INTRODUCTION
Section:
ChooseTop of pageABSTRACTINTRODUCTION <<RESULTSDISCUSSIONMETHODSSUPPLEMENTARY MATERIALUsing a breast cancer model, we focus on studying the changes in EC chirality caused by paracrine signaling and physical contact of TCs. Any changes or disruptions of intrinsic CW chirality of EC into non-chiral (NC) or counterclockwise (CCW) chirality may compromise the overall endothelial integrity which potentially increases the risk of metastasis. We hypothesized that either physical contact or paracrine signaling between TCs and ECs disrupts the chirality of the endothelial barrier. We hope to provide insight into how TC–EC interaction modulated endothelial morphogenesis may affect trans-endothelial migration during extravasation and aid the identification of therapeutic strategies to stop metastasis.
RESULTS
Section:
ChooseTop of pageABSTRACTINTRODUCTIONRESULTS <<DISCUSSIONMETHODSSUPPLEMENTARY MATERIALThe intrinsic clockwise chirality of endothelial cells is disrupted in TC–EC co-culture
Oncogenic activation of GTPase HRas (HRAS) and human epidermal growth factor receptor 2 (HER2) has been frequently found in breast cancers, leading to tumor initiation, progression, and metastasis [Fig. S1(a)].33–3533. M. Galiè, “ RAS as supporting actor in breast cancer,” Front. Oncol. 9, 1199 (2019). https://doi.org/10.3389/fonc.2019.0119934. E. Y. H. P. Lee and W. J. Muller, “ Oncogenes and tumor suppressor genes,” Cold Spring Harbor Perspect. Biol. 2, a003236 (2010). https://doi.org/10.1101/cshperspect.a00323635. M. M. Moasser, “ The oncogene HER2: Its signaling and transforming functions and its role in human cancer pathogenesis,” Oncogene 26, 6469–6487 (2007). https://doi.org/10.1038/sj.onc.1210477 In this study, the MCF10A human breast epithelial cell line and its malignant mutants, MCF10A-HRAS (or MCF10AT1, HRAS overexpression) and MCF10A-HER2 (HER2 overexpression) [Fig. S1(b)], were co-cultured with ECs in a transwell system to simulate the non-contact or contact TC–EC interactions during the metastasis [Fig. 1(a)]. In addition, each type of TCs was used at a high, mid, or low concentration to simulate different degrees of tumorous conditions. The ECs in co-culture were seeded on a ring-shaped micropattern, and their chirality, including the numbers and percentages of CW, CCW, NC rings [Figs. 1(b)–1(e) and S2], the chiral factor [Fig. 1(d)], and the mean circumferential angle of cell alignment on the micropattern [Fig. 1(f)], were calculated using a custom-written MATLAB (MathWorks) program as described previously.1818. L. Q. Wan et al., “ Micropatterned mammalian cells exhibit phenotype-specific left-right asymmetry,” Proc. Natl. Acad. Sci. U. S. A. 108, 12295–12300 (2011). https://doi.org/10.1073/pnas.1103834108FIG. 1. (a) Models of co-cultured ECs and TCs used to study the chirality change of ECs caused by TC paracrine signaling and physical contact during extravasation. (b) Top row: typical images of clockwise (CW), non-chiral (NC) micropatterns formed by ECs under the control condition, or counterclockwise (CCW) micropattern under the physical contact of HRAS TCs; bottom row: the color bars, generated by a pre-developed MATLAB program, show the aligning direction of each cell on the micropattern. (Scale bar: 100 μm). (c) The definition of circumferential angle of each bar [blue bar, cell alignment in (b)] deviated from the circumferential direction (red line) and its relationship to chirality. (d) Numbers of the CW, CCW, and NC rings analyzed in each group and the corresponding chiral factors of ECs treated by paracrine signaling or physical contact of TCs. Chiral factor: defined by comparing ring numbers from the formula: (number of CW rings − number of CCW rings)/number of total rings, with CF = +1 standing for complete CW and CF = −1 for complete CCW. (e) Percentage of CW, NC, and CCW rings of ECs from table (d). (f) Mean circumferential angle of ECs on the ring-shaped micropatterns [illustrated in (c)]. The “low,” “mid,” and “high” in figures. (d)–(f) represent the TC densities (20, 100, and 200k/ml) introduced to ECs in the different co-culture groups. Data in (f) are represented as mean ± SD. Under the control condition, 633 rings of ECs were analyzed; under the non-contact condition of MCF10A cells, 50 (low), 44 (mid), and 84 (high) rings were analyzed; under the contact condition of MCF10A cells, 62 (low), 67 (mid), and 76 (high) rings were analyzed; under the non-contact condition of HRAS cells, 115 (low), 114 (mid), and 134 (high) rings were analyzed; under contact condition of HRAS cells, 64 (low), 69 (mid), and 65 (high) rings were analyzed; under the non-contact condition of HER2 cells, 24 (low), 52 (mid), and 35 (high) rings were analyzed; under contact condition of HER2 cells, 70 (low), 74 (mid), and 78 (high) rings were analyzed. “*” in (f) indicates significant differences between the experimental group and control; “*” or “#” indicates P < 0.05; “**” or “##” indicates P <0.01; and “***” or “###” indicates P <0.001 by one-way analyses of variance (ANOVAs) with the Tukey method between groups.
Normally, human umbilical vein endothelial cells (hUVECs) are strongly CW dominant with a chiral factor above 0.90 [Figs. 1(d) and 1(e)], consistent with previous reports.3131. J. Fan et al., “ Cell chirality regulates intercellular junctions and endothelial permeability,” Sci. Adv. 4, eaat2111 (2018). https://doi.org/10.1126/sciadv.aat2111 Non-contacting co-culture with wildtype MCF10A and HRAS cells did not change the strong CW chirality of ECs, while increasing concentration of HER2 cells induces a 27% decrease in CW chirality of EC shown as the chiral factor decreasing from 0.95 to 0.69 [Figs. 1(d) and 1(e)]. For TC–EC contact groups, the ECs formed notable NC and CCW rings when co-cultured with TCs. In addition, their chiral factor is decreased to 0.83 when co-cultured with MCF10A, and down to a 0.60 level with HRAS or HER2 cells [Figs. 1(d) and 1(e)]. The results from the mean circumferential angle give details about the chiral alignment of cells on the micropattern [Fig. 1(f)]. In both TC–EC non-contacting groups and contacting groups, it shows an overall similar decreasing trend along with the increase in TC density. [Note that the circumferential angle for CW cell alignment is negative in Fig. 1(f).] Under non-contacting co-culture, it requires a high TC concentration to induce a significant decrease in the chiral alignment of ECs; however, under TC–EC contacted co-culture, low density of HRAS or HER2 overexpression cells was adequate to make differences [Fig. 1(f)], but not the wildtype. Particularly for HRAS cells, the physical contact caused a significantly stronger interruption of the CW chirality of ECs than paracrine signaling.Lung is one of the most common target organs for metastasis of malignant breast cancer.3636. W. Chen, A. D. Hoffmann, H. Liu, and X. Liu, “ Organotropism: New insights into molecular mechanisms of breast cancer metastasis,” npj Precis. Oncol. 2, 4 (2018). https://doi.org/10.1038/s41698-018-0047-0 Therefore, we also examined the effects of breast cancer cells on the chirality of the human lung microvascular endothelial cells (hLMVECs). Under the normal condition, the hLMVECs exhibit a strong CW chirality with a chiral factor of 0.88 [Fig. S3(a)], comparable to 0.95 of the hUVECs. Co-cultured with different TCs, the hLMVECs exhibit quite similar patterns of CW chirality weakening [Figs. S3(a) and S3(b)]. Particularly under the physical contact of the HRAS and HER2 cells, the chiral factors of hLMVECs dropped to 0.66 and 0.50 [Fig. S3(a)], along with significant decreases in the mean circumferential angles [Fig. S3(c)], again consistent with the observations from hUVECs. These together suggest the TCs could cause the weakness of the CW chirality of ECs by interacting with them under a physical contact condition. This effect could be significantly enhanced by the oncogenic expression of HRAS and HER2, which may contribute to their malignancy in metastasis.Tumor secretion plays a limited role in endothelial cell chirality
In the TC–EC contact condition, the cell interactions are not only from physical bindings but also from the paracrine signaling of TCs simultaneously. In addition, the TC–EC communications are bi-directional even without the physical contact of the two cell types.99. D. Quail and J. Joyce, “ Microenvironmental regulation of tumor progression and metastasis,” Nat. Med. 19, 1423–1437 (2013). https://doi.org/10.1038/nm.3394 It means TC secretions are modifying ECs, while EC secretions are simultaneously stimulating TCs in a cycle. Therefore, we further investigated the EC chirality changes in the TC conditioned medium (CM) to understand whether the weakness of EC chirality results from inherent tumor secretions which do not require the EC induction [Fig. 2(a)]. When cultured in TC conditioned medium, the ECs formed an increased percentage of NC rings, and their chiral cell alignment was significantly decreased compared with control [Figs. 2(b)–2(d)]. However, the decreases of either overall EC chiral factors or the chiral cell alignment caused by the inherent tumor secretions are not comparable with the physical contact groups in Figs. 1(d)–1(f). In addition, neither HRAS nor HER2 conditioned medium induces significantly weaker CW chiral alignment of ECs compared with the wildtype [Fig. 2(d)]. Considering the highest density of TCs was used to generate this conditioned medium, therefore, the inherent secretions of TCs play a limited role in EC chirality change under the TC–EC physical contact condition.FIG. 2. (a) Schematic representation of testing the chirality of ECs treated by TC conditioned medium. (b) Numbers of the CW, CCW, and NC rings and the corresponding chiral factors of ECs treated by TC conditioned medium. “CM” is an abbreviation for “conditioned medium” in group names. (c) Percentage of CW, NC, and CCW rings of ECs from table (b), showing the increased non-chiral chirality of ECs treated with HRAS or HER2 TC conditioned medium. (d) Mean circumferential angle of ECs on the ring-shaped micropatterns, which, however, does not show significant differences. Data are represented as mean ± SD. Under normal medium, 633 rings of ECs were analyzed; under MCF10A, HRAS, or HER2 cell conditioned medium, mean circumferential angles of ECs from 89, 102, or 117 rings were analyzed, respectively. (e) Schematic diagram of endothelial monolayer permeability measurement using a transwell model. (f) Permeability of the hUVEC monolayer with TC conditioned medium or TC physical contact. Data are presented as mean ± SD (n = 4). “*” in (d) and (f) indicates significant differences between the experimental group and control; “*” or “#” indicates P <0.05; “**” or “##” indicates P <0.01; and “***” or “###” indicates P <0.001 by one-way analyses of variance (ANOVAs) with the Tukey method between groups. “n.s.” stands for no significant difference.
The cell chirality is closely associated with the integrity of the EC monolayer. Its permeability increases with the weakening of CW chirality and peaks when ECs become completely non-chiral.3131. J. Fan et al., “ Cell chirality regulates intercellular junctions and endothelial permeability,” Sci. Adv. 4, eaat2111 (2018). https://doi.org/10.1126/sciadv.aat2111 The results from a permeability assay show that the TC conditioned medium is not sufficient to induce a significant change in EC permeability [Figs. 2(e) and 2(f)], suggesting the limited influence of TC secretions on the EC chirality. On the other hand, in the physical TC–EC co-culture groups where EC chirality was weakened, significant increases in EC permeability were also observed, as expected [Fig. 2(f)].Physical contact with tumor cells disrupts the chiral alignments of endothelial cells
The integrity of cell–cell junctions plays an important role in the collective migration of planar cells to form multicellular chiral alignment on the micropattern.2525. K. E. Worley, D. Shieh, and L. Q. Wan, “ Inhibition of cell-cell adhesion impairs directional epithelial migration on micropatterned surfaces,” Integr. Biol. 7, 580–590 (2015). https://doi.org/10.1039/c5ib00073d Previously, we have found significant local misalignment of ECs occurred on the ring-shaped micropattern when they became non-chiral.3131. J. Fan et al., “ Cell chirality regulates intercellular junctions and endothelial permeability,” Sci. Adv. 4, eaat2111 (2018). https://doi.org/10.1126/sciadv.aat2111 In this study, it is shown in fluorescent images that the ECs form a well-arranged monolayer on the micropatterns with intact cell–cell junctions between adjacent cells under the control condition (Fig. 3). With only tumor secretion in TC–EC non-contact groups, the ECs kept relatively intact and continuous cell–cell junctions, while the endothelial monolayer is significantly interrupted with physical involvement of TCs, and misalignment of ECs occurs at the boundary of TCs, suggesting the ECs with direct contact with TCs may possess inconsistent cell chirality.Physical contact of tumor cells compromises the left–right biases of endothelial cells
The individual cell chirality can be reflected by the biased positioning of cell organelles.24,27,29,3124. J. Xu et al., “ Polarity reveals intrinsic cell chirality,” Proc. Natl. Acad. Sci. U. S. A. 104, 9296–9300 (2007). https://doi.org/10.1073/pnas.070315310427. K. Taniguchi et al., “ Chirality in planar cell shape contributes to left-right asymmetric epithelial morphogenesis,” Science 333, 339–341 (2011). https://doi.org/10.1126/science.120094029. P. Ray et al., “ Intrinsic cellular chirality regulates left-right symmetry breaking during cardiac looping,” Proc. Natl. Acad. Sci. U. S. A. 115, E11568–E11577 (2018). https://doi.org/10.1073/pnas.180805211531. J. Fan et al., “ Cell chirality regulates intercellular junctions and endothelial permeability,” Sci. Adv. 4, eaat2111 (2018). https://doi.org/10.1126/sciadv.aat2111 We have demonstrated CW rings contain mainly right-biased cells, while CCW rings contain significantly more left-biased cells.2828. J. Fan, H. Zhang, T. Rahman, D. N. Stanton, and L. Q. Wan, “ Cell organelle-based analysis of cell chirality,” Commun. Integr. Biol. 12, 78–81 (2019). https://doi.org/10.1080/19420889.2019.1605277 Each fluorescent image of micropatterned EC monolayer was further segmented into cell borders, cell centroids, nuclear centroids, and centrosomes as described in Methods section [Figs. 4(a)–4(d)]. The left–right biases of individual ECs were then determined by the positional bias of the cell centroid relative to the nucleus-centrosome axis to study the role of TC physical contact in EC chirality [Figs. 4(e) and 4(f)]. The ECs spaced from TCs (without direct contact with TCs) show a significant right bias and a positive chiral factor which is consistent with their CW chirality [Figs. 4(g) and 4(h)]. A neighboring relationship with MCF10A cells did not affect the right bias of ECs; however, those ECs neighbored with HRAS or HER2 cells are lack of left–right bias, and their chiral factors change into negative values close to zero, indicating a slight reversal of the intrinsic chiral bias of ECs [Figs. 4(g) and 4(h)]. These results together suggest the chirality of ECs is significantly altered by the physical contact of malignant TCs with HRAS and HER2 overexpression.FIG. 4. (a) Immunofluorescence of ECs on a ring-shaped micropattern showing cell adherens junctions (VE-cad, red), cell nuclei (DAPI, blue), and centrosomes (pericentrin, green). (b) Cell borders segmented from the red channel in (a), shown with the calculated cell centroids (yellow). (c) Cell nuclei (blue) segmented from the blue channel in (a), shown with nuclear centroids (cyan). (d) Merged image for cell bias analysis, including cell borders (red), centrosomes (green), nuclear centroids (blue), and cell centroids (yellow). (e) A schematic diagram of determination of the left (L) or right (R) cell bias according to the positioning of the cell centroid relative to the nucleus-centrosome vector. (f) Color-coded cells by their biases on the micropattern. (g) and (h) L–R biases and the corresponding chiral factors of ECs with or without direct TC contact. Data are represented as mean ± SE. In the MCF10A-WT group, 16 rings and 1222 TC-spaced ECs were analyzed; 16 rings and 382 TC-neighbored ECs were analyzed; in the MCF10A-HRAS group, 15 rings and 1094 TC-spaced ECs were analyzed; 15 rings and 327 TC-neighbored ECs were analyzed; in MCF10A-HER2 group, 15 rings and 1417 TC-spaced ECs were analyzed; 15 rings and 325 TC-neighbored ECs were analyzed. “*” indicates P <0.05; “**” indicates P <0.01; and “***” indicates P <0.001 by student t-test. “n.s.” stands for no significant difference. Scale bar in (a): 100 μm.
TC–EC physical contact promotes the binding of CD44 and E-selectin, activates PKCα, and induces pseudopodial movement of EC toward TC
CD44 is a multifunctional cell surface adhesion receptor that is commonly expressed in MCF10A cell series.37–4037. L. T. Senbanjo and M. A. Chellaiah, “ CD44: A multifunctional cell surface adhesion receptor is a regulator of progression and metastasis of cancer cells,” Front. Cell Dev. Biol. 5, 00018 (2017). https://doi.org/10.3389/fcell.2017.0001838. C. Sheridan et al., “ CD44+/CD24− breast cancer cells exhibit enhanced invasive properties: An early step necessary for metastasis,” Breast Cancer Res. 8, R59 (2006). https://doi.org/10.1186/bcr161039. C. M. Fillmore and C. Kuperwasser, “ Human breast cancer cell lines contain stem-like cells that self-renew, give rise to phenotypically diverse progeny and survive chemotherapy,” Breast Cancer Res. 10, R25 (2008). https://doi.org/10.1186/bcr198240. J. Y. So, H. J. Lee, P. Kramata, A. Minden, and N. Suh, “ Differential expression of key signaling proteins in MCF10 cell lines, a human breast cancer progression model,” Mol. Cell. Pharmacol. 4, 31–40 (2012). It has been reported that CD44 binding to its EC surface ligands, E-selectin, activates the PKCα signaling in ECs.3232. P. Zhang, C. Goodrich, C. Fu, and C. Dong, “ Melanoma upregulates ICAM-1 expression on endothelial cells through engagement of tumor CD44 with endothelial E-selectin and activation of a PKCα–p38-SP-1 pathway,” FASEB J. 28, 4591–4609 (2014). https://doi.org/10.1096/fj.11-202747 We have demonstrated previously that PKCα signaling is involved in EC chirality regulation.3131. J. Fan et al., “ Cell chirality regulates intercellular junctions and endothelial permeability,” Sci. Adv. 4, eaat2111 (2018). https://doi.org/10.1126/sciadv.aat2111 In this study, we found a significant elevation of phosphorylated PKCα in ECs when co-cultured with HRAS or HER2 overexpressed TCs rather than the wild type, but the PKCα was inactivated by TC secretions, suggested by undetectable pPKCα bands in any groups of the TC conditioned medium [Fig. 5(a)]. Thus, PKCα activation in ECs requires the physical contact of TCs during the interactions.DISCUSSION
Section:
ChooseTop of pageABSTRACTINTRODUCTIONRESULTSDISCUSSION <<METHODSSUPPLEMENTARY MATERIAL
Comments (0)