The Application of Light-Assisted Drying to the Thermal Stabilization of Nucleic Acid Nanoparticles

Abstract

Background:

Cold-chain storage can be challenging and expensive for the transportation and storage of biologics, especially in low-resource settings. Nucleic acid nanoparticles (NANPs) are an example of new biological products that require refrigerated storage. Light-assisted drying (LAD) is a new processing technique to prepare biologics for anhydrous storage in a trehalose amorphous solid matrix at ambient temperatures. In this study, LAD was used to thermally stabilize four types of NANPs with differing structures and melting temperatures.

Methods:

Small volume samples (10 μL) containing NANPs were irradiated with a 1064 nm laser to speed the evaporation of water and create an amorphous trehalose preservation matrix. Samples were then stored for 1 month at 4°C or 20°C. A FLIR C655 mid-IR camera was used to record the temperature of samples during processing. The trehalose matrix was characterized using polarized light imaging (PLI) to determine if crystallization occurred during processing or storage. Damage to LAD-processed NANPs was assessed after processing and storage using gel electrophoresis.

Results:

Based on the end moisture content (EMC) as a function time and the thermal histories of samples, a LAD processing time of 30 min is sufficient to achieve low EMCs for the 10 μL samples used in this study. PLI demonstrates that the trehalose matrix was resistant to crystallization during processing and after storage at 4°C and at room temperature. The native-polyacrylamide gel electrophoresis results for DNA cubes, RNA cubes, and RNA rings indicate that the main structures of these NANPs were not damaged significantly after LAD processing and being stored at 4°C or at room temperature for 1 month.

Conclusions:

These preliminary studies indicate that LAD processing can stabilize NANPs for dry-state storage at room temperature, providing an alternative to refrigerated storage for these nanomedicine products.

Introduction

A challenge in the development of a range of new biologics, including protein-based drugs, assays, vaccines, and nanomedicine products, is that these items require cold-chain storage to maintain potency and/or functionality. Cold storage strategies can be challenging and expensive for the transportation of biologics, especially in low resource settings due to a lack of available infrastructure. The long-term preservation of biologics at near ambient temperatures is desirable for minimizing the cost and complexity of transportation and storage. Light-assisted drying (LAD) is a new, light-based processing technique to prepare biologics for dry-state (anhydrous) storage at near ambient temperatures.

Freeze-drying has achieved long-term preservation for some biological products.^1–4 However, freeze-drying is a costly and complex technique. In addition, the first step of the freeze-drying process involves freezing, which can damage biologics, limiting the applicability of this process.⁵ Foam-drying and spray drying have also been used to stabilize biologics. Foam drying produces a dried product by boiling the solution under reduced vapor pressure followed by rapid evaporation. Foam drying, much like freeze drying, is a complex process that requires long processing times, and biologics are exposed to extreme pressure conditions during processing that can be damaging.⁶ Spray drying produces a dehydrated powder by atomizing a liquid into a drying environment consisting of a heated, dry inert gas. Spray drying exposes biologics to a hot (100°C) drying gas, and aseptic processing poses a significant challenge for industrial-scale use of this technique.⁷

Research has demonstrated that anhydrous preservation in a trehalose amorphous solid matrix may be an alternative to freeze drying for the preservation of biological samples. Trehalose is used to form the sugar matrix because it can form an amorphous solid at room temperature and can also act as a bioprotectant.^8,9 A substantial reduction of molecular mobility is necessary to ensure an extended shelf life for anhydrous samples. To ensure that this is the case, samples need to be stored below the glass transition temperature, T_g, of the amorphous matrix to prevent degradation. The glass transition temperature for an amorphous trehalose solid formed by dehydration depends on the amount of water remaining in the sample after processing. Low moisture contents are necessary for storage at higher temperatures.

We have developed a new optical processing technique, LAD, to create trehalose amorphous solids for the preservation of biologics.^10,11 LAD uses illumination with near-infrared laser light to assist in the formation of trehalose amorphous solids. LAD allows for the precise deposition of energy during processing not offered by other drying techniques. The precise energy deposition gives control over the sample temperature during processing, which is important for avoiding injury to thermally sensitive biologics.¹⁰ LAD processing also results in a uniform distribution of the trehalose matrix and uniform water content throughout the sample. Precise energy deposition enables repeatable rapid attainment of the desired end moisture content (EMC) of the sample, which dictates sample storage temperature.^10,11

In this study, LAD is used to stabilize therapeutically relevant nucleic acid nanoparticles (NANPs).^12–17 NANPs are an attractive material for diverse applications in biomedical sciences because of their programmable multitasking and ability to respond dynamically to environmental changes. Confirmed practical applications of NANPs include in vivo imaging and coordinated delivery of multiple therapeutic agents.^18–24 Currently, the standard for stabilization and storage of NANPs after synthesis is refrigeration in a buffer solution.

Four types of representative NANPs (Fig. 1), RNA cubes,²⁵ RNA fibers,^26–29 RNA rings,^30–35 and DNA cubes^25,36–39 were LAD processed and then stored at either room temperature (20°C) or 4°C for 1 month. We present drying curves and thermal histories that allow for the determination of appropriate LAD processing parameters and polarized light imaging (PLI) to access the quality of the trehalose matrix. Damage to LAD-processed NANPs was assessed after storage using gel electrophoresis. These preliminary studies indicate that LAD processing can stabilize these NANPs for dry-state storage at room temperatures.

FIG. 1.

NANPs used in this study. Upper panel schematically shows the connectivity rules of assembled RNA and DNA NANPs that were confirmed by AFM with corresponding images shown in the lower panel. AFM, atomic force microscopy; NANPs, nucleic acid nanoparticles.

Materials and Methods

Preparation of NANPs

The DNA templates for individual RNA strands were amplified using polymerase chain reaction (PCR) with MyTaq™ Mix (Bioline). PCR products were purified with the DNA Clean and Concentrator™ Kit (Zymo Research). Production of the RNA was completed using in vitro transcription starting with incubation of the DNA templates at 37°C for 3.5 h with T7 RNA polymerase (Promega), 80 mM HEPES-KOH (pH 7.5), 2.5 mM spermidine, 50 mM DTT, 25 mM MgCl₂, and 5 mM of each rNTP. To stop the reaction, samples were incubated with RQ1 RNase-free DNase for an additional 30 min and then purified using denaturing 8 M urea polyacrylamide gel electrophoresis (PAGE, 15%). After visualizing the bands under UV light, they were cut out and eluted overnight in a crush and soak buffer (300 mM NaCl, 89 mM tris-borate (pH 8.2), 2 mM ethylenediaminetetraacetic acid). For precipitation of the RNAs, the elution was first mixed with 2.5 volumes of 100% ethanol, incubated at −2°C, centrifuged to remove supernatant, and the pellet was rinsed with 90% ethanol, vacuum dried, and dissolved in double-deionized water (17.8 MΩ·cm).

Six-stranded NANPs were assembled at 0.5 μM final concentration by mixing all six RNA strands in equimolar concentrations along with doubled deionized water and assembly buffer in one-pot thermal annealing. For that, samples were heated to 95°C for 2 min, mixed with assembly buffer (89 mM tris-borate (pH 8.2), 2 mM MgCl₂, 50 mM KCl), and incubated at 45°C for 30 min. All samples were stored at 4°C after preparation.

LAD processing

The LAD processing system is shown in Figure 2a. An IPG Photonics continuous wave ytterbium fiber laser at 1064 nm (YLR-5-1064) was used for LAD processing (maximum power output of 5 W). The laser has a factory collimated Gaussian beam with a full width at half maximum spot size of 4.5 mm, which was measured using a Beam Track 10A-PPS thermal sensor (Ophir Photonics). A FLIR SC655 mid-IR camera was used to record the temperature of samples during processing. All studies were performed in a humidity-controlled environment that was kept at ∼11% relative humidity (RH). This was achieved by pumping dry air into a chamber containing the experimental setup and monitoring the RH with a temperature and RH logger (ONSET UX100-011).

FIG. 2.

(a) LAD experimental setup enclosed in the low relative humidity chamber. (b) PLI setup. Samples were placed between the polarizer and analyzer. LAD, light-assisted drying; PLI, polarized light imaging.

Samples consisted of 10 μL droplets of NANPs (concentration 0.25 μM) suspended in a drying solution (DS) consisting of 0.2 M disaccharide trehalose in 0.33 × phosphate buffer solution. For each test, a 10 μL droplet of the NANP/drying solution was deposited onto an 18 mm diameter borosilicate glass coverslip (Fisher brand 12–546). The glass coverslips allow for easy recovery and rehydration of the NANPs after LAD processing. The initial mass of the sample was determined gravimetrically using a 0.01 mg readability balance (RADWAG AS 82/220.R2).

RNA fibers (n = 6), RNA rings (n = 6), and RNA cubes (n = 6) were processed for 40 min at 5 W (26.9 W/cm²), and the DNA cubes (n = 6) were processed for 40 min at 4 W (21.5 W/cm²). DNA cubes were processed at a lower power because higher powers caused thermal damage due to the lower melting temperature of these NANPs.³⁷ The thermal damage to the DNA cubes when processed at 5 W was confirmed using gel electrophoresis. The temperature of the sample was monitored during processing using the thermal camera. The maximum temperatures reached during processing for the RNA fibers, RNA rings, RNA cubes, and DNA cubes were 35.4 ± 0.8°C, 35.8 ± 0.2°C, 35.8 ± 0.6°C, and 31.3 ± 1.1°C, respectively. After irradiation, the sample was removed from the humidity chamber and immediately massed again. EMC, which is a measure of the amount of water relative to the dry mass of a sample, was calculated using Equation 1:

$$EMC=\frac{m_f-m_s-m_{dw}}{m_{dw}}$$ (1)

where m_f is the final mass of the sample, including the mass of the substrate, m_s is the mass of the substrate, and m_dw is the measured dry weight of the initial sample. The dry weight was calculated by multiplying the initial mass of the sample by the percent dry weight (%DW), which was determined using the bake-out method. The %DW is dominated by trehalose and the buffer solutions, therefore is the same for all types of NANPs. The average %DW of all the NANPs was 5.9 ± 0.2%. After LAD processing, samples were stored individually in small volume containers inside moisture barrier bags (ULine) for 1 month. The RH inside the bags was 11.0 ± 0.5% RH and was measured at the start and end of storage. Three samples for each type of NANP were stored at 4°C, and three samples were stored at room temperature (∼20°C).

Polarized light imaging

To investigate crystal content in the samples, PLI was used. The PLI experimental setup (Fig. 2b) consisted of a white light fiber optic illuminator (41720, Cole Palmer), two linear polarizers (LPVISE050-A; Thorlabs), with the second polarizer acting as an analyzer, and a digital camera (Nikon D100) aligned in the vertical direction. The camera was equipped with a Nikon 28–105 mm f/3.5–4.5 lens and manually focused on the image plane. The spatial resolution of the setup was 10 μm/pixel. Samples were placed on a glass microscope slide in between the polarizers and imaged from above. Two images were taken: the first with the analyzer oriented at 0° to the polarizer and the second with the analyzer oriented at 90° to the polarizer. For each sample, images were taken immediately after processing and after storage.

Gel electrophoresis

Structures of assembled NANPs were verified on nondenaturing native polyacrylamide gels (native-PAGE, 8%, 37.5:1) immediately after initial assembly, after LAD processing, and after storage. Samples were rehydrated by first pipetting water (10 μL) directly onto the center of the dried droplet and then mixing with the pipette tip. After 1–2 min, the solution was transferred into an Eppendorf tube. These tubes were left on ice or in the cold room (4°C) until used. The gels were run in a Mini-PROTEAN Tetra system (Bio-Rad) with running buffer (89 mM tris-borate (pH 8.2), 2 mM MgCl2) for 30 min at 300 V in a 4°C cold room. Afterward, the gels were stained with ethidium bromide for 5 min and washed with water. Gel imaging was completed using a ChemiDoc MP system (Bio-Rad).

Results and Analysis

Drying curves

Figure 3 shows the drying curves of the four types of NANPs (RNA fibers, RNA cubes, RNA rings, and DNA cubes; n = 6 for each type of particle). Drying curves show how the EMC of a sample changes with processing time. Measurements of EMC were taken at processing times of 0, 5, 10, 20, 25, 30, and 40 min. The EMC decreases approximately exponentially as the processing time increases, indicating that the majority of sample drying occurs during the early stages of the LAD process. The EMC reaches a plateau at ∼30 min. The average EMC of samples after 30 and 40 min of processing is given in Table 1. There is no significant change in the EMC between 30 and 40 min. This suggests that a processing time of 30 min is sufficient to achieve low EMCs for the 10 μL samples used in this study.

FIG. 3.

Average EMC as a function of processing time for all types of NANPs. Panel (a) shows the drying curve for RNA cubes, panel (b) is for RNA fibers, panel (c) is for RNA rings and panel (d) is for DNA cubes. EMC, end moisture content.

Table 1.

End Moisture Content (EMC) of Nucleic Acid Nanoparticles After 30 and 40 Min of Light-Assisted Drying Processing and EMCs After Processing for 40 Min Followed by 1-Month Storage

Type of NANP	EMC after 30 min	EMC after 40 min	EMC after 1 month storage
RNA cubes	0.63 ± 0.09	0.62 ± 0.03	0.54 ± 0.07
RNA fibers	0.50 ± 0.2	0.54 ± 0.04	0.59 ± 0.05
RNA rings	0.69 ± 0.06	0.85 ± 0.04	0.88 ± 0.05
DNA cubes	0.49 ± 0.03	0.51 ± 0.03	0.50 ± 0.03

EMC, end moisture content.

Table 1 also provides the average EMC of samples processed for 40 min and then stored for 1 month in a low humidity environment. No significant change is observed in the EMC after storage. The DNA cube NANPs were processed at a lower laser power of 4 W to prevent thermal damage to these particles. Again, no significant difference between processing times of 30–40 min is noted.

Thermal histories

Figure 4 shows the thermal histories of RNA cubes (a), RNA fibers (b), RNA rings (c), and DNA cubes (d) (n = 6 for each) and the average curves for all samples processed for 40 min. All graphs show the change in the droplet temperature compared to the initial droplet temperature as a function of time. All RNA NANPs were processed at 5 W; therefore, their thermal curves are similar. Slight variations in the sample temperatures are likely due to fluctuations in the ambient temperature in the processing environment. The DNA cubes were processed at lower laser power than the other NANPs because they have a lower melting temperature than the RNA NANPs. As expected, the maximum sample temperature is lower for the DNA cubes. However, the change in the temperature of the samples was similar to that seen for the RNA NANPs.

FIG. 4.

Thermal histories of NANPs during 40 min of LAD processing. Panel (a) shows the thermal history for RNA cubes, panel (b) is for RNA fibers, panel (c) is for RNA rings and panel (d) is for DNA cubes.

The overall shape of the thermal history is the same for all types of NANPs processed. The initial rise in the temperature is the result of laser heating of the water in the sample. A maximum temperature is reached during the first minute of processing. After this peak in temperature, evaporative cooling reduces the sample temperature, indicating that LAD is effectively removing water from the sample. Near 20 min, the temperature reaches a minimum value and then again starts to increase. By 30 min, the temperature of the sample plateaus. On this plateau, the heating and cooling are balanced resulting in a stabilization of sample temperature. This plateau marks the end of significant rapid evaporation of water from the sample. This is consistent with the EMCs seen at 30 and 40 min. There was no significant decline in the water content between these processing times. A higher peak temperature results in the sample reaching the plateau more quickly. This is consistent with the idea that at a higher temperature, evaporation will drive water out of the sample more quickly so that the sample will reach the end of significant evaporative cooling (the plateau) faster.

A comparison of EMC versus the peak, minimum, and plateau temperatures reached by the samples during LAD processing reveals that there is little/no correlation between EMC and peak/minimum/plateau temperature. These results suggest that the final EMC does not depend on the peak temperature as long as samples are processed until they reach the temperature plateau. These results demonstrate that the thermal history can be used to determine the processing time that will maximize the amount of water removal from the sample during LAD processing.

Polarized light imaging

PLI was used to investigate the stability of the trehalose matrix against crystallization immediately after LAD processing and after storage. This is significant as crystallization of the matrix can damage embedded biologics. As polarized light travels through the amorphous matrix, the plane of polarization of the light remains constant and the light cannot propagate through the crossed polarizer. Any crystalline inclusions in the matrix are birefringent and will rotate the plane of polarization so that light can travel through the crossed polarizer resulting in a bright spot in crossed-polarizer images.

Figure 5 shows PLI for RNA cubes after processing (a–b) and after storage (c–d) at room temperature for one month. PLI for the other NANPs was similar, and results were similar at both room temperature and 4°C storage. The images at left were taken with the polarizer and analyzer at the same angle providing a detailed view of the droplet after LAD processing and storage. The wrinkled appearance seen in some samples immediately after processing is the result of water absorption by the surface layers of the droplet when exposed to a high humidity laboratory environment. This effect was noted in previous studies and does not adversely affect the matrix or embedded biologic.¹¹ All samples continued to dry during low humidity storage, and wrinkling was not seen in these samples as water absorption was not large enough to cause wrinkling.

FIG. 5.

Representative polarized light images of LAD processed RNA cubes. Panels (a, b) were taken immediately after LAD processing, and panels (c, d) were taken after 1 month of storage at room temperature. Panels (a, c) were taken with the polarizer and analyzer at the same angle and reveal the morphology of the droplet. Panels (b, d) show images taken through crossed polarizers, and regions of crystallization should appear as bright spots in these images. For this sample, no crystals were detected indicating that the matrix is stable against crystallization when stored at room temperature.

In the crossed-polarizer images shown in Figure 5 regions of crystallization should appear as bright spots in these images. For these samples, no crystals were detected in the matrix immediately after LAD processing or after 1 month of storage. The amount of crystallization in droplets was characterized by determining a pixel area in each cross-polarizer image that was brighter than the background. Table 2 shows the average crystal area after 1-month storage at 4°C and at room temperature for all types of NANPs. The average crystal area of all NANPs both before and after storage is small suggesting that the trehalose matrix was resistant to crystallization during processing and after storage at 4°C and at room temperature.

Table 2.

Average Crystal Area Measured in Pixels of Light-Assisted Drying Processed Samples Before and After 1-Month Storage

Type of NANP	Before storage	After storage
Storage at 4°C
RNA cubes	20 ± 10	30 ± 60
RNA fibers	20 ± 20	16 ± 10
RNA rings	50 ± 40	70 ± 60
DNA cubes	90 ± 60	30 ± 20
Storage at 20°C
RNA cubes	2 ± 1	2 ± 1
RNA fibers	2 ± 1	2 ± 1
RNA rings	2 ± 1	10 ± 6
DNA cubes	150 ± 120	120 ± 30

Gel electrophoresis

Figure 6 shows the native-PAGE results for LAD processed DNA cubes, RNA cubes, RNA rings, and RNA fibers. Each gel contains three sets of samples. From left to right for each sample set, the figures contain the results of an unprocessed control stored at 4°C, a LAD processed sample stored at 4°C, an unprocessed control stored at room temperature, and a LAD processed sample stored at room temperature. The controls were NANPs suspended in the LAD processing buffer containing trehalose. The native-PAGE results for DNA cubes, RNA cubes, and RNA rings are free of stray fragments and all bands are uniform. This indicates that the main structures of these NANPs were not damaged significantly after LAD processing and being stored at 4°C or at room temperature for 1 month. No bands are evident in the gel results for the RNA fibers. The RNA fibers exhibit a range of sizes, and this method is not ideal for detecting these NANPs.

FIG. 6.

Ethidium bromide total staining native-PAGE visualizing the retention of structures after one month of storage for (a) DNA cubes, (b) RNA cubes, (c) RNA rings, and (d) RNA fibers. PAGE, polyacrylamide gel electrophoresis.

Discussion and Conclusions

A processing time of 30 min at 5 W for RNA NANPs and 4 W for DNA cubes can dehydrate the embedded biologics without causing thermal damage. Processing beyond 30 min does not significantly reduce the EMC. Using the thermal histories in conjunction with the drying curves allows for a determination of the optimum processing time for samples. PLI shows that the trehalose matrix is stable against crystallization for all NANPs when stored at room temperature and 4°C for 1 month. Gel electrophoresis shows that RNA cubes, RNA rings, and DNA cubes in trehalose buffer are not damaged after 40 min of LAD processing and 1-month storage (at 4°C and at room temperature). Gel electrophoresis yielded inconclusive results for the RNA fibers.

In a recent study, Tran et al. compared LAD stabilization of a panel of four representative DNA and RNA NANPs to dehydration using lyophilization and a protocol that centrifuged samples while exposing them to a combination of heat and infrared radiation.⁴⁰ Lyophilization and LAD proved to be better methods for stabilization of NANPs compared to SpeedVac. The droplets used in this study were small volume (10 μL). Typical doses of biologics such as vaccines are larger (0.25–1 mL). Recently, we have successfully applied LAD to larger volume droplets (0.25 mL).⁴¹ Processing time for these larger droplets are longer because a larger volume of water must be removed but were still much shorter than typical lyophilization times.

Authors' Contributions

All authors contributed to the study conception and experimental design. The first draft of the article was written by P.A.L. and S.R.T. All authors commented on the previous versions of the article. P.A.L., D.P.F., A.T., R.Q.M., D.B., and M.C. performed the laboratory work. P.A.L., S.R.T., and K.A.A. produced the figures. All authors read and approved the article.

Disclaimer

The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.

Author Disclosure Statement

No conflicting financial interests exist.

Funding Information

This work was supported, in part, by funds provided by The University of North Carolina at Charlotte Faculty Research Grants Program (S.R.T.). The research reported in this publication was also supported by the National Institute of General Medical Sciences of the National Institutes of Health under award nos. R01GM120487 and R35GM139587 (to K.A.A.).

Supplementary Material

Supplementary Table S1