TGT surface is an innovative platform to study growth factor-integrin crosstalk. The flexible probe design, specificity of the adhesion ligand, and precise modulation of stimulation conditions allow robust quantitative assessments of EGFR-integrin interplay. The results highlight EGFR as a 'mechano-organizer' tuning integrin mechanics, influencing focal adhesion assembly and cell spreading.
Multicellular organisms rely on interactions between membrane receptors and cognate ligands in the surrounding extracellular matrix (ECM) to orchestrate multiple functions, including adhesion, proliferation, migration, and differentiation. Mechanical forces can be transmitted from the cell via the adhesion receptor integrin to ligands in the ECM. The amount and spatial organization of these cell-generated forces can be modulated by growth factor receptors, including epidermal growth factor receptor (EGFR). The tools currently available to quantify crosstalk-mediated changes in cell mechanics and relate them to focal adhesions, cellular morphology, and signaling are limited. DNA-based molecular force sensors known as tension gauge tethers (TGTs) have been employed to quantify these changes. TGT probes are unique in their ability to both modulate the underlying force threshold and report piconewton scale receptor forces across the entire adherent cell surface at diffraction-limited spatial resolution. The TGT probes used here rely on the irreversible dissociation of a DNA duplex by receptor-ligand forces that generate a fluorescent signal. This allows quantification of the cumulative integrin tension (force history) of the cell. This article describes a protocol employing TGTs to study the impact of EGFR on integrin mechanics and adhesion formation. The assembly of the TGT mechanical sensing platform is systematically detailed and the procedure to image forces, focal adhesions, and cell spreading is outlined. Overall, the ability to modulate the underlying force threshold of the probe, the adhesion ligand, and the type and concentration of growth factor employed for stimulation make this a robust platform for studying the interplay of diverse membrane receptors in regulating integrin-mediated forces.
Cells have the intrinsic ability to sense, generate, and respond to mechanical forces, leading to changes in cellular phenotype and remodeling of the local microenvironment1,2. Forces play a crucial role in regulating many aspects of cell behavior, including adhesion, migration, proliferation, differentiation, and wound healing3,4. Aberrations in the bi-directional mechanical exchange between a cell and the microenvironment can lead to diseased states, including cancer5. Numerous membrane receptors are involved in maintaining cell-matrix homeostasis; of these, integrins and epidermal growth factor receptor (EGFR) have robust synergy6,7. Classically, integrins establish the mechanical link between the microenvironment and intracellular cytoskeleton while EGFR regulates cell growth, proliferation, and survival8,9. EGFR is a highly studied therapeutic target, focused on outside-in regulation facilitating intracellular signaling. EGFR-integrin crosstalk has been established genetically and biochemically to regulate the progression of multiple diseases, including cancer10,11. While studies indicate the existence of EGFR-integrin interplay, the outcomes are attributed to signaling pathways away from the plasma membrane7,12,13,14. The impact of EGFR, or other growth factors, on cell mechanics remains largely unexplored in part due to the lack of tools to measure cellular forces and signaling outcomes. The challenge lies in identifying appropriate tools to study the communication between these parallel signaling paradigms and to quantify their specific contributions to cell mechanics.
Several approaches have been developed to measure forces generated by cell adhesion receptors, and the reader is directed to in-depth reviews of these techniques15,16. Briefly, traction force microscopy and micro-pillar array detection rely on the deformation of an underlying substrate to infer nanonewton (nN) forces, an order of magnitude more than individual receptor forces17,18. Single-molecule techniques, including AFM and optical tweezers, are sensitive to single protein piconewton (pN) forces but measure only one receptor at a time and do not offer good (or any) spatial resolution. DNA-based molecular tension probes and tension gauge tether (TGT) probes offer pN force resolution with diffraction-limited (or better) spatial resolution, giving them a unique role in studying single-cell forces19,20 from diverse cell types, including fibroblasts, cancer cells, platelets, and immune cells21,22,23,24. While molecular tension probes have an extendable "spring" element, ideal for real-time imaging, TGTÂ probes irreversibly rupture, leaving behind a fluorescent "force history". TGTs additionally modulate the tension threshold of the underlying substrate; a series of probes with similar chemical compositions but different rupture forces, or tension tolerances (Ttol), can be used to quantify the minimum tension required for focal adhesion formation and cell spreading. TGT probes consist of two complementary DNA strands, one anchored to the surface and the other presenting a ligand to the cell. If a receptor binds the ligand and exerts a force greater than the Ttol of the probe, the strands will be separated. Ttol is defined as the constant force needed to rupture 50% of the probes in a 2 s interval under ideal conditions. In "turn-on" TGT probes, a quencher on the top strand can be separated from a fluorophore on the bottom strand. Only where the TGT probe has been ruptured, presumably by forces greater than or equal to Ttol, will a fluorescent signal be generated. TGT probes can also be fixed, allowing for easy manipulation of biological systems and testing of multiple conditions. For these reasons, TGT probes were used in this work.
TGT probes were employed to study how integrin-dependent cell adhesion and mechanical forces are modulated by activated EGFR21. This work established EGFR as a 'mechano-organizer', tuning focal adhesion organization and tension generation. Additionally, it was found that EGF stimulation influenced the distribution and maturity of focal adhesions and enhanced cell spreading. This approach could be used in future studies to investigate how growth factors influence mechanical forces in tumor progression and dynamics. While the role of EGFR-integrin crosstalk in regulating the epithelial to mesenchymal transition is established, the role of mechanical forces in this process remains under-explored10.
Here, a detailed protocol is presented for these experiments covering the synthesis and assembly of 56 pN TGT probes, generation of TGT surfaces on glass coverslips, application of Cos-7 cells on the TGT surface and stimulation with EGF, fixation, and staining of cells with phalloidin, and an anti-paxillin antibody, high-resolution total internal reflection fluorescence (TIRF) and reflection interference contrast microscopy (RICM) imaging, and image quantification. This protocol, though written to investigate EGF stimulation of Cos-7 cells, is readily adaptable for many TGT based experiments. Different ligands, Ttol, cell types, stimulation parameters, proteins labeled following fixation, and quantitative analysis can be easily substituted in, making this protocol robust and widely useful.
1. TGT oligonucleotide preparation
NOTE: The details of oligonucleotide probe synthesis are outlined here. Please note some modifications and purification steps can be outsourced for custom synthesis.
2. Surface preparation
Day 1:
Day 2:
3. Cell preparation and staining
4. Image acquisition
5. Data analysis
NOTE: Perform quantitative image analysis using the Fiji software and the analysis using statistics software.
Turn-on TGT probes were used to investigate the effect of ligand-activated epidermal growth factor receptor (EGFR) on integrin-mediated cell mechanics and adhesion formation in Cos-7 cells21. The probes present the ligand cyclic Arg-Gly-Asp-Phe-Lys (cRGDfK)21,23,25,26, which is selective for the integrin heterodimer αVβ3 with only a small affinity for α5β1 integrins27,28,29,30. The TGT probe comprises a duplex DNA functionalized onto a glass coverslip surface via the bottom strand using biotin-streptavidin binding. The top strand displays the integrin ligand and is available to bind to the cognate integrin receptor on the cell membrane (Figure 5A). The bottom strand is labeled with a fluorophore and the top strand with a quencher, leading to minimal background fluorescence when the duplex TGT is intact. If an integrin binds the ligand and applies a force with a magnitude larger than the Ttol of the probe, the DNA duplex will separate leading to fluorescence (Figure 5A). Any TGT probe that has not been ruptured by a mechanical force will remain non-fluorescent. This force selective turn-on fluorescence allows systematic and quantitative mapping of pN scale integrin-generated forces at diffraction-limited resolution. TGT probes additionally modulate the tension threshold of the substrate.
Shown here is a representative example of a TGT surface with a Ttol of 56 pN. Cos-7 cells were plated on this TGT surface with or without EGF stimulation to study the impact of EGFR activation with ligand stimulation on cell adhesion and integrin mechanics (Figure 5A,B). The cells were incubated with or without EGF on the TGT surfaces for 60 mins, fixed, and immuno-stained to display the focal adhesion distribution (paxillin) and the organization of the cytoskeleton (F-actin) (Figure 5B). The cells were then imaged using RICM and TIRF microscopy. As clearly seen in the RICM image, Cos-7 cell spreading on the 56 pN TGT surface was significantly enhanced with EGF stimulation compared to without stimulation. This was quantified for 50 cells in each condition by measuring the size of the cell-substrate contact region from the RICM image (Figure 5C). Stimulation with EGF resulted in a more circular morphology, representative of Cos-7 cells spreading and growing in their natural physiological environment (Figure 5D). The fluorescence from open probes is also higher with EGF stimulation as observed in the tension fluorescence image. The integrated intensity of open probes, which is proportional to the number of open probes, was much higher with EGF stimulation compared to without (Figure 5B,E). This is a representation of all the receptor-ligand engagements where integrins applied a force greater than Ttol (56 pN).
Staining with paxillin showed that the distribution, number, maturation (size), and organization of focal adhesions were also impacted by EGF stimulation. Focal adhesions in EGF stimulated cells appeared more mature and radially oriented compared to no EGF controls. The F-actin cytoskeletal organization was also enhanced with EGF stimulation, as assessed by the phalloidin staining (Figure 5B). These qualitative assessments were made by visual comparison of images from both treatment groups. Quantitative analysis of focal adhesion can be done but is beyond the scope of this protocol. In this experiment, the TGT surface provided a platform to systematically detail the effect of EGFR activation on cell spreading, integrin mechanics, and focal adhesion formation.
Figure 1: Schematic for Day 1 of the TGT surface preparation. (A) Clean the coverslips. (B) Etch the coverslip surface. (C) Neutralize the Piranha solution. (D) Silanize the surface to make reactive amine groups. (E) Equilibrate the coverslips to the organic phase. (F) Dry the coverslips with an inert gas. (G) Biotinylate the surface amine groups. Please click here to view a larger version of this figure.
Figure 2: Schematic for Day 2 of the TGT surface preparation. (A) Clean and dry the coverslips to remove any residual biotin from the day before. (B) Passivate with BSA to prevent non-specific binding of reagent in subsequent steps. (C) Functionalize the coverslips with streptavidin. (D) Hybridize the probes in a thermo cycler. (E) Apply the synthesized probes to the coverslips (F) Assemble the coverslip in the cell imaging chamber. Please click here to view a larger version of this figure.
Figure 3: General workflow highlighting the broad steps across the entire experimental setup. (A) Process for cell detachment and plating on the TGT surface in basal media (DMEM) with or without EGF stimulation. (B) Flowchart of the steps involved in fixation and immuno-staining post-attachment and spreading on the TGT surface. (C) Post-staining, cells are imaged on an inverted fluorescence microscope with RICM and TIRF microscopy. Please click here to view a larger version of this figure.
Figure 4: Example of data processing and quantitative analysis. (A) Step-by-step breakdown of the analysis pipeline employed in Fiji (ImageJ) for RICM and tension image quantification. (B) A representative example for cell morphometric outcomes analyzed using the above pipeline. (C) Representative examples for cell mechanical outcomes analyzed using the above-mentioned pipeline. Please click here to view a larger version of this figure.
Figure 5: Example data from a TGT experiment. (A) Diagram highlighting the contact zone at the cell membrane-TGT surface interface. Inset projects integrins interacting with its cognate ligand cRGDfK with (right) or without (left) EGF stimulation. (B) RICM and TIRF images of Cos-7 cells spread on the 56 pN TGT surface. The images are obtained 60 min post-plating with or without EGF stimulation. Individual RICM (as acquired), integrin tension (grayscale), paxillin (orange hot), and F-actin (blue-green) images are shown with overlays for both stimulation conditions. Scale Bar: 10 μm. The inset highlights a zoomed-in ROI (region of interest) detailing the colocalization of the generated integrin tension at sites of adhesion formation marked by paxillin, and the underlying subcellular cytoskeletal organization marked by actin. Scale Bar: 5 μm. (C-E) Scatter plots for the spread area (RICM cell footprint) (C), circularity (D), and integrated tension (E) for Cos-7 cells with or without EGF stimulation. Bars indicate mean ± s.d. Differences between the groups were assessed statistically with Student's t-test; ****P < 0.0001. n = 50 cells across three independent experiments. Please click here to view a larger version of this figure.
Figure 6: Example TGT surfaces with different possible problems. (A) Tension and RICM images of an ideal TGT surface with assembled probe quenched before cell adhesion. (B) Tension and RICM images of a TGT surface where the TGT probe lacks the top strand (quencher). Tension image shows uniform fluorescence from open fluorophore in the bottom strand. (C) Tension and RICM images for cells spread on an ideal TGT surface. (D) Tension and RICM images for cells spread on a poorly made TGT surface with limited passivation or degraded probe. (E) Tension, RICM, and brightfield images for cells plated on an ideal surface with cRGDfK ligand indicating cRGDfK-integrin interactions are vital for cell attachment and tension generation. (F) Tension, RICM, and brightfield images for cells plated on a surface without cRGDfK ligand on the TGT. While the cells are visible in the brightfield image, no cell attachment or generated integrin tension is observed. Scale Bar: 10 μm. Please click here to view a larger version of this figure.
With the detailed step-by-step procedure outlined above, one can prepare TGT surfaces to quantify cell morphology and integrin tension generated by adherent cells during cell attachment and spreading following treatment with EGF. The straightforward probe design and synthesis and surface preparation along with the simple experimental setup provided a stable platform to study the interaction of EGFR and integrins. Overall, the results validate that ligand-dependent activation of EGFR enhances cell spreading, tunes the force-bearing properties of integrin receptors, and promotes focal adhesion organization and maturation. The results obtained using TGT probes support the overarching hypothesis that growth factors, such as EGFR, act as 'mechano-organizers', increasing the amount and spatial organization of integrin tension and regulating the orientation and mechanics of focal adhesions.
Upon application onto the TGT surface, the cells land, attach, and spread as the integrin (αVβ3) receptors sense and bind to the cRGDfK ligand. In doing so the TGT probes can be mechanically ruptured, generating fluorescence at the site of ligand engagement. The readout is the cumulative "force history" of the cell interacting with the surface. There are some common issues with the TGT surfaces that can be present during these experiments. High surface background fluorescence (Figure 6A,B), patchy surface appearance, failure of the cells to generate tension signal (Figure 6C,D), and failure of cells to spread (Figure 6E,F) may be due to technical shortcomings with the TGT probe or surface synthesis. Solutions to these common issues are presented in Table 1.
The straightforward design of TGT probes provides cell biologists with a powerful tool to study specific growth factor-integrin signaling outcomes in isolation without interference from other cell surface receptors by providing only specific ligands and stimulations. Additionally, TGT probes allow investigation of the tension threshold underlining individual integrin receptors during cell adhesion at pN sensitivity. Alternate approaches fail to report forces exerted by individual receptors with high spatial resolution in fixed samples31. Traction force microscopy is only sensitive to nN forces, an order of magnitude higher than the forces applied by individual integrin receptors15, and molecular tension probes measure pN forces, but because they are reversible, they do not robustly withstand fixation. For these reasons, TGT probes are an attractive tool to study the mechanics of growth factor-integrin interactions.
There are several technical nuances associated with TGT probes that should be considered before designing an experiment. The tension image is a snapshot in time, representing the force history and not an indicator of the receptor-ligand engagements at any given time point. Since signal generation is dependent on probe separation, the TGT fluorescence results from open probes not under active tension from receptor-ligand engagement. This means the readout for integrin tension obtained on the TGT surface is historical and cumulative in nature representing where there were forces larger than Ttol; the locations of current receptor-ligand forces less than Ttol are not reported19,32. Because TGT rupture results in termination of the receptor-ligand engagement, cell spreading is due to integrin-ligand interactions that experience forces lower than Ttol. The user must therefore be careful when defining the time post-plating to estimate the mechanical outcomes associated with integrin-based adhesions. Finally, the meaning of Ttol must be considered. The TGT probes employed here have a Ttol of 56 pN, where Ttol is the constant force needed to rupture 50% of the probes when applied for 2 s. When considering complicated biological systems, TGTs likely experience a heterogeneous and diverse force gradation with varying time dependencies. If TGTs are ruptured by forces larger than Ttol, the fluorescence would be an underestimation of the total tension. Alternatively, forces below Ttol applied for longer durations can rupture a similar number of probes as high threshold forces applied for shorter times. Both these scenarios may result in the same fluorescence intensity readout, making it difficult to resolve the exact tension magnitude or dynamics using TGT probes33,34.
Overall, assessments of integrin tension with growth factor stimulation should be made carefully by designing experiments with internal controls, comparing spreading profiles on other matrix-coated surfaces, making parallel assessments of TGT fluorescence in cells in the presence or absence of growth factor stimulation, and using TGTs with different Ttol. TGTs allow quantification of the role of growth factor signaling in regulating the mechanics of integrin receptors, focal adhesion dynamics, and cell spreading. This protocol can be used as a template for many TGT-based experiments using probes with different Ttol, different ligands, different cell types, or different stimulation conditions. Any proteins of interest can be labeled following fixation, and any type of quantitative image analysis may be implemented. As such, we present a template for numerous TGT experiments.
The use of TGT probes is not limited to studying integrins but can be extended to a diverse array of cell membrane receptors across different cell types by modifying the ligand. TGT probes have been used to investigate the role of forces in regulating various receptor signaling cascades, including identifying the mechanical role of Notch receptor mechanics in embryonic development and neurogenesis35, the forces mediating identification and internalization of antigens by B cell receptors36, and the mechanical proof-reading ability of T-cell surface receptors to detect changes in forces to boost the strength and specificity of signal transfer37. Together, these findings highlight the immense potential of TGT probes in a variety of experimental settings.
The authors would like to recognize the members of the Mattheyses laboratory for fruitful discussions and critiques. We acknowledge funding to A.L.M. from NSF CAREER 1832100 and NIH R01GM131099.
Name | Company | Catalog Number | Comments |
(3-Aminopropyl)triethoxysilane | Millipore Sigma | 440140 | Surface Preparation |
3-hydroxypicolinic acid (3-HPA) | Millipore Sigma | 56197 | Maldi-TOF-MS matrix |
Acetic Acid, Glacial | Fisher Scientific | A38S | Diluting EGF |
Acetonitrile (HPLC) | Fisher Scientific | A998SK | Oligonucleotide Preparation |
Alexa Fluor 488 Phalloidin | Cell Signaling Technology | 8878S | Immunocytochemistry |
Ammonium Chloride | Fisher Scientific | A687 | Immunocytochemistry |
Anti-Paxillin antibody [Y113] | Abcam | ab32084 | Immunocytochemistry |
BD Syringes only with Luer-Lok | BD bioscience | 309657 | Surface Preparation |
Bio-Gel P-2 | Bio-Rad | 1504118 | Oligonucleotide Preparation |
Bovine Serum Albumin (BSA) Protease-free Powder | Fisher Scientific | BP9703100 | Surface Preparation |
Cos-7 cells | ATCC | CRL-1651 | Cell Culture, Passage numbers 11-20 |
Coverslip Mini-Rack, for 8 coverslips | Fisher Scientific | C14784 | Surface Preparation |
c(RGDfK(PEG-PEG)), PEG=8-amino-3,6-dioxaoctanoic acid | Vivitide | PCI-3696-PI | Oligonucleotide Preparation |
Cy3B NHS ester | GE Healthcare | PA63101 | Oligonucleotide Preparation |
Dimethylformamide | Millipore Sigma | PHR1553 | Oligonucleotide Preparation |
DMEM with L-Glutamine, 4.5g/L Glucose and Sodium Pyruvate | Fisher Scientific | MT10013C | Cell Culture |
Epidermal Growth Factor human EGF | Millipore Sigma | E9644 | Cell Culture |
Ethanol, 200 proof (100%) | Fisher Scientific | 22032601 | Surface Preparation |
Falcon Standard Tissue Culture Dishes | Fisher Scientific | 08-772E | Surface Preparation |
Fetal Bovine Serum | Fisher Scientific | 10-438-026 | Cell Culture |
Flurobrite DMEM | Fisher Scientific | A1896701 | Cell Culture |
Goat anti-Rabbit IgG (H+L) Cross-Adsorbed Secondary Antibody, Alexa Fluor 647 | Invitrogen | A-21244 | Immunocytochemistry |
Goat Serum | Fisher Scientific | 16-210-064 | Immunocytochemistry |
Hank’s balanced salts (HBSS) | Fisher Scientific | 14-170-161 | Cell Culture |
Horse Serum | Fisher Scientific | 16050130 | Immunocytochemistry |
Hydrogen Peroxide | Fisher Scientific | H325-500 | Surface Preparation |
Nanosep MF centrifugal devices | Pall laboratory | ODM02C35 | Oligonucleotide Preparation |
NHS-azide | Fisher Scientific | 88902 | Oligonucleotide Preparation |
Nitrogen Gas Cylinder | Airgas | Surface Preparation | |
No. 2 round glass coverslips - 25 mm | VWR | 48382-085 | Surface Preparation |
Parafilm M Laboratory Film | Fisher Scientific | 13-374-10 | Surface Preparation |
Paraformaldehyde 16% | Fisher Scientific | 50-980-487 | Immunocytochemistry |
PBS, 1X | Fisher Scientific | 21-030-CV | Surface Preparation/Immunocytochemistry |
Penicillin-Streptomycin (5,000 U/mL) | Fisher Scientific | 15-070-063 | Cell Culture |
PYREX Low Form Griffin Beakers | Fisher Scientific | 02-540G | Surface Preparation |
Sodium Ascorbate | Fisher Scientific | 18-606-310 | Oligonucleotide Preparation |
Sodium Bicarbonate | Fisher Scientific | S233 | Oligonucleotide Preparation |
Sodium Chloride | Fisher Scientific | BP358 | Surface Preparation |
Streptavidin | Fisher Scientific | 434301 | Surface Preparation |
Sulfo-NHS-LC-Biotin | Fisher Scientific | 21335 | Surface Preparation |
Sulfuric Acid | Fisher Scientific | A300-500 | Surface Preparation |
TEAA | Fisher Scientific | NC0322726 | Oligonucleotide Preparation |
Triethylamine | Millipore Sigma | 471283 | Oligonucleotide Preparation |
Trifluoroacetic Acid (TFA) | Fisher Scientific | PI28901 | Oligonucleotide Preparation |
THPTA | Fisher Scientific | NC1296293 | Oligonucleotide Preparation |
Triton X 100 Detergent Surfact Ams Solution | Fisher Scientific | 85111 | Immunocytochemistry |
Water, DNA Grade, DNASE, Protease free | Fisher Scientific | BP24701 | Oligonucleotide Preparation |
Equipment | |||
Agilent AdvanceBio Oligonucleotide C18 column, 4.6 x 150 mm, 2.7 μm | Agilent | 653950-702 | Oligonucleotide Preparation |
High-performance liquid chromatography | Agilent | 1100 | Oligonucleotide Preparation |
Low Speed Orbital Shaker | Fisher Scientific | 10-320-813 | Immunocytochemistry |
Matrix-assisted laser desorption/ ionization time-of-flight mass spectrometer (MALDI-TOF-MS) | Voyager STR | Oligonucleotide Preparation | |
Molecular Probes Attofluor Cell Chamber | Fisher Scientific | A7816 | Surface Preparation |
Nanodrop 2000 UV-Vis Spectrophotometer | Thermo Fisher | Oligonucleotide Preparation | |
Nikon Eclipse Ti inverted microscope | pe Nikon | Microscopy | |
Nikon Perfect Focus System | Nikon | Microscopy | |
NIS Elements software | Nikon | Microscopy | |
ORCA-Flash4.0 V3 Digital CMOS camera | Hamamatsu | Microscopy | |
Quad band TIRF 405/488/561/647 cube | CHROMA | Microscopy | |
RICM Cube | CHROMA | Microscopy | |
SOLA v-nIR Light Engine | Lumencor | Microscopy | |
Thermo Forma Steri Cycle 370 CO2 Incubator | Fisher Scientific | Cell Culture | |
VWR 75D Ultrasonic Cleaner | VWR | 13710 | Surface Preparation |
Data Analysis | Use | ||
Fiji (Image J) | https://imagej.net/software/fiji/downloads | Quantitative Analysis | |
Graph Pad Prism | Graph Pad | Statistical Analysis | |
Oligo name | 5'modification/ 3' modification | Sequence (5' to 3') | Use |
Alkyne-21-BHQ2 | 5' Hexynyl/ 3' BHQ_2 | GTGAAATACCGCACAGATGCG | Top strand TGT probe |
56 pN TGT | 5' Biosg/TTTTTT/iUniAmM | CGCATCTGTGCGGTATTTCACTTT | Bottom strand TGT probe |
12 pN TGT | 5' AmMC6/ 3' BioTEG | CGCATCTGTGCGGTATTTCACTTT | Bottom strand TGT probe |
This article has been published
Video Coming Soon
ABOUT JoVE
Copyright © 2024 MyJoVE Corporation. All rights reserved