Impact of soft tissue homogenization methods on RNA quality

Abstract

Background

High-quality RNA isolation from tissues is crucial for transcriptomic analysis. The tissue disruption influences the RNA quality. The aim of this study was to compare different methods of tissue homogenization.

Methods

Three homogenization methods were used (mortar and pestle, ball mill, and tissue homogenizer) to disrupt head and neck cancerous tissues, healthy tissues from free margin of head and neck cancer patients, and breast skin from breast cancer patients. The comparison of isolated RNA quantity and quality by measuring the concentration and absorbance, RIN, and Ct values of reference genes were performed.

Results

RNA isolated after tissue homogenizer usage has the highest 260/230 ratio (p = 0.02) and concentration (p = 0.02) also RIN values tend to be highest across all studied tissues. There are no significant differences between Ct values across all tissues processed with different homogenization methods; however, the Ct values of GAPDH and S18 are negatively correlated with RIN number (p = 0.002, p = 0.003).

Conclusion

The tissue homogenizer is the most suitable for obtaining high-quality RNA across all examined tissues, which is essential in various RNA-based analyses. GAPDH and S18 Ct can indicate the RNA quality measured by RIN values.

Supplementary Information

The online version contains supplementary material available at 10.1007/s11033-025-10508-0.

Background

Extracting high-quality genetic material from human tissue is crucial for various molecular applications, including standard laboratory procedures and high-throughput analysis [1, 2]. Proper methods of tissue homogenization must allow for the recovery of as much genetic material as possible with simultaneous preservation. The clinical samples are limited in size, are one-time collected; thus, selecting a established homogenization method is essential. Tissues delivered from different solid tumors, including head and neck cancers (HNC), are challenging in molecular processing caused by histologic and genetic heterogeneity. Also, skin needs a specific approach due to the presence of hyaluronic acid, collagen fibers, and surface nucleases [3]. There are several methods of tissue homogenization, including mechanical disruption or enzymatic digestion [4, 5].

Previous works emphasize challenges of patients’ heterogeneous tissue homogenization. Ivanov et al. used mortar and pestle to ground very hard tissue such as cartilage, and isolated RNA presents satisfaction RIN values [6].

Usage of tissue homogenizator is common and various protocols are publicized [7]. Another method that could be used for tissue homogenization includes cryotome rotor/stator-based homogenization. The cryotome methods are more suitable for grinding the elastic tissues with high collagen content in the skin. Moreover the vortexer bead beating method is similarly effective to cryotome but with a disadvantage in sample heating [8]. However, the cryosectioning snap-frozen tisuue seems to be a good alternative for the extraction of high-integrity RNA from the skin (RIN > 8), and it highlights the role of cold-inactivating ribonucleases, while mechanically breaking the skin [9]. There is no data capturing comparison of homogenization methods on patients’ tissue in the context of the quality and quantity of isolated material.

Materials and methods

Biological material

Primary oral cavity tumor and pair-matched tissues from free margin were collected from four HNC patients while breast skin samples from five patients with breast cancer (BC) who underwent tumor surgical resection in The Greater Poland Cancer Centre. The procedures were approved by the Local Ethical Committee of Poznan University of Medical Sciences (no. 452/20 and 283/21). Samples were immediately snap-frozen after surgical resection in liquid nitrogen and stored at -4ºC with RNA-later. Tissues from the same patient were cut in three and homogenized using mortar and pestle, ball mill, and Ultra Turrax tissue homogenizer. Skin fragments were treated with collagenase IV (%) for 1 h at 37 ºC to allow the hyaluronic acid-collagen matrix degradation.

Mortar and pestle

Tissues were transferred to liquid nitrogen-cooled mortar and ground with a pestle to obtain fine tissue powder. Then, it was transferred to 600 µl RLT lysis buffer, supplemented with 2-mercaptoethanol (BME). The samples were frozen at -20ºC for an hour and then stored at -80ºC until RNA extraction.

Ball mill

Tissue samples were ball-milled using a Retsch MM400 mixer mill. Samples were milled in cryogenic conditions using Retsch CryoKit for grinding with liquid nitrogen. One grinding cycle lasted one minute at an oscillation frequency of 30 Hz. In case of incomplete grinding, the milling was repeated.

Tissue homogenizer

Tissues were transferred to the cytometric tube and filled with an RLT lysis buffer with BME. Subsequently, tissues were mechanically disrupted with T10 basic ULTRA-TURRAX Homogenizer at the highest speed until complete homogenization (app. three minutes). The lysate was transferred to a tube and centrifuge at 16,000 rcf for three minutes. The supernatant was used for RNA extraction.

RNA isolation, quality, and quantity assessment

Total RNA was isolated with RNeasy Mini Kit due to standard protocol guidelines. The concentration and purity of RNA were determined by absorbance measuring in wavelengths: 230, 260, and 280 nm. The quality of isolated RNA was determined by RIN values (RNA Integration Number) and electrophoresis data obtained from the 2100 Bioanalyzer System using Agilent RNA 6000 Nano Kit and standard protocol.

Reverse transcription and real-time quantitative polymerase chain reaction (RT-qPCR)

RNA samples were reverse-transcribed into cDNA with RevertAid First Strand cDNA Synthesis Kit using 100 ng of total RNA. RT-qPCR was carried out in CFX96 Real-Time System using the PowerTrack SYBR Green Master Mix and primers specific for GAPDH, SDHA, and S18 reference genes [Supplementary Table 1].

Statistical analysis

The statistical analysis was performed with GraphPad Prism Version 10.0.2 (2023). The normal distribution of data was assessed with the Shapiro-Wilk test. The differences between more than two groups were estimated by one-way ANOVA Friedman or Kruskal-Wallis test. The comparison between the two groups was assessed with a t-test. A p-value less than 0.05 was considered statistically significant.

Results & discussion

Influence of tissue disruption method on RNA integration, purity, and concentration

We compared RNA quality and quantity by spectrophotometric measurements and RIN values determination after tissue homogenization [Supplementary Table 2]. All methods gave satisfactory results of RNA purity (Fig. 1B) however usage of tissue homogenizer gives proper values of 260/230 ratio (p = 0.02) (Fig. 2A), while mortar and pestle provide the highest 260/280 ratio (p = 0.01). Moreover, we demonstrate that median RIN values tend to be highest in RNA samples after tissue disruption with homogenizer (Fig. 1C) and the highest RNA concentration across all tissues (p = 0.02) (Fig. 1D). There are no differences in mortar and pestle versus ball mill usage in RNA concentrations, spectrophotometrically measured purity, and RIN values.

Fig.1.

The comparison of RNA purity A- 260/230 ratio, B– 260/280 ratio, C - RNA integrity D– concentration isolated from HNC, normal and skin tissues after homogenization with three different methods. The p-value was calculated using the one-way ANOVA for multiple comparisons. The p < 0.05 was considered statistically significant. All the tissues had the same weight before cutting in three equal parts to undergo homogenization

Fig. 2.

RNA quality visualization after electrophoresis using 2100 Bioanalyzer and RNA 6000 Nano chips after extraction RNA from HNC tissues (a), normal (b), and breast skin (c). The graph depicts RIN values and RNA concentration. M– M-mortar, B-ball mill, and T - T-tissue homogenizer

Subsequently, we assessed the same parameters for all tissue types separately. The RNA concentrations in HNC tissues significantly differ between homogenization methods (p = 0.049) (Table 1). There are no significant differences between 260/280, 260/230 ration and RIN values in all homogenization methods across examined tissues (Table 1).

Table 1.

Comparison of extracted RNA concentrations and quality in different tissue types after homogenization using three methods

Tissue	Method	RNA concentration [ng/µl]	p-value	RNA purity 260/280	p-value	RNA purity 260/230	p-value	RIN	p-value
HNC tumor	mortar and pestle	139.92 ± 91.45	0.049	2.19 ± 0.12	0.085	0.74 ± 0.45	0.099	6 ± 2.86	0.091
	ball mill	309.68 ± 178.33		2.05 ± 0.02		1.25 ± 0.71		4.38 ± 2.06
	tissue homogenizer	685.73 ± 420.83		2.11 ± 0.04		1.63 ± 0.29		5.37 ± 0.65
HNC normal	mortar and pestle	120.80 ± 201,87	0.459	2.13 ± 0.09	0.085	0.34 ± 0.42	0.073	6.83 ± 3.12	0.278
	ball mill	58.41 ± 40.32		2.06 ± 0.04		1.17 ± 0.58		3.88 ± 2.03
	tissue homogenizer	100.85 ± 47.83		2.06 ± 0.06		1.15 ± 0.92		5.23 ± 1.53
Breast skin	mortar and pestle	15.97 ± 9.67	0.311	2.65 ± 0.83	0.402	0.14 ± 0.09	0.207	2.6 ± 2.1	0.146*
	ball mill	10.64 ± 11.74		2.21 ± 0.27		0.10 ± 0.07		1.8 ± 1.9
	tissue homogenizer	108.58 ± 135.99		2.23 ± 0.28		0.68 ± 0.70		-

The p-values were calculated using the one-way ANOVA, with Tukey’s test for multiple comparisons or the t-test*. The p < 0.05 was considered statistically significant. Results are presented as mean ± SD

From the practical view, mortar and pestle usage is low-cost, and grinding lasts app. 5 to 10 min. It could be sterilized and reused. Ball milling takes about 2 to 5 min, depending on tissue hardness, but it requires special equipment and an experienced person. Liquid nitrogen usage is similar in both methods while no needed for tissue homogenizer. The tissue dissociation lasts app. 3 min, depending on tissue type. The tips used for dispensing the tissue are sterile and not reusable.

Tissue homogenizer is the most suitable method for RNA extraction from human tissues

The 2100 Bioanalyzer and RNA 6000 Nano chips were used to estimate the quality of isolated RNA. The visible bands corresponding to the 28 S rRNA and 18 S rRNA. The RIN value informs about RNA integrity, and the higher the RIN value is, the better the RNA quality [6, 10, 11]. Our results indicate all three methods of tissue homogenization are suitable for RNA isolation from HNC and normal tissues. The low RIN values in breast skin samples could result from the presence of surface RNases and difficulties in cell disruption due to the presence of collagen and elastin fibers (Fig. 1). The results from electropherograms were shown only for HNC and pair-matched tissues from free margin (normal tissues) (Fig. 3) due to the poor quality of breast skin samples.

Fig. 3.

Electropherograms for different homogenization methods and tissues: HNC (T) and HNC-healthy tissue (H), obtained using 2100 Bioanalyser and RNA 6000 Nanochips

The tissue homogenization method does not influence the Ct values of reference genes

We compared the mean Ct values obtained from three reference genes (GAPDH, S18, and SDHA, respectively) to assess the influence of the tissue disruption method on gene expression analysis by RT-qPCR. Our studies revealed that Ct do not statistically differ between compared tissue homogenization methods (Fig. 4).

Fig. 4.

Comparison of mean Ct values of GAPDH, S18, and SDHA and tissue homogenization methods. The p-values were obtained with the one-way ANOVA, with Tukey’s correction for multiple comparisons, and p < 0.05 is considered statistically significant. The mean Ct values are shown with standard deviation. Pearson’s correlation of RIN and the Ct for GAPHD, S18, and SDHA genes.

In two of three housekeeping genes (GAPDH and S18), we observe a strong negative correlation between RIN and Ct values (r=-0.6, p = 0.0031; r=-0.69, p = 0.0005, respectively). RNA samples with a higher degradation rate are responsible for the higher Ct values (3). Surprisingly, the SHDA gene has a strong positive correlation between RIN and Ct values (r = 0.63, p = 0.0022) (Fig. 4). However, the more degraded RNA is, the higher the Ct values in RT-qPCR [3, 8].

In the majority of cases, when RNA is intact, reference genes are fully transcribed, leading to a higher RNA input in qPCR, which results in lower Ct values. Highet et al. study demonstrated a strong negative correlation between Ct and RIN values for six tested reference genes in fresh-frozen post-mortem human brain tissue [12]. Similar results were obtained for the reference genes S18, S28, and β-actin, where the authors showed a negative correlation between RIN and Ct values [13]. In RNA samples derived from the heart, the expression of beta-2-microglobulin (β2M), glyceraldehyde-3-phosphate dehydrogenase (GAPDH), transferrin receptor (TfR), and β-actin was also negatively correlated with RIN values, clearly showing that RNA integrity affects gene expression levels [11]. Our results support this hypothesis for the two genes we examined, GAPDH and S18. However, Miyahara et al. demonstrated that the expression of the reference gene ACTB was negatively correlated with RIN values (r = − 0.90). They also showed that reference gene expression could be affected not only by RIN values but also by tissue pH [14]. Considering these data, the positive correlation between Ct values and RIN for the SDHA gene is surprising, making this gene unsuitable for predicting RNA integrity. A possible explanation for the negative correlation between RIN values and Ct for certain reference genes could be the increased formation of secondary RNA structures, which may influence reverse transcription efficiency. Additionally, gene length may play a role, as shorter genes may be less affected by RNA degradation, resulting in more stable Ct values. Overall, the negative correlation between RIN values and Ct is rare and not yet fully understood.

Conclusion

These brief study revealed that tissue homogenizer is the most suitable approach for tissue disruption to obtain good-quality RNA samples with high 230/260, RIN, and concentration. We demonstrate that the method of tissue homogenization does not affect the Ct values of selected reference genes (GAPHD, S18, and SDHA) in all examined tissues. Furthermore, we determined negative correlation between Ct of GAPDH and S18 and RIN values that may indicate RNA integrity. The novel ball milling approach of tumor tissue is an alternative method for quick and effective tissue disruption without additional reagents. The limitations of our study include the small sample size and only one considered method of RNA isolation however, the amount of clinical material after surgical resection is restricted and limits in taking account all possible variables.

Electronic supplementary material

Below is the link to the electronic supplementary material.

Author contributions

J.O., B.M., K.O., B.K., N.P., D.F. conceived and designed the analysis and collected the data. J.O., B.M., prepared the figures. K.O., K.K., performed the analysis. M.F., W.G., W.S. supervised the project.

Funding

The authors declare no financial interests.

Data availability

No datasets were generated or analysed during the current study.

Declarations

Ethical approval

The procedures used in this study were approved by the Bioethics Committee of Poznan University of Medical Sciences for head and neck tissues (protocol code 452/20) and skin tissues (protocol code 283/21).

Consent to participate

Informed consent was obtained from all individual participants included in the study.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.