Materials

2-Ethyl-1-hexanol (98%, Sigma); 1-(3-dimethylaminopropyl)-3-ethylcarbodiimide HCl (Carbosynth); 1,6-diaminohexane (98%, Sigma); PEG-bis-(N-succinimidyl succinate) (M_w = 2,000, Sigma); DyLight 405 NHS ester (ThermoFisher); FITC NHS ester (ThermoFisher); BSA (heat-shock fraction, pH 7.0, ≥98%; Sigma); streptavidin from Streptomyces avidinii (Sigma); Tamavidin 2-HOT, recombinant (Wako Chemicals); Dynabeads M-270 amine (Invitrogen); Dynabeads MyOne carboxylic acid (Invitrogen); 1 M MgCl₂ (Invitrogen); 1 M Tris pH 8.0 RNase free (Invitrogen); 5 M NaCl (Invitrogen); EvaGreen (Biotium); KAPA HiFi HotStart PCR kit (Roche); Micellula DNA emulsion and purification kit (EURx); ibidi anti-evaporation oil (ibidi); 30% (19:1 monomer:bis) acrylamide solution (Bio-Rad); and SYBR Gold (ThermoFisher) were used as received. All the other chemicals used were purchased from Sigma. The enzymes were purchased from New England Biolabs, unless noted otherwise.

Synthesizing BSA-NH₂–PNIPAm nanoconjugates

Cationized BSA (BSA-NH₂) was synthesized according to a previously reported method³⁶. Typically, a solution of diaminohexane (1.5 g, 12.9 mmol in 10 ml Milli-Q water) was adjusted to pH 6.5 using 5 M HCl and added dropwise to a stirred solution of BSA (200 mg, 3 μmol in 10 ml Milli-Q water). The coupling reaction was initiated by adding 100 mg of 1-(3-dimethylaminopropyl)-3-ethylcarbodiimide HCl immediately and then another 50 mg after 5 h. If needed, the pH value was readjusted to 6.5 and the solution was stirred for another 6 h and then centrifuged to remove any precipitate. The supernatant was dialysed (Medicell dialysis tubing; molecular weight cutoff (MWCO) of 12−14 kDa) overnight against Milli-Q water and freeze dried.

End-capped mercaptothiazoline-activated PNIPAm (M_n = 13,284 g mol⁻¹, 4 mg in 5 ml Milli-Q water) was synthesized according to the previously reported method¹ and added to a stirred solution of BSA-NH₂ (10 mg in 5 ml PBS buffer at pH 8.0). The molecular weight and polydispersity of activated PNIPAm were determined using gel permeation chromatography (Supplementary Fig. 20). The solution was stirred for 10 h and then purified using a centrifugal filter (Millipore, Amicon Ultra; MWCO, 50 kDA) and freeze dried. After freeze drying, the obtained BSA-NH₂–PNIPAm conjugate was characterized by matrix-assisted laser desorption/ionization–mass spectrometry and zeta potentiometry (Supplementary Fig. 20).

FITC- and DyLight-405-labelled BSA-NH₂–PNIPAm conjugates were prepared using the same method, except that labelled BSA was used as the starting material.

Labelling BSA with fluorescent dyes

BSA was labelled with FITC as follows: 200 mg of BSA was dissolved in 10 ml of 50 mM sodium carbonate buffer (pH 9). Then, 2.36 mg FITC was dissolved in 590 μl DMSO and added to the stirred BSA solution. The resulting solution was stirred for 5 h, purified by dialysing (Medicell dialysis tubing; MWCO, 12−14 kDa) overnight against Milli-Q water and freeze dried. BSA was labelled with DyLight 405 as follows: 30 mg of BSA was dissolved in 6 ml of 50 mM sodium carbonate buffer (pH 9). Then, 1 mg of DyLight 405 NHS ester was dissolved in 100 μl DMF and added to the stirred BSA solution. The solution was stirred for 2 h, purified by dialysing (Medicell dialysis tubing; MWCO, 12−14 kDa) overnight against Milli-Q water and freeze dried. Using ultraviolet–visible spectrophotometry, we determined the average number of dyes per protein. For DyLight 405 labelling, we measured, on average, 1.4 dyes per BSA molecule. For FITC labelling, we measured 1.3 dyes per BSA molecule.

Preparing Tamavidin 2-HOT-containing proteinosomes

Proteinosomes containing Tamavidin 2-HOT and magnetic particles (Dynabeads M-270 amine) were prepared similar to previous³⁷ descriptions for streptavidin-containing proteinosomes. Proteinosomes used for the experiments (Fig. 1c–e) did not contain magnetic particles. BSA–PNIPAm nanoconjugates (6 mg ml^–1 total, 1 mg ml^–1 of which was fluorescently labelled) were mixed with 4 µM Tamavidin 2-HOT and 4 mg ml^–1 Dynabeads in 7.5 µl aqueous phase; 0.6 mg of PEG-bis(N-succinimidyl succinate) (M_w = 2,000) was dissolved in 7.5 µl of 50 mM sodium carbonate buffer (pH 9) and added to the mix, which was then briefly vortexed. Next, 300 µl of 2-ethyl-1-hexanol was added, followed by vortexing to yield the Pickering emulsion. The resulting mixture was left at room temperature for 2 h to crosslink the nanoconjugates. The oil phase was removed by pipetting away the upper oil layer and 300 µl of 70% ethanol was added to resuspend the sediment. The dispersion was then sequentially dialysed (Medicell dialysis tubing; MWCO, 12−14 kDa) against 70% and 50% ethanol for 2 h each and finally overnight against Milli-Q water to yield proteinosomes in water. Proteinosomes were then stored at 4 °C for later use.

DNA oligonucleotide synthesis

Except DNA encoding files from Twist Bioscience, all the DNA oligonucleotides were purchased from Integrated DNA Technologies. Modified oligonucleotides that were purified with high-performance liquid chromatography were purchased, and desalted non-modified oligonucleotides were purchased. Stock solutions (100 and 10 μM) were made using nuclease-free TE buffer (10 mM Tris, 0.1 mM EDTA, pH 7.5; Integrated DNA Technologies) and stored at −30 °C. DNA encoding for the files was ordered from Twist Bioscience. These files were individually PCR amplified in a 20 µl reaction containing 1 ng DNA pool, 0.5 µM forward and reverse primers, and KAPA HiFi HotStart polymerase. The amplification protocol was as follows: denaturing at 95 °C for 3 min, denaturing at 98 °C for 20 s, annealing at 65 °C for 15 s and extending at 72 °C for 15 s. The second denaturing, annealing and extending steps were repeated 8–10 times, followed by a final extension at 72 °C for 30 s before cooling down to 4 °C. The resulting amplicons were then purified using the Qiagen PCR extraction kit following the manufacturer’s instructions. The files that were obtained this way were then mixed in equal ratios in 10 ng and shipped at room temperature from Seattle to Eindhoven. These templates were then used similar to ssDNA oligonucleotides ordered from Integrated DNA Technologies.

Preparing dsDNA with biotin of fluorophores

Double-stranded complexes consisting of strands shorter than 100 nt were formed by thermal annealing. Biotinylated strands were mixed with non-biotinylated strands at 12 and 10 µM and heated to 95 °C in a thermocycler for 3 min. The samples were subsequently cooled to room temperature at a rate of −0.5 °C min^–1.

Here dsDNA strands longer than 100 bp, with fluorescent and/or biotin modifications, were prepared from either single-stranded ultramer templates or dsDNA files and modified primers using PCR since these constructs could not be directly ordered. Typically, reactions were performed at the 100 µl scale using 5 µl (1 nM) diluted template, primers (0.5 µM each) and KAPA HiFi HotStart polymerase. The amplification protocol was as follows: denaturing at 95 °C for 3 min, denaturing at 98 °C for 20 s, annealing at 65 °C for 15 s and extending at 72 °C for 15 s. The second denaturing, annealing and extending steps were repeated 16 times, followed by a final extension at 72 °C for 30 s before cooling down to 4 °C. The resulting amplicons were then purified using the Qiagen PCR extraction kit following the manufacturer’s instructions.

Localizing DNA in proteinosomes

Biotin-labelled DNA was initially localized in 10.0 mM Tris (pH 8.0) with 11.5 mM MgCl₂ and 0.1% vol/vol Tween 20. Typically, 10 µl of proteinosome-containing solution was added to 5 µl of 4× buffer solution and 5 µl of DNA to be localized. The mixture was kept at 4 °C overnight. The following day, 500 µl wash buffer consisting of 10.0 mM Tris (pH 8.0), 1 M NaCl, 11.5 mM MgCl₂ and 0.1% vol/vol Tween 20 was added, left at 4 °C overnight and removed the following day. Secondary washing steps were performed similar to the first steps, except that no overnight step was used. Instead, proteinosomes were separated from the solution by placing the mixture in a magnetic separation rack (DynaMag, Invitrogen) for 3 min, after which the supernatant was removed by a pipette.

Initial localization stability testing

Proteinosomes containing localized DNA were heated to 95 °C in a 10 µl solution using a MiniPCR thermocycler for at least 5 min. After cooling to room temperature, a 2 µl drop was placed on a glass microscopy coverslip and confocal micrographs were taken.

Temperature-dependent fluorescence microscopy

Fluorescence data were acquired using a confocal laser scanning microscope (Leica SP8) equipped with solid-state lasers (405 nm for DyLight 405, 552 nm for Alexa 546 and 638 nm for Cy5) and a hybrid detector in the photon-counting mode. The time-lapse measurements were performed with a ×10/0.40 numerical aperture (field of view, 1.55 × 1.55 mm²; slice thickness, 7 μm) at a resolution of 512 × 512 pixels. High-temperature confocal laser scanning microscopy data were obtained using a VAHEAT micro-heating system (Interherence) with SmS-r substrates. A 50 µl solution of DNA-loaded proteinosomes was pipetted onto the sample cell and covered with 150 µl ibidi anti-evaporation oil to prevent evaporation. Room-temperature diffusion experiments were conducted in our previously³⁷ described microfluidic trapping array. Data processing was done using a custom Python code similar to what we have previously described³⁷.

Statistics

All the results reporting statistical values were obtained from independent triplicates. The analysis was performed using Python’s SciPy (Python 3.6.5, SciPy version 1.1.0) library. Welch’s two-sided t-test was used to compare two populations. Statistical significance between more than two samples was determined using one-way analysis of variance, followed by a post-hoc analysis using Tukey’s multiple comparison testing. Only values of p < 0.05 were considered to be statistically significant.

SDA

The SDA of DNA localized in proteinosomes was performed using a protocol adapted from another work⁴⁶. The reaction mixture consisted of 1× NEB buffer 2, 0.5 µM primer, 250.0 µM dNTP each, 0.125 U µl^–1 Klenow Fragment (Exo-), 0.250 U µl^–1 Nt.BspQI, 0.2 mg ml^–1 BSA, 4 µM T4 gene 32 protein and 1× EvaGreen. The reaction volume was 25 µl, 2 µl of which consisted of proteinosomes in a buffer (10.0 mM Tris with 11.5 mM MgCl₂ and 0.1% vol/vol Tween 20). The reaction was kept at 37 °C and recorded using a CFX96 Touch real-time PCR detection system (Bio-Rad). To prevent evaporation, the plate was sealed with a transparent sticker. The production rate was determined using Python by fitting a linear function to the fluorescence intensity and cycle number.

qPCR

qPCR was performed using the CFX96 Touch real-time PCR detection system (Bio-Rad). The total reaction volume was 25 µl and consisted of KAPA HiFi HotStart, 0.5 µM primers, 1× EvaGreen and 2 µl template solution (DNA or proteinosomes). The initial denaturation was set to 3 min at 95 °C, and then 40 denaturation cycles at 98 °C for 20 s, annealing at 65 °C for 15 s and extension at 72 °C for 15 s were performed, followed by a final extension at 72 °C for 30 s before cooling down to 4 °C. Fluorescence was measured during each annealing step. To prevent evaporation, the plate was sealed with a transparent plate sealer. CFX Maestro software version 3.1.1517.0823 (Bio-Rad) was used to perform baseline correction and calculate the threshold cycles (C_t).

Chimera formation determination using PAGE gel analysis

The total reaction volume was 25 µl and consisted of KAPA HiFi HotStart, 0.5 µM primers and 2.5 µl proteinosome solution. Thermocycling was performed in a T1000 Touch thermocycler (Bio-Rad). Following initial denaturation for 3 min at 95 °C, 20 cycles of denaturation at 98 °C for 20 s, annealing at 65 °C for 15 s and extension at 72 °C for 15 s were performed, followed by a final extension at 72 °C for 30 s before cooling down to 4 °C. To the PCR mixtures, 5 µl of 6× loading dye (ThermoFisher) was added before loading 12 µl on 10% TB-Mg PAGE gels. The gels were cast using 30% (19:1 monomer:bis) acrylamide solution. Running and gel buffers were 44.5 mM Tris, 44.5 mM boric acid and 11.5 mM MgCl₂. The gels were run for 1 h 15 min at 150 V in a Criterion vertical cell electrophoresis device (Bio-Rad) and stained for 10–15 min using SYBR Gold (ThermoFisher). The images were taken using an ImageQuant 400 Digital Imager (GE Healthcare). Bands-of-interest image intensities were determined using ImageJ’s gel analysis plug-in. In some gels, a signal is observed at the top of the gel, which we attribute to larger complexes (such as polymerase–DNA complexes) present in unpurified reaction mixtures. These signals were not considered in the analysis. Statistical analysis was performed using Python.

Emulsion PCR

Emulsion PCR of the library was performed using a Micellula DNA emulsion and purification kit. A 50 µl reaction mixture containing 25 µl KAPA HiFi HotStart 2×, 50 ng template DNA, 4 µM primers (0.16 µM per pair) and 1.25 mg ml^–1 BSA was used. The emulsion was formed by adding 300 µl premixed inorganic phase and vortexing at the maximum speed in a fridge at 4 °C for 5 min, per the manufacturer’s instructions. The resultant emulsion was split into four tubes and thermocycled as follows: initial denaturation for 3 min at 95 °C, 18 cycles of denaturation at 95 °C for 20 s, annealing at 65 °C for 15 s and extension at 72 °C for 15 s were performed, followed by a final extension at 72 °C for 30 s before cooling down to 4 °C. The reactions were pooled, the emulsion was broken and DNA was purified according to the manufacturer’s instructions.

The relative concentrations of individual files in the purified DNA were then quantified using qPCR.

Quantifying multiplex PCR concentration

DNA localized inside a mixed pool of proteinosomes or in bulk was amplified in 25 µl reactions consisting of 2 µl template DNA and primers (0.4 µM of each forward and reverse pair; Supplementary Table 4 lists the primer sequences) using KAPA HiFi HotStart polymerase. Thermocycling was performed in a C1000 Touch thermocycler (Bio-Rad). Following initial denaturation for 3 min at 95 °C, 18 cycles of denaturation at 98 °C for 20 s, annealing at 65 °C for 15 s and extension at 72 °C for 15 s were performed, followed by a final extension at 72 °C for 30 s before cooling down to 4 °C. The reaction mixtures were purified using the Qiagen PCR extraction kit following the manufacturer’s instructions.

The relative concentrations of individual files in the purified DNA were then quantified using qPCR.

Library preparation and sequencing

Multiplex PCR reactions were performed at the Eindhoven University of Technology and shipped at room temperature to the University of Washington. On receipt, the samples were validated using an Implen nanophotometer. Subsequently, the samples were prepared for sequencing following the Illumina TruSeq Nano DNA Library Prep protocol. The ends were blunted with the End Repair buffer (ERP2) and then purified with Beckman Coulter AMPure XP beads; an ‘A’ nucleotide was annealed to the 3’ end with an A-tailing ligase. Ligation was performed using Illumina sequencing adapters from Illumina’s TruSeq DNA CD Indexes kit, with each sample ligated to a unique Illumina index. Finally, the samples were cleaned using Illumina sample purification beads and enriched using a 12 cycle PCR. The final product length and purity were qualified using a QIAxcel Bioanalyser. Then, the samples were individually quantified using qPCR and mixed to create an equal-mass library.

A final library was prepared for sequencing by following the Illumina NextSeq Denature and Dilute Libraries Guide. The sequencing libraries were loaded in the Illumina NextSeq at 1.3 pM with a 20% control spike-in of the ligated PhiX genome.

Analysis of Illumina sequencing data

Basecalling and demultiplexing of the sequenced samples were performed using bcl2fastq. The generated FASTQ files were then aligned against the reference sequences using Burrows–Wheeler Aligner⁵⁹. Next, the coverage for each sequence was determined using SAMtools⁶⁰.

Repeated-access PCR in bulk

Three files were mixed in equal molar ratios to a final concentration of 0.5 nM. Three aliquots were amplified in 100 µl reactions consisting of 5 µl template mix and primers (0.5 µM) using KAPA HiFi HotStart polymerase. The PCR cycling protocol comprised initial denaturation for 3 min at 95 °C, 10 cycles of denaturation at 98 °C for 20 s, annealing at 65 °C for 15 s and extension at 72 °C for 15 s, followed by a final extension at 72 °C for 30 s before cooling down to 4 °C. A 5 µl aliquot was taken in which primers and dNTPs were then inactivated using Exo-CIP Rapid PCR Cleanup Kit (NEB). From the resulting mixture, 5 µl was used in the next PCR as a template. The remaining 95 µl of the PCR reaction was stored at −30 °C.

Repeated access using emulsion PCR

Three files were mixed in equal molar ratios and emulsion PCR of the library was performed using a Micellula DNA emulsion and purification kit. A 50 µl reaction mixture containing 25 µl KAPA HiFi HotStart 2×, 50 ng template DNA, 0.5 µM primers and 1.25 mg ml^–1 BSA. The emulsion was formed by adding 300 µl premixed inorganic phase and vortexing at the maximum speed in a fridge at 4 °C for 5 min, per the manufacturer’s instructions. The resultant emulsion was split into four tubes and thermocycled as follows: initial denaturation for 3 min at 95 °C, 10 cycles of denaturation at 95 °C for 20 s, annealing at 65 °C for 15 s and extension at 72 °C for 15 s were performed, followed by a final extension at 72 °C for 30 s, before cooling down to 4 °C. The reactions were pooled, the emulsion was broken and DNA was purified according to the manufacturer’s instructions. From this purified reaction, 1 µl was used in the next reaction.

After four rounds of repeated access, the final purified mixture was amplified using bulk PCR to generate enough DNA for sequencing experiments. The protocol for this amplification used a total reaction volume of 25 µl and consisted of KAPA HiFi HotStart, 0.5 µM primers and 2 µl purified reaction mixture. Thermocycling was performed in a T1000 Touch thermocycler (Bio-Rad). Following initial denaturation for 3 min at 95 °C, 20 cycles of denaturation at 98 °C for 20 s, annealing at 65 °C for 15 s and extension at 72 °C for 15 s were performed, followed by a final extension at 72 °C for 30 s before cooling down to 4 °C. The reaction mixture was then purified using the Qiagen PCR extraction kit following the manufacturer’s instructions.

Repeated-access PCR in proteinosomes

Three files were localized in individual proteinosome populations, washed five times and mixed to create the final pool. Three aliquots were amplified in 100 µl reactions consisting of 5 µl of the proteinosome library and primers (0.5 µM), using KAPA HiFi HotStart polymerase. The PCR cycling protocol comprised initial denaturation for 3 min at 95 °C, 10 cycles of denaturation at 98 °C for 20 s, annealing at 65 °C for 15 s and extension at 72 °C for 15 s, followed by a final extension at 72 °C for 30 s before cooling down to 4 °C. The reaction mixtures were then placed on a magnetic separation rack (DynaMag, Invitrogen) for 3 min to recover the proteinosomes. Next, 95 µl of the reaction mixture was pipetted off and stored at −30 °C. The remaining 5 µl was washed three times using 100 µl wash buffer (10.0 mM Tris (pH 8.0), 1 M NaCl, 11.5 mM MgCl₂ and 0.1% vol/vol Tween 20). After the magnetic recovery of proteinosomes, a new reaction mix was added to the washed 5 µl solution of proteinosomes to make a final volume of 100 µl.

Preparing Tamavidin 2-HOT-containing proteinosomes for sorting

Due to the smaller sizes required by the nozzle used and interference from the magnetic particles normally employed, proteinosomes utilized in the sorting experiment were prepared according to a slightly modified protocol. Proteinosomes containing Tamavidin 2-HOT and magnetic particles (Dynabeads MyOne carboxylic acid) were prepared using methods similar to those described above. BSA–PNIPAm nanoconjugates (6.00 mg ml^–1 total, 1.00 mg ml^–1 of which were fluorescently labelled) were mixed with 4 µM Tamavidin 2-HOT and 0.75 mg ml^–1 Dynabeads in 20 µl total aqueous phase. Next, 0.6 mg of PEG-bis(N-succinimidyl succinate) (M_w = 2,000) was dissolved in 40 µl of 50 mM sodium carbonate buffer (pH 9), added to the mix and briefly vortexed. Then, 1 ml 2-ethyl-1-hexanol was added and the mixture was subsequently vortexed for 30 min to yield the Pickering emulsion. The resulting mixture was left at room temperature for 1.5 h to crosslink the nanoconjugates. The oil phase was removed by pipetting away the upper oil layer and 500 µl of 70% ethanol was added to resuspend the sediment. The dispersion was then sequentially dialysed (Medicell dialysis tubing; MWCO, 12−14 kDa) against 70% and 50% ethanol for 2 h each and finally overnight against Milli-Q water to yield proteinosomes in water, before filtering using a 30 µm CellTrics cell strainer (Sysmex). We verified these proteinosomes’ reduced size using confocal microscopy, the results of which are shown in Supplementary Fig. 2. Proteinosomes were then stored at 4 °C for later use.

Fluorescence-assisted sorting

A FACS Aria III flow cytometer (BD Biosciences), operating at low–middle pressure, was used to interrogate a mix of proteinosomes through a 100 μm nozzle. DyLight-405- and FITC-labelled proteinosomes were prepared for FACS, and the files encoded in DNA were localized overnight in the proteinosomes. These proteinosomes were subsequently washed with 500 µl wash buffer (10.0 mM Tris (pH 8.0), 1 M NaCl, 11.5 mM MgCl₂ and 0.1% vol/vol Tween 20) and stored overnight at 4 °C. The next day, 500 µl supernatant was removed, and fluorescently barcoded DNA (labelled with biotin and either Cy5 or Cy3) was added to the proteinosomes and allowed to localize for 15 min at room temperature. After barcode localization, the proteinosomes were washed four more times by the addition of 500 µl wash buffer, followed by magnetic separation and supernatant removal. Proteinosome populations were mixed just before sorting. A total of 1,000–2,000 events were recorded, from which two-dimensional plots of the forward-scattered light height (FSC-H) versus forward-scattered light area (FSC-A) were obtained. FITC fluorescence was interrogated using a 488 nm laser and a 530/30 nm detector; Dylight 405 was excited at 405 nm and detected at 460/55 nm. Cy5 (ex, 633; em, 660/20). Cy3 (ex, 561; em, 582/15). The gating was performed with BD FACSDiva software (BD Biosciences) and consisted of initial gating in the FSC-H versus FSC-A plot to select the proteinosomes against background and unincorporated magnetic beads. From this initial gate, we defined a high FITC and high DyLight 405 gate, each of which were subsequently split into high Cy3 and high Cy5 gates. These were the final gates used to sort the proteinosomes. Supplementary Fig. 16 shows the detailed gating strategy. Sorting was performed in BSA-coated tubes and the sorted populations were reanalysed using BD Aria III flow cytometer, without sorting. The final graphs were plotted using flowCore and flowViz packages in R.

• SEO Powered Content & PR Distribution. Get Amplified Today.
• PlatoAiStream. Web3 Data Intelligence. Knowledge Amplified. Access Here.
• Minting the Future w Adryenn Ashley. Access Here.
• Buy and Sell Shares in PRE-IPO Companies with PREIPO®. Access Here.
• Source: https://www.nature.com/articles/s41565-023-01377-4