribution is localized nearby focal adhesion conditions(FA) in (A MCF-7,D) and cultured microgravity in the on-ground conditions condition (B,E,C (A,F–)D for, arrows). 24 h. Vinculin In MCF-7 distributionRPM adherent is cells, localized vinculin nearby was recogniza- focal adhesion (FA) inble MCF-7 within cultured the cytosol in theand on-grounddiminished conditionin proximity (A –plasmaD, arrows). membrane In MCF-7 (B,E). RPMIn MCF-7 adherent RPM-CLUMP cells, vinculin cells, vinculin was recognizable was withindramatically the cytosol reduced and diminished and disappeared in proximity almost comp plasmaletely from membrane the cytosol (B, Eand). In plasma MCF-7 membrane RPM-CLUMP (C,F). Scale cells, bars vinculin 20 µm. was dramaticallyImmunoblot reduced bar chart and disappeared(G,H) showing almost the expression completely of vinculin fromthe in MCF-10A cytosol and (G) plasmaat 24 h (in membrane correlation ( Cwith,F). the Scale addition bars20 µm. of p-ERK inhibitor PD89059) and in MCF-7 (H) in on-ground control cells, RPM adherent cells, and RPM cell clumps at 24 Immunoblot bar chart (G,H) showing the expression of vinculin in MCF-10A (G) at 24 h (in correlation with the addition of and 72 h. Data show that in MCF10A in microgravity, inhibition of phosphorylated ERK (pERK) leads to an increase in p-ERKvinculin, inhibitor while PD89059) no effects and are in recorded MCF-7 (inH 1) ing conditions. on-ground In control MCF-7 cells, vinculin RPM adherent is severely cells, reduced and RPMonly in cell RPM-CLUMP clumps at 24 and 72 h. Datacells. showColumns that and in MCF10Abars represent in microgravity, densitometric inhibition quantification of phosphorylated of the optical density ERK (pERK) (OD) of leads a specific to an increaseprotein signal in vinculin, whilenormalized no effects arewith recorded the OD values in 1 gof conditions. the GAPDH Inserving MCF-7 as the cells, loading vinculin control. is severely Each column reduced represents only inthe RPM-CLUMP mean value ± cells. SD of three independent experiments. In MCF-10A, ## p < 0.01 vs. RPM by ANOVA followed by Bonferroni post-test. In Columns and bars represent densitometric quantiﬁcation of the optical density (OD) of a speciﬁc protein signal normalized MCF-7, *** p < 0.001 vs. OG; ### p< 0.001 vs. RPM-AD by ANOVA followed by Bonferroni post-test. with the OD values of the GAPDH serving as the loading control. Each column represents the mean value ± SD of three independent experiments. In MCF-10A, ## p < 0.01 vs. RPM by ANOVA followed by Bonferroni post-test. In MCF-7, *** p < 0.001 vs. OG; ### p< 0.001 vs. RPM-AD by ANOVA followed by Bonferroni post-test.

Int. J. Mol. Sci. 2021, 22, x FOR PEER REVIEW 10 of 26 Int. J. Mol. Sci. 2021, 22, 862 10 of 26

It is notable that ERK activity and vinculin levels and the distribution behind the cell membraneIt is notable are inversely that ERK correlated, activity and i.e., vinculin a reduction levels in vinculin and the distributionappears to be behind instrumental the cell membranein enhancing are ERK inversely activation correlated, [27]. Indeed, i.e., a reduction ERK was inunaffected vinculin appearsin adherent to be MCF-7 instrumental cells— in enhancingwhich no significant ERK activation changes [27]. in Indeed, vinculin ERK were was recorded. unaffected However, in adherent ERK MCF-7 activation cells—in was whichunexpectedly no significant reduced changes in floating in vinculin MCF-7 were cells, recorded. the dramatic However, decrease ERK in activation vinculin was notwith- unex- pectedlystanding reduced(Figure in7G,H). floating Given MCF-7 that cells, ERK the activa dramatiction is decrease tightly incoupled vinculin with notwithstanding vinculin, we (Figureshould 7haveG,H). expected Given that a “compensatory” ERK activation isincrease tightly in coupled ERK activation with vinculin, following we should the vinculin have expecteddecrease. aThe “compensatory” lack of ERK activation increase in in ERK the activationpresence of following the decreased the vinculin vinculin decrease. values Thecan lackbe explained of ERK activation by the fact in thethat presence ERK activation of the decreased requires vinculinthe integrity values of canFAK–integrin be explained com- by theplexes, fact thatwhich ERK instead activation are severely requires theimpaired integrity in of RPM-CLUMP FAK–integrin complexes,cells (see above). which insteadConse- arequently, severely in RPM-CLUMP impaired in RPM-CLUMP cells, the almost cells (seecomplete above). loss Consequently, of adhesion in inhibits RPM-CLUMP the increase cells, thein ERK almost activity complete that, loss in principle, of adhesion should inhibits had the been increase triggered in ERK by activity the concomitant that, in principle, disap- shouldpearance had of beenvinculin. triggered by the concomitant disappearance of vinculin. 2.8. Integrins 2.8. Integrins Integrins play a key role in adhesion processes and, conversely, in anoikis, i.e., the Integrins play a key role in adhesion processes and, conversely, in anoikis, i.e., the programmed cell death occurring because of detachment from the substrate [28]. FAK programmed cell death occurring because of detachment from the substrate [28]. FAK becomes activated by autophosphorylation upon integrin-mediated cell attachment. Loss becomes activated by autophosphorylation upon integrin-mediated cell attachment. Loss of integrin and/or vinculin may hamper their function. The resulting FAK–integrin com- of integrin and/or vinculin may hamper their function. The resulting FAK–integrin com- plex interacts with both ERK and Akt in modulating anti-apoptotic and pro-survival plex interacts with both ERK and Akt in modulating anti-apoptotic and pro-survival path- pathways [29]. When compared to cells growing in normal gravity, immunostaining of ways [29]. When compared to cells growing in normal gravity, immunostaining of MCF- MCF-7 adherent cells (RPM-AD) shows an increased deposition of β1-integrin on the cell membrane7 adherent cells (Figure (RPM-AD)8A,B). On shows the contrary, an increased in the deposition RPM-CLUMP of β1-integrin cluster, onβ1-integrin the cell mem- was almostbrane (Figure completely 8A,B). reduced On the (Figure contrary,8C), in as the previously RPM-CLUMP observed cluster, in thyroid β1-integrin cells growingwas almost in simulatedcompletely weightlessness reduced (Figure and 8C), forming as previously organoid-like observed structures in thyroid [30]. cells growing in simulated weightlessness and forming organoid-like structures [30].

Figure 8. Confocal images of β1-integrin (in green) distribution in MCF-7 cultured in normal gravity (A) and microgravity conditions (B,C) for 24 h. β1-integrin in MCF-7 cells appears localized on the cell membrane both in the on-ground condi- conditions (B,C) for 24 h. β1-integrin in MCF-7 cells appears localized on the cell membrane both in the on-ground condition tion and in RPM-cultured cells. β1-integrin shows an increased deposition in RPM-AD (B) when compared to control cells and in RPM-cultured cells. β1-integrin shows an increased deposition in RPM-AD (B) when compared to control cells cultured on ground. In the RPM-CLUMP cluster, β1-integrin is almost completely absent (C). Nuclei (in blue) were stained β C culturedwith TO-PRO-3 on ground. iodide. In the Scale RPM-CLUMP bars 20 µm. cluster, 1-integrin is almost completely absent ( ). Nuclei (in blue) were stained with TO-PRO-3 iodide. Scale bars 20 µm. 2.9. Apoptosis 2.9. Apoptosis A mild, albeit slight increase in apoptosis rate was recorded in MCF10A cells exposed to simulatedA mild, albeitmicrogravity slight increase at both in 24 apoptosis and 72 h rate (Figure was recorded 9A). Surprisingly, in MCF10A apoptosis cells exposed was tohardly simulated noticed microgravity in cancerous at MCF-7 both 24cells and at 7224 h (Figure 99B).A). However, Surprisingly, at 72 apoptosis h, even MCF- was hardly7 cells showed noticed ina significant cancerous increase MCF-7 cellsin apoptosis, at 24 h (Figure but this9B). effect However, was limited at 72 h,to evenRPM- MCF-7CLUMP cells cells. showed This somewhat a significant puzzling increase behavior in apoptosis, indicates but thisthat effectbreast was cancer limited cells to can RPM- ef- CLUMPficiently cope cells. with This the somewhat microgravity puzzling stress, behavior at leastindicates in the short that term. breast However, cancer cells at 72 can h, efficientlyapoptosis rate cope increases with the steadily, microgravity but only stress, in the at leastRPM-CLUMP in the short cluster. term. This However, finding at is 72 not h, apoptosissurprising, rate as it increases is well-known steadily, that but programmed only in the RPM-CLUMP cell death can cluster. be successfully This finding induced is not surprising, as it is well-known that programmed cell death can be successfully induced by

Int. J. Mol. Sci. 2021, 22, 862 11 of 26 Int. J. Mol. Sci. 2021, 22, x FOR PEER REVIEW 11 of 26

disruptionby disruption of the of the interactions interactions between between normal norm epithelialal epithelial cells cells and and the the extracellular extracellular matrix, ma- despitetrix, despite the fact the that fact this that anchorage this anchorage dependence dependence is reduced is reduced in malignantly in malignantly transformed trans- cells [31]. These observations seem to suggest that some kind of anti-apoptotic processes formed cells [31]. These observations seem to suggest that some kind of anti-apoptotic are enacted in cancerous cells exposed to stressful physical environments, and, at least processes are enacted in cancerous cells exposed to stressful physical environments, and, in the short term, these survival adaptive mechanisms are successful in counteracting at least in the short term, these survival adaptive mechanisms are successful in counter- cell-death-related processes. acting cell-death-related processes.

FigureFigure 9.9. ApoptosisApoptosis analysisanalysis inin MCF-10AMCF-10A ((AA)) andand MCF-7MCF-7 ((BB)) cellscells atat 2424 andand 7272 h h after after exposition exposition to to weightlessness. weightlessness. BarBar charts/graphs show the percentage of apoptotic cells (Annexin V+/7-AAD-); each column represents the mean value ± SD charts/graphs show the percentage of apoptotic cells (Annexin V+/7-AAD-); each column represents the mean value ± SD of three independent experiments. ** p < 0.01 vs. OG by ANOVA followed by Bonferroni post-test. OG—on ground. of three independent experiments. ** p < 0.01 vs. OG by ANOVA followed by Bonferroni post-test. OG—on ground. 2.10. Survival Pathways: Cyclin D1 2.10. Survival Pathways: Cyclin D1 Undoubtedly, cells cultured in simulated microgravity underwent a dramatic re- modelingUndoubtedly, of the cytoskeleton cells cultured architecture, in simulated involving microgravity almost underwent all CSK acomponents. dramatic remodel- These ingprocesses of the cytoskeletonpromote an outstanding architecture, phenotypic involving almost transition: all CSK cells components. adopt a new These morphological processes promoteshape and an establish outstanding entirely phenotypic new connections transition: between cells adopt them, a new as morphologicalshown by the emergence shape and establish entirely new connections between them, as shown by the emergence of organoid- of organoid-like structures detached from the substrate. In this context, it is not surprising like structures detached from the substrate. In this context, it is not surprising that the that the apoptosis rate increases because of the relevant changes involving the CSK. There- apoptosis rate increases because of the relevant changes involving the CSK. Therefore, the fore, the right question is not “why do cells in RPM undergo programmed cell death?”, right question is not “why do cells in RPM undergo programmed cell death?”, but “why is but “why is the apoptosis rate ultimately so limited?”. There are indeed several indices the apoptosis rate ultimately so limited?”. There are indeed several indices suggesting that suggesting that cells in microgravity adopt a “survival strategy” by activating some spe- cells in microgravity adopt a “survival strategy” by activating some specific biochemical cific biochemical cascades. Indeed, cyclin D1—a molecule amplified in the G1 phase that cascades. Indeed, cyclin D1—a molecule amplified in the G1 phase that coordinates the coordinates the entry, not the S-phase and the subsequent DNA synthesis—is highly ex- entry, not the S-phase and the subsequent DNA synthesis—is highly expressed in MCF10A pressed in MCF10A exposed to weightlessness and is significantly downregulated in exposed to weightlessness and is significantly downregulated in MCF-7 RPM-CLUMP MCF-7 RPM-CLUMP cells (Figure 10A,B). It is notable that cyclin D1 levels correlate with cells (Figure 10A,B). It is notable that cyclin D1 levels correlate with the trend we observed inthe the trend proliferation we observed curve in (Figurethe proliferation3A,B), thus curve supporting (Figure the3A,B), hypothesis thus supporting of a “survival the hy- program”pothesis of triggered a “survival by microgravity.program” triggered To obtain by microgravity. insights into theseTo obtain processes, insights we into decided these toprocesses, investigate we ERKdecided and to Akt investigate pathways, ERK which and are Akt critically pathways, involved which inare triggering critically involved survival andin triggering apoptosis survival resilience. and apoptosis resilience.

Int. J. Mol. Sci. 2021, 22, 862 12 of 26 Int. J. Mol. Sci. 2021, 22, x FOR PEER REVIEW 12 of 26

FigureFigure 10.10. ImmunoblotImmunoblot barbar chartchart showingshowing the the expressionexpression of of cyclincyclin D1D1 inin MCF-10AMCF-10A ((AA)) andand MCF-7MCF-7 ((BB)) inin on-groundon-ground controlcontrol cells,cells, RPM RPM adherent adherent cells, cells, and and RPM RPM cell cell clumps clumps at at 24 24 and and 72 72 h. h. Columns Columns and and bars bars represent represent densitometric densitometric quantification quantification of of the optical density (OD) of a specific protein signal normalized with the OD values of the GAPDH serving as the loading the optical density (OD) of a specific protein signal normalized with the OD values of the GAPDH serving as the loading control. Each column represents the mean value ± SD of three independent experiments. In MCF-10A, * p < 0.05 vs. OG by control. Each column represents the mean value ± SD of three independent experiments. In MCF-10A, * p < 0.05 vs. OG by t test; in MCF-7, * p < 0.05, ** p < 0.01 vs. OG by ANOVA followed by Bonferroni post-test. t test; in MCF-7, * p < 0.05, ** p < 0.01 vs. OG by ANOVA followed by Bonferroni post-test. 2.11. ERK and Akt 2.11. ERKWe investigated and Akt the levels of activated ERK (phosphorylated ERK, pERK) and we foundWe that investigated pERK increases the levels significantly of activated at both ERK 24 (phosphorylatedand 72 h in MCF10A ERK, cells pERK) (Figure and 11A). we foundOn the that contrary, pERK pERK increases was significantly reduced in MCF-7 at both ce 24lls. and Conversely, 72 h in MCF10A activated cells Akt (Figure (pAkt) 11 doesA). Onnot thechange contrary, in MCF10A, pERK waswhile reduced it increases in MCF-7 in MCF-7 cells. cells Conversely, exposed to activated microgravity, Akt (pAkt) espe- doescially not in the change RPM-CLUMP in MCF10A, clus whileter of itcells increases (Figure in 11B). MCF-7 cells exposed to microgravity, especiallyOverall, in the these RPM-CLUMP data seem clusterto suggest of cells that (Figure weightlessness 11B). triggers a pro-survival re- sponseOverall, in both these cell datalines seemby activating to suggest differ thatent weightlessness pathways, i.e., triggers those related a pro-survival to pAkt and re- sponsepERK in in MCF-7 both cell and lines MCF10A by activating cells, respectively. different pathways, To assess i.e., if the those relative related increase to pAkt in andboth pERKfactors in could MCF-7 explain and MCF10A how cells cells, in respectively.microgravity To modulate assess if thethe relativeapoptotic increase response, in both we factorstreated couldMCF10A explain and MCF-7 how cells cells in with microgravity specific inhibitors modulate of pERK the apoptotic and pAkt, response, respectively. we treatedApoptosis MCF10A increased and MCF-7up to two- cells and with three-fold specific inhibitorsin 1g and ofmicrogravity pERK and pAkt, exposed respectively. MCF10A Apoptosiscells (Figure increased 12A), respectively, up to two- andupon three-fold addition of in PD89059, 1g and microgravity a pERK inhibitor. exposed Conversely, MCF10A cellsaddition (Figure of the12A), pAkt respectively, inhibitor uponto MCF-7 addition exposed of PD89059, to weightlessness a pERK inhibitor. increased Conversely, apoptosis additionup to three-fold of the pAkt in both inhibitor RPM-AD to MCF-7 and RPM-CLUMP exposed to weightlessness cells (Figure 12B). increased apoptosis up to three-foldOverall, in these both findings RPM-AD demonstrated and RPM-CLUMP that enhancement cells (Figure 12of B).ERK and Akt activity ef- ficientlyOverall, damped these the findings apoptotic demonstrated response seco thatndary enhancement to microgravity of ERK exposure, and Akt while activity the efficientlyinhibition dampedof both pathways the apoptotic significantly response increased secondary apoptosis. to microgravity The two exposure, pathways while are thenot inhibitioncross-linked of bothin MCF-7 pathways cells, significantlygiven that the increased pAkt inhibitor apoptosis. does The not two modify pathways pERK are levels not cross-linked(Figure 13A). in Instead, MCF-7 in cells, MCF10A, given thatthe inhibi the pAkttion inhibitorof pERK doesactivity not induces modify a pERK “compensa- levels (Figuretory” increase 13A). Instead, in pAkt in MCF10A, activity the(Figure inhibition 13B). ofThese pERK findings activity inducesshow that a “compensatory” noncancerous increaseMCF10A in cells pAkt may activity paradoxically (Figure 13B). choose These betw findingseen showdifferent that survival noncancerous alternatives MCF10A by cellsswitching may paradoxically from one pathway choose (pERK) between to differentanother survivalone (pAkt), alternatives while MCF-7 by switching cells (can) from ex- oneclusively pathway activate (pERK) Akt to anotherto counteract one (pAkt), the mi whilecrogravity-related MCF-7 cells (can) pro-apoptotic exclusively activate effects. AktWe tohave counteract previously the microgravity-relatedobserved a similar pattern pro-apoptotic in breast effects. cancer We cell have line previously MDA-MB-231 observed ex- aposed similar to patternweightlessness, in breast in cancer which cell activated line MDA-MB-231 Akt and survivin exposed both to increased weightlessness, at 72 h inin whichadherent activated cells [10]. Akt and survivin both increased at 72 h in adherent cells [10].

Int. J. Mol. Sci. 2021, 22, 862 13 of 26 Int. J. Mol. Sci. 2021, 22, x FOR PEER REVIEW 13 of 26

FigureFigure 11. 11.ImmunoblotImmunoblot bar bar chart chart showing showing the the expression expression of of the the phosphorylated AKT AKT (p-AKT)/AKT (p-AKT)/AKT ratio ratio and and p-ERK/ERK p-ERK/ERK ratioratio in MCF-10A in MCF-10A (A, (BA), Band) and MCF-7 MCF-7 (C (,CD,D) in) in on-ground on-ground control control cells, cells, RPM adherentadherent cells,cells, and and RPM RPM cell cell clumps clumps at at 24 24 and and 72 h.72 Columns h. Columns and and bars bars represent represent densitom densitometricetric quantification quantiﬁcation of of the the optical density (OD)(OD) ofof a a speciﬁc specific protein protein signal signal normalizednormalized with with the the OD OD values values of of the the GAPDH servingserving as as the the loading loading control. control. Each Ea columnch column represents represents the mean the mean value ±valueSD ± SD of three independent experiments. In MCF-10A, *** p < 0.001 vs. OG by t test; in MCF-7, * p < 0.05, ** p < 0.01 vs. OG by of three independent experiments. In MCF-10A, *** p < 0.001 vs. OG by t test; in MCF-7, * p < 0.05, ** p < 0.01 vs. OG by ANOVA followed by Bonferroni post-test. ANOVA followed by Bonferroni post-test.

Int. J. Mol. Sci. 2021, 22, 862 14 of 26

Int. J. Mol. Sci. 2021, 22, x FOR PEER REVIEW 14 of 26

FigureFigure 12. Apoptosis 12. Apoptosis analysis analysis in in MCF-10A MCF-10A (A(A)) andand MCF-7 (B (B) )cells cells at at24 24h after h after exposition exposition to weightlessness to weightlessness w/o the w/op- the ERK inhibitor, PD89059, and w/o the p-AKT inhibitor, LY294002, respectively. Bar charts/graphs show the percentage of p-ERKapoptotic inhibitor, cells PD89059, (Annexin and V+/7-AAD-); w/o the p-AKT each column inhibitor, represents LY294002, the mean respectively. value ± SD Bar of charts/graphsthree independent show experiments. the percentage In of apoptoticMCF-10A,Figure cells 12. (Annexin * pApoptosis < 0.05, V+/7-AAD-);** panalysis < 0.01 vs. in OG;MCF-10A each # p column < 0.05 (A) vs.and represents RPM MCF-7 by ANOVA(B the) cells mean at followed 24 value h after ±Bonferroni expositionSD of three post-test. to independentweightlessness In MCF-7 experiments. *w/o p < 0.05,the p- In ERK inhibitor, PD89059, and w/o the p-AKT inhibitor, LY294002, respectively. Bar charts/graphs show the percentage of MCF-10A,** p < *0.01p < vs. 0.05, OG; ** §p p << 0.010.05 vs. vs. RPM-AD; OG; # p <# 0.05p < 0.05 vs. vs. RPM RPM-CLUMP by ANOVA by followedANOVA followed Bonferroni by Bonferroni post-test. post-test. In MCF-7 * p < 0.05, apoptotic cells (Annexin V+/7-AAD-); each column represents the mean value ± SD of three independent experiments. In ** p < 0.01MCF-10A, vs. OG; * §p p< <0.05, 0.05 ** vs.p < RPM-AD;0.01 vs. OG; # #p p< < 0.050.05 vs. RPM RPM-CLUMP by ANOVA by followed ANOVA Bonferroni followed post-test. by Bonferroni In MCF-7 post-test. * p < 0.05, ** p < 0.01 vs. OG; § p < 0.05 vs. RPM-AD; # p < 0.05 vs. RPM-CLUMP by ANOVA followed by Bonferroni post-test.

Figure 13. Immunoblot bar chart showing the expression of the p-AKT/AKT ratio and p-ERK/ERK ratio in MCF-10A (A) and MCF-7 (B) in on-ground control cells, RPM adherent cells, and RPM cell clumps at 24 h w/o the p-ERK inhibitor. PD89059, and w/o the p-AKT inhibitor, LY294002, respectively. Columns and bars represent densitometric quantification Figureof 13.Figure theImmunoblot optical 13. Immunoblot density bar (OD) chartbar of chart the showing p-AKT showing and the the expressionp-ERK expression signal ofof normalized thethe p-AKT/AKT p-AKT/AKT with the ratio OD ratio and values andp-ERK/ERK p-ERK/ERKof AKT andratio ERK in ratio MCF-10A serving in MCF-10A as(A ) and MCF-7 (B) in on-ground control cells, RPM adherent cells, and RPM cell clumps at 24 h w/o the p-ERK inhibitor. (A) andthe MCF-7 loading ( Bcontrol.) in on-ground Each column control represents cells, RPMthe me adherentan value ± cells, SD of and three RPM independent cell clumps experiments. at 24 h w/o * p < the0.05, p-ERK ** p < 0.01 inhibitor. PD89059, and w/o the p-AKT inhibitor, LY294002, respectively. Columns and bars represent densitometric quantification vs. OG; ## p < 0.01 vs. RPM by ANOVA followed by Bonferroni post-test. PD89059,of andthe optical w/o thedensity p-AKT (OD) inhibitor, of the p-AKT LY294002, and p-ERK respectively. signal normalized Columns with and the bars OD represent values of densitometric AKT and ERK quantificationserving as of the opticalthe loading density control. (OD) Each of column the p-AKT represents and p-ERKthe mean signal value normalized± SD of three withindependent the OD experiments. values ofAKT * p < 0.05, and ** ERK p < 0.01 serving These findings were further confirmed when effectors of apoptosis was considered. vs. OG; ## p < 0.01 vs. RPM by ANOVA followed by Bonferroni post-test.± as the loading control. Each columnThe enzyme represents poly(ADP-ribose) the mean value polymeraseSD of (PARP), three independent whose expression experiments. is triggered * p < 0.05,by ** p < 0.01 vs. OG; ## p < 0.01 vs. RPM by ANOVA followed by Bonferroni post-test. DNA strandThese findingsbreaks, represents were further a main confirmed substrat whene for theeffectors activity of ofapoptosis caspases, was which considered. inactivate PARP through selective cleavage, releasing cleaved PARP (Cl-PARP). In fact, Cl- TheThese enzyme findings poly(ADP-ribose) were further polymerase confirmed when(PARP), effectors whose expression of apoptosis is triggered was considered. by TheDNA enzyme strand poly(ADP-ribose) breaks, represents polymerase a main substrat (PARP),e for whosethe activity expression of caspases, is triggered which inacti- by DNA vate PARP through selective cleavage, releasing cleaved PARP (Cl-PARP). In fact, Cl- strand breaks, represents a main substrate for the activity of caspases, which inactivate PARP through selective cleavage, releasing cleaved PARP (Cl-PARP). In fact, Cl-PARP

Int. J. Mol. Sci. 2021, 22, 862 15 of 26

Int. J. Mol. Sci. 2021, 22, x FOR PEER REVIEW 15 of 26

levels are usually considered a valuable marker for assessing apoptosis [32]. As expected, Cl-PARPPARP levels levels are were usually only considered slightly increased a valuable in marker both MCF10Afor assessing and apoptosis MCF-7 cells, [32]. whereasAs weexpected, observed Cl-PARP a four-fold levels increase were only in slightly Cl-PARP increased values in after both 72 MCF10A h of weightlessness and MCF-7 cells, in RPM- CLUMPwhereas MCF-7 we observed cells, mirroring a four-fold the increase apoptosis in Cl-PARP values estimatedvalues after in 72 those h of weightlessness cells (Figure 14 A,B). Byin adding RPM-CLUMP pERK andMCF-7 pAkt cells, inhibitors mirroring tothe MCF10A apoptosis andvalues MCF-7 estimated cells, in those respectively, cells (Fig- we ob- servedure 14A,B). that Cl-PARP By adding increased pERK and signiﬁcantly pAkt inhibitors in both to cases, MCF10A conﬁrming and MCF-7 that cells, activation respec- of ERK tively, we observed that Cl-PARP increased significantly in both cases, confirming that and Akt pathways markedly decreases PARP cleavage enacted by microgravity exposure activation of ERK and Akt pathways markedly decreases PARP cleavage enacted by mi- (Figure 14C,D). crogravity exposure (Figure 14C,D).

Figure 14. Immunoblot bar chart showing the expression of cleaved poly(ADP-ribose) polymerase (cl-PARP) in MCF-10A Figure 14. Immunoblot bar chart showing the expression of cleaved poly(ADP-ribose) polymerase (cl-PARP) in MCF-10A (A) and MCF-7 (B) in on-ground control cells, RPM adherent cells, and RPM cell clumps at 24 and 72 h. Columns and bars (A) andrepresent MCF-7 densitometric (B) in on-ground quantification control of cells, the RPMoptical adherent density (OD) cells, of and a specific RPM protein cell clumps signal at normalized 24 and 72 with h. Columns the OD and bars representvalues of densitometricthe GAPDH serving quantiﬁcation as the loading of the control. optical Each density column (OD) represents of a speciﬁc the mean protein value signal ± SD of normalized three independent with the OD valuesexperiments. of the GAPDH * p

Int. J. Mol. Sci. 2021, 22, x FOR PEER REVIEW 16 of 26

Bonferroni post-test. In MCF-7, ** p < 0.01 vs. OG; ## p < 0.01 vs. RPM-CLUMP by ANOVA followed by Bonferroni post- test.

2.12. Proteins That Regulate Apoptosis: Bad, Bax, Bcl-2 Members of the Bcl-2 family of regulator proteins that regulate cell death, by either Int. J. Mol. Sci. 2021, 22, 862 inhibiting (anti-apoptotic) or inducing (pro-apoptotic) apoptosis, include16 of 26 Bax, Bad and Bcl-2 proteins, to mention just a few [33]. Namely, Bax and Bak are believed to initiate

apoptosis2.12. Proteinsby forming That Regulate a pore Apoptosis: in the Bad,mitochondr Bax, Bcl-2 ial outer membrane that allows cytochrome c to escapeMembers into the of thecytoplasm Bcl-2 family and of regulator activate proteins the pro-apoptotic that regulate cell caspase death, by eitherwhen the Bad/Bcl-2 and/orinhibiting Bax/Bcl-2 (anti-apoptotic) ratios increase or inducing up to (pro-apoptotic) 1 [34]. Their apoptosis, involvement include Bax, allows Bad andus to suggest that the intrinsicBcl-2 proteins, apoptotic to mention pathway just afew is enacted [33]. Namely, duri Baxng andmicrogravity. Bak are believed The to Bad/Bcl-2 initiate ratio is not apoptosis by forming a pore in the mitochondrial outer membrane that allows cytochrome significantlyc to escape modified into the cytoplasm in MCF10A and activate cells the at pro-apoptotic 24 h when caspase the apoptotic when the Bad/Bcl-2rate is “unexpectedly” low. However,and/or Bax/Bcl-2 after ratios the increaseaddition up toof 1 the [34]. pERK Their involvement inhibitor allows PD98059, us to suggest Bad increases that steadily, whilethe Bcl-2 intrinsic remains apoptotic unaffected pathway is (Figure enacted during 15A), microgravity. suggesting The that Bad/Bcl-2 ERK ratioactivity is not is instrumental significantly modified in MCF10A cells at 24 h when the apoptotic rate is “unexpectedly” in downregulatinglow. However, after Bad the additionrather than of the pERKin stimulating inhibitor PD98059, Bcl-2 Bad activity. increases In steadily, MCF-7 cells, Bax is significantlywhile Bcl-2 reduced remains unaffected at 24 h, (Figurethus explaining 15A), suggesting why that no ERK relevant activity is apoptosis instrumental can be found at this stage.in downregulating However, Badat 72 rather h, when than in a stimulating dramatic Bcl-2 increase activity. in In apoptosis MCF-7 cells, is Bax observed is in RPM- significantly reduced at 24 h, thus explaining why no relevant apoptosis can be found CLUMPat this MCF-7 stage. cells, However, a concomitant at 72 h, when arise dramatic in Bax increase and in in the apoptosis Bax/Bcl-2 is observed ratio incan be observed (FigureRPM-CLUMP 15B). It is MCF-7 worth cells, noting a concomitant that the rise an inti-apoptotic Bax and in the Bcl-2 Bax/Bcl-2 protein—when ratio can be not inhibited by Baxobserved or Bad—antagonizes (Figure 15B). It is worth cytochrome noting that the c anti-apoptoticrelease through Bcl-2 protein—when the mitochondrial not pore and inhibited by Bax or Bad—antagonizes cytochrome c release through the mitochondrial pore inhibitsand activation inhibits activation of the of cytoplasmic the cytoplasmic caspase caspase cascade. cascade. This This explains explains why we why did not we did not notice an increasenotice an in increase apoptosis in apoptosis at 24 at 24h, h,when when thethe Bax/Bcl-2 Bax/Bcl-2 ratio ratio is significantly is significantly lower than lower than in controlin controlor in orRPM-AD in RPM-AD cells. cells. Conversely, when when the apoptosis the apoptosis rate increases rate at 72increases h, the at 72 h, the Bax/Bcl-2Bax/Bcl-2 ratio ratio significantly significantly increases.increases.

Figure 15. ImmunoblotFigure 15. Immunoblot bar chart bar showing chart showing the expr the expressionession of of the the Bad/Bcl-2 ratio ratio w/o w/o the the p-ERK p-ERK inhibitor, inhibitor, PD89059, PD89059, in in MCF- 10A (A) andMCF-10A the Bax/Bcl-2 (A) and theratio Bax/Bcl-2 MCF-7 ratio (B MCF-7) in on-ground (B) in on-ground control control cells, cells, RPM RPM adherent adherent cells, cells, and RPM and cell RPM clumps cell at clumps at 24 h. In MCF-10A,24 columns h. In MCF-10A, and columns bars represent and bars represent densitometric densitometric quantification quantiﬁcation of theof opticalthe optica densityl (OD)density of a speciﬁc (OD) protein of a specific protein signal normalized with the OD values of the GAPDH serving as the loading control. In MCF-7, Bax and Bcl-2 were blotted signal normalizedonto the with same the membrane; OD values thus, Bcl-2 of the was GAPDH used as the serving loading control.as the Eachloading column control. represents In MCF-7, the mean valueBax and± SD Bcl-2 of were blotted onto the samethree membrane; independent thus, experiments. Bcl-2 Inwas MCF-10A, used *asp

2.13. Survival Cascades Downstream of ERK/Akt Activation Activation of ERK and Akt-related cascades does not only inhibit apoptosis processes but also enhances survival pathways. The increase in NF-kB and survivin activity plays a pivotal role in favoring cell survival under a variety of chemical and physical stresses [35].

Int. J. Mol. Sci. 2021, 22, 862 17 of 26

Int. J. Mol. Sci. 2021, 22, x FOR PEER REVIEW 17 of 26

2.13. Survival Cascades Downstream of ERK/Akt Activation Activation of ERK and Akt-related cascades does not only inhibit apoptosis processes Therefore,but also enhanceswe investigated survival pathways. NF-kB and The survivin increase in release NF-kB and downstream survivin activity of ERK plays and a Akt acti- vationpivotal in roleMCF10A in favoring and cell MCF-7 survival cells, under respectively. a variety of chemical In MCF10A, and physical NF-kB stresses increases [35]. up to three-foldTherefore, in wemicrogravity, investigated NF-kBwhile, and after survivin the addition release downstream of the pERK of ERK inhibitor, and Akt NF-kB de- activation in MCF10A and MCF-7 cells, respectively. In MCF10A, NF-kB increases up to creases below the control value recorded in MCF10A cells cultured in 1 g (Figure 16A). In three-fold in microgravity, while, after the addition of the pERK inhibitor, NF-kB decreases MCF-7below cells, the controlas no relevant value recorded activity in MCF10A of ERK cells is observed, cultured in 1NF-kBg (Figure is 16slightlyA). In MCF-7 reduced in both RPM-ADcells, as and no relevant more significantly activity of ERK in is observed,RPM-CLUMP NF-kB iscells slightly (Figure reduced 16B). in bothNevertheless, RPM- a fur- therAD prominent and more decrease significantly can in RPM-CLUMPbe recorded cellsif th (Figureese cell 16 clustersB). Nevertheless, are treated a further with the pAkt inhibitor.prominent These decrease findings can be suggest recorded that if these while cell clusters microgravity are treated per with se the strives pAktinhibitor. in limiting NF-kB These findings suggest that while microgravity per se strives in limiting NF-kB activation activationin MCF-7 in cells, MCF-7 the concurrent cells, the increaseconcurrent in pAkt incr activityease efficientlyin pAkt activity antagonizes efficiently such effect. antagonizes suchOverall, effect. theseOverall, data these show thatdata ERK show and that Akt supportERK and NF-kB Akt activationsupport underNF-kB stressful activation under stressfulconditions, conditions, although although the overall the impacts overall reach impacts different reach levels different of effectiveness levels in of the effectiveness two in the twocell lines cell consideredlines considered here. here.

Figure 16.Figure Immunoblot 16. Immunoblot bar chart bar chart showing showing the the ex expressionpression of of NF-kB NF-kB in MCF-10Ain MCF-10A (A) and (A in) and MCF-7 in (MCF-7B) in on-ground (B) in on-ground control control cells, RPMcells, adherent RPM adherent cells, cells,and andRPM RPM cell cell clumps clumps at at 2424 hh w/o w/o the the p-ERK p-ERK inhibitor, inhibitor, PD89059, PD89059, and w/o and the p-AKTw/o the inhibitor, p-AKT inhibitor, LY294002,LY294002, respectively. respectively. Columns Columns and and bars bars repr representesent densitometricdensitometric quantification quantification of the of optical the optical density (OD)density of a specific(OD) of a specific protein signalprotein normalized signal normalized with with the theOD OD values values of of the the GAPDHGAPDH serving serving as the as loadingthe loading control. control. Each column Each represents column therepresents the ± mean valuemean ± valueSD of threeSD of threeindependent independent experiments. experiments. In MCF-10A,MCF-10A, ** p**< p 0.01 < 0.01 vs. OG; vs. ###OG;p < ### 0.001 p vs.< 0.001 RPM byvs. ANOVA RPM by ANOVA followed by Bonferroni post-test. In MCF-7, * p < 0.05 vs. OG by ANOVA followed by Bonferroni post-test. followed by Bonferroni post-test. In MCF-7, * p < 0.05 vs. OG by ANOVA followed by Bonferroni post-test. Similarly, survivin levels were almost unchanged at 24 h in MCF10A cells exposed toSimilarly, microgravity, survivin while upon levels addition were ofalmost the ERK unchanged inhibitor, survivin at 24 h values in MCF10A significantly cells exposed to microgravity,decreased (−60%, while Figure upon 17A). addition In MCF-7 of cells, the at ERK 24 h, survivininhibitor, increased survivin steadily values upon significantly decreasedexposure (− to60%, weightlessness, Figure 17A). while In dramatically MCF-7 cells, decreasing at 24 belowh, survivin the control increased values after steadily upon the addition of the pAkt inhibitor (Figure 17B). In agreement with the observed apoptotic exposurerate, survivin to weightlessness, decreased at 72 h,while mainly dramatically in RPM-CLUMP decreasing MCF-7 cells, below while the no significantcontrol values after the additionchanges were of the observed pAkt at inhibitor this time in(Figure MCF10A 17B). cells In (data agreement not shown). with the observed apoptotic rate, survivin decreased at 72 h, mainly in RPM-CLUMP MCF-7 cells, while no significant changes were observed at this time in MCF10A cells (data not shown).

Int. J. Mol. Sci.Int. 2021 J. Mol., 22 Sci., x2021 FOR, 22 PEER, 862 REVIEW 18 of 26 18 of 26

Figure 17.Figure Immunoblot 17. Immunoblot bar chart bar chartshowing showing the the expr expressionession of of survivinsurvivin in in MCF-10A MCF-10A (A) and(A) inand MCF-7 in MCF-7 (B) in on-ground (B) in on-ground control cells,control RPM cells, adherent RPM adherent cells, cells,and andRPM RPM cell cell clumps clumps at 2424 h h w/o w/o the the p-ERK p-ERK inhibitor, inhibitor, PD89059, PD89059, and w/o and the p-AKTw/o the p-AKT inhibitor, LY294002,inhibitor, LY294002, respectively. respectively. Columns Columns and and bars bars represent represent densitometric densitometric quantiﬁcation quantification of the of opticalthe optical density density (OD) (OD) of a specific proteinof a speciﬁc signal protein normalized signal normalized with the with OD the values OD values of the of theGAPDH GAPDH serving serving as thethe loading loading control. control. Each Each column column rep- represents the mean value ± SD of three independent experiments. In MCF-10A, * p < 0.05 vs. OG; # p < 0.05 vs. RPM by resents the mean value ± SD of three independent experiments. In MCF-10A, * p < 0.05 vs. OG; # p < 0.05 vs. RPM by ANOVA followed by Bonferroni post-test. In MCF-7, * p < 0.05 vs. OG; § p < 0.05 vs. RPM-AD; # p < 0.05 vs. RPM-CLUMP ANOVA followed by Bonferroni post-test. In MCF-7, * p < 0.05 vs. OG; § p < 0.05 vs. RPM-AD; # p < 0.05 vs. RPM-CLUMP by ANOVA followed by Bonferroni post-test. by ANOVA followed by Bonferroni post-test.

3. Discussion3. Discussion Living cells exposed to microgravity undergo dramatic changes in morphology and Livingbiological cells properties, exposed ultimately to microgravity culminating un intodergo appreciable dramatic apoptosis changes [36– 39in], morphology despite and biologicalsome controversialproperties, results.ultimately Indeed, culminating it has been reported into appreciable that apoptosis apoptosis can occur at[36–39], very despite somedifferent controversial rates, while results. in some Indeed, cases, cellsit has appear been to reported be paradoxically that apoptosis protected fromcan occur pro- at very differentgrammed rates, cell while death in during some weightlessness cases, cells a [5ppear,40]. These to be discrepancies paradoxica canlly be protected explained from pro- by looking at differences in between cell types, and especially by considering the notion grammedthat the cell behavior death ofduring adherent weightlessness cells dramatically [5,4 diverges0]. These in respect discrepancies to that of detachedcan be explained by lookingcells organized at differences into organoids-like, in between floating cell types, structures. and especially Two main apoptotic by considering pathways the notion that thehave behavior been described: of adherent an intrinsic cells pathway dramatical (mostlyly relying diverges upon in events respect occurring to that in theof detached cells mitochondria)organized into and organoids-like, a death-receptor orfloating extrinsic structures. pathway [41 ],Two both main of them apoptotic leading to pathways activation of caspases that are responsible for the dismantlement of cells committed to have been described: an intrinsic pathway (mostly relying upon events occurring in the death [42]. However, this canonical model has not been implemented in recent years, as mitochondria)other molecular and factorsa death-receptor have been recognized or extrinsic to be efficient pathway substitutes [41], both for caspases of them [43 ],leading to activationincluding of caspases calpains [44 that,45], whileare responsible increasing evidence for the indicates dismantlement that remodeling of cells of the committed cy- to deathtoskeleton [42]. However, may act as this a pivotal canonical trigger model in apoptosis has initiationnot been upstream implemented of caspases, in bothrecent in years, as otherthe molecular extrinsic and factors intrinsic have pathway been [recognized46]. Moreover, to further be efficient apoptotic substitutes mechanisms—albeit for caspases [43], scarcely investigated—have been suspected to play a significant role in triggering cell death includingin simulated calpains microgravity [44,45], while [27,47]. increasing evidence indicates that remodeling of the cy- toskeletonIn may our experimental act as a pivotal model, trigger we observed in apop thattosis apoptosis initiation affects upstream both MCF10A of caspases, and both in theMCF-7 extrinsic cultured and inintrinsic microgravity pathway in different [46]. Moreover, ways. While further no significant apoptotic cell death mechanisms—al- was beit scarcelynoticed at investigated—have 24 h, a statistically significant been suspected increase occurs to play at 72a significant h in spite of role the factin triggering that cell death in simulated microgravity [27,47]. In our experimental model, we observed that apoptosis affects both MCF10A and MCF-7 cultured in microgravity in different ways. While no significant cell death was noticed at 24 h, a statistically significant increase occurs at 72 h in spite of the fact that such finding is limited to MCF-7 cells detached from their substrate and organized into organoid-like, floating structures. This finding implies that a critical factor in triggering programmed cell death in simulated microgravity exposed cells is represented by cell detachment and the subsequent events that lead to cytoskeleton and shape remodeling. Indeed, we observed that after exposure to microgravity, the architecture of F-actin, microtubules,

Int. J. Mol. Sci. 2021, 22, 862 19 of 26

such finding is limited to MCF-7 cells detached from their substrate and organized into organoid-like, floating structures. This finding implies that a critical factor in triggering programmed cell death in simulated microgravity exposed cells is represented by cell detachment and the subsequent events that lead to cytoskeleton and shape remodeling. Indeed, we observed that after exposure to microgravity, the architecture of F-actin, microtubules, and intermediate filaments was severely disorganized in organoids-like cell clusters. Minor, albeit significant, changes were also noticed in microgravity-exposed cells growing adherent to the substrate. Looking at those data, we would have expected a higher rate of apoptosis at both early (24 h) and later times (72 h). As this did not happen, we focused on survival pathways to investigate if cell clusters adopt a “survival strategy” to counteract the cytoskeleton disruption induced by weightlessness. It is worth noting that an increase in ERK activity was found to damp apoptosis in cancer cells treated with drugs that disrupt microtubule dynamics and cytoskeleton architecture [48,49]. Moreover, modulation of the cytoskeleton using perturbing drugs promotes mechanically induced signal activation of both ERK and AKT as compensatory mechanisms [50]. Briefly, both Akt and ERK are suspected to be activated to counterbalance the apoptosis induced after cytoskeleton disruption promoted by several chemical/physical factors [51]. Therefore, we investigated such biochemical cascades, and we found that ERK and Akt pathways were selectively and significantly elicited in MCF10A and MCF-7 cells, respectively. Conse- quently, several molecular factors involved in the apoptotic process were downregulated at early times after weightlessness exposure. Indeed, it has previously been shown that the activation of the ERK pathway has a dominant protecting effect over apoptotic signaling from the death receptors and acts by suppressing activation of caspase machinery [52]. The slowdown of programmed cell death is associated with the concomitant increase in key factors triggering survival pathways, i.e., survivin and NF-kB. On the contrary, by inhibiting ERK in MCF10A cells and Akt activation in MCF-7 cells, we observed that apoptosis was unleashed and increased significantly in these cell types at both early and later times after microgravity exposure (Figure 13A,B). Accordingly, inhibition of ERK/Akt promoted a significant increase in the release of pro-apoptotic markers: Cl-PARP, Bad and Bax. Consequently, survival effectors (NF-kB and survivin) were downregulated after the addition of pAkt and pERK inhibitors. However, the increase in Akt and ERK is not enough, per se, to successfully antagonize the apoptotic processes driven by weightlessness at later times, given that the loss of adherence may overcome the protective effect provided by the activation of the aforementioned pathways. The integrity of both FAK and integrins is needed to allow the Akt-dependent pathway to display an efficient anti-apoptotic effect, as highlighted by the apoptotic trend that we observed in MCF-7 cells. Indeed, in RPM-AD cells, FAK and integrin can counteract the apoptotic process through an appreciable increase in Akt activity, as apoptosis is still kept at bay at 72 h; on the contrary, in RPM-CLUMP cells—in which integrins are dramatically reduced—apoptosis is markedly increased despite the concomitant extraordinary increase in Akt. However, Akt plays a role in damping the apoptotic increase, given that the treatment with the Akt inhibitor triggers a further relevant apoptotic boost in both RPM-AD and RPM-CLUMP cells. This finding suggests that the current paradigm should be reversed: integrity of the cytoskeleton and of the FAK–vinculin–integrin complexes is probably not necessary to activate the Akt pathway, which ultimately can be triggered even in the absence of such stimuli, but they are mandatory to enable Akt in providing a “protective” effect. In fact, the observed increase in Akt activation is insufficient for blocking the apoptotic process, at least in RPM-CLUMP cells, in which the integrin/cytoskeleton organization is severely compromised. It is recognized that loss of adhesion, namely when experimentally obtained by blocking integrin binding or depriving the cell membrane of their integrin content [53], can suffice per se in inducing cell apoptosis. In our study, the transmission of environmental signals through the FAK–integrin complexes seems mandatory in allowing Akt to properly exert its anti-apoptotic function. In fact, in absence Int. J. Mol. Sci. 2021, 22, 862 20 of 26

of the substratum anchorage, even higher levels of Akt are insufficient in bestowing cells with an adequate survival shield. On the contrary, in MCF10A cells, in which substrate adherence is preserved, no significant increase in apoptosis was observed, although the cytoskeleton exhibited remarkable disorganization after exposure to weightlessness. These data suggest that in MCF10A cells, changes in cytoskeleton architecture are in some way counterbalanced by the persistence of integrin-mediated anchorage to the substratum, while the concomitant drop in vinculin significantly enhances ERK activity. Moreover, it is of note that between vinculin and ERK, a further feedback loop may exist, given that—upon microgravity exposure—ERK is enhanced by the reduction in vinculin expression and, in turn, ERK could reinforce vinculin downregulation, as the addition of the pERK inhibitor allows vinculin to be significantly re-expressed. Overall, such findings outline how relevant the adhesion process is in modulating the survival strategy adopted by cells to antagonize cell death in microgravity [54,55]. The extent to which cytoskeleton and other molecular factors are involved in bestowing a proper anti-apoptotic shield likely depends on cell types and on microenvironmental cues [56,57]. In our model, a critical role is undoubtedly provided by the interplay between integrins, FAK, cytoskeleton filaments, and vinculin. Vinculin reduction is required for activating the ERK-dependent pathway, but the functionality of integrin-FAK complexes is mandatory for enabling both Akt and ERK in counteracting cell death-dependent processes. The fact that different breast cells enact differentiated survival pathways when exposed to a biophysical stress like microgravity deserves further investigations and provides useful suggestions. Furthermore, it is intriguing that cancerous breast cells strive to antagonize cell death by preferentially activating the Akt pathway, whereas normal breast cells activate ERK-dependent biochemical cascades. In both cases NF-kB and survivin, values were enhanced; however, we cannot ignore the fact that other unexplored mechanism could have been selectively promoted. This implies that survival strategies do not overlap in between the two cell types. This statement has huge implications for anticancer treatment, as it suggests that we should focus on the Akt pathway (and upon PI3K, its upstream- driving controller [58]), rather than on ERK-related cascades when pro-apoptotic strategies for controlling breast cancer are envisaged. Finally, the disruption of the cytoskeleton promoted by microgravity can overcome the pro-survival strategy enacted by cancerous nonadherent MCF-7 cells. This observation reinforces the belief that attempts of harnessing the cytoskeleton and the connectivity in between cells and their microenvironment could represent a useful strategy of anticancer treatments.

4. Material and Methods 4.1. Random Positioning Machine (RPM) Microgravity conditions were simulated using a Desktop RPM, a particular kind of 3D clinostat manufactured by Dutch Space (B.V. P.O. Box 32070 2303 DB, Leiden, The Netherlands). The angular speed and the inclination of the disk inﬂuence the degree of microgravity simulation. RPMs do not actually eliminate the gravity but allow a stimulus rather than a unidirectional or omnidirectional 1 g force to be applied. The effects generated by the RPM are comparable to those of real microgravity when the direction changes are faster than the response time of the system to gravity ﬁeld. The RPM was located in a ◦ standard incubator at 37 C and 5% CO2 and connected to the control console through standard electric cables.

4.2. Cell Culture The nontumorigenic epithelial cell line MCF-10A (ATCCCRL-10317) was obtained from LGC Standards S.r.l, MI, Italy; the human hormone-sensitive breast adenocarcinoma cell line MCF-7 (ECACC Cat# 86012803) was obtained from Sigma-Aldrich (St. Louis, MO, USA). Both MCF-10A and MCF-7 were seeded in 25 cm2 ﬂasks. MCF-10A cells were grown in Dulbecco’s modiﬁed Eagle’s medium/nutrient mixture F12 Ham (Sigma-Aldrich, Int. J. Mol. Sci. 2021, 22, 862 21 of 26

Merck, Darmstadt, Germany) supplemented with 10% horse serum (Euroclone Ltd., Cram- lington, UK) and EGF 500 µg/5mL (Santa Cruz Biotechnologies, Dallas, TX, USA), Hy- drocortisone (50 µM), cholera toxin (0.5 mg/mL), insulin (10 mg/mL) (all from Sigma Chemical Co, 3050 Spruce St., St. Louis, MO, USA), and antibiotics (penicillin 100 IU/mL, streptomycin 100 µg/mL, gentamycin 200 µg/mL), all from Euroclone Ltd., Cramlington, UK. MCF-7 were grown in Dulbecco’s modified Eagle’s medium (DMEM) supplemented with 10% fetal bovine serum (FBS) and antibiotics (penicillin 100 IU/mL, streptomycin 100 µg/mL, gentamycin 200 µg/mL; all from Euroclone Ltd., Cramlington, UK). For microgravity experiments, cells were seeded into Nunc OptiCell Cell Culture Systems, gas-permeable cell culture disks (Thermo Scientific, Rochester, 75 Panorama Creek Dr, Rochester, NY, USA). OptiCells containing subconfluent monolayers were fixed onto the RPM as close as possible to the center of the platform, which was rotated at a speed of 60◦/s using the random mode of the machine. On-ground control (1 g static cultures) and RPM cultures were kept in the same humidified incubator at 37 ◦C in an atmosphere of 5% CO2 in air. Experiments were performed for 24 and 72 h. After 24 and 72 h of microgravity exposure, cell clumps swimming in culture supernatants were found, in addition to adherent cells, and separately collected. The three cell populations (on-ground control cells, RPM adherent cells, and RPM cell clumps) were characterized separately. Similar experiments were performed in the presence of 10 µM PD98059, an inhibitor of ERK phosphorylation (Cell Signaling Technology, Inc., Boston, MA, USA), or 10 µM LY294002, an inhibitor of AKT phosphorylation (Santa Cruz, Biotechnology, Inc. Santa Cruz, CA, USA).

4.3. Cell Proliferation Cells were treated for 24 and 72 h in RPM or kept on ground, and then they were counted with a particle count and size analyzer (Beckman-Coulter, Inc., Fullerton, CA, USA). Three independent experiments in duplicate were performed.

4.4. Optical Microscopy Cell clumps were collected, washed in PBS, and deposited onto a clearly defined area of a glass slide using a Shandon CytoSpin 4 Cytocentrifuge, Thermo Scientific, while maintaining cellular integrity. Cell clumps and adherent and on-ground control cells were fixed in 4% paraformaldehyde for 10 min at 4 ◦C and photographed with a Nikon Coolpix 995 digital camera coupled with a Zeiss Axiovert optical microscope. The images were obtained with a 100 × or 320 × magnification, saved as TIFF files, and used for image analysis.

4.5. Image Analysis Image analysis was performed on 10 images for each group of cells. As the analysis was performed blindly, the image groups were classified as follows: A (on-ground cells, 24 h), B (RPM adherent cells, 24 h), C (RPM cell clumps, 24 h). In each image, single randomly chosen cells (50 for each group) were contoured with a fine black marker by different researchers, simply scanned, and cataloged according to the time of study. This method was chosen because pathologists are used to correlate the shape the cells acquire with their malignancy by means of morphological, qualitative, and subjective observations. Thus, we decided to perform a semiautomatic analysis, coupling the expertise of researchers with a computerized parameterization method. All the images were processed by Adobe Photoshop CS4. All the pictures (i.e., all the sheets of the groups for each time point) were resized at 2560 × 1920 pixels according to original scale of image acquisition. For each black contoured cell, the edges were refined. Then cells were filled in black, and the threshold was adjusted in order to exclude other cells and backgrounds from the image. For each time point, a single sheet of all of the cells considered was created. To obtain single cell shape parameters (area A, roundness, solidity, and fractal dimension FD), ImageJ v1.47h Int. J. Mol. Sci. 2021, 22, 862 22 of 26

software was used. Then, the software analyzed single cells with the “shape descriptor”. In addition to area A, the following were calculated: √ Roundness = 4Aπma Solidity = ACA, (1)

where A is the area of the cell, ma is the major axis, and CA is the convex area, namely, the area of the convex hull of the region. The convex hull of a region is the smallest region that satisﬁes two conditions: (a) it is convex and (b) it contains the original region. As for FD, it was obtained by means of the box counting method using the FracLac plugin:

FD = limε→0[1 − log[Lε(C)]logε] (2)

where C is the considered curve, L is the length of the curve C, and ε is the length of the segment used as the unit to calculate L.

4.6. Annexin V/7-AAD Staining Cell clumps were collected and centrifuged, and pellets were trypsinized and washed twice with PBS. Adherent cells and ground control cells were trypsinized and washed twice with PBS. The cells were stained with FITC-labeled annexin V/7-AAD (7-aminoactinomycin- D) according to the manufacturer’s instructions (annexin-V/7-AAD kit; Beckman Coulter, Marseille, France). Brieﬂy, a washed cell pellet (5 × 104 cells/mL) was resuspended in 500 µL binding buffer; 10 µL of annexin-V together with 20 µL 7-AAD was added to 470 µL cell suspension. The cells were incubated for 15 min on ice in the dark. The samples were analyzed by ﬂow cytometry. Apoptosis assay was performed three times.

4.7. Confocal Microscopy Cells were ﬁxed with 4% paraformaldehyde for 10 min at 4 ◦C and washed twice for 10 min with PBS. The cells were permeabilized for 30 min using PBS, 3% BSA, and 0.1% Triton X-100, followed by anti-vinculin (7F9) sc-73614, anti-integrin-β1 sc-8978 (all from Santa Cruz Biotechnology10410 Finnell Street Dallas, Texas 75220 U.S.A.), anti-cytokeratin 8 (DC10) (from Thermo Fisher Scientiﬁc 75 Panorama Creek Dr, Rochester, NY 14625, USA), and anti-α-tubulin T5168 (from Sigma Aldrich 3050 Spruce St. Louis, MO, USA). Staining in PBS and 3% BSA was conducted at 4 ◦C overnight. The cells were washed with PBS and incubated for 1 h at room temperature with the appropriate secondary antibody FITC conjugated (Invitrogen Molecular Probes Eugene, OR, USA). Negative controls were processed in the same conditions as those used in primary antibody staining. For F-actin visualization, rhodamine phalloidin (Invitrogen Molecular Probes Eugene, 1: 40 dilution) was used. Cells were then washed in PBS and mounted in buffered glycerol (0.1 M, pH 9.5). Finally, analysis was conducted using a Leica confocal microscope TCS SP2 (Leica Microsystems Heidelberg GmbH, Mannheim, Germany) equipped with Ar/ArKr and He/Ne lasers. Laser line were at 543 and 488 nm for TRITC and FITC excitation, respectively. The images were scanned under a 40 × oil objective.

4.8. Western Blot Cell clumps were washed twice with ice-cold PBS and resuspended in RIPA lysis buffer (Sigma Chemical Co 3050 Spruce St. Louis, MO, USA). Adherent and ground control cells were washed twice with ice-cold PBS and scraped in RIPA lysis buffer (Sigma Chemi- cal Co.). A mix of protease inhibitors (Complete-Mini Protease Inhibitor Cocktail Tablets, Roche, Mannheim, Germany) and phosphatase inhibitors (PhosStop; Roche, Mannheim, Germany) was added just before use. Cellular extracts were then centrifuged at 8000× g for 10 min. The protein content of supernatants was determined using the Bradford assay. For Western blot analysis, cellular extracts were separated on SDS polyacrylamide gels with a concentration of acrylamide speciﬁc to the studied proteins. Proteins were blotted onto nitrocellulose membranes (BIO-RAD, Bio-Rad Laboratories, Hercules, CA, USA) and probed with the following antibodies: anti-cyclin D1 (AB-90009) from Immunological Sci- Int. J. Mol. Sci. 2021, 22, 862 23 of 26

ences; anti-survivin (number 2808), anti-phospho-AKT (ser473) (number 9271S), anti-AKT (number 9272S), anti-phospho-ERK1/2 (number 9106), anti-cleaved PARP (number 9541), anti-NF-kB #8242, and anti-GAPDH (number 2118), all from Cell Signaling Technology; anti-Bad sc-8044, anti-Bax (sc-493), anti-Bcl-2 (sc-492), anti-ERK1 (sc-94), sc-58326, anti- vinculin (7F9): sc-73614, and anti-integrinβ1 sc-8978, all from Santa Cruz Biotechnology. Antigens were detected with an enhanced chemiluminescence kit (Amersham Biosciences, Little Chalfont Buckinghamshire, UK) according to the manufacturer’s instructions. All Western blot images were acquired and analyzed through Imaging Fluor S densitometer and ChemiDoc Imaging System (Biorad-Hercules). In the graphs, the columns and bars represent densitometric quantiﬁcation of the optical density (OD) of a speciﬁc protein signal normalized with the OD values of the loading control, and they are expressed as fold increase in the control value considered as 1.

4.9. Statistical Analysis Data were expressed as mean ± standard deviation (SD). Data were statistically analyzed with the analysis of variance (ANOVA) followed by the Bonferroni post-test or unpaired two-tailed t test. Differences were considered signiﬁcant at the level of p < 0.05. Statistical analysis was performed by using GraphPad Instat software (GraphPad Software, Inc.; San Diego, CA, USA).

5. Conclusions Removal of gravity-related constraints constitutes a relevant systemic stress for living cells. Generally, exposure to simulated microgravity enhances apoptosis, although the cell death rate greatly varies depending on cell types, times of exposition, and culture conditions. Herein. we showed that preserving cell adhesivity to the substrate represents a critical factor in enabling cells to escape from apoptosis. Noticeably, in response to microgravity, both normal and cancerous breast cells activated distinct survival pathways—ERK and Akt in normal and cancer cells, respectively. However, pro-survival countermeasures showed to be only partially effective in cancer cells growing in organoids-like structures detached from the substrate and with dramatically altered cytoskeleton architecture. Not only do these results provide insights regarding the notion that survival pathways enacted by cancer cells differ from those activated by normal cells, but they also contribute to furthering our understanding of how apoptosis in normal tissues [59] can be antago- nized and, ultimately, to preserving the health and performance of astronauts exposed to space microgravity.

Author Contributions: Conceptualization: M.B.; formal analysis: A.C. (Alessandra Cucina); funding acquisition: M.B., A.C. (Alessandra Cucina), A.H.H.; investigation: N.M., S.P., M.G.M., A.C. (Alessandra Cucina), A.C. (Angela Catizone), G.R.; methodology: A.C. (Alessandra Cucina), G.R., A.C. (Angela Catizone), A.H.H., S.H.A.; resources: A.H.H., S.H.A.; supervision: N.M., M.B.; visualization: S.P.; writing—original draft: M.B.; writing—review and editing: A.C. (Alessandra Cucina), S.P., A.H.H. All authors have read and agreed to the published version of the manuscript. Funding: This work has been partly funded by the EPIREPAIR project (grant number 2014-010-R.0), COREA project (grant number 2013-084-R.0), and grant n. 2015-009-R.0, received from the Italian Space Agency (ASI). Abdel Halim Harrath, Saleh H. Alwasel, and Mariano Bizzarri extend their appreciation to the International Scientific Partnership program ISPP at King Saud University for funding this research work through ISPP-122. Data Availability Statement: The data that support the findings of this study are available from the corresponding author upon reasonable request. Conflicts of Interest: The authors declare no conflict of interest. Int. J. Mol. Sci. 2021, 22, 862 24 of 26

Abbreviations

RPM random positioning machine RO calculated roundness FD fractal dimension CSK cytoskeleton OG on ground CKs cytokeratins pERK phosphorylated ERK pAkt phosphorylated AKT Cl-PARP cleaved PARP

References 1. Bizzarri, M.; Monici, M.; van Loon, J.J. How microgravity affects the biology of living systems. Biomed. Res. Int. 2015, 2015, 863075. [CrossRef] [PubMed] 2. Bauer, J. Microgravity and Cell Adherence. Int. J. Mol. Sci. 2020, 21, 2214. [CrossRef][PubMed] 3. Sharma, C.S.; Sarkar, S.; Periyakaruppan, A.; Ravichandran, P.; Sadanandan, B.; Ramesh, V.; Thomas, R.; Hall, J.C.; Wilson, B.L.; Ramesh, G.T. Simulated microgravity activates apoptosis and NF-kappaB in mice testis. Mol. Cell. Biochem. 2008, 313, 71–78. [CrossRef][PubMed] 4. Pietsch, J.; Bauer, J.; Egli, M.; Infanger, M.; Wise, P.; Ulbrich, C.; Grimm, D. The effects of weightlessness on the human organism and mammalian cells. Curr. Mol. Med. 2011, 11, 350–364. [CrossRef] 5. Beck, M.; Tabury, K.; Moreels, M.; Jacquet, P.; Van Oostveldt, P.; De Vos, W.H.; Baatout, S. Simulated microgravity decreases apoptosis in fetal ﬁbroblasts. Int. J. Mol. Med. 2012, 30, 309–313. [CrossRef][PubMed] 6. Svejgaard, B.; Wehland, M.; Ma, X.; Kopp, S.; Sahana, J.; Warnke, E.; Aleshcheva, G.; Hemmersbach, R.; Hauslage, J.; Grosse, J.; et al. Common Effects on Cancer Cells Exerted by a Random Positioning Machine and a 2D Clinostat. PLoS ONE 2015, 10, e0135157. [CrossRef] 7. Po, A.G.A.; Masiello, M.G.; Cucina, A.; Catizone, A.; Ricci, G.; Chiacchiarini, M.; Tafani, M.; Ferretti, E.; Bizzarri, M. Phenotypic transitions enacted by simulated microgravity do not alter coherence in gene transcription proﬁle. NPJ Microgravity 2019, 5. [CrossRef] 8. Dietz, C.; Infanger, M.; Romswinkel, A.; Strube, F.; Kraus, A. Apoptosis Induction and Alteration of Cell Adherence in Human Lung Cancer Cells under Simulated Microgravity. Int. J. Mol. Sci. 2019, 20, 3601. [CrossRef] 9. Grimm, D.W.M.; Pietsch, J.; Aleshcheva, G.; Wise, P.; van Loon, J.; Ulbrich, C.; Magnusson, N.E.; Infanger, M.; Bauer, J. Growing tissues in real and simulated microgravity: New methods for tissue engineering. Tissue Eng. Part B Rev. 2014, 20, 555–566. [CrossRef] 10. Masiello, M.G.; Cucina, A.; Proietti, S.; Palombo, A.; Coluccia, P.; D’Anselmi, F.; Dinicola, S.; Pasqualato, A.; Morini, V.; Bizzarri, M. Phenotypic switch induced by simulated microgravity on MDA-MB-231 breast cancer cells. Biomed. Res. Int. 2014, 2014, 652434. [CrossRef] 11. Riwaldt, S.; Bauer, J.; Wehland, M.; Slumstrup, L.; Kopp, S.; Warnke, E.; Dittrich, A.; Magnusson, N.E.; Pietsch, J.; Corydon, T.J.; et al. Pathways Regulating Spheroid Formation of Human Follicular Thyroid Cancer Cells under Simulated Microgravity Conditions: A Genetic Approach. Int. J. Mol. Sci. 2016, 17, 528. [CrossRef] 12. Testa, F.A.P.; Dinicola, S.; D’Anselmi, F.; Proietti, S.; Pasqualato, A.; Masiello, M.G.; Coluccia, P.; Cucina, A.; Bizzarri, M. Fractal analysis of shape changes in murine osteoblasts cultured under simulated microgravity. Rend. Fis. Acc. Lincei 2014, 25, 39–47. [CrossRef] 13. Mingo-Sion, A.M.; Marietta, P.M.; Koller, E.; Wolf, D.M.; Van Den Berg, C.L. Inhibition of JNK reduces G2/M transit independent of p53, leading to endoreduplication, decreased proliferation, and apoptosis in breast cancer cells. Oncogene 2004, 23, 596–604. [CrossRef][PubMed] 14. Corydon, T.J.; Kopp, S.; Wehland, M.; Braun, M.; Schutte, A.; Mayer, T.; Hulsing, T.; Oltmann, H.; Schmitz, B.; Hemmersbach, R.; et al. Alterations of the cytoskeleton in human cells in space proved by life-cell imaging. Sci. Rep. 2016, 6, 20043. [CrossRef][PubMed] 15. Tabony, J. Morphological bifurcations involving reaction-diffusion processes during microtubule formation. Science 1994, 264, 245–248. [CrossRef] [PubMed] 16. Gershovich, P.M.; Gershovich, J.G.; Zhambalova, A.P.; Romanov, Y.A.; Buravkova, L.B. Cytoskeletal proteins and stem cell markers gene expression in human bone marrow mesenchymal stromal cells after different periods of simulated microgravity. Acta Astronaut. 2012, 70, 36–42. [CrossRef] 17. Zhao, B.; Li, L.; Wang, L.; Wang, C.Y.; Yu, J.; Guan, K.L. Cell detachment activates the Hippo pathway via cytoskeleton reorganization to induce anoikis. Genes Dev. 2012, 26, 54–68. [CrossRef] 18. Ong, M.S.; Deng, S.; Halim, C.E.; Cai, W.; Tan, T.Z.; Huang, R.Y.; Sethi, G.; Hooi, S.C.; Kumar, A.P.; Yap, C.T. Cytoskeletal Proteins in Cancer and Intracellular Stress: A Therapeutic Perspective. Cancers 2020, 12, 238. [CrossRef] Int. J. Mol. Sci. 2021, 22, 862 25 of 26

19. Kopp, S.; Warnke, E.; Wehland, M.; Aleshcheva, G.; Magnusson, N.E.; Hemmersbach, R.; Corydon, T.J.; Bauer, J.; Infanger, M.; Grimm, D. Mechanisms of three-dimensional growth of thyroid cells during long-term simulated microgravity. Sci. Rep. 2015, 5, 16691. [CrossRef] 20. Portet, S.; Vassy, J.; Beil, M.; Millot, G.; Hebbache, A.; Rigaut, J.P.; Schovart, D. Quantitative analysis of cytokeratin network topology in the MCF7 cell line. Cytometry 1999, 35, 203–213. [CrossRef] 21. Vassy, J.; Portet, S.; Beil, M.; Millot, G.; Fauvel-Lafeve, F.; Karniguian, A.; Gasset, G.; Irinopoulou, T.; Calvo, F.; Rigaut, J.P.; et al. The effect of weightlessness on cytoskeleton architecture and proliferation of human breast cancer cell line MCF-7. FASEB J. 2001, 15, 1104–1106. [CrossRef][PubMed] 22. Portet, S.; Arino, O.; Vassy, J.; Schoëvaërt, D. Organization of the cytokeratin network in an epithelial cell. J. Theor. Biol. 2003, 223, 313–333. [CrossRef] 23. Townson, D.H.; Putnam, A.N.; Sullivan, B.T.; Guo, L.; Irving-Rodgers, H.F. Expression and distribution of cytokeratin 8/18 intermediate filaments in bovine antral follicles and corpus luteum: An intrinsic mechanism of resistance to apoptosis? Histol. Histopathol. 2010, 25, 889–900. [CrossRef][PubMed] 24. Lee, J.C.; Schickling, O.; Stegh, A.H.; Oshima, R.G.; Dinsdale, D.; Cohen, G.M.; Peter, M.E. DEDD regulates degradation of intermediate filaments during apoptosis. J. Cell Biol. 2002, 158, 1051–1066. [CrossRef][PubMed] 25. Goldmann, W.H.; Ingber, D.E. Intact vinculin protein is required for control of cell shape, cell mechanics, and rac-dependent lamellipodia formation. Biochem. Biophys. Res. Commun. 2002, 290, 749–755. [CrossRef][PubMed] 26. Nassef, M.Z.; Kopp, S.; Wehland, M.; Melnik, D.; Sahana, J.; Kruger, M.; Corydon, T.J.; Oltmann, H.; Schmitz, B.; Schutte, A.; et al. Real Microgravity Influences the Cytoskeleton and Focal Adhesions in Human Breast Cancer Cells. Int. J. Mol. Sci. 2019, 20, 3156. [CrossRef][PubMed] 27. Fukazawa, T.; Tanimoto, K.; Shrestha, L.; Imura, T.; Takahashi, S.; Sueda, T.; Hirohashi, N.; Hiyama, E.; Yuge, L. Simulated microgravity enhances CDDP-induced apoptosis signal via p53-independent mechanisms in cancer cells. PLoS ONE 2019, 14, e0219363. [CrossRef] [PubMed] 28. Frisch, S.M.; Ruoslahti, E. Integrins and anoikis. Curr. Opin. Cell Biol. 1997, 9, 701–706. [CrossRef] 29. Schlaepfer, D.D.; Hanks, S.K.; Hunter, T.; van der Geer, P. Integrin-mediated signal transduction linked to Ras pathway by GRB2 binding to focal adhesion kinase. Nature 1994, 372, 786–791. [CrossRef] 30. Grosse, J.; Wehland, M.; Pietsch, J.; Schulz, H.; Saar, K.; Hubner, N.; Eilles, C.; Bauer, J.; Abou-El-Ardat, K.; Baatout, S.; et al. Gravity-sensitive signaling drives 3-dimensional formation of multicellular thyroid cancer spheroids. FASEB J. 2012, 26, 5124–5140. [CrossRef] 31. Frisch, S.M.; Francis, H. Disruption of epithelial cell-matrix interactions induces apoptosis. J. Cell Biol. 1994, 124, 619–626. [CrossRef][PubMed] 32. Mullen, P. PARP cleavage as a means of assessing apoptosis. Methods Mol. Med. 2004, 88, 171–181. [CrossRef][PubMed] 33. Cleary, M.L.; Smith, S.D.; Sklar, J. Cloning and structural analysis of cDNAs for bcl-2 and a hybrid bcl-2/immunoglobulin transcript resulting from the t(14;18) translocation. Cell 1986, 47, 19–28. [CrossRef] 34. Helmreich, E.J.M. The Biochemistry of Cell Signalling; Oxdord University Press: Oxford, UK, 2001; pp. 238–243. 35. Piva, R.; Belardo, G.; Santoro, M.G. NF-kappaB: A stress-regulated switch for cell survival. Antioxid. Redox Signal 2006, 8, 478–486. [CrossRef][PubMed] 36. Infanger, M.; Kossmehl, P.; Shakibaei, M.; Baatout, S.; Witzing, A.; Grosse, J.; Bauer, J.; Cogoli, A.; Faramarzi, S.; Derradji, H.; et al. Induction of three-dimensional assembly and increase in apoptosis of human endothelial cells by simulated microgravity: Impact of vascular endothelial growth factor. Apoptosis 2006, 11, 749–764. [CrossRef] 37. Grimm, D.; Bauer, J.; Kossmehl, P.; Shakibaei, M.; Schoberger, J.; Pickenhahn, H.; Schulze-Tanzil, G.; Vetter, R.; Eilles, C.; Paul, M.; et al. Simulated microgravity alters differentiation and increases apoptosis in human follicular thyroid carcinoma cells. FASEB J. 2002, 16, 604–606. [CrossRef] 38. Qian, A.; Zhang, W.; Xie, L.; Weng, Y.; Yang, P.; Wang, Z.; Hu, L.; Xu, H.; Tian, Z.; Shang, P. Simulated weightlessness alters biological characteristics of human breast cancer cell line MCF-7. Acta Astronaut. 2008, 63, 947–958. [CrossRef] 39. Sokolovskaya, A.A.; Korneeva, E.A.; Virus, E.D.; Kolesov, D.V.; Moskovtsev, A.A.; Kubatiev, A.A. Inhibition of Cell Cycle Progression, Induction of Apoptosis, and Changes in Surface Markers of MEG-01 Megakaryoblastic Cells Exposed to a Random Positioning Machine. Microgravity Sci. Technol. 2019, 32, 35–45. [CrossRef] 40. Risin, D.; Pellis, N.R. Modeled Microgravity Inhibits Apoptosis in Peripheral Blood Lymphocytes. Vitr. Cell. Dev. Biol. Anim. 2001, 37.[CrossRef] 41. MO, H. The biochemistry of apoptosis. Nature 2000, 407, 770–776. 42. Slee, E.A.; Adrain, C.; Martin, S.J. Serial killers: Ordering caspase activation events in apoptosis. Cell Death Differ. 1999, 6, 1067–1074. [CrossRef] [PubMed] 43. Lorenzo, H.K.; Susin, S.A. Mitochondrial effectors in caspase-independent cell death. FEBS Lett. 2004, 557, 14–20. [CrossRef] 44. HR, M. Role of calpain in apoptosis. Cell J. 2011, 13, 65–72. 45. Gasperi, V.; Rapino, C.; Battista, N.; Bari, M.; Mastrangelo, N.; Angeletti, S.; Dainese, E.; Maccarrone, M. A functional interplay between 5-lipoxygenase and mu-calpain affects survival and cytokine profile of human Jurkat T lymphocyte exposed to simulated microgravity. Biomed. Res. Int. 2014, 2014, 782390. [CrossRef][PubMed] Int. J. Mol. Sci. 2021, 22, 862 26 of 26

46. Desouza, M.; Gunning, P.W.; Stehn, J.R. The actin cytoskeleton as a sensor and mediator of apoptosis. Bioarchitecture 2012, 2, 75–87. [CrossRef] [PubMed] 47. Zhao, T.; Tang, X.; Umeshappa, C.S.; Ma, H.; Gao, H.; Deng, Y.; Freywald, A.; Xiang, J. Simulated Microgravity Promotes Cell Apoptosis Through Suppressing Uev1A/TICAM/TRAF/NF-kappaB-Regulated Anti-Apoptosis and p53/PCNA- and ATM/ATR-Chk1/2-Controlled DNA-Damage Response Pathways. J. Cell Biochem. 2016, 117, 2138–2148. [CrossRef] 48. Stadheim, T.A.; Xiao, H.; Eastman, A. Inhibition of extracellular signal-regulated kinase (ERK) mediates cell cycle phase independent apoptosis in vinblastine-treated ML-1 cells. Cancer Res. 2001, 61, 1533–1540. 49. Li, J.; Zhang, S.; Chen, J.; Du, T.; Wang, Y.; Wang, Z. Modeled microgravity causes changes in the cytoskeleton and focal adhesions, and decreases in migration in malignant human MCF-7 cells. Protoplasma 2009, 238, 23–33. [CrossRef] 50. Muller, P.; Langenbach, A.; Kaminski, A.; Rychly, J. Modulating the actin cytoskeleton affects mechanically induced signal transduction and differentiation in mesenchymal stem cells. PLoS ONE 2013, 8, e71283. [CrossRef] 51. Asnaghi, L.; Calastretti, A.; Bevilacqua, A.; D’Agnano, I.; Gatti, G.; Canti, G.; Delia, D.; Capaccioli, S.; Nicolin, A. Bcl-2 phosphorylation and apoptosis activated by damaged microtubules require mTOR and are regulated by Akt. Oncogene 2004, 23, 5781–5791. [CrossRef] 52. Tran, S.E.; Holmstrom, T.H.; Ahonen, M.; Kahari, V.M.; Eriksson, J.E. MAPK/ERK overrides the apoptotic signaling from Fas, TNF, and TRAIL receptors. J. Biol. Chem. 2001, 276, 16484–16490. [CrossRef][PubMed] 53. Bates, R.C.; Lincz, L.F.; Burns, G.F. Involvement of integrins in cell survival. Cancer Metastasis Rev. 1995, 14, 191–203. [CrossRef] [PubMed] 54. Lin, X.; Zhang, K.; Wei, D.; Tian, Y.; Gao, Y.; Chen, Z.; Qian, A. The Impact of Spaceflight and Simulated Microgravity on Cell Adhesion. Int. J. Mol. Sci. 2020, 21, 3031. [CrossRef][PubMed] 55. Lewis, M.L.; Reynolds, J.L.; Cubano, L.A.; Hatton, J.P.; Lawless, B.D.; Piepmeier, E.H. Spaceflight alters microtubules and increases apoptosis in human lymphocytes (Jurkat). FASEB J. 1998, 12, 1007–1018. [CrossRef][PubMed] 56. Schatten, H.; Lewis, M.L.; Chakrabarti, A. Spaceflight and clinorotation cause cytoskeleton and mitochondria changes and increases in apoptosis in cultured cells. Acta Astronaut. 2001, 49, 399–418. [CrossRef] 57. Ricci, G.; Cucina, A.; Proietti, S.; Dinicola, S.; Ferranti, F.; Cammarota, M.; Filippini, A.; Bizzarri, M.; Catizone, A. Microgravity Induces Transient EMT in Human Keratinocytes by Early Down-Regulation of E-Cadherin and Cell-Adhesion Remodeling. Appl. Sci. 2020, 11, 110. [CrossRef] 58. Hemmings, B.A.; Restuccia, D.F. The PI3K-PKB/Akt pathway. Cold Spring Harb. Perspect. Biol. 2015, 7.[CrossRef] 59. Yoo, Y.M.; Han, T.Y.; Kim, H.S. Melatonin Suppresses Autophagy Induced by Clinostat in Preosteoblast MC3T3-E1 Cells. Int. J. Mol. Sci. 2016, 17, 526. [CrossRef]