Optimizing RNA quality in cryopreserved tissues without preservatives: impact of preservatives, thawing methods and tissue aliquot sizes

Abstract

Background

RNA degradation during freeze‒thaw cycles in cryopreserved tissues is a major challenge for biomedical research, particularly when tissues are stored without preservatives. While agents such as TRIzol and RNALater are effective for fresh tissues, their utility for archival frozen tissues remains unclear.

Results

We evaluated various RNA preservation strategies for frozen rabbit kidney tissues without preservatives, considering key variables such as thawing temperatures (ice vs. room temperature (RT)), preservatives (RNALater, TRIzol, and RL lysis buffers), processing delay (time before disruption), tissue aliquot sizes (ranging from 70–100 mg, 100–150 mg, to 250–300 mg), and freeze‒thaw cycles. Compared with RT-treated frozen rabbit tissues, preservative-treated tissues presented significantly greater RNA integrity when thawed on ice (p < 0.01). The RNALater group performed best in maintaining high-quality RNA (RIN ≥ 8). Although a significant difference in the RIN was observed between the 120-minute and 7-day processing delays (9.38 ± 0.10 vs. 8.45 ± 0.44), all the samples ≤ 30 mg maintained a RIN ≥ 8. For tissues ≤ 100 mg, thawing overnight on ice or at -20 °C maintained a marginally higher RIN (RIN ≥ 7). However, larger tissue aliquots (250–300 mg) presented significantly lower RINs with ice thawing than with thawing at -20 °C (5.25 ± 0.24 vs. 7.13 ± 0.69). After 3–5 freeze‒thaw cycles, tissues thawed at -20 °C presented notably greater variability in the RIN, particularly in larger tissue aliquots. In validation experiments involving cryopreserved human and murine kidney tissues, the RNALater-treated murine kidney tissues ≤ 30 mg consistently maintained high-quality RNA integrity (RIN ≥ 8), whereas the frozen human kidney tissues resulted in marginally reduced RINs compared with those of the LN grinding control (7.76 ± 0.54).

Conclusions

Preservatives, tissue aliquot sizes, and thawing methods significantly impact the RNA quality of frozen tissues originally stored without preservatives. Key recommendations include (1) adding RNALater during thawing, (2) thawing on ice for small aliquots (≤ 100 mg) or at -20 °C for larger samples, and (3) minimizing freeze‒thaw cycles, despite observed variations among species.

Introduction

The quality of RNA extracted from cryopreserved tissues determines the reliability of downstream applications, including quantitative reverse transcription PCR (qRT-PCR), microarray hybridization, and next-generation sequencing platforms [1–3]. Immersive cryopreservation in liquid nitrogen (LN) has been widely adopted as an effective method for tissue stabilization, achieving vitrification temperatures below -150 °C within seconds to effectively inhibit RNase activity and preserve RNA integrity [4]. Standard operating procedures typically recommend that tissues should be divided into aliquots measuring 0.5 × 0.5 × 0.4 cm³ (GB/T 40352.1–2021) or weighing 0.5–1 g [5, 6]. However, these conventional aliquot sizes exist practical challenges, since most commercial RNA extraction kits (e.g., the Qiagen RNeasy Mini Kit, Roche High Pure RNA, Promega ReliaPrep™ systems, Qiagen AllPrep^® and Zymo Research Quick-RNA MiniPrep Kit) are optimized for ≤ 30 mg of tissue inputs [1, 7–10]. Reducing aliquot sizes to meet the specifications of kits substantially increases the processing time, storage costs, and potential quality variability. Consequently, frozen tissue aliquots prepared according to standard operating procedures require either being smashed via LN in a prechilled mortar or thawing followed by aseptic dissection into ≤ 30 mg aliquots.

The adverse effects of freeze‒thaw cycles on RNA integrity have been well reported [6, 11, 12]. A critical challenge lies in maintaining RNA quality in residual tissue following partial retrieval of frozen samples. While pretreatment with preservatives such as TRIzol [13–15] or RNALater [16, 17] immediately upon sample collection can effectively decrease RNA degradation, strategies for rescuing RNA quality in archival tissues originally frozen without preservatives remain poorly investigated. Poutoglidou et al. (2021) demonstrated the application of RNAlater-ICE for RNA extraction from bone samples of Wistar rats cryopreserved without preservatives [18]. This method simply submerged frozen tissue in 10 × volumes of RNAlater-ICE at -20 °C overnight. However, the high costs limit its practical utility, especially for scenarios such as biobanks where large quantities of unprotected tissues are preserved. Biobanks serve as critical infrastructures for biomedical research by ensuring the quality management and long-term preservation of biological samples [19–21]. It is necessary to develop optimized protocols for handling unprotected archival tissues, particularly when partial retrieval is required for quality control or pilot studies.

This study aimed to establish an optimized workflow that maximizes both RNA quality and efficient utilization of precious biobanked samples, particularly focusing on archival tissues stored without preservatives. We first evaluated multiple preanalytical variables affecting RNA quality in frozen rabbit kidney tissues: thawing temperature, application of preservatives, processing delay (time before disruption), tissue aliquot sizes and number of freeze‒thaw cycles (Fig. 1). Subsequently, experimental validation was performed in both frozen human and murine kidney tissues.

Fig. 1.

Schematic workflow of frozen rabbit tissue preprocessing for RNA isolation

Materials and methods

Rabbit tissue preparation

The rabbit kidney study was conducted on a healthy, male rabbit (12 weeks old) weighing approximately 2.5 kg, which was procured from Hubei Yizhi-cheng Biotechnology Co., Ltd. The experiment received approval from the Experimental Animal Welfare and Ethics Committee of Zhongnan Hospital, Wuhan University (No. 2025109). The rabbit was initially anesthetized via intravenous injection of 3% sodium pentobarbital (30 mg/kg) via the marginal ear vein. Once anesthesia was achieved, blood samples were collected for further study. Subsequently, euthanasia was performed by administering 10% sodium pentobarbital (100 mg/kg) through the same route. After euthanasia was confirmed, the visceral organs, including the kidneys, were harvested. The fresh tissue samples were promptly sectioned into fragments ranging from 50 to 300 mg, aliquoted into cryovials, and subsequently stored in vapor-phase LN for one week. These procedures adhered to the standards of China’s guidelines for the treatment of laboratory animals.

Thawing temperature and preservatives

The frozen large rabbit tissues were put into the LN-precooled mortar and gently smashed into small aliquot sizes by pestle under LN. The smashed tissue aliquots were weighed, with 10–30 mg aliquots randomly allocated to four groups: (A) neat control group without preservatives; (B) RL lysis buffer (Magen Biotechnology, China); (C) RNALater stabilization solution (Beyotime Biotechnology, China); and (D) TRIzol reagent (Thermo Fisher Scientific, USA). Since the Hipure Total RNA Mini Kit (Magen Biotechnology, China) protocol requires tissue lysis in RL buffer, this solution was also used as a preservative. Prior to tissue processing, 750 µL of each preservative was added to sterile 2 mL microcentrifuge tubes. Each treatment group was further divided into two subgroups and subjected to distinct thawing protocols: thawing on ice for 15 min or thawing at RT for 10 min (Fig. 2A). Tissue softening was confirmed by visual inspection and mechanical probing with sterile pipette tips. All the experimental conditions were performed in six biological replicates (n = 6).

Fig. 2.

Evaluation of RNA quality under different thawing temperatures and preservatives. (A) Schematic workflow of rabbit tissue pre-processing. (B) RIN comparison between tissues thawed on ice and those thawed at RT. (C) RINs of tissues thawed on ice across different preservatives. (D) RINs of tissues thawed at RT across different preservatives. (E) RNA yields of tissues thawed on ice across different preservatives. (F) RNA yields of tissues thawed at RT across different preservatives. RT, room temperature; ●, tissues thawed on ice; ○, tissues thawed at RT; normalized RNA yield, ng RNA/mg tissue; n.s., nonsignificant; *, p values < 0.05; **, p values < 0.01. Statistical significance was determined with unpaired t tests (B), Kruskal‒Wallis tests (C, E), and ordinary one‒way ANOVA (D, F)

Processing delay

Frozen rabbit kidney tissues were cryogenically smashed into 10–30 mg aliquots as described above. Prior to tissue processing, 300 µL of RNALater stabilization solution was added to sterile 2 mL microcentrifuge tubes and maintained on ice. To assess the impact of processing delay (time before disruption) on RNA quality, the smashed tissue aliquots were transferred into microcentrifuge tubes and randomly allocated to seven groups: 15 min, 30 min, 45 min, 90 min, 120 min, 1 day and 7 days. Short-term delays (15–120 min) and long-term delays (1 day and 7 days) were designed to simulate realistic laboratory or clinical processing delays. For long-term delays, the tissues were put into an ice box and placed in a refrigerator at 4 °C to maintain stability. All experimental conditions were performed in four biological replicates (n = 4).

Tissue aliquot sizes and number of freeze‒thaw cycles

Prior to tissue processing, 1.5 mL of RNALater stabilization solution was added to sterile 2 mL microcentrifuge tubes and maintained on ice. Twenty-four smashed tissue aliquots were selected and stratified into three mass-based groups: (A) 70–100 mg, (B) 100–150 mg, and (C) 250–300 mg (Fig. 4A). Moreover, one frozen tissue sample was cryogenically smashed into 10–30 mg aliquots and used as a control group. Each group was further divided into two thawing conditions: thawing on ice overnight or thawing at -20 °C overnight, followed by a 30 min incubation on ice. All experimental conditions were performed in four biological replicates (n = 4). After overnight, RNALater was carefully removed, and then 10–30 mg portions were aseptically excised from each sample via RNase-free scissors and tweezers for RNA extraction as freeze‒thaw cycle 0. The remaining tissue samples were flash-frozen at -80 °C overnight (Fig. 4A). The frozen tissues were fully thawed by ≥ 30 min of incubation on ice. Owing to variations in tissue aliquot size, the tissues were subjected to 3 freeze‒thaw cycles (70‒100 mg) or 5 cycles (100‒150 mg and 250‒300 mg). For each freeze‒thaw cycle, 10–30 mg aliquots were excised for RNA extraction.

Fig. 4.

Effects of aliquot size and thawing conditions on the RNA integrity of RNALater-treated tissues. (A) Schematic workflow of tissue preprocessing. (B) RINs of tissues thawed on ice across different aliquot sizes. (C) RINs across different aliquot sizes of tissues thawed at -20 °C. (D) Comparative RIN analysis between tissues thawed on ice and those thawed at -20 °C. ●, tissues thawed on ice; ○, tissues thawed at -20 °C; n.s., nonsignificant; *, p values < 0.05; **, p values < 0.01. Statistical significance was determined with ordinary one-way ANOVA (B, C) and unpaired t test (D)

Validation of the optimal preservation protocol for human kidney tissue

Five human kidney tissue samples cryopreserved in vapor-phase LN tanks for 2–5 years were acquired from the certified biobank of Zhongnan Hospital of Wuhan University, which was approved by the Medical Ethics Committee of Zhongnan Hospital of Wuhan University (No. 2017038). Tissues were collected immediately after routine surgical resection and confirmed by pathologists before cryopreservation. The frozen kidney tissues were cryogenically smashed via the protocol described and allocated into mass-stratified cohorts, which included one aliquot of 10–30 mg with immediate LN grinding as the control, three aliquots of 10–30 mg each treated with RNALater (thawing on ice for 45 min, thawing on ice overnight and thawing at -20 °C overnight), and two aliquots of 70–100 mg each treated with RNALater (thawing on ice overnight and thawing at -20 °C overnight). Since the human kidney tissues were archival tissues collected from different patients and stored in vapor-phase LN tanks for several years, a control group using 10–30 mg aliquots were grinded into powder by LN grinding, which is a recognized technique for RNA extraction.

Validation of the optimal preservation protocol for murine kidney tissue

The efficacy of RNALater in maintaining RNA integrity in cryopreserved murine kidney tissues was assessed. The murine kidney tissues utilized in this study were residual samples obtained in 2019 from our previous investigation [22]. The mice were euthanized through intravenous administration of an overdose of 3% sodium pentobarbital (150 mg/kg). Considering the naturally small size of murine kidneys, RNA integrity was compared in tissue aliquots weighing between 10 and 30 mg, which were cryogenically smashed as described. Similar to the validation experiment of human kidney tissue, these samples were subjected to cryogenic grinding in LN as a control or treated with RNALater under various thawing conditions: short-term (45 min) or long-term (overnight) thawing on ice and thawing at -20 °C overnight.

RNA extraction and quality assessment

The preservation solution (when present) was carefully discarded and replaced with 750 µL of RL lysis buffer (Magen Biotechnology, China). For the untreated control group, 750 µL of RL Buffer was added directly. Each sample received two nuclease-free homogenization beads (4 mm and 2 mm diameter). Tissue disruption was performed via a high-throughput TissueLyser (#Scientz-48; Ningbo SCIENTZ Biotechnology Ltd., China) under the following conditions: frequency at 50 Hz and continuous operation for 90 s. Subsequent RNA purification was performed according to the manufacturer’s protocol for the HiPure Total RNA Mini Kit. The extracted RNA was eluted in 100 µL of elution buffer, and the final volume was adjusted to 200 µL with additional nuclease-free water (Ambion, Thermo Fisher Scientific).

The RNA concentration and purity (A260/280 ratio) were determined via a NanoDrop One spectrophotometer (Thermo Fisher Scientific Ltd.). The RNA yields were normalized to the tissue mass (ng RNA/mg tissue). RIN was determined via an Agilent 4200 TapeStation system with RNA ScreenTape (Agilent Biotechnologies Ltd., USA).

Statistics

All the statistical analyses were performed via GraphPad Prism (version 8.0.1; GraphPad Software, San Diego, CA). The normality of distribution and homogeneity of variance for all values were first assessed. For the freeze‒thaw cycle experiments and validation experiments on human kidney tissue, paired tests were applied. Other experimental data were analyzed via unpaired t tests or one-way ANOVA with Tukey’s post hoc test. If values violated the assumptions of homogeneity of variance or normality, the Kruskal‒Wallis U/H test with Dunn’s multiple comparisons was used. The level of significance was set at p < 0.05.

Results

Effects of thawing temperature and preservatives

To determine the optimal thawing conditions, we evaluated three preservatives (RL buffer, RNALater, and TRIzol) and two temperatures (ice vs. RT). All samples exhibited high purity (A260/280 ≈ 2.10). There was no significant difference in the RINs between the neat group (without preservative) and the other groups (Fig. 2B). In contrast, the preservative-treated samples presented significantly higher RINs when thawed on ice than did the RT-treated samples (p < 0.01, Fig. 2B). Under thawing on ice, the RNALater group outperformed the neat group in terms of the RIN (p < 0.01), although no significant differences were observed among the three preservatives (Fig. 2C). RNALater-thawed samples had a greater proportion of high-quality RNA, with 100% of the samples achieving a RIN ≥ 8, significantly outperforming both the RL buffer (100% RIN ≥ 7, 16.7% RIN ≥ 8) and the TRIzol treatments (83.3% RIN ≥ 7, 16.7% RIN ≥ 8) (Table 1). The RL buffer group and the TRIzol group presented more variable quality profiles (Fig. 2C). Under thawing at RT, the RNALater group showed significantly superior RNA integrity compared to both the neat group and TRIzol group (Fig. 2D). As to RNA yield, regardless of the thawing temperatures, the TRIzol group yielded significantly less RNA than the other groups (Fig. 2E and F).

Table 1.

RINs under different thawing temperatures and preservatives

Thawing temperature	Preservatives
	Neat			RL			RNALater			TRIzol
	Mean ± SD	RIN ≥ 7 (%)	RIN ≥ 8 (%)	Mean ± SD	RIN ≥ 7 (%)	RIN ≥ 8 (%)	Mean ± SD	RIN ≥ 7 (%)	RIN ≥ 8 (%)	Mean ± SD	RIN ≥ 7 (%)	RIN ≥ 8 (%)
Ice	6.35 ± 0.71	33.33	0	7.76 ± 1.11	100	16.67	8.50 ± 1.06	100	100	7.47 ± 0.66	83.33	16.67
RT	5.47 ± 0.92	0	0	6.47 ± 0.73	33.33	0	7.57 ± 0.54	83.33	16.67	5.85 ± 0.87	16.67	0

The data represent the percentage of samples with RIN ≥ 7 and RIN ≥ 8, RT, room temperature

Effects of processing delay

Since RNALater demonstrated optimal effects in preserving the RNA integrity of frozen tissue samples, this reagent was selected for subsequent experiments. To evaluate the impact of delayed processing on RNA integrity, frozen tissue aliquots (10–30 mg) treated with RNALater were maintained on ice for durations ranging from 15 min to 7 days. RIN remained exceptionally stable for up to 1 day (Fig. 3A). Although a statistical difference in the RIN was observed between the 120-minute and 7-day groups (9.38 ± 0.10 vs. 8.45 ± 0.44, respectively), all processed delayed samples maintained sequencing-quality thresholds (RIN ≥ 8) [23, 24]. Furthermore, RNA yields showed no time-dependent variation (p = 0.14), demonstrating that RNALater effectively preserves RNA quantity during typical processing delays (Fig. 3B).

Fig. 3.

Evaluation of RNA quality under different processing delays (time before disruption) on ice. (A) RINs. (B) RNA yields. Normalized RNA yield, ng RNA/mg tissue; n.s., nonsignificant. Statistical significance was determined via ordinary one-way ANOVA

Effects of tissue aliquot size

Considering that tissue aliquots may exhibit irregular morphology during partitioning, we evaluated the impact of tissue aliquot sizes ranging from 70 to 300 mg on RNA quality under different thawing conditions. After overnight thawing on ice, no significant differences in the RIN were detected between group A (70–100 mg) and the control group (10–30 mg) (p = 0.45) or between group B (100–150 mg) and group C (250–300 mg) (p = 0.12, Fig. 4B). However, both group A and the control group presented significantly greater RINs than groups B and C did, suggesting that smaller tissue aliquots better preserved RNA integrity. When the tissues were thawed at -20 °C overnight, groups A, B and C presented significantly fewer RINs than did the control group, although groups A, B and C presented comparable RINs (Fig. 4C). Notably, group A (70–100 mg) maintained slightly greater RINs (RIN ≥ 7) than larger aliquots did, confirming the advantage of smaller sample aliquots in RNA preservation.

A direct comparison of thawing conditions revealed that, compared with those of samples thawed on ice, the RINs of group C samples were significantly greater when the samples were thawed at -20 °C (7.13 ± 0.69 vs. 5.25 ± 0.24, p < 0.01; Fig. 4D). In contrast, no significant differences were observed between thawing methods for groups A and B. These findings indicate that while smaller aliquots (≤ 100 mg) generally maintain superior RNA integrity regardless of thawing conditions, thawing at -20 °C may be beneficial for larger samples.

Effects of freeze‒thaw cycles

To assess the effects of freeze‒thaw cycling on RNA integrity, a longitudinal comparison was conducted on divided aliquots originating from the same tissue block. Samples of varying sizes were subjected to either three or five freeze‒thaw cycles under two thawing conditions (on ice or at -20 °C) (Fig. 4A). Due to the high variability in thawing on ice group B and the reduced RIN in thawing on ice group C, these two groups were excluded from the subsequent freeze‒thaw cycle experiments (Fig. 4). After 3 freeze‒thaw cycles, no significant differences in the RIN were detected among 70–100 mg of tissues thawed on ice (Fig. 5A). The groups of 70–100 mg and 100–150 mg of tissue thawed at -20 °C presented similar results (Fig. 5B and C). Compared with those in the 0 freeze‒thaw cycle, the RIN of tissues weighing 250–300 mg and thawed at -20 °C decreased after 5 freeze‒thaw cycles (Fig. 5D). Furthermore, samples thawed at -20 °C presented notably greater variability in the RIN, particularly in larger tissue aliquots.

Fig. 5.

Effects of freeze‒thaw cycles on the RNA integrity of RNALater-treated tissues. (A) RINs of 70–100 mg of tissue subjected to three freeze‒thaw cycles on ice. (B) RINs of 70–100 mg of tissue subjected to three freeze‒thaw cycles at -20 °C. (C) RINs of 100–150 mg of tissue subjected to five freeze‒thaw cycles at -20 °C. (D) RINs of 250–300 mg of tissue subjected to five freeze‒thaw cycles at -20 °C. 0–5, 0–5 freeze-thaw cycles, solid circles, tissues thawed on ice, hollow circles, and tissues thawed at -20 °C. For all the conditions, each freeze‒thaw cycle was compared with 0 freeze‒thaw cycles (n = 4). *, p values < 0.05. Statistical significance was determined with repeated measures one-way ANOVA

Validation of the optimal preservation protocol for human kidney tissue

To validate the protective efficacy of RNALater in cryopreserved human frozen kidney tissues, five archival specimens were selected for evaluation of the optimal thawing conditions. Given the inherent variability in precryopreservation factors during clinical sample collection, cryogenic pulverization of LNs was performed on an aliquot of each sample to establish matched controls. Compared with the LN-ground control (7.76 ± 0.54, p = 0.017), only the 70–100 mg of tissue thawed on ice overnight presented a significantly lower RIN (5.42 ± 0.78) (Fig. 6A). While no statistically significant differences in the RINs were detected among the other experimental groups, thawing the frozen human kidney tissues with the RNA preservative resulted in a marginally reduced RIN compared with that of the LN-ground control (Table 2). Notably, the preservation of tissues thawed at -20 °C was superior to that of those thawed on ice. With respect to RNA yield, all thawing conditions resulted in a significantly greater RNA yield than did the LN-ground control (Fig. 6B).

Fig. 6.

Quality of RNA isolated from frozen human kidney tissues. (A) RINs. (B) RNA yields. All RNALater-treated groups were compared to LN grinding control (n = 5). Normalized RNA yields, ng RNA / mg tissue, LN, liquid nitrogen grinding. *, p values < 0.05. Statistical significance was determined with repeated measures one-way ANOVA

Table 2.

RINs subjected to various treatments in human kidney tissue

Treatments	10–30 mg LN	10–30 mg Ice 45 min	10–30 mg Ice overnight	70–100 mg Ice overnight	10–30 mg -20 °C overnight	70–100 mg -20 °C overnight
Mean ± SD	7.76 ± 0.54	5.94 ± 0.10	5.94 ± 1.24	5.42 ± 0.78	7.24 ± 0.94	6.72 ± 0.70
RIN ≥ 7 (%)	100	20	20	0	40	60
RIN ≥ 8 (%)	60	0	0	0	40	0

Data show percentage of samples with RIN ≥ 7 and RIN ≥ 8, LN, liquid nitrogen

Validation of the optimal preservation protocol for murine kidney tissue

The RNALater-treated 10–30 mg murine kidney tissue samples consistently maintained high-quality RNA integrity (RIN ≥ 8) under all thawing conditions examined, including short-term (45 min) and long-term (overnight) thawing on ice, as well as overnight thawing at -20 °C. No statistically significant differences were observed compared with those of the LN grinding group, demonstrating that RNALater treatment effectively preserved RNA quality in small murine kidney tissues regardless of thawing duration or temperature conditions (Fig. 7A). Although all the groups maintained excellent RNA integrity, the LN grinding method resulted in significantly lower RNA quantities than thawing both on ice and at -20 °C overnight (Fig. 7B).

Fig. 7.

Quality of RNA isolated from frozen murine kidney tissues. (A) RINs. (B) RNA yields. All RNALater-treated groups were compared to LN grinding control (n = 4). Normalized RNA yields, ng RNA / mg tissue, LN, liquid nitrogen grinding. n.s., non-significant; *, p values < 0.05; **, p values < 0.01. Statistical significance was determined with Kruskal-Wallis test (A) and ordinary one-way ANOVA (B)

Discussion

Quick-frozen tissue preservation remains a cornerstone of biobanking, providing valuable resources for biomedical research. However, RNA is highly prone to degradation during tissue thawing [6, 11, 25], despite its essential role in downstream molecular analyses [26, 27]. RNA degradation can severely compromise the reliability of sequencing data and other downstream applications, promoting the need for optimized tissue handling protocols. This study was designed to establish an optimal thawing protocol for archival tissues originally frozen without preservatives to ensure both high RNA yield and integrity. For RNA extraction, we employed a high-throughput tissue homogenization method demonstrated in our previous study [22], achieving a RIN comparable to that of conventional LN grinding. This approach also provides greater practicality, enhanced safety and improved scalability for parallel processing of multiple samples.

First, we investigated the effects of common preservatives and thawing temperatures. While RNA preservatives such as RNALater and TRIzol are well established for prefreezing tissue stabilization, their efficacy in cryopreserved tissues remains less explored [14, 28–31]. Prior studies have shown that RNALater effectively maintains RNA integrity across diverse fresh tissues, including skeletal muscle, placenta, and brain [28–30]. Similarly, TRIzol treatment significantly improved the RIN in samples processed for red blood cell removal (RIN 9.2 vs. 4.5 in untreated samples, p = 0.002) [31], with comparable benefits observed in peripheral blood mononuclear cells (mean RIN 8.2 ± 0.37) [14]. Therefore, we hypothesized that cryopreserved tissues might similarly benefit from protective additives to prevent RNA degradation. Given the well-documented temperature dependence of RNase activity, we evaluated the effects of thawing at both RT [32, 33] and low temperatures (ice bath) [34, 35]. Our findings establish that the optimal and most stable protocol for RNA preservation involves the addition of RNALater to frozen tissues, followed by immediate processing on ice. This method consistently yielded high-quality RNA with a RIN ≥ 8. In accordance with established quality standards, RNA with a RIN ≥ 7 is generally considered suitable for routine molecular applications, including conventional RT‒qPCR and next-generation sequencing [36, 37]. For more sensitive techniques such as microarray analyses, higher RNA integrity thresholds (RIN ≥ 7.8) are typically needed [1, 2, 38].

Processing delay is inevitable in actual operations, particularly when managing large sample batches. To determine safe handling parameters, we conducted a stability assessment of RNALater-treated frozen rabbit tissues maintained on ice (0–4 °C). Our results revealed no detectable degradation in tissues stored for 1 day, and RNA integrity remained above the critical threshold for up to 7 days. The RNA yield showed no time-dependent variation, demonstrating that RNALater effectively preserves both RNA quantity and integrity during typical processing delay. These findings validate its reliability for clinical workflows and batch-processing scenarios where time before disruption intervals may vary.

Our initial experiments followed the recommended tissue aliquot size (≤ 30 mg) of the RNA Kit, but given that frozen rabbit tissues involve variable aliquot sizes, we evaluated three mass-based groups (70–100 mg, 100–150 mg and 250–300 mg) under different preservation conditions. For 70–100 mg aliquots, both ice (0–4 °C) and -20 °C overnight thawing achieved optimal RNA quality, demonstrating complete RNALater penetration [39], whereas larger aliquots (100–300 mg) stored on ice showed suboptimal preservation, with thawing at -20 °C maintaining superior integrity. These findings align with those of Samadani et al. [40], confirming that low-temperature processing and smaller tissue aliquots yield optimal RNA quality, where low-temperature conditions prevent nucleic acid degradation and reduce sample size limits RNase quantity.

While freeze‒thaw cycles are known to compromise RNA quality [25, 33], tissues treated with RNALater before freezing maintain excellent stability (RIN >8) through multiple freeze‒thaw cycles [41]. We conducted a freeze‒thaw cycle assessment on the remaining tissues from the tissue-size experiment. Although no statistically significant differences were detected among the tested groups across 3‒5 freeze‒thaw cycles, the three groups that were thawed at -20 °C presented marked fluctuations in the RIN. We observed a partial freezing state (ice-slush formation) in the RNALater-treated groups during thawing at -20 °C, which likely created physical barriers that impeded full preservative penetration into the tissue interstitium. While surface tissue yielded excellent results in terms of RNA integrity, the repeated sampling protocol during freeze‒thaw cycling could not differentiate between external and internal tissue sections, potentially introducing significant variability. Owing to the limited availability of 250–300 mg of tissue, we could not assess whether prolonged incubation at -20 °C would permit complete RNALater penetration. When 70–100 mg of frozen tissue fully penetrated after being incubated overnight on ice, RNA integrity remained high even after three freeze‒thaw cycles.

Overall, for rabbit kidney tissues ≤ 100 mg, RNALater treatment combined with thawing on ice overnight provides optimal RNA preservation, maintaining high integrity even through multiple freeze‒thaw cycles. For larger rabbit tissues > 100 mg, we recommend either employing LN smashing into small aliquots or thawing at -20 °C followed by processing superficial tissue sections (where preservative penetration is complete). These protocols ensure high RNA quality, sample reuse, and high-throughput processing of frozen rabbit tissues.

On the basis of the experimental results from rabbit kidney tissues, several optimal thawing conditions applicable to human archived frozen kidney tissues were investigated. To prevent the formation of an ice slush subsequent to prolonged storage at -20 °C, we attempted to double the quantity of RNALater (RNALater without tissue does not solidify at -20 °C), which was effective. For human frozen tissues treated with RNALater, except for the 70–100 mg samples that were thawed on ice, which presented significantly lower RINs than did the LN control group, the remaining groups presented no statistically significant differences. However, when evaluated on the basis of the RIN criterion (RIN ≥ 7), only a portion of the human tissues achieved RNA quality comparable to that of the LN control group, whereas all rabbit tissues maintained high-quality RNA under identical processing conditions. We speculate that the following reasons may have contributed to this phenomenon: (1) interspecies variation, (2) the limited efficacy of RNAlater in cryopreserved human tissue, and (3) discrepancies in RNase activity or abundance among cells in different regions of the same kidney [42–44].

We subsequently validated the protective effect of RNAlater in frozen murine kidney tissues. The results were consistent with those for rabbit kidney tissues, which yielded high-quality RNA in all thawing treatments (RIN ≥ 8), confirming our first hypothesis regarding interspecies variation, specifically the closer similarity between rabbits and mice in the Glires. Similarly, RNAlater-ICE (Invitrogen™, Carlsbad, CA, USA) is a unique RNA stabilizing solution specifically designed for unprotected frozen tissues. The verification experiments detailed in its procedural manual and the few studies conducted were exclusively on animal samples, and documentation regarding its efficacy on human tissues is notably scarce [18, 45]. Although rabbits and mice play important roles in medical research and this study also aims to apply findings from animal kidney tissues to human kidney tissues, the significant interspecies differences in kidney tissue increase the variability of the protective effects of RNALater [42]. Nevertheless, how to improve the RNA preservation effect in cryopreserved human tissues still warrants further investigation. Potential strategies may include the use of mixtures of RNA-stabilizing reagents or the supplementation of potent RNase inhibitors to achieve increased protective efficiency. In addition, methods for extracting high-quality RNA from frozen tissues of other human organs are worthy of study.

Conclusions

This study provides practical protocols for preserving RNA in archival frozen tissues, addressing critical challenges in biobanking and molecular research. We demonstrated that treatment with RNALater followed by immediate thawing on ice or at -20 °C effectively protected the RNA from all tissues ≤ 30 mg. For larger tissues, the efficacy of RNAlater in cryopreserved tissues varies among different species. It is suggested that fresh tissues be divided into small aliquots prior to cryopreservation, which consistently yields superior RNA stability. Our results provide flexible experimental workflows for archival tissue samples.

Abbreviations

RT

Room temperature

RIN

RNA integrity number

LN

Liquid nitrogen

Author contributions

K.Q. and S.Z. assisted in designing the experiments, interpreting the data, and supervised the entire study. C.Z., S.Z., M.Y., H.P., X.Z., Z.Z. and W.L. conducted the experiments and wrote the initial draft. ZD was involved in the animal experiments. C.Z., M.Y., K.Q. and S.Z. analyzed the data and revised the manuscript. All the authors reviewed and approved the final manuscript.

Funding

This work was supported by grants from the Research Fund of Zhongnan Hospital of Wuhan University (CXPY2020031 and ZLYNXM20200213), and Hubei Provincial Science and Technology Plan Project (2025CFC012).

Data availability

All the data used or analyzed during the current study are available from the corresponding author upon reasonable request.

Declarations

Ethics approval and consent to participate

All experiments involving the use of human kidney tissues were performed in compliance with the Declaration of Helsinki. All experimental procedures, including animal and human procedures, were approved by the Experimental Animal Welfare and Ethics Committee of Zhongnan Hospital, Wuhan University (Approval No. 2025109) and the Medical Ethics Committee of Zhongnan Hospital of Wuhan University (Approval No. 2017038). Informed consent was obtained from all participants in this study.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.