Direct Observation of Wet Biological Samples by Graphene Liquid Cell ...
SUPPORTING INFORMATION
Direct Observation of Wet Biological Samples by Graphene Liquid Cell Transmission Electron Microscopy Jungwon Park 1, 2, Hyesung Park 3, Peter Ercius 4, Adrian F. Pegoraro 1, 2, Chen Xu 5, Jin Woong Kim 6, 7, Sang Hoon Han 8, and David A. Weitz 1, 2 * 1
Department of Applied Physics, Harvard University, Cambridge, MA 02138 USA.
2
School of Engineering and Applied Sciences, Harvard University, Cambridge, MA 02138
USA. 3
School of Energy and Chemical Engineering, Ulsan National Institute of Science and
Technology, Ulsan 689-798, South Korea. 4
5
The Molecular Foundry, Lawrence Berkeley National Laboratory, Berkeley, CA 94720 USA. Rosenstiel Basic Medical Sciences Research Center, Brandeis University, Waltham, MA
02454, USA. 6
Department of Applied Chemistry, Hanyang University, Ansan 426-791, South Korea.
7
Department of Bionano Technology, Hanyang University, Ansan 426-791, South Korea.
8
Amore-Pacific Co. R&D Center, Yongin, Gyeonggi-do 446-729, South Korea.
* To whom correspondence should be addressed: [email protected]
S1
Supporting Text Chemical vapor deposition of a multi-layer graphene sheet. A multi-layer graphene sheet is synthesized by atmospheric chemical vapor deposition (APCVD) on 25 μm thick copper foil (99.8 %, Alfa Aesar, Ward Hill, MA). The copper foil is inserted into a quartz tube and heated to 1,000 °C under 600 sccm Ar, 400 sccm H2 followed by annealing for 30 min. Then, a gas mixture of 1600 sccm Ar, 400 sccm H2, and 30 sccm CH4 is introduced for 15 min. to synthesize the graphene sheet. After the synthesis is completed, fast cooling to room temperature with 300 sccm Ar is performed. Graphene transfer onto the amorphous carbon film TEM grid and preparation of freestanding multi-layer graphene. Graphene transfer onto the amorphous holey carbon film TEM grids begins with selective removal of graphene from the one side of copper foil by applying mild oxygen plasma. The Quantifoil TEM grid is placed onto a graphene-covered copper foil with the Quantifoil carbon film side facing the graphene. Then 20 μl of isopropyl alcohol (IPA) is dropped onto the sample to wet the interface, and the sample is dried at 85 °C on a hot plate for 30 min to promote adhesion between the Quantifoil carbon film and the graphene. The grid is floated on an 0.05 g/ mL aqueous solution of ammonium persulfate, (NH4)2SO4, to etch the underlying copper foil, and rinsed several times by floating the graphene transferred grid on deionized water. Graphene surface is further cleaned by annealing graphene coated TEM grids at 500 oC for 2 hours under H2 (700sccm) and Ar (400 sccm).
S2
Preparation procedures of multi-layer graphene free-standing on the surface of solutions are described in Fig. S1. Copper foil with the multi-layer graphene grown on both sides is flatten onto a glass slide and exposed to O2 plasma to remove multi-layer graphene on one side. Then, it is floated on an aqueous solution of 113 mM ammonium persulfate, (NH4)2SO4, with plasma cleaned side of the copper foil facing the etching solution, to remove the underlying copper foil. Once the etching process is completed, the free-standing graphene is lifted out with the edge of a glass slide and the glass slide with graphene is carefully immersed in deionized water. Since the multi-layer graphene is floated on the surface of deionized water during this immersion step, the etching solution which wets the back side of the graphene sheet can be rinsed. These processes are repeated multiple times. Since each layer of the graphene sheet is composed of many domains with micrometer scale, grain boundaries and defects from different layers are overlaid, as a result, it holds its integrity on the surface of aqueous solution. Due to an extinction of light by multi-layer graphene, it is easily visible on the surface of water, the buffer solution, and the culture media solution. MDCK cell culture, Au nanoparticle incubation, and GLC preparation. Madin-Darby Canine Kidney (MDCK) cells are cultured on graphene and Si3N4 TEM grids placed in the culture flask and incubated at 37 oC with 5% CO2 with Modified Eagles’s Medium (MEM) with Earle's salts supplemented with 5% FBS, 2mM l-glutamine, 100 U/ml penicillin, and 100 g/mL streptomycin. 40 nm Au nanoparticles with a concentration of 20 mM are incubated in the aqueous solution of serum for 2 hours. MDCK cells pre-cultured on graphene TEM grids are incubated in culture media with serum-treated Au nanoparticles for 4 hours. Then, MDCK cell cultured TEM grids are used to lift out free-standing graphene on the surface of the culture media solution. S3
GLC preparation of MDCK cells with fixation, cell membrane extraction. The simultaneous extraction and fixation of the MDCK cells on TEM grids are executed with 0.5% Triton X-100 and 0.25% glutaraldehyde in a cytoskeleton buffer (10mM MES buffer, 150mM NaCl, 5mM EGTA, 5mM glucose, and 5mM MgCl2, at pH 6.1). An initial fixation of 3 min in this mixture is followed by gentle cleaning with PBS buffer. Post fixation is completed in the solution mixture of 2% glutaraldehyde and 10 μg/mL phalloidin in cytoskeleton buffer, to stabilize the actin filaments. The grids are then stored at 4 oC. MDCK cells on grids for whole cell imaging are fixed in the solution mixture of 2% glutaraldehyde and 10 μg/mL phalloidin in a cytoskeleton buffer. Fluorescence confocal microscopy and transmission electron microscopy. A Leica DMI 6000 confocal microscope is used for FLM. TEM experiments are performed in JEOL JEM 2100 TEM operated at 200 kV, FEI F20 TEM operated at 120 kV. Low-dose and low-accelerating voltage imaging was performed using the aberration-corrected TEAM I microscope operated at 80 kV, operated by the National Center for Electron Microscopy facility within The Molecular Foundry at Lawrence Berkeley National Laboratory. Fabrication of SGLC. Grids are fabricated using ultra-thin silicon wafers (100 μm, 4-inches, p-doped) purchased from Virginia Semiconductor (Fredericksburg, VA). The fabrication process follows growing lowstress Si3N4 membranes on the silicon wafers (20 nm in thickness), lithographic patterning for 200 μm by 300 μm windows, and KOH silicon wafer etching. The size and number of windows in the TEM grid can be varied in the photolithography, depending on the dimension of specimens. Lifting out a free-standing multi-layer graphene sheet over the wet samples on the S4
Si3N4 films results in conformal and tight encapsulation; however, we do expect increased background scattering by Si3N4 film compared to multilayer graphene windows. (Fig. S2). TEM of non-adherent Bacillus subtilis using GLCs. GLC is not limited to TEM observation of adherent cells that are pre-cultured on the TEM grids. Since a multi-layer graphene sheet can be floated onto the surface of a variety of aqueous solutions, we can simply transfer a multi-layer graphene sheet onto a solution that contains specimens of interest and lift it out with the graphene coated TEM grid to encapsulate suspension samples. We successfully encapsulate a solution of Bacillus subtilis and obtain a TEM image of Bacillus subtilis in GLC (Fig. S4a). If Bacillus subtilis are exposed to typical TEM vacuum conditions, volume shrinkage is significant, yet we do not observe such changes which implies the cellular integrity is preserved by GLC. Nonetheless, it is obvious that the degradation of susceptible structures is inevitable as cells are exposed to the imaging electron beam for a long time within the confinement of the GLC. We show one obvious example that Bacillus subtilis are damaged by a high electron dose (Fig. S4b). We expose Bacillus subtilis in GLC to the continuous electron beam with 200 kV acceleration voltage and a beam current of 106 µA at 5,000 magnification for more than five minutes. The mass loss shown as bubbles with a bright contrast and the collapse of cellular membrane structures are notable. TEM of fixed MDCK cells in GLCS. While cellular membranes are extracted prior to encapsulation in GLCs to expose internal cytoskeleton structure reveals, we can also study exterior morphology that covers such internal structures by using GLC of the whole fixed MDCK cells that have intact cellular membranes (Fig. S5) in TEM. Cells are fixed and treated with phalloidin to stabilize cytoskeleton. We observe the morphology of the whole cell with extending filopodia along the cellular boundary.
S5
We also visually discern thick cytoskeleton filaments protruding out parallel to the direction of filopodia.
S6
Supporting Figures
Figure S1. Graphene transfer to the surface of solutions. (a) Schematic illustration of preparation for multi-layer graphene sheet free-standing on the surface of solution sample to be encapsulated. (b) Optical images of the multi-layer graphene free-standing on the surface of Cu etching solution (left), water (middle), and cell culture media solution (right). The same multi-layer graphene is sequentially transferred from the solution on the left to the right.
S7
Figure S2. SGLC sample preparation for TEM observation. Procedures for SGLC of cultured cells are accomplished by lifting out the free-standing multi-layer graphene floated on the surface of the culture media solution by graphene transferred TEM girds with cells. SGLC can be used to encapsulate suspended cells and viruses for TEM imaging when a more rigid supporting film is needed.
S8
Figure S3. TEM contrast difference of H3N2 viruses in GLC and GLC on amorphous carbon film. Viruses encapsulated in GLC on the amorphous carbon film support (left top region in (a) and right top region in (b)) show reduced contrast while viruses in GLC with no carbon film support (bright region within the arc in (a) and (b)) show distinct contrast. The electron beam accelerated by 80 kV and a camera with pixel resolution of 0.074 nm2/pixel for (a) and 0.14 nm2/pixel for (b) are used.
S9
Figure S4. TEM images of Bacillus subtilis in GLC and electron beam damage. (a) TEM image of Bacillus subtilis in GLC. Wrinkles of multilayer graphene are also visible along with Bacillus subtilis. (b) TEM image of Bacillus subtilis in GLC whose cellular integrity is damaged by the electron beam after a prolonged observation. The electron beam accelerated by 200 kV and a camera with pixel resolution of 1.80 nm2/pixel are used.
S10
Figure S5. TEM observation of fixed whole MDCK cells encapsulated in GLCs. TEM images of fixed whole MDCK cells in different magnifications are shown in (a) and (b). Cells are cultured on a Si3N4 TEM grid, fixed, and covered with graphene. The electron beam accelerated by 200 kV and a camera with pixel resolution of 472.12 nm2/pixel for (a) and 20.49 nm2/pixel for (b) are used.
S11
Direct Observation of Wet Biological Samples by Graphene Liquid Cell Transmission Electron Microscopy Jungwon Park 1, 2, Hyesung Park 3, Peter Ercius 4, Adrian F. Pegoraro 1, 2, Chen Xu 5, Jin Woong Kim 6, 7, Sang Hoon Han 8, and David A. Weitz 1, 2 * 1
Department of Applied Physics, Harvard University, Cambridge, MA 02138 USA.
2
School of Engineering and Applied Sciences, Harvard University, Cambridge, MA 02138
USA. 3
School of Energy and Chemical Engineering, Ulsan National Institute of Science and
Technology, Ulsan 689-798, South Korea. 4
5
The Molecular Foundry, Lawrence Berkeley National Laboratory, Berkeley, CA 94720 USA. Rosenstiel Basic Medical Sciences Research Center, Brandeis University, Waltham, MA
02454, USA. 6
Department of Applied Chemistry, Hanyang University, Ansan 426-791, South Korea.
7
Department of Bionano Technology, Hanyang University, Ansan 426-791, South Korea.
8
Amore-Pacific Co. R&D Center, Yongin, Gyeonggi-do 446-729, South Korea.
* To whom correspondence should be addressed: [email protected]
S1
Supporting Text Chemical vapor deposition of a multi-layer graphene sheet. A multi-layer graphene sheet is synthesized by atmospheric chemical vapor deposition (APCVD) on 25 μm thick copper foil (99.8 %, Alfa Aesar, Ward Hill, MA). The copper foil is inserted into a quartz tube and heated to 1,000 °C under 600 sccm Ar, 400 sccm H2 followed by annealing for 30 min. Then, a gas mixture of 1600 sccm Ar, 400 sccm H2, and 30 sccm CH4 is introduced for 15 min. to synthesize the graphene sheet. After the synthesis is completed, fast cooling to room temperature with 300 sccm Ar is performed. Graphene transfer onto the amorphous carbon film TEM grid and preparation of freestanding multi-layer graphene. Graphene transfer onto the amorphous holey carbon film TEM grids begins with selective removal of graphene from the one side of copper foil by applying mild oxygen plasma. The Quantifoil TEM grid is placed onto a graphene-covered copper foil with the Quantifoil carbon film side facing the graphene. Then 20 μl of isopropyl alcohol (IPA) is dropped onto the sample to wet the interface, and the sample is dried at 85 °C on a hot plate for 30 min to promote adhesion between the Quantifoil carbon film and the graphene. The grid is floated on an 0.05 g/ mL aqueous solution of ammonium persulfate, (NH4)2SO4, to etch the underlying copper foil, and rinsed several times by floating the graphene transferred grid on deionized water. Graphene surface is further cleaned by annealing graphene coated TEM grids at 500 oC for 2 hours under H2 (700sccm) and Ar (400 sccm).
S2
Preparation procedures of multi-layer graphene free-standing on the surface of solutions are described in Fig. S1. Copper foil with the multi-layer graphene grown on both sides is flatten onto a glass slide and exposed to O2 plasma to remove multi-layer graphene on one side. Then, it is floated on an aqueous solution of 113 mM ammonium persulfate, (NH4)2SO4, with plasma cleaned side of the copper foil facing the etching solution, to remove the underlying copper foil. Once the etching process is completed, the free-standing graphene is lifted out with the edge of a glass slide and the glass slide with graphene is carefully immersed in deionized water. Since the multi-layer graphene is floated on the surface of deionized water during this immersion step, the etching solution which wets the back side of the graphene sheet can be rinsed. These processes are repeated multiple times. Since each layer of the graphene sheet is composed of many domains with micrometer scale, grain boundaries and defects from different layers are overlaid, as a result, it holds its integrity on the surface of aqueous solution. Due to an extinction of light by multi-layer graphene, it is easily visible on the surface of water, the buffer solution, and the culture media solution. MDCK cell culture, Au nanoparticle incubation, and GLC preparation. Madin-Darby Canine Kidney (MDCK) cells are cultured on graphene and Si3N4 TEM grids placed in the culture flask and incubated at 37 oC with 5% CO2 with Modified Eagles’s Medium (MEM) with Earle's salts supplemented with 5% FBS, 2mM l-glutamine, 100 U/ml penicillin, and 100 g/mL streptomycin. 40 nm Au nanoparticles with a concentration of 20 mM are incubated in the aqueous solution of serum for 2 hours. MDCK cells pre-cultured on graphene TEM grids are incubated in culture media with serum-treated Au nanoparticles for 4 hours. Then, MDCK cell cultured TEM grids are used to lift out free-standing graphene on the surface of the culture media solution. S3
GLC preparation of MDCK cells with fixation, cell membrane extraction. The simultaneous extraction and fixation of the MDCK cells on TEM grids are executed with 0.5% Triton X-100 and 0.25% glutaraldehyde in a cytoskeleton buffer (10mM MES buffer, 150mM NaCl, 5mM EGTA, 5mM glucose, and 5mM MgCl2, at pH 6.1). An initial fixation of 3 min in this mixture is followed by gentle cleaning with PBS buffer. Post fixation is completed in the solution mixture of 2% glutaraldehyde and 10 μg/mL phalloidin in cytoskeleton buffer, to stabilize the actin filaments. The grids are then stored at 4 oC. MDCK cells on grids for whole cell imaging are fixed in the solution mixture of 2% glutaraldehyde and 10 μg/mL phalloidin in a cytoskeleton buffer. Fluorescence confocal microscopy and transmission electron microscopy. A Leica DMI 6000 confocal microscope is used for FLM. TEM experiments are performed in JEOL JEM 2100 TEM operated at 200 kV, FEI F20 TEM operated at 120 kV. Low-dose and low-accelerating voltage imaging was performed using the aberration-corrected TEAM I microscope operated at 80 kV, operated by the National Center for Electron Microscopy facility within The Molecular Foundry at Lawrence Berkeley National Laboratory. Fabrication of SGLC. Grids are fabricated using ultra-thin silicon wafers (100 μm, 4-inches, p-doped) purchased from Virginia Semiconductor (Fredericksburg, VA). The fabrication process follows growing lowstress Si3N4 membranes on the silicon wafers (20 nm in thickness), lithographic patterning for 200 μm by 300 μm windows, and KOH silicon wafer etching. The size and number of windows in the TEM grid can be varied in the photolithography, depending on the dimension of specimens. Lifting out a free-standing multi-layer graphene sheet over the wet samples on the S4
Si3N4 films results in conformal and tight encapsulation; however, we do expect increased background scattering by Si3N4 film compared to multilayer graphene windows. (Fig. S2). TEM of non-adherent Bacillus subtilis using GLCs. GLC is not limited to TEM observation of adherent cells that are pre-cultured on the TEM grids. Since a multi-layer graphene sheet can be floated onto the surface of a variety of aqueous solutions, we can simply transfer a multi-layer graphene sheet onto a solution that contains specimens of interest and lift it out with the graphene coated TEM grid to encapsulate suspension samples. We successfully encapsulate a solution of Bacillus subtilis and obtain a TEM image of Bacillus subtilis in GLC (Fig. S4a). If Bacillus subtilis are exposed to typical TEM vacuum conditions, volume shrinkage is significant, yet we do not observe such changes which implies the cellular integrity is preserved by GLC. Nonetheless, it is obvious that the degradation of susceptible structures is inevitable as cells are exposed to the imaging electron beam for a long time within the confinement of the GLC. We show one obvious example that Bacillus subtilis are damaged by a high electron dose (Fig. S4b). We expose Bacillus subtilis in GLC to the continuous electron beam with 200 kV acceleration voltage and a beam current of 106 µA at 5,000 magnification for more than five minutes. The mass loss shown as bubbles with a bright contrast and the collapse of cellular membrane structures are notable. TEM of fixed MDCK cells in GLCS. While cellular membranes are extracted prior to encapsulation in GLCs to expose internal cytoskeleton structure reveals, we can also study exterior morphology that covers such internal structures by using GLC of the whole fixed MDCK cells that have intact cellular membranes (Fig. S5) in TEM. Cells are fixed and treated with phalloidin to stabilize cytoskeleton. We observe the morphology of the whole cell with extending filopodia along the cellular boundary.
S5
We also visually discern thick cytoskeleton filaments protruding out parallel to the direction of filopodia.
S6
Supporting Figures
Figure S1. Graphene transfer to the surface of solutions. (a) Schematic illustration of preparation for multi-layer graphene sheet free-standing on the surface of solution sample to be encapsulated. (b) Optical images of the multi-layer graphene free-standing on the surface of Cu etching solution (left), water (middle), and cell culture media solution (right). The same multi-layer graphene is sequentially transferred from the solution on the left to the right.
S7
Figure S2. SGLC sample preparation for TEM observation. Procedures for SGLC of cultured cells are accomplished by lifting out the free-standing multi-layer graphene floated on the surface of the culture media solution by graphene transferred TEM girds with cells. SGLC can be used to encapsulate suspended cells and viruses for TEM imaging when a more rigid supporting film is needed.
S8
Figure S3. TEM contrast difference of H3N2 viruses in GLC and GLC on amorphous carbon film. Viruses encapsulated in GLC on the amorphous carbon film support (left top region in (a) and right top region in (b)) show reduced contrast while viruses in GLC with no carbon film support (bright region within the arc in (a) and (b)) show distinct contrast. The electron beam accelerated by 80 kV and a camera with pixel resolution of 0.074 nm2/pixel for (a) and 0.14 nm2/pixel for (b) are used.
S9
Figure S4. TEM images of Bacillus subtilis in GLC and electron beam damage. (a) TEM image of Bacillus subtilis in GLC. Wrinkles of multilayer graphene are also visible along with Bacillus subtilis. (b) TEM image of Bacillus subtilis in GLC whose cellular integrity is damaged by the electron beam after a prolonged observation. The electron beam accelerated by 200 kV and a camera with pixel resolution of 1.80 nm2/pixel are used.
S10
Figure S5. TEM observation of fixed whole MDCK cells encapsulated in GLCs. TEM images of fixed whole MDCK cells in different magnifications are shown in (a) and (b). Cells are cultured on a Si3N4 TEM grid, fixed, and covered with graphene. The electron beam accelerated by 200 kV and a camera with pixel resolution of 472.12 nm2/pixel for (a) and 20.49 nm2/pixel for (b) are used.
S11
