Light Activates Cdc42-Mediated Needle-Shaped Filopodia Formation via the Integration of Small GTPases

Abstract

Introduction

Cdc42 has been linked to multiple human cancers and is implicated in the migration of cancer cells. Cdc42 could be activated via biochemical and biophysical factors in tumor microenvironment, the precise control of Cdc42 was essential to determine its role to cell behaviors. Needle-shaped protrusions (filopodia) could sense the extracellular biochemical cues and pave the path for cell movement, which was a key structure involved in the regulation of cancer cell motility.

Methods

We used the photoactivatable Cdc42 to elucidate the breast cancer cell protrusions, the mutation of Cdc42 was to confirm the optogenetic results. We also inhibit the Cdc42, Rac or Rho respectively by the corresponding inhibitors.

Results

We identified that the activation of Cdc42 by light could greatly enhance the formation of filopodia, which was positive for the contribution of cell movement. The expression of Cdc42 active form Cdc42-Q61L in cells resulted in the longer and more filopodia while the Cdc42 inactive form Cdc42-T17N were with the shorter and less filopodia. Moreover, the inhibition of Cdc42, Rac or Rho all significantly reduced the filopodia numbers and length in the co-expression of Cdc42-Q61L, which showed that the integration of small GTPases was necessary in the formation of filopodia. Furthermore, photoactivation of Cdc42 failed to enhance the filopodia formation with the inhibition of Rac or Rho. However, with the inhibition of Cdc42, the photoactivation of Cdc42 could partially recover back the filopodia formations, which indicated that the integration of small GTPases was key for the filopodia formations.

Conclusions

Our work highlights that light activates Cdc42 is sufficient to promote filopodia formation without the destructive structures of small GTPases, it not only points out the novel technique to determine cell structure formations but also provides the experimental basis for the efficient small GTPases-based anti-cancer strategies.

Introduction

Tumor is a persistent disease threatening human health, and its metastasis is the main reason for the poor prognosis of patients. Cancer cell motility is controlled by the activity of protrusive and contractile forces, which is mainly generated by actin filaments. The actin-based protrusions are structures such as lamellipodia, filopodia and plasma membrane blebs.²¹ The actin filaments sense and respond to the tumor microenvironment and regulate the cell behaviors, however, how does the actin filament response and translate signal to different protrusions engaged in migration is unknown. As the driving force for cell motility,¹⁹ lamellipodia are detected during cell migration arising at the leading cell edge.² In the lamellipodia at the front, multiple signal pathways can be activated such as the accumulation of motor proteins myosin. Moreover, the activation of myosin regulatory light chain also contributes to the detachment of the rear part of the cell.¹⁴ Filopodia are needle-shaped structures that are dynamic, narrow and sharp.^{4, 10} The roles of filopodia are multiple, such as the transferring structure for cell–matrix interaction and participate in the cancer cell movement. Due to its special structure, it can sense the chemical signals from extracellular, as well as transmit the signal to cells around. Filopodia generally participating in various biological processes, such as cell migration, wound closure, etc.^{3, 12, 13} Although the contributions of lamellipodia are relatively clear, further research is needed on how filopodia are formed and how does the needle-shaped structure affect the cancer cell behavior and its molecular mechanism.

Cancer cell motility are actin based, they are regulated by Rho-GTPse family, they serve as oncogenes in several human cancers and are involved in cancer cell invasion.^{5, 17} Small-GTPases are composed of Rho, Rac and Cdc42, in which Rac/Rho are positive associated with the EMT and cell malignancy.²⁶ All the Rho-GTPases are with an inactive GDP state and active GTP bound state.⁸ Compare with the Rho/Rac contributions to lamellipodia, Cdc42 might be closely associated with filopodia.⁴ The lamellipodia formation are with the fast-growing of “barbed ends” and slow growing of “pointed ends”.¹⁸ This kind of structure can modulate the cell adhesions and cytoskeleton reorganization. Meanwhile, Cdc42 are recruited after the cytoskeleton reorganization. However, inhibition of Cdc42 cannot block the lamellipodia formations.⁷ Therefore, Cdc42-mediated cancer cell migration might be different from the other two small GTPases. Cdc42 overexpression is mainly mediated by cell surface receptors,^{1, 22, 23} which functions as an EGFR-signaling regulator. Cdc42 also positively involved in cell morphology and polarity,^{6, 28} aberrant Cdc42 activity directly affect the tumor metastasis in cancer cell migration.²⁰ Moreover, Cdc42 could regulate single cell polarity by microtubule-based intracellular vesicle trafficking.²³ Therefore, how Cdc42-triggered cytoskeleton remolding have yet to be fully elucidated. Cdc42 might be closely associated with the needle-shaped structure and initiate the sense of tumor cell to the extracellular and the signal transmissions between cells. Hence, targeting Cdc42 represents a promising strategy for precise cancer therapy.

Optogenetics is a biological technique that combines the genetic and optical method. The precise spatio-temporal controls of protein activity are a promising method for cell tracking and cell behavior studies. There are multiple ways for light control, either by activation or inhibition. Light can induce the protein associations by light, which induce the intracellular signals. Meanwhile, by controlling the gene expression or clustered-base activation, the signaling cascade could be activated with the photosensitive protein LOV, PHYB or CRY2. Inhibition can be also conduct by optogenetics as the sequestration may lead to the loss of function of proteins. The conformational changes of proteins are another key strategy for protein or signaling activation.²⁵ This is based on the conjunction with LOV domains to regulate formins, Cdc42 and Rac in control of cell migration.^{15, 29} Zhou et al. also used the optogentics to regulate Cdc42 and proteinase K activity in conjunction with Dronpa.³⁰ By using the genetically encoded photoactivatable derivative of Cdc42, it is possible to control the activities of Cdc42 within seconds.

In this study, we investigated whether photoactivation of Cdc42 could promote breast cancer cell needle-shaped structure formation. We demonstrated that the activation of Cdc42 significantly enhanced the filopodia formation both in the optogenetics and gene manipulation ways. The integration of small GTPases were necessary for the filopodia formation as the inhibition of either of Cdc42, Rac or Rho failed to accumulate filopodia on the membrane. These findings provide valuable evidence for Cdc42 roles during the protrusion formations and confirm the novel structure by spatial–temporal control of protein activities within living cells.

Materials and Methods

Reagents and Antibodies

Cell culture medium of L15, penicillin, streptomycin and fetal bovine serum (FBS) was purchased from Gibco (Grand Island, NY, USA). Phalloidin was purchased from Sigma-Aldrich (Saint Louis, MO, USA). Lipofectamine LTX were purchased from Thermo Fisher Scientific (Waltham, MA, USA), DAPI was purchased from Beyotime Biotechnology (Shanghai, CHINA). Rhosin and ML141 were purchased from MedChemExpress (Monmouth Junction, NJ, USA). NSC23766 was purchased from Selleck (Houston, TX, USA). All other reagents were used as received without additional purification unless otherwise noted.

Cell Culture

The triple negative human breast cancer cell MDA-MB-231 was obtained from the American Type Culture Collection (ATCC, Manassas, VA, USA). Cells were cultured in L15 medium composed of 10% fetal bovine serum and penicillin (100 U/mL) and streptomycin (100 μg/mL) at 37 °C. The culture media were replaced every 3 days. Confluent cells were trypsinized with 0.25% trypsin in 0.02% EDTA (Invitrogen, Carlsbad, USA) and resuspended in the culture medium.

Inhibition Treatment

MDA-MB-231 cells were treated with the selective inhibitor of Rho Rhosin, which is an effective and specific RhoA subfamily Rho GTPases inhibitor. It can specifically bind RhoA and inhibit RhoA GEF interaction without interact with Cdc42, Rac1 or GEF. Ml141 is a highly efficient, allosteric, selective and reversible Cdc42 GTPase non-competitive inhibitor. NSC 23766 is a RAC GTPase inhibitor. It targets Rac activation through guanine nucleotide factors (GEFs). It does not inhibit closely related targets, Cdc42 or RhoA. The ML141, Rhosin and NSC stock solution was further diluted in cell culture medium to generate working concentrations, and DMSO was also used for all control. Prior to the photoactivation treatment, inhibitor was added to the medium 60 min with 100 μM (ML141), 100 μM (NSC23766) and 30 μM (Rhosin), respectively, and then the images were acquired for filopodia analysis.

Plasmids and Transfection

Plasmides encoding Cdc42-WT, PA-Cdc42 was a gift from Klaus Hahn (Addgene plasmid # 12599, Addgene plasmid # 75263), Cdc42-T17N and Cdc42-Q61L were gifts from Gary Bokoch Addgene (plasmid # 12976, Addgene plasmid # 12986, respectively). Cells were trypsinized and quantified before plating on a glass. Cell density should be 50–80% confluent on the day of transfection. Cells were transfected with Lipofetamine LTX (Thermofisher, USA) according to the manufacturer’s protocol.

Immunofluorescence Staining

For the preparation of immunofluorescence staining, the cells were washed with PBS and then fixed with 4% paraformaldehyde in PBS for 15 min. After rinsing with PBS for three times, the cells were permeabilized by treatment with 0.1% Triton X-100 in PBS and rinsed with PBS, followed by incubation in 1% BSA (bovine serum albumin) in PBS for 0.5 h. Phalloidin was used according to the manufacturer’s protocol. The nuclei were stained with DAPI for 5 min. The fluorescence of the cells was imaged using a confocal laser scanning microscope (FV1000, Olympus, Japan). Optical density was used to measure the contents of specific cellular components. Fiji graphical processing software was used to quantitatively analyze the length of filopodia, and the cells with filopodia longer than 2 μm were considered as the filopodia positive cells.

Time Lapse Microscopy

Cultured cells were exposed to time-lapse microscopy. After the transfection, the cells were ready for the photo-excitation. The time lapses were recorded by confocal laser scanning microscope. The dashed regions were exposed to light with the interval of 5 s for 84 times (with the duration of 420 s), images were obtained from several randomly selected fields for each insert after the first light exposure. We used 0, 60, 120, 180, 240, 300, 360, 420 s to analyze the filopodia length and cell area after the photo-excitation.

Statistical Analysis

T test was used in each experiment, and the results for each experiment were repeated at least three times. The comparisons of significant differences between the two groups were detected and collected as means ± SDs. A value of p < 0.001 (**) were considered to be remarkably statistically significant, N.S was considered as no statistically significant between groups.

Results

Photoactivation of Cdc42 Promoted Breast Cancer Cell Needle-Shaped Protrusions’ Length

Needle-shaped protrusions are essential structure for cancer cell movement, both for suspended and adherent cells. After the transfection of photoactivatable Cdc42, membrane Cdc42 was detected around the cells. We selected a specific region to light, and Fig. 1 showed that the filopodia began to grow and increased significantly within minutes after the excitation, suggesting that Cdc42 by optogenetic activation promoted the formation of filopodia in breast cancer cells. We also examined the filopodia length, there was a significantly longer filopodia after photoexcitation compared with before photoexcitation. However, the cell areas were stable during the photo-excitation times, which indicated that the activation of Cdc42 by optogenetic way could enhance the formation of filopodia.

Figure 1.

Photo-excitation of Cdc42 promoted needle-shaped structure formation of breast cancer cell. (a) Time lapse images of Cdc42 (B&W) from 0 to 420 s, dashed area is the region that are exposed to light for the excitation, and the fluorescence is green embedded in the plasmid, which can be used to mark the membrane filopodia. (scale bar = 10 μm). (b) Region of interest at 0 and 420 s for the filopodia (scale bar = 5 μm). (c, d) Quantitative analysis of cells’ filopodia length in one cell and cells’ areas (N = 8) before and after photo-excitation. Data are expressed as the mean ± SD, N.S. means no significant difference, while **p < 0.001 means significant difference.

Gene Manipulation of Cdc42 Controls the Filopodia Formations

Optogenetics determines the effect of Cdc42 to filopodia in a short-time way, we further conduct the gene manipulation way to control the activity of Cdc42. The transfection of active mutant of Cdc42 (Cdc42-Q61L) showed a more and longer filopodia compared with the control group, meanwhile, the inactivation mutant of Cdc42 (Cdc42-T17N) failed to promote the filopodia formation and even cause the loss of filopodia both in numbers and length (Fig. 2).Similar stable cell areas were detected in these conditions, which indicated the role Cdc42 in the formation of filopodia in breast cancer cells.

Figure 2.

Cdc42 mutants regulate filopodia formations. (a) Immunofluorescence of F-actin (red), DAPI (blue) and Cdc42 (Green) in different Cdc42 genetic background (Cdc42 wild type, Cdc42 Q61L constitutively active mutant and Cdc42 T17N dominant negative mutant). Dashed regions are the ROI (Region of interest). (scale bar = 10 μm). (b) Quantitative analysis of cells’ with filopodia (N = 15), filopodia length in one cell and cell areas (N = 15) in different genetic background. Data are expressed as the mean ± SD, N.S. means no significant difference, while **p < 0.001 means significant difference.

The Inhibition of Small GTPases Failed to Accumulate Filopodia

Small GTPases serve as oncogenes in human cancers by affecting cytoskeletal dynamics. We further identified filopodia formation upon different treatments. We used different inhibitors of small GTPases Cdc42, Rac and Rho. Figure 3 showed that compare with control which the filopodia is not prominent, the expression of Cdc42-Q61L treatment group showed a more and longer filipodia. Meanwhile, the addition of Cdc42 inhibtor ML141, Rac inhibitor NSC or Rho inhibitor Rhosin all failed to accumulate filopodia with cells, which suggested the either of them was indispensable to accelerate the filopodia formation. The stable cell areas also indicated the incoherence between cell area and filopodia formation.

Figure 3.

The inhibition of small GTPases failed to rescue the filopodia formation. (a) Immunofluorescence of F-actin (red), DAPI (blue) and Cdc42 (Green) in different treatments, (scale bar = 10 μm). (b) Quantitative percentage analysis of cells with filopodia (N = 15), filopodia length in one cell and cell areas (N = 15) in different treatments. Data are expressed as the mean ± SD, N.S. means no significant difference, while **p < 0.001 means significant difference.

Co-expression of Photoactivatable Cdc42 and Small GTPases Inhibition Exhibited the Indispensable Integration of Small GTPases for Needle-Shaped Structure Formation

We concluded that the inhibition of small GTPases could block breast cancer cell filopodia formation. We further rescue back the Cdc42 by optogenetics to elucidate the role of small GTPases. We identified that co-expression of Cdc42 activation by photoexcitation with Rac or Rho inhibition failed to rescue the filopodia in breast cancer cells. However, the cells co-expression of Cdc42 activation by photoexcitation with Cdc42 inhibition by ML141 largely rescued back the filopodia formation. The results means that the activity of Cdc42 directly controls the formation of needle-shaped protrusions without the change of cell areas (Fig. 4). Cdc42 act as the major receptor for external display of cancer cell malignant progression, which is typically based on the integrity of small GTPases.

Figure 4.

Photo-excitation of Cdc42 solely rescued filopodia formation in ML141 treatment. (a) Time lapse images of Cdc42 (B&W) from 0 to 420 s, dashed area is the region that are exposed to light for the excitation and the fluorescence is green embedded in the plasmid, which can be used to mark the membrane filopodia. (scale bar = 10 μm). Region of interest at 420 s for the filopodia (scale bar = 5 μm). (b, c) Quantitative analysis of cells’ filopodia length in one cell and cells’ areas (N = 7) before and after photo-excitation under the ML141, NSC23766 or Rhosin treatments. Data are expressed as the mean ± SD, N.S. means no significant difference, while **p < 0.001 means significant difference.

The Filopodia is Positively Linked with Cell Migration

We performed the wound healing assay to check the migration ability of cells, and we found that the over-expression of Cdc42 resulted in a longer migration distance, while the inhibition of either of the small GTPases blocked the cell migration, which indicated the indispensable role of small GTPases for cell motility (Figs. 5a and 5b). We further identified the small GTPases inhibitor effect to Rho, Rac and Rho. We found that the addition of ML141, NSC, Rhosin mainly decreased Cdc42, Rac and Rho, respectively (Figs. 5c and 5d).

Figure 5.

Small GTPases are essential for the cell migrations. (a) Images of cell migration distances at 0 and 4 h in control, Cdc42-Q61L, Cdc42-Q61L + ML141, Cdc42-Q61L + NSC, Cdc42-Q61L + Rhosin treatments groups (scale bar = 50 μm). (b) Quantitative analysis of cells migration distance in different treatments. N = 6 and data are expressed as the mean ± SD. (c) Representative images of Cdc42, Rac1 and RhoABC expressions in control, + ML141, + NSC, + Rhosin groups (scale bar = 10 μm). (d) Quantitative analysis of relative small GTPases intensity in different treatments. N = 12 cells and data are expressed as the mean ± SD. **p < 0.001 means significant difference.

Discussion

Here, we have demonstrated that the activation of Cdc42 by light governs the formation of filopodia in the MDA-MB-321 breast cancer cells. Filopodia are connected between the cells that act as the messenger to transmit the signal among cells. The specific structure also insets in the neighboring cells, which allows the supracellular cytoskeleton formations in all cells. We also identified that Cdc42 controls the formation of filopodia by optogenetic way in a short time, which provides a mechanism of Cdc42 that is associated with cancer cell motility.

Because of the needle-shaped structure, filopodia senses the chemical and mechanical environment in a structure-based way. However, whether Cdc42 is the direct upstream of the structure is not clear. Cdc42 is the member of small GTPases and could be hyperactivated by switching the GDP/GTP phases.¹¹ Although it is well-known that Cdc42 could affect the filopida, however, the genetic manipulation lacks the direct identification and may have some post-translational effect of proteins. Cdc42 also coordinates microtubule and actin interaction, which indicates the roles of Cdc42 are essential for the cytoskeleton network. The exact role of Cdc42 towards microtubule and actin needs further investigations. The downstream of Cdc42 is also a complex cascade for the biological behaviors. LIMK is known to be the downstream effector of Cdc42.²⁴ WASP and N-WASP are also involved in the formation of filopodia, which could binds to activate the Arp2/3 complex.¹⁶ However, the exact role of Cdc42 and its downstream effectors still needs further investigations to identify the path for biological responses.

Our work also shows that the roles of small GTPases are integrated. The rescue of Cdc42 inhibition will recover back of the filopodia formation, while the inhibition of either Rho or Rac fails to rescue filopodia formation by Cdc42 activation. Although the functions of small GTPases are diverse, the integration of them is necessary for the biological responses. Our findings supports the hypothesis of the crosstalk among the small GTPases, the blockage of either of them will lead to the failure of filopodia formation, and the formation is positively linked with cancer cell malignancy. Small GTPases are all involved in cell adhesion, migration etc., they are all associated with cell motility.⁹ The filopodia formation might be the first step that cells sense the microenvironment and begin to progress without control. Therefore, targeting the integration of small GTPases might be a way to inhibit the malignancy of tumor cells, the exact role of small GTPases for the cell malignancy initiation needs more ways to confirm. Although the inhibition of one type of small GTPases will lead to the slightly decrease of the other two small GTPases, the tendency of the specifical small GTPases inhibition is maintained. For cancer cells, the roles of small GTPases might be different, even Cdc42 contributes to most cell for the filopodia formation, whether the small GTPases are all integrated is unknown in this field. We confirmed the interactions in the breast cancer cells, but for general type of cancer cells, whether it is universal or specific-type oriented, this still needs further investigations. For the effectors of small GTPases, the Rho-associated protein kinases (ROCK) and formin mDia1 (mammalian homolog of the Drosophila Diaphanous protein) are the two Rho effectors. ROCK is divided as ROCK1 and ROCK2, which shared 65% overall homology and 92% homology in the kinase domain.²⁷ Formins regulates actin and microtubule to control cell functions. Serum response factor (SRF) transcriptional activity also involved in the formin activities. Although we give the overview that the integration of small GTPases are important for the filopidia formation, more investigations are needed for more physiological and pathological processes.

The precise spatio-temporal control of protein activity allows us to directly see the effect of Cdc42. Given the substantial structure similarity of Rac1 and Cdc42, the LOV domain is used to cage Cdc42. The irradiation of Cdc42 promotes the filopodia formation, which shows that the technique can be used for engineering caged Cdc42 GTPase. It is reported that localized Rac activation or deactivation was sufficient to generate polarized cell movement.²⁹ The results sheds light on the optogentics to control cell proteins in a timely manner. Although the optogenetic ways for protein activation or deactivation are diverse, it is necessary to first identify some protein roles and then expand for more tagged proteins or kinases. The non-invasive method to modify small GTPases activity provides us the feasibility for signal pathway in a spatio-temporal resolution. It would be interesting to check whether and how the optogentic tools can be used in a broad range of tagged proteins, which will facilitate our understanding of signal transductions in development processes of living organisms.

Conclusion

We elucidated Cdc42-induced cancer cell needle-shaped structure formation by both optogenetics and gene manipulation. Cdc42 greatly enhanced the filopodia numbers and length upon activation in a spatio-temporal manner. Further, Cdc42 alone was not enough to initiate filopodia formation, the combination effect of Cdc42 with Rac and Rho would finally start the enhancement of the needle-shaped structures. The integration of small GTPases is essential for filopodia formation, as the direct (Cdc42) or indirect (Rho or Rac) block both leads to the failure of filopodia formation, which makes the contribution to the complex interactions among small GTPases. In addition, the application of optogenetics to control protein spatial–temporal dynamics suggests the novel technique for exploring cell structures and behaviors, thus accelerating anti-cancer research.