Direct Synthesis of EM-Visible Gold Nanoparticles in Cells for Protein Localization Analysis with Well-Preserved Ultrastructure

Summary

The present protocol describes a clonable electron microscopy labeling technology for detecting metallothionein-tagged proteins in cells using a novel autonucleation suppression mechanism-based gold nanoparticle synthesis technique.

Abstract

Analyzing the precise localization of protein molecules in cells with ultrastructural resolution is of great significance for the study of various physiological or pathological processes in all living organisms. Therefore, the development of clonable tags that can be used as electron microscopy probes is of great value, just as fluorescent proteins have played a crucial role in the field of optical imaging. The autonucleation suppression mechanism (ANSM) was recently uncovered, which allows for the specific synthesis of gold nanoparticles (AuNPs) on cysteine-rich tags, such as metallothionein (MT) and antifreeze protein (AFP).

Based on the ANSM, an electron microscopy labeling technology was developed, which enables the specific detection of tagged proteins in prokaryotic and eukaryotic cells with an unprecedented labeling efficiency. This study illustrates a protocol for the detection of MTn (an engineered MT variant lacking aldehyde-reactive residues) fusion proteins in mammalian cells with well-preserved ultrastructure. In this protocol, high-pressure freezing and freeze-substitution fixation were performed using non-aldehyde fixatives (such as tannic acid, uranyl acetate) to preserve near-native ultrastructure and avoid damage to the tag activity caused by aldehyde crosslinking.

A simple one-step rehydration was used prior to the ANSM-based AuNP synthesis. The results showed that the tagged proteins targeted various organelles, including the membranes and the lumen of the endoplasmic reticulum (ER), and mitochondrial matrices were detected with high efficiency and specificity. This research provides biologists with a robust protocol to address an enormous range of biological questions at the single-molecule level in cellular ultrastructural contexts.

Introduction

In the postgenomic era, the development of green fluorescent protein (GFP) as a single-molecule reporter for light microscopy has revolutionized the field of modern life science research^1,2. For decades, electron microscopy (EM) has been a powerful tool for intuitively observing the cellular ultrastructure with nanoscale resolution³; however, the precise identification and localization of protein molecules remain challenging.

The most commonly used EM labeling technique is the immunoelectron microscopy (IEM) labeling technique, which is based on the antigen-antibody reaction. However, although many techniques have been developed in the field of IEM labeling, including pre-embedding IEM and post-embedding IEM (on resin sections or hydrated cryosections), it still suffers from low labeling efficiency (<10%)^4,5, which is related to the sample preparation and antibody quality. To overcome these limitations, developing genetically encoded tags has great application potential.

Two main types of EM tags have been thoroughly explored in recent years. One type is the DAB staining method, which utilizes tags such as APEX2 to oxidize 3,3'-diaminobenzidine (DAB) to osmiophilic polymers for EM visualization^{6,7,8,9,10,11,12}. It enables the labeling of high-abundance proteins in subcellular regions but is not suitable for single-molecule counting. The other type utilizes metal-binding proteins, such as ferritin¹³ and metallothionein (MT)^{14,15,16,17,18,19,20,21,22,23,24,25,26}, to generate electron-dense metal deposits in situ for EM visualization. Only the latter has real potential for single-molecule visualization and counting. The molecular size of ferritin is too large (~450 kD) for it to be used as a promising tag, whereas the small size (~5 kD) of MT and its ability to bind various ions through its 20 cysteines have attracted great attention. Several labs have tried to label purified MT-fusion proteins or MT-expressing cells by incubating directly with Au⁺. These attempts have initially proved that MT tags can bind gold ions to form high-contrast signals, but none have really achieved the effective identification of individual proteins in cells, and they are not widely applicable^{14,15,16,17,18,19,20,21,22,23}.

The ANSM-based AuNP synthesis technique, which involves synthesizing 2-6 nm-sized AuNPs directly on cysteine-rich tags (e.g., MT, the MT variants MTn and MTα, AFP) as electron-dense labels for EM visualization, is the first reliable and applicable approach for protein labeling and single molecule detection in cells^24,25,26. It allows for the specific synthesis of AuNPs on isolated tag fusion proteins and has achieved an unprecedented labeling efficiency in nonfixed or chemically fixed prokaryotic (E. coli) and eukaryotic (S. pombe) cells. However, implementing the same protocol in more advanced systems such as mammalian cells or even tissues involves additional challenges, such as the more complex intracellular redox homeostasis and more fragile cellular structure.

This study presents a clonable EM labeling technology, which combines the novel ANSM-based AuNP synthesis technique for labeling genetically encoded cysteine-rich tags (MT) with the HPF/FSF-rehydration-HPF/FSF sample preparation method, enables the unambiguous single-molecule identification of tagged proteins in the ER membrane, ER lumen, and mitochondrial matrix in HeLa cells. The current method combines the characteristics of high labeling efficiency, a high signal-to-noise ratio, single-molecule labeling, and strong universality, and this method has broad application prospects in life science research.

Protocol

All the supplies used in this experiment are listed in the Table of Materials. The step-by-step workflow of the current protocol is shown in Figure 1.

1. Cell culture on sapphire discs

1. Transfer 3 mm x 0.16 mm sapphire discs to a 2 mL centrifuge tube containing 1 mL of ethanol, and sonicate in an ultrasonic cleaner for 10 min.
2. Burn each sapphire disc in an alcohol lamp flame until there are no visible deposits on the surface.
   NOTE: Through the above method, sapphire discs can be recycled many times.
3. Label one side of each sapphire disc with a number using a solvent-resistant pen (Figure 1A).
   NOTE: The marker pen used needs to be tested to determine whether it is resistant to acetone and methanol solvents in advance to prevent the loss of marks.
4. Place the labeled sapphire discs on the bottom of a 35 mm culture dish with the labeled side facing up.
5. Sterilize the cell culture dish with sapphire discs under UV light for 30 min.
6. Flip the sapphire discs with tweezers (the labeled side facing down), and sterilize under UV light for an additional 30 min. Now, the culture dish containing the sapphire discs is ready for cell culture.
   NOTE: The sapphire discs can also be sterilized by autoclaving.
7. Seed the cells on the sapphire discs prepared above, and grow to 80%-90% confluency in a cell incubator at 37 °C, 95% humidity, and 5% CO₂ (Figure 1B).
   NOTE: Stable HeLa cell lines expressing MTn tags were generated as described in Jiang et al.²⁶.

2. High-pressure freezing (HPF)

CAUTION: Liquid nitrogen is used in this experiment. When working with liquid nitrogen, use proper safety procedures and personal protective equipment to prevent frostbite and asphyxiation.

1. Prepare the high-pressure freezing machine at least 2 h before the experiment according to the manufacturer's instructions.
2. Transfer type A HPF aluminum carriers (recess: 0.025/0.275 mm) to a 2 mL centrifuge tube containing 1 mL of ethanol, and sonicate in an ultrasonic cleaner for 10 min.
3. Dry the carriers in a Petri dish covered with qualitative filter paper (medium speed).
4. Soak the precleaned carriers in 1-hexadecene.
   NOTE: BSA is not recommended as a cryoprotectant in the current experiment as it will interfere with the subsequent gold nanoparticle synthesis.
5. Use fine tweezers to pick up a sapphire disc with cells from the culture dish; touch the labeled surface with qualitative filter paper (medium speed) to remove the excess medium.
   NOTE: The thermal conductivity of water is very low, and too much residual water will affect the freezing effect. Retain minimal water to prevent the cells from drying out.
6. Mount the sapphire disc into the HPF specimen holder, and quickly cap the disc with a 0.025 mm deep aluminum carrier containing 1-hexadecene (Figure 1C).
7. Aspirate excess solution with qualitative filter paper (medium speed).
8. Load the specimen holder for high-pressure freezing (Figure 1D).
   NOTE: The specimen loading process needs to be as fast as possible (within 1 min) to minimize physiological alterations due to changes in the temperature, humidity, pH, and oxygen content.
9. Unload the sapphire disc-carrier assembly under liquid nitrogen in a foam cryobox, and store the assembly in a cryovial in liquid nitrogen before use.
   ​NOTE: The protocol may be paused here. The specimens in cryovials can be stored in a liquid nitrogen dewar for years.

3. Freeze-substitution fixation (FSF) and rehydration (Figure 1E)

1. Fill the automated freeze-substitution machine (AFS) with liquid nitrogen, and cool the chamber to −90 °C.
   CAUTION: Liquid nitrogen is used in this experiment. When working with liquid nitrogen, use proper safety procedures and personal protective equipment to prevent frostbite and asphyxiation.
2. Prepare 2 mL of 0.01% tannic acid (w/v) in acetone (FSF solution 1) in a 20 mm round polypropylene container (Figure 1E) in a fume hood and freeze it in liquid nitrogen.
3. Load the specimens (the sapphire disc-carrier assemblies) into the container with frozen FSF solution 1 under LN2 using a pair of precooled tweezers.
4. Transfer the container to the precooled AFS chamber for FSF processing at −90 °C.
5. Keep the specimens at −90 °C for 1 h; then, separate the sapphire discs from the carriers with precooled tweezers, and make sure the marked sides of the sapphire discs are facing down.
   NOTE: The tweezers must be sufficiently precooled to prevent temperature changes. The sapphire discs should be easily separated.
6. Keep at −90 °C for a further 8-10 h; then, warm to −60 °C within 3 h.
7. Replace the FSF solution 1 in the container with precooled acetone, and incubate for 1 h. Repeat this acetone wash 2x.
8. Warm to −30 °C within 3 h. During this period, prepare and precool 2 mL of 0.01% uranyl acetate in acetone (FSF solution 2, diluted from 10% uranyl acetate in methanol) in a 2 mL centrifuge tube.
   CAUTION: Uranyl acetate, used in the current experiment, is radioactive and highly toxic; it must be handled according to proper safety procedures and disposed of as hazardous chemical waste.
9. Replace the acetone with FSF solution 2, and incubate at −30 °C for 3 h.
10. Replace the FSF solution 2 in the container with precooled acetone, and incubate for 30 min at −30 °C. Repeat this acetone wash 2x.
11. Warm from −30 °C to 4 °C within 2 h.
12. Replace the acetone with 2 mL of 0.2 M HEPES buffer containing 1 mM CaCl₂ and 1 mM MgCl₂ at pH 5.5.
13. Change the buffer once, and incubate at room temperature for 1 h or at 4 °C overnight.
    ​NOTE: The protocol may be paused here for one night or continued for at least 6 h until the specimens are frozen again in step 5.

4. ANSM-based AuNP synthesis for mammalian cells (Figure 1F)

1. Wash with PBS-A buffer 3x for 5 min each time.
2. Transfer the sapphire discs to a 35 mm culture dish containing 1 mL of PBS-A buffer at room temperature.
   NOTE: Always keep the marked sides of the sapphire discs facing down, and avoid overlapping the sapphire discs and rubbing off the cells.
3. Prepare the reducing solution by adding 4.28 µL of 2-mercaptoethanol into a 2 mL centrifuge tube containing 1 mL of PBS-A buffer and mixing well in a fume hood.
   CAUTION: 2-Mercaptoethanol is toxic and has a pungent odor; it must be handled according to proper safety procedures.
4. Replace the PBS-A buffer in the culture dish with 1 mL of reducing solution, and incubate for 1 h at room temperature.
5. Prepare the gold precursor before use.
   1. Add 80 µL of 10 mM HAuCl₄ in ddH₂O into a 2 mL centrifuge tube containing 1 mL of reducing solution, and immediately vortex.
      NOTE: The solution may become cloudy.
   2. Add 80 µL of 500 mM D-penicillamine in ddH2O into the above solution, and immediately vortex.
      NOTE: The solution may become clear again.
6. Replace the solution in the culture dish with the 1 mL of gold precursor solution prepared in step 4.5, and incubate for 2 h at 4 °C.
   NOTE: One milliliter of the gold precursor is sufficient to react with approximately 1 × 10⁶ cells (confluent on a 35 mm dish). Larger volumes of the precursor are needed when the AuNPs become significantly smaller.
7. Weigh 0.0038 g of NaBH₄ into a 1.5 mL centrifuge tube, add 1 mL of ice-cold ddH₂O, and vortex to make fresh 100 mM NaBH₄ just prior to use.
8. Add 20-100 µL of 100 mM NaBH₄ to the solution from step 4.6, immediately shake to mix well, and incubate for 5 min at room temperature.
   NOTE: Prepare 100 mM NaBH₄ freshly for immediate use. After adding the NaBH₄, the color of the solution will gradually turn red and then deepen. If no color change is observed, it largely indicates that the experiment has failed.
9. Replace the solution with 2 mL of PBS-A buffer to stop the reaction.
   NOTE: Stopping the reaction within 30 min is critical, as a prolonged reaction will gradually break down the AuNPs.

5. High-pressure freezing and freeze-substitution fixation (Figure 1C-F)

NOTE: The specimens are HPF and FSF again, and the procedure is almost the same as that previously described in section 2 and section 3 with a few modifications.

1. Prepare 1% osmium tetroxide and 0.1% uranyl acetate in acetone (FSF solution 3, diluted from 4% osmium tetroxide in acetone and 10% uranyl acetate in methanol).
   CAUTION: Osmium tetroxide and uranyl acetate are highly toxic chemicals; they must be handled according to proper safety procedures and disposed of as hazardous chemical waste.
2. Load the specimens (the sapphire disc-carrier assemblies) into the container with frozen FSF solution 3 under LN2 using a pair of precooled tweezers.
3. Transfer the container to the precooled AFS chamber for FSF processing at −90 °C, and run the FSF program as follows: keep at −90 °C for 8-10 h; then, warm to −60 °C within 3 h; keep at −60 °C for 3 h; then, warm to −30 °C within 3 h; keep at −30 °C for 3 h; then, warm to 4 °C within 2 h.
4. When the temperature reaches 4 °C, transfer the specimens to the fume hood, and maintain at room temperature for 15 min.
5. Wash with acetone 3x for 15 min each time.
6. Maintain in acetone at room temperature for at least 2 h.

6. Resin infiltration, embedding, and polymerization

1. Prepare the epoxy resin mixture according to the manufacturer's instructions before use.
   CAUTION: The epoxy resin used in the current experiment is toxic prior to polymerization; it must be handled according to proper safety procedures. Unpolymerized resin should be disposed of as hazardous chemical waste or polymerized prior to disposal.
2. Transfer the sapphire discs to flat-bottom embedding capsules, and infiltrate with resin for at least 30 min at room temperature.
   NOTE: Keep the marked sides of the sapphire discs facing down.
3. Polymerize the specimen at 60 °C in an oven for at least 18 h.
   ​NOTE: The protocol can be paused here after polymerization. Polymerized specimen blocks can be stored in a dry container for years.

7. Ultrathin sectioning

1. Trim the polymerized specimen blocks carefully using a razor to expose the sapphire discs.
2. Dip the block tips with sapphire discs into liquid nitrogen for several seconds, and thaw at room temperature. Repeat the freeze-thaw action several times to detach the discs from the blocks.
   NOTE: Avoid prolonged freezing, as this may crack the specimen blocks. Gently detach the sapphire discs with a blade, avoiding excessive force, which may damage the samples.
3. Trim the block face to a trapezoidal shape for ultrathin-sectioning (Figure 1G).
4. Obtain 70-100 nm ultrathin sections with an ultramicrotome (Figure 1H).
5. Pick the sections on 200-mesh hexagonal copper grids.
   ​NOTE: The protocol can be paused here after sectioning. The sections on girds can be stored in a dry container for years. If desired, the sections on grids can be stained with 2% uranyl acetone to obtain better membrane contrast. Lead staining should be avoided as it will mask the nanogold particle signals.

8. TEM imaging (Figure 1I)

1. Examine the ultrathin sections on copper grids using a transmission electron microscope. Typically, a magnification range from 11 kX to 30 kX is suitable for visualizing AuNPs in cells, and an appropriate defocus value is needed to ensure 2-3 nm AuNPs are visible.

Results

The ANSM-based AuNP synthesis technique is an extremely useful tool for labeling and detecting MT-tagged proteins with TEM²⁶. To validate its robustness in mammalian cells, three stable cell lines expressing EGFP-MTn-KDEL, Ost4-EGFP-MTn, or Mito-acGFP-MTn in Hela cells were generated. KDEL is a canonical C-terminal endoplasmic reticulum (ER) retention/retrieval sequence, which maintains the fusion protein EGFP-MTn-KDEL within the ER lumen or the perinuclear space of the nuclear envelope (NE). Ost4...

Discussion

The study presents here a robust clonable EM labeling technology for the single-molecule visualization of protein molecules within the cellular environment with ultrastructural resolution. The AuNPs directly synthesized on genetically encoded cysteine-rich tags provide unambiguous and precise localization of the target proteins. High-pressure freezing and freeze-substitution technique excellently preserve the ultrastructure of biological samples. Taken together, the clonable electron microscopy labeling technology presen...

Materials

Name	Company	Catalog Number	Comments
0.025 mm/0.275 mm Aluminum carrier	Beijing Wulundes Biotech Ltd., or Engineering Office of M. Wohlwend
0.2 M HEPES buffer			Dissolve HEPES (0.2 M) in 980 mL of ddH2O, then add 10 mL of 100 mM MgCl2 and 10 mL of 100 mM CaCl2 (final concentration 1 mM), respectively, adjust pH to 5.5
1.5 mL MaxyClear snaplock microtube	Axygen Scientific	MCT150C
2 mL polypropylene screw cap microtubes	Biologix	81-0204
200 mesh hexagonal copper grid	Tedpella inc	G200HEX
2-mercaptoethanol	Amresco	0482-250ML
35 mm cell culture dishes	Corning	430165
50 mL polypropylene centrifuge tubes	Corning	430928
Acetone	Beiijng Tong Guang Fine Chemicals Company	31025
Automated freeze substitution machine	Leica	AFS2
Customized 3.05 mm x 0.66 mm specimen holders for HPF	Beijing Wulundes Biotech Ltd.
D-penicillamine	TCI	P0147
Dulbecco’s modified Eagle medium	GBICO	C11965500BT
Fetal bovine serum	GBICO	10099-141C
Flat bottom embedding capsule	Tedpella inc
Foam cryobox
Formvar 15/95 resin	Electron Microscopy Sciences	15800
HAuCl4	Sigma	4022-1G
HEPES	sigma	H3375-500G
HPF machine	Wohlwend	HPF compact01
Methonal	Beiijng Tong Guang Fine Chemicals Company	12397
NaBH4	Sigma	480886-25G
OsO4	Electron Microscopy Sciences	19110
PBS-A buffer			Dissolve NaH2PO4 (1.125 mM), Na2HPO4 (3.867 mM), NaCl (100 mM) in 1 L of ddH2O, adjust to pH 7.4
Qualitative filter paper (medium speed)	Beyotime Biotechnology	FFT08
Sapphire discs	Beijing Wulundes Biotech Ltd., or Engineering Office of M. Wohlwend
Solvent resistant pen	Electron Microscopy Sciences	62053-B
SPI-Pon 812 resin	SPI Inc	02659-AB
Transmission electron microscopy	FEI	Tecnai G2 spirit
Trypsin-EDTA	GBICO	C25200-056
Tweezers	Dumont
Ultramictotome	Leica	FC7
Uranyl acetate	Electron Microscopy Sciences	22400