1. Introduction
Bone homeostasis orchestrates the delicate balance between bone formation and resorption, which is pivotal for maintaining bone density, strength, and overall skeletal integrity [
1]. Moreover, bone is sensitive to a spectrum of mechanical stimuli [
2]. Among these stimuli, low-intensity vibration (LIV) has emerged as a potent modality for triggering bone remodeling, prompting cellular responses from osteocytes, osteoblasts, osteoclasts, and mesenchymal stem cells (MSCs) [
3], [
4]. While the beneficial effects of LIV on bone health are extensively documented, its implications in cancer progression and bone metastasis are multifaceted and not entirely understood [
5], [
6], [
7]. One challenge lies in understanding the differential responses of tumor cells and nontumor cells to LIV. Vibrations may directly impede the progression of tumor cells while indirectly compromising nontumor cells and diminishing their tumor-suppressive capabilities.
In this study, our focus was on exploring the potential of LIV-induced conversion of nontumor cells into tumor-suppressing cells. Previous research has demonstrated the capacity to convert both tumor and nontumor cells, including osteoblasts, osteocytes, osteoclasts, and MSCs, into induced tumor-suppressing (iTS) cells through the activation of various signaling pathways, such as the Wnt, phosphoinositide 3-kinase (PI3K), and protein kinase A (PKA) pathways [
8], [
9], [
10]. Leveraging our prior success in generating iTS cells via genetic engineering and chemical treatment, our objective herein was to extend this methodology to investigate the efficacy of LIV in converting lymphocytes into iTS cells.
Lymphocytes are pivotal in immune responses, underscoring their critical role in disease combat [
11], [
12]. T lymphocytes are the primary regulators of immunotherapeutic responses in malignant tumors [
13]. Unlike bone cells such as osteoblasts and osteocytes, which adhere to substrates, lymphocytes are typically grown in suspension. T cells are exposed to biomechanical environments, which regulate their development [
14]. Here, we pose an intriguing question: Can LIV reprogram suspended cells such as lymphocytes and peripheral blood mononuclear cells (PBMCs) into iTS cells? The biophysical effect may represent paracrine-like signaling for immune cells to indirectly regulate tumor progression [
15]. The iTS cell approach shows promise in augmenting cancer immunotherapy modalities such as chimeric antigen receptor (CAR) T-cell therapy, particularly in overcoming challenges associated with infiltrating solid tumors such as breast cancer [
16], [
17]. LIV could expand the repertoire of lymphocyte-based therapeutic interventions, offering an additional avenue for combatting cancer with profound clinical implications.
In addition to exploring LIV, we investigated whether mechanical stimulation through shaking could effectively induce the conversion of lymphocytes into iTS cells. Unlike custom-made vibration tables, tube shakers are commonplace tools in standard medical laboratories. Leveraging LIV for immune cell reprogramming in such a familiar laboratory setting facilitates seamless translation from laboratory experimentation to clinical application. Our objective was to impede cancer cell progression, particularly in breast cancer, through the indirect mechanism of generating iTS cells from lymphocytes and PBMCs.
To determine the LIV-driven mechanism underlying the generation of iTS cells, we investigated the involvement of Sad1 and UNC-84 domain containing 1 (SUN1), a protein localized within the inner nuclear membrane. The SUN protein is part of the linker of the nucleoskeleton and cytoskeleton (LINC) complex, which transmits intracellular forces by connecting the nucleus to the cytoskeleton [
18], [
19]. While the role of SUN1 may not be universally pivotal in iTS generation, our mass spectrometry-based proteomic analysis revealed the enrichment of numerous tumor-suppressing proteins [
8], [
20], irrespective of their generation method. These findings have significant implications for iTS cell-associated cancer therapeutics, shedding new light on leveraging mechanical cues to fortify the body’s antitumor defenses. Future endeavors, including further research and clinical trials, may corroborate and refine this approach, offering an unconventional option for patients contending with the challenges of breast cancer bone metastasis.
2. Materials and methods
2.1. Cell culture
MDA-MB-231 breast cancer cells (ATCC; USA), MDA-MB-436 breast cancer cells (ATCC), 4T1.2 mouse mammary tumor cells (obtained from Dr. R. Anderson at Peter MacCallum Cancer Institute, Australia), MC3T3-E1 osteoblasts (Sigma, USA) and murine MSCs isolated from the bone marrow of C57BL/6 mice were cultured in DMEM. TRAMP-C2ras prostate tumor cells (ATCC) were cultured in DMEM/F-12, and PC-3 human prostate cancer cells (ATCC), Jurkat T lymphocytes (ATCC), and PBMCs (Lonza, Switzerland) were cultured in RPMI1640 (Gibco, USA). RAW264.7 preosteoclast cells (ATCC) were grown in αMEM. The culture medium used was supplemented with 10% fetal bovine serum and antibiotics (100 units∙mL−1 penicillin and 100 µg∙mL−1 streptomycin; Life Technologies, USA). All of the cells were maintained under standard conditions at 37 °C with a 5% CO2 atmosphere.
2.2. Chemical agent and plasmid transfection
The cells were treated with cisplatin (#2251; Tocris Bioscience, UK) as a positive control chemotherapeutic agent. Methylcellulose (#428430500; Thermo Fisher Scientific, USA) was used to increase the viscosity of the culture medium. Overexpression of low-density lipoprotein receptor-related protein 5 (Lrp5), SUN1, cellular Myc (cMyc), and Kristen rat sarcoma viral oncogene homolog (K-Ras) was accomplished with plasmids (#115907, #125851, #17758, #159554; Addgene, USA), while blank plasmids (FLAG-HA-pcDNA3.1; Addgene) served as the control. Transfection was conducted via a K4 transfection system (Biontex Laboratories, Germany) according to the manufacturer's instructions.
2.3. Application of LIV and shaking
Both adherent cells and cells in suspension were grown in culture dishes (35 mm × 10 mm), and LIV was applied using a custom-made vibration table. The primary vibration condition was 90 Hz with an intensity of 0.7g (where 1g = 9.8 m∙s−2). The control cells were placed on the vibration table without any exposure to vibrations. For in vitro assays, the cells were subjected to two 20-min sessions of LIV exposure, each separated by a 3 h interval, for 3 d. Notably, adherent cells shrank after LIV treatment, but their shape recovered within approximately 3 h. Shaking of cells in suspension (Jurkat T lymphocytes) in a 2 mL tube was performed with a standard tube shaker (ThermoMixer R; Eppendorf, Germany). The cells were shaken for 20 min twice a day at 800 r∙min−1, with a 3 h break, for 3 d. The shaking procedure resulted in greater acceleration (1.3g) than did the standard table vibration procedure. The viscosity of the culture medium was adjusted to 10 centipoise (cP, viscosity unit) by adding methylcellulose. Notably, the viscosity of the control medium was 0.8 cP.
2.4. 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT), 5-ethynyl-2'-deoxyuridine (EdU), and clonogenic assays
MTT-based assessment of metabolic activity was conducted using ∼2000 cells in each well of a 96-well plate (Corning, USA). The response to CM was tested after a 2-d incubation. Metabolic activity was quantified by measuring the optical density of thiazolyl blue tetrazolium bromide (#M5655; Sigma) at 570 nm. In the EdU assay, ∼1000 cells were seeded in each well of a 96-well plate. The effect of CM on cellular proliferation was assessed after 2 d using a fluorescence-based cell proliferation kit (Click-iT EdU Alexa Fluor 488 Imaging Kit; Thermo Fisher Scientific). Fluorescently labeled cells were counted, and the ratio of the labeled cells to the total cells was determined. For the clonogenic assay, approximately 1000 cells were seeded in each well of a 35 mm tissue culture dish (Corning). The cells were allowed to grow for 2 weeks to form colonies. Afterward, the colonies were fixed and stained with Giemsa Stain (#4201457; RICCA Chemical, USA), and the number of colonies was quantified via ImageJ (National Institutes of Health, USA).
2.5. Two-dimensional motility scratch and Transwell invasion assays
A two-dimensional motility scratch assay was performed to assess two-dimensional migratory behavior. Approximately 4 × 105 cells were seeded into each well of a 12-well plate. Upon cell attachment, a plastic pipette tip was used to generate a scratch within the cell layer. The dislodged cells were removed, and CM was added. Images of the cell-free scratch zone were captured at 0 h. Subsequently, the areas newly occupied by the cells were evaluated 24 h postscratching. These areas were quantified via ImageJ. In a Transwell invasion assay, ∼5 × 104 cells were suspended in 200 µL of serum-free DMEM and placed within the upper Matrigel-coated chamber (Thermo Fisher Scientific). The lower chamber was supplied with 800 µL of CM, and after 2 d, the cells that had penetrated the underside of the membrane were stained with crystal violet. A minimum of five randomly selected images were captured, and the average count of the stained cells was determined.
2.6. Osteoclast and osteoblast differentiation assays
The differentiation assay for RAW264.7 preosteoclasts was conducted in a 12-well plate. The preosteoclast cells were incubated in medium containing 40 ng∙mL−1 receptor activator of nuclear factor-kappa B ligand (RANKL) for 6 d. During this incubation period, the culture medium was replaced once on the fourth day. Adherent cells were then fixed and subjected to staining via a tartrate-resistant acid phosphate (TRAP)-staining kit (Sigma) following the instructions provided by the manufacturer. As mature osteoclasts, TRAP-positive multinucleated cells, with more than three nuclei, were counted. To evaluate the effect of CM on the differentiation of osteoblasts, MC3T3 osteoblasts were cultured in osteogenic medium supplemented with 50 µg∙mL−1 ascorbic acid and 10 mmol∙L−1 sodium β-glycerophosphate with 10% fetal bovine serum (FBS) and antibiotics. The medium was changed every 3 d, and the cells were fixed after 4 weeks and stained with Alizarin Red to visualize calcium deposits.
2.7. Western blot analysis
The cells were lysed in radioimmunoprecipitation assay buffer, after which the proteins were separated using 10%-15% sodium dodecyl sulfate (SDS) gels and subsequently transferred onto polyvinylidene difluoride membranes (Millipore, USA). The membrane was incubated overnight with primary antibodies, followed by incubation with secondary antibodies conjugated with horseradish peroxidase (Cell Signaling, USA). Antibodies against snail family transcriptional repressor 1 (Snail), p-Src, Src, aldolase A (ALDOA), cleaved caspase 3, caspase 3, alkaline phosphatase (ALP), RANKL, low-density lipoprotein receptor-related protein 5 (Lrp5), c-Myc, K-Ras, p-Akt, enolase 1 (Eno1), moesin (MSN), Hsp90ab1 (Cell Signaling), and SUN1 (Proteintech, Rosemont, IL, USA) were used, with β-actin (Sigma) serving as a control. The protein levels were quantified via the SuperSignal West Femto Maximum Sensitivity Substrate (Thermo Fisher Scientific). The signal intensities were then quantified using a luminescent image analyzer (LAS-3000; Japan).
2.8. Ex vivo assays
Human breast and prostate cancer tissues were obtained from the Simon Cancer Center Tissue Procurement Core with approval from the Indiana University Institutional Review Board (#1911155674). Each tissue sample (∼1 g) was manually fragmented into small pieces (0.5-0.8 mm in length) using a scalpel. These fragments were cultured for 1 d in DMEM supplemented with 10% fetal bovine serum and antibiotics. The CM was subsequently added for four more days, and changes in the size of at least four fragments were monitored and recorded.
In an ex vivo mouse bone assay, femurs from female BALB/c mice (∼10 weeks old) were harvested and cleaned of surrounding connective tissue. Each femur was bisected at the mid-diaphysis and perforated with a 25-gauge needle. A total of 2.5 × 105 4T1.2 mouse mammary tumor cells suspended in 10 μL of culture medium were injected into the bone marrow cavity. The bones were then cultured in 2 mL of DMEM supplemented with 10% FBS and 1% antibiotics at 37 °C with 5% CO2. Every day, half of the medium was replaced with fresh medium. After two weeks of daily vibration, the cells growing inside the bones were subjected to Western blot analysis.
2.9. Mass spectrometry-based global proteomics
Mass spectrometry-based global proteomics was conducted via a previously described procedure [
10]. The three groups of Jurkat cell-derived CMs, each consisting of three samples, were control CM, CW008-treated positive control CM, and LIV-treated CM. CW008 is an activator of the PKA signaling pathway. Proteins with two or more tandem mass spectrometry (MS/MS) counts and label-free quantitation (LFQ) values above 0 were retained and used for downstream bioinformatics analysis. We used MS/MS counts to quantify the protein levels, and the relative abundance was compared with that of the control CM group without the generation of iTS cells.
2.10. Gene set enrichment analysis (GSEA)
To understand the biological mechanisms downstream of CM, we performed pathway analysis on the results of MS/MS. We first normalized the MS/MS counts using quantile normalization. We subsequently used a linear model to determine the relative differential protein abundance between the PKA and control groups and between the LIV and control groups. We performed GSEA using fgsea (v1.20.0) in R (v4.1.1) with the hallmark gene sets from MSigDB. The analysis was performed on the list of sorted, signed log10p values, and any proteins whose average count was less than 15 were excluded. p values were adjusted for multiple comparisons using the Benjamini-Hochberg procedure. Pathway analysis revealed several significant pathways in both the PKA and LIV groups, including genes related to Myc signaling and K-Ras downregulation.
2.11. Finite element modal analysis
To estimate the natural frequencies and deformation modes of LIV-treated lymphocytes, modal analysis was carried out using the finite element procedure, which was described previously [
21]. Using COMSOL Multiphysics software (v6.2; Altsoft, USA), we assumed a spherical cell in a suspension that was modeled as an outer fluid using the boundary condition of an infinite domain. We evaluated the effects of varying the cell size (5 to 15 μm in radius) and Young’s modulus. The infinite domain considers that the cell culture medium in which the vibration occurs extends infinitely in all directions.
2.12. Ethics statement
The animal procedures were approved by the Indiana University Animal Care and Use Committee (SC345R) and complied with the Guiding Principles in the Care and Use of Animals endorsed by the American Physiological Society and the European Union Directive on Animal Experimentation. The use of human tissues was approved by the Indiana University Institutional Review Board (#1911155674) and performed following the World Medical Association Code of Ethics (Declaration of Helsinki).
2.13. Statistical analysis
For the cell-based experiments, three or four independent experiments were conducted, and the data are expressed as the mean ± standard deviation (SD). In the ex vivo assay, two or three independent experiments were performed. Statistical significance was evaluated via one-way analysis of variance (ANOVA). Post hoc statistical comparisons with control groups were performed using Bonferroni correction, with p < 0.05 indicating statistical significance. The single and double asterisks and pound signs in the figures indicate p < 0.05 and p < 0.01, respectively.
3. Results
3.1. LIV of adherent cells vs suspended cells
This study explored the application of LIV on various cell types, including adherent cells such as breast cancer cells and prostate cancer cells, as well as suspended cells such as lymphocytes and PBMCs. The primary question was whether these distinct cell types would exhibit differential responses under LIV conditions. In previous investigations involving adherent cells, we utilized vibration frequencies ranging from 10 to 100 Hz, approximating their natural frequencies. Finite element analysis of the suspended cells revealed distinctive modal patterns and dependences on the Young’s modulus and cell size (
Figs. 1(a) and (b)). For a typical lymphocyte with a radius of 7.5 μm, the natural frequency ranged from 10 to 100 Hz with a small Young’s modulus of 0.001 kPa and from 1 000 to 10 000 Hz with an anticipated Young’s modulus of 1 kPa. We initially evaluated the inhibitory effect of LIV at frequencies ranging from 30 to 150 Hz (Fig. S1(a) in Appendix A). The MTT results, indicating metabolic activity, revealed that MDA-MB-436 was most sensitive to LIV at 90 Hz. In this study, our focus was primarily on employing a frequency of 90 Hz, which effectively suppresses the motility of adherent cancer cells. This frequency significantly differed from the natural frequency of lymphocytes.
3.2. Effects of LIV on the viability and motility of breast cancer and prostate cancer cells
We then investigated the impact of LIV on the tumorigenic behaviors of breast cancer cell lines (MDA-MB-231 and MDA-MB-436) and prostate cancer cell lines (PC-3 and TRAMP). Upon applying LIV at a frequency of 90 Hz with an intensity of 0.7
g (amplitude of ∼7 m∙s
−2 acceleration), we discerned a trend toward diminished MTT-based cell viability within 3 d (Figs. 1(c)−(f)). An inhibitory effect was evident in all three cell lines except for the MDA-MB-231 cell line. A scratch-based cell motility assay revealed that the suppressive effect of LIV was significant across all of these cell lines (
Figs. 1(g) and (h); Figs. S1(b) and (c) in Appendix A). In addition, the application of LIV downregulated the oncogenic proteins p-Src and Snail, which are considered to promote tumor cell metastasis and invasion, in MDA-MB-436 and PC-3 cells (
Figs. 1(i) and (j)). LIV also induced the activation of cleaved caspase 3, the key marker of programmed cell death. We also examined the effects of different LIV conditions on tumor cells. The results of the MTT assay revealed that cell viability inhibition was dependent on the amplitude or frequency (Figs. S2(a)−(d) in Appendix A). Interestingly, the viability of MSCs was not significantly affected by LIV of 0.7
g at 90 Hz (Fig. S2(e) in Appendix A).
3.3. Suppression of tumorigenic behaviors by LIV-treated Jurkat CM
Our previous study demonstrated that LIV can convert MSCs into iTS cells [
21]. On this basis, we explored the feasibility of generating iTS cells from Jurkat cells exposed to LIV. We applied LIV to the Jurkat cells twice daily for 20 min, with a 3-hour interval in between. The culture medium was replaced on the 3rd day, and the CM was collected after 24 h. Jurkat-derived CM without mechanical stimulation did not exhibit antitumor properties, but LIV-treated Jurkat-derived CM showed significant tumor suppressive ability. Specifically, the application of CM from LIV-treated cells (90 Hz, 0.7
g) reduced MTT-based viability and diminished the scratch-based motility of MDA-MB-231 breast cancer cells (
Figs. 2(a) and (b)), MDA-MB-436 breast cancer cells (
Figs. 2(c)and (d) ), PC-3 prostate cancer cells (
Figs. 2(e) and (f)), and TRAMP prostate cancer cells (
Figs. 2(g) and (h)). In addition, LIV-treated Jurkat-derived CM significantly reduced EdU-based proliferation and colony-forming ability and inhibited the Transwell invasion of MDA-MB-436 and PC-3 cells (Figs. 3(a−f)). We also examined whether CM from LIV-treated cancer cells affects noncancerous cells. Notably, MTT assays revealed that CM from LIV-treated Jurkat cells did not decrease MSC viability (Fig. S2(f)).
3.4. Inhibition of the growth of human cancer tissue fragments
After demonstrating the antitumor ability of LIV-induced Jurkat CM, to further evaluate its antitumor efficacy, we applied it to freshly obtained human breast cancer tissue fragments (triple-negative breast cancer) and prostate cancer tissue fragments. In the
ex vivo tissue assay, the size of the tissue fragments in the LIV-treated Jurkat CM group was significantly reduced after 48 and 96 h of culture compared with that in the standard medium group or the CM control group without LIV (
Figs. 3(g) and (h)).
3.5. Effects of a chemotherapeutic drug on the differentiation of osteoclasts and osteoblasts
We next studied the effects of the coadministration of LIV CM with cisplatin, a chemotherapeutic drug widely used in clinical treatment [
22]. The results showed that CM from LIV-treated Jurkat cells in combination with cisplatin enhanced the reduction in the MTT-based viability of both MDA-MB-231 and MDA-MB-436 breast cancer cells and PC-3 and TRAMP prostate cancer cells (Figs. 4(a−d)).
Since breast and prostate cancer frequently metastasize to bone, our next examination was the effect of CM from LIV-treated cells on bone homeostasis. We observed that the introduction of CM from LIV-treated Jurkat cells significantly hindered the differentiation of RANKL-stimulated osteoclasts, resulting in a reduced number of multinucleated TRAP-positive osteoclasts (with more than three nuclei) (
Fig. 4(e)). Furthermore, in MC3T3 osteoblasts cultured in CM from LIV-treated cells, Alizarin Red staining, a mineralized calcium deposit detection assay, increased over four weeks (
Fig. 4(f)). Interestingly, Alizarin Red staining of MC3T3 osteoblasts cultured in CM from Jurkat cells not treated with LIV also revealed a significant increase, suggesting that lymphocyte secretomes may promote osteoblast differentiation independent of mechanical stimulation.
3.6. Inhibitory effects of h-PBMC CM on human cancer tissue fragments
After demonstrating the antitumor potential of LIV-induced Jurkat CM, we engaged human peripheral blood mononuclear cells (h-PBMCs) to study the possibility of producing antitumor CM. CM obtained from LIV-treated h-PBMCs inhibited MTT-based viability and scratch-based motility of MDA-MB-436 breast cancer cells and PC-3 prostate cancer cells (Figs. 5(a)−(d)). We then evaluated its antitumor efficacy using human breast cancer (triple-negative breast cancer) and prostate cancer tissue fragments.
In vitro morphometric measurements revealed that breast cancer tissue fragments as well as prostate cancer tissue fragments shrank significantly after culture in LIV-treated h-PBMC CM for 48 and 96 h (
Figs. 5(e) and (f)).
3.7. Generation of iTSCs by shaking in highly viscous culture medium
To facilitate the application of mechanical stimulation in a standard laboratory, we examined the possible replacement of the custom-made vibration table with a routine-use tube shaker. We used methylcellulose to increase the viscosity of the culture medium [
23]. Interestingly, the same antitumor efficacy of CM from LIV-treated cells was obtained by shaking a tube containing Jurkat cells that were suspended in high-viscosity medium (10 cP). While the modulation of viscosity did not affect the growth of MDA-MB-436 cells (
Fig. 6(a)), increasing the viscosity was necessary to generate iTS cells that produced CM with potent antitumor activity against MDA-MB-436 cells (
Figs. 6(b) and (c)), PC-3 cells (Figs. 6(d)-(f)), and 4T1.2 breast cancer cells (Fig. S3 in Appendix A).
3.8. Tumor-suppressive effects of LIV-treated CM in an ex vivo mouse model.
To further investigate the impact of LIV on the bone tumor microenvironment, we performed an
ex vivo bone assay. A pair of femurs from immunocompetent female mice were harvested, and subsequently, 4T1.2 mammary tumor cells were inoculated into the bone marrow cavity. The bone was treated with or without LIV daily for 2 weeks. LIV decreased Src phosphorylation and Snail expression, whereas the level of the cytotoxic marker cleaved caspase 3 (c-Cas) increased in the cells cultured inside the bone. Additionally, alkaline phosphatase (ALP) [
24] expression increased, whereas the receptor activator of RANKL [
25] decreased, indicating a bone-protective effect of LIV (
Fig. 7(a)). Consistently, an increase in p-Src and Snail and a decrease in c-Cas were observed in MDA-MB-436 breast cancer cells and PC-3 prostate cancer cells in response to LIV-treated Jurkat CM (
Figs. 7(b) and (c)).
3.9. Differential effects of SUN1 and Lrp5 on the tumor-suppressing ability of Jurkat iTSC CM
As one of the components of the LINC complex, we examined the function of SUN1 in the LIV-driven generation of iTS cells. The role of Lrp5, a coreceptor of Wnt signaling, was also evaluated as a control mediator of flow-induced mechanotransduction (
Fig. 7(d)). The results of the MTT-based viability assay revealed that the overexpression of SUN1 increased the anticancer ability of CM, but that of Lrp5 did not (
Figs. 7(e) and (f)). The results of the EdU-based cell proliferation assay with MDA-MB-436 cells also supported the involvement of SUN1 in the response to LIV (
Fig. 7(g)).
3.10. Mass spectrometry-based proteomic analysis of conditioned media
Using PKA activation as a control, mass spectrometry-based proteomic analysis of LIV-treated lymphocyte-derived CM was performed. To identify any similarities among the various iTS CMs, we included proteomics results previously obtained in response to Wnt signaling [
8], PI3K signaling [
9], and Oct4 overexpression [
10]. Notably, the results revealed the number of proteins enriched in common in the five different iTSC CMs, which are linked to Wnt, PI3K, PKA, Oct4, and LIV (
Fig. 8(a); Figs. S4(a) and (b) in Appendix A). Among them, nine proteins were highly enriched in all five CMs, including EEF2, PKM, PPIA, ENO1, PGAM1, MSN, FLNA, HSPA8, and HSP90AB1 (
Fig. 8(b)). In addition to the LIV-treated CM, sixteen proteins were enriched in the three CMs (PKA, PI3K, and Oct4), whereas fourteen proteins were upregulated in the other set of three CMs (Wnt, PI3K, and Oct4) (Figs. 8(c)−(e)). Among these commonly enriched proteins in three to four different CMs, proteins such as ENO1, MSN, HSP90AB1, ARHGDIA, ALDOA, and MYH9 were previously shown to act as extracellular tumor suppressor proteins [
9], [
10], [
16], [
20].
3.11. Myc and K-Ras signaling as the predicted underlying pathways for PKA and LIV
Using global proteomics data for the activation of PKA and application of LIV, we conducted GSEA to predict signaling pathways that are active in the secretion of tumor-suppressing proteins (Fig. S5 in Appendix A). The highest enrichment score was 1.5 (both LIV and PKA) for Myc signaling, whereas the lowest enrichment scores were −1.81 (LIV) and −1.88 (PKA) for the downregulation of K-Ras signaling (
Fig. 9(a)). These predictions are consistent with our previous finding that MSCs can be converted into iTS cells through the overexpression of c-Myc and K-Ras [
10].
In line with the anticipated roles of c-Myc and K-Ras, as indicated by gene enrichment analysis, LIV treatment led to increased expression levels of c-Myc, K-Ras, and SUN1 in Jurkat cells (
Fig. 9(b)). Additionally, the upregulation of c-Myc and K-Ras resulted in increased levels of p-Akt, a crucial enzyme in PI3K signaling (
Fig. 9(c)). Moreover, the overexpression of c-Myc and K-Ras induced elevated levels of tumor suppressor proteins such as Eno1, MSN, ALDOA, and Hsp90ab1 in CM derived from Jurkat cells overexpressing c-Myc and K-Ras (
Fig. 9(d)). Finally, assessments through an MTT-based cell viability assay and a scratch-based motility assay demonstrated the tumor-suppressive potential of CM derived from Jurkat cells overexpressing c-Myc and K-Ras (
Figs. 9(e) and (f); Fig. S6 in Appendix A).
4. Discussion
This study presented a biophysical approach, such as LIV and shaking, for converting lymphocytes and PBMCs into iTS cells (
Fig. 9(d)). The CM derived from iTS cells demonstrated multifaceted anticancer properties, including suppression of the viability, proliferation, migration, and invasion of breast and prostate cancer cells, as well as shrinkage of human cancer tissue fragments, along with notable bone-protective effects. Notably, achieving strong transformation efficiency by shaking necessitated an increase in media viscosity. The LINC complex, which acts as a mechanosensor at the nuclear membrane, particularly implicates SUN1, a constituent protein, in the generation of iTS cells. Through mass spectrometry-based proteomic analysis of iTS cell-harvested CM, we revealed that tumor-suppressing proteins, including ENO1, MSN, Hsp90ab1, and ALDOA, were enriched irrespective of the methodology used to generate iTS cells, such as the application of LIV, the activation of PKA signaling, or the upregulation of oncogenic signaling, such as c-Myc and K-Ras.
Through a counterintuitive procedure involving the activation of oncogenic signaling, iTS cells were originally developed [
8]. Hence, it is not surprising that among the five different procedures for generating iTS cells, two signaling pathways associated with c-Myc and K-Ras emerged as foundational. Additionally, iTS cells can be derived from both MSCs and lymphocytes, revealing remarkable diversity yet common features. Despite originating from distinct cell types, iTS cells exhibit commonalities in signaling pathways such as the Wnt, PI3K, PKA, and Oct4 pathways. A prevailing consensus suggests that proliferative signals bolster cells, empowering them to combat cancer. This shared characteristic is evident across the five different procedures, as well as in the secretome found in the CM of both MSCs and lymphocytes.
The proper LIV conditions vary depending on the intended application [
3], [
26], [
27]. While LIV can effectively impede the progression of cancer cells, its efficacy varies across different cell lines [
6], [
7], [
28]. While decreased viability may not be uniformly observed, the inhibition of migration tends to be more consistent. In this study, LIV was applied to nontumor cells, such as lymphocytes, to generate iTS cells. Notably, solid tumor cells, such as breast cancer cells, typically adhere to surfaces, whereas lymphocytes remain in suspension. The LIV frequency utilized in this study, 90 Hz, falls within the natural frequency range of adherent cells [
21], whereas the natural frequency of lymphocytes may vary significantly from 10 Pa to 1 kPa depending on their size and Young’s modulus [
29], making it challenging to identify the appropriate frequency for lymphocytes. The efficacy of LIV was also significantly greater in the 3-day treatment group than in the single-day treatment group. This may be due to the accumulation of tumor-suppressing proteins in the CM. Moreover, when in suspension, the culture medium contributes to added mass in response to LIV, and viscosity emerges as a crucial factor. Further investigations are warranted to elucidate the frequency dependence and identify the appropriate frequencies for generating iTS cells from lymphocytes.
In devising an effective strategy for treating bone metastases, especially those linked with breast cancer, it is crucial to target multiple objectives, including halting tumor progression and preventing bone loss [
30]. Lymphocyte-derived CM has three key functions: anticancer activity, anti-bone resorption properties, and promotion of bone formation. The multifaceted nature of these capabilities positions iTS cells as promising candidates for therapeutic interventions, particularly in addressing bone metastasis associated with breast cancer.
Another highly promising application lies in CAR T-cell immunotherapy, where CAR T cells are genetically engineered to target specific cancer cell antigens [
31], [
32]. CAR T-cell therapy has proven highly effective in treating hematological malignancies. However, its efficacy in solid tumors remains limited, primarily due to challenges in selecting appropriate CARs and achieving sufficient infiltration into solid tumor tissues [
33], [
34], [
35]. Leveraging the attributes of iTS cells, CAR T-cell therapy could be enhanced. For example, the broad, systemic anticancer effect offered by the activity of iTS cells could synergize with the targeted approach of CAR T cells. Notably, iTS cells secrete tumor-suppressing proteins, eliminating the need for direct contact with tumor cells, a feature not shared by CAR T cells. Additionally, the anti-bone resorption and bone formation-promoting capabilities of iTS cells represent unique features that distinguish them from CAR T cells.
Research indicates that the nucleus serves as a crucial mechanosensory organelle [
36]. Transmission of LIV necessitates connections between the nucleus and the cytoskeleton, facilitated by the LINC complex [
36], [
37]. Within the LINC complex, SUN proteins establish connections between the actin cytoskeleton and the nucleus [
38]. The LINC complex has been implicated in regulating cancer cell responses to LIV; deletion of SUN1 and SUN2 impedes the ability of LIV to influence cell stiffness [
6]. Disruption of LINC function via siRNA-mediated deletion of SUN1 and SUN2 diminished mechanical responsiveness in both nontumor cells [
39] and breast cancer cells [
6]. In this study, we evaluated SUN1 as an enhancer of mechanical signal amplification to generate vibration-induced iTS cells. However, the membrane receptor Lrp5, which plays a role in the response to fluid flow-induced shear stress, does not act as a mediator.
This study has several limitations. The mechanical environment surrounding bone cells in the body is notably intricate. Different cell types, including lymphocytes, MSCs, osteoblasts, and osteocytes, may respond differently to biophysical stimuli [
40], [
41], [
42]. Moreover, the bone microenvironment is complex and consists of a collagen matrix and hydroxyapatite substrates, which could influence the response to LIV. Notably, however, the conversion of lymphocytes into iTS cells can be accomplished in a standard laboratory setting. In conclusion, the accessibility of lymphocytes and PBMCs from patients with breast or prostate cancer renders them easy to harvest for therapeutic applications. The conversion of lymphocytes into iTS cells, coupled with their versatile functionalities, reveals promising prospects for increasing the efficacy of current treatments for bone metastases linked to breast cancer. Moreover, this breakthrough is anticipated to increase the effectiveness of CAR T-cell immunotherapy, representing a significant advancement in cancer treatment modalities. These developments signify the emergence of novel avenues in the battle against cancer, offering renewed hope for more effective and targeted therapeutic interventions.
Acknowledgments
This research was funded by the Indiana Clinical and Translational Science Institute and Showalter Trust (to Uma K. Aryal), the Human Resources Program in Energy Technology of the Korea Institute of Energy Technology Evaluation and Planning (20214000000140 (to HeeChang Lim)), the National Institutes of Health R35GM147412 (to Jing Liu) and R01AR074473 (to William R. Thompson), the National Natural Science Foundation of China, 81971326 (to Bai-Yan Li), and the 100 Voices of Hope (to Hiroki Yokota).
Compliance with ethics guidelines
Xue Xiong, Qingji Huo, Changpeng Cui, Uma K. Aryal, BonHeon Ku, Chin-Suk Hong, HeeChang Lim, Jing Liu, Andy Chen, William R. Thompson, Bai-Yan Li, Xue-Lian Li, and Hiroki Yokota declare that they have no conflict of interest or financial conflicts to disclose.
PDF
(6599KB)
