Abstract
Introduction
A
Materials and Methods
Patient samples
FFPE renal biopsy tissues were obtained from the Renal Biobank of National Clinical Research Center of Kidney Diseases and approved by the local committee of human subjects at Jinling Hospital. The size of each biospecimen is as follows: diameter ≤1.2 mm and length ∼10 mm. All biospecimens were fixed with 10% buffered formalin for 4–6 hours. All paraffin blocks were stored at room temperature in a dark room. Twenty-four blocks from patients with IgA nephropathy were studied: 12 paraffin blocks that were 5 years old and 12 that were 10 years old. Then, serial sections of 5 μm were cut from the blocks (initial sections were discarded). For comparison between the kits, the 5-year-old blocks were used, with RNA being extracted from one section per block for each kit. For comparison between the two storage times, RNA was extracted from two sections per block.
RNA extraction, quantification, and reverse transcription of RNA
Two kits were used to extract RNA from 5-year sections: RecoverAll Total Nucleic Acid Isolation kit for FFPE (AM1975; Ambion) and RNeasy FFPE kit (73504; Qiagen). The Ambion kit was used for comparison of RNA extracted from 10- and 5-year tissue blocks. We extracted RNA manually using the two kits. Both kits incorporated a DNase digest step, and we eluted the RNA in the same volume of 60 μL. Quantification of RNA was performed using Nanodrop 2000. RNA integrity, determined by RNA integrity number (RIN), was assessed with Agilent Bioanalyzer 2100 using the RNA 6000 Nano and Pico kit. Reverse transcription of RNA was performed using High Capacity cDNA Reverse Transcription kit (Life Technologies). Random primers were used for the reverse transcription (RT) reaction. There was 25 ng (for comparison of the two kits) or 50 ng (for comparison of the two storage times) RNA in 20 μL volume of the reverse transcription reaction.
Gene expression by quantitative PCR
Real time quantitative PCR (RT-qPCR) using the TaqMan system (Life Technologies) was carried out using oligonucleotide primers, then probes for the following genes: SYNPO (Hs00355049_m1, amplification length 67 bp), CTNNB1 (Hs00702468_s1, amplification length 70 bp), and the 18S ribosomal RNA gene. For the 18S rRNA gene, two amplicons of different lengths were used: an 18S-1 amplicon of 86 bp (5′-GTTCAGCCACCCGAGATTGA-3′, 5′-CTGAGCCAG TCAGTGTAGCG-3′) and an 18S-2 amplicon of 136 bp (5′-CGCAAATTACCCACTCCCGA-3′, 5′-ACCAGACTTGC CCTCCAATG-3). The 20 μL volume of qPCR containing 2 μL cDNA was run in triplicate by the ABI HT 7900 and analyzed by SDS 2.4 software (Life Technologies).
Statistical analysis
Experimental data are expressed as the mean or median values and standard deviation (SD). The performance of the two RNA extraction kits and the RNA from tissue blocks of 10 or 5 years was compared with the Student's t-test or the Mann–Whitney Rank Sum Test using SigmaPlot software (v 12.5), with p < 0.05 denoting statistical significance.
Results
Comparison of the two RNA extraction kits
RNA was isolated from FFPE renal biopsy biospecimens using Ambion and Qiagen kits with the same quantity of tissues. As shown in Figure 1A and B, the extracted RNA fragments were mostly less than 200 bp. There were no significant differences in mean RIN values (2.3 [SD 0.2] vs. 2.4 [SD 0.4]) or purity (A260/A280, 1.54 [SD 0.14] vs. 1.50 [SD 0.18]) between the two kits. As shown in Figure 1C, median RNA concentrations from Ambion kit extractions were higher, but not statistically significant, than those from Qiagen kit extractions (4.35 [SD 1.9] ng/μL vs. 2.65 [SD 0.72] ng/μL, equating to 261 vs. 159 ng/5 μm section, p = 0.09).
Comparison of the extracted RNA from two commercial kits and extracted RNA from FFPE renal biopsy biospecimens at 10- and 5-year storage time points.
Mean Ct values of 18S-1 and 18S-2 were 21.66 (SD 1.12) and 22.02 (SD 0.49) for the Ambion kit and 22.71 (SD ±1.03) and 24.42 (SD 0.79) for the Qiagen kit, so the longer 18S-2 fragment returned higher Ct values than the shorter 18S-1 fragment. When the two kits were compared, for both 18S-1 and 18S-2, the higher Ct values (indicating more degraded RNA) with the Qiagen kit compared with the Ambion kit were statistically significant (p < 0.03 [18S-1] and p < 0.001 [18S-2]). For each sample, ΔCt was calculated (18S-2 Ct−18S-1 Ct), which could be used as a measure of RNA degradation (a less degraded sample will have a lower ΔCt). Mean ΔCt values were 0.36 (SD 1.4) for the Ambion kit and 1.71 (SD 1.5) for the Qiagen kit (p = 0.03; Fig. 1D), indicating that RNA quality was better when using the Ambion extraction method than when using the Qiagen extraction method.
Comparison of extracted RNA from 5- and 10-year FFPE renal tissues
There were no significant differences in the quality of RNA extracted from 10- and 5-year renal biopsy tissues, whether the measurement was by RIN (mean 2.4 [SD 0.2] vs. 2.5 [SD 0.4], Fig. 1E), 18S-1 RT-qPCR Ct value (mean 17.3 [SD 1.1] vs. 17.4 [SD 0.7]), or RNA purity (mean 260/280 1.33 [SD 0.1] vs. 1.33 [SD 0.04]). However, the concentrations of RNA extracted from 10-year-old tissues were lower than those from 5-year-old tissues (median 10.9 ng/μL [SD 0.3] vs. 12.7 ng/μL [SD 2.2], p = 0.01, equating to 327 vs. 381 ng/5 μm section, p = 0.01, Fig. 1F). Mean Ct values of amplified SYNPO and CTNNB1 genes were 35.97 (SD 0.3) and 35.10 (SD 0.4) for 10-year-old blocks and 35.73 (SD 2.0) and 35.87 (SD 1.8) for 5-year-old blocks, the difference between 5- and 10-year storage not being statistically significant.
Discussion
Even though there are reports comparing the performance of commercial kits for RNA isolation from FFPE tumor biospecimens of larger size,5,6 comparisons of different kits for small needle biopsy samples are scarce. For the FFPE renal biopsy biospecimens, our data showed that there were no differences in RNA purity or RIN scores by the two commercial RNA extraction kits from Ambion and Qiagen. Although the lower yield found with the Qiagen kit compared with that of the Ambion kit was not statistically significant (p = 0.09), it may prove to be so with a larger sample size. However, the mRNA from the Qiagen kit was more degraded than that from the Ambion kit (ΔCt). In addition, 18S-1 Ct and 18S-2 Ct values were improved (i.e., lower) in the Ambion kit. The Qiagen and Ambion protocols have similar proteinase k digest steps, comparable heating steps (15 min at 80°C), and both incorporate a DNase digest. It is possible that the deparaffinization steps (which are different in the two protocols) may drive some of the differences. The lower Ct values found using the Ambion kit may indicate that the Ambion kit enriches the total RNA extraction for mRNA than the Qiagen kit. The ΔCt between the two amplicon lengths in the 18S gene is improved in the Ambion extractions, which demonstrates that the integrity of the mRNA is better than that resulting from the Qiagen kit. Therefore, although either kit will return RNA from FFPE tissues, for biopsies we prefer the Ambion kit.
The RNA fragments were mostly within 200 bp isolated from FFPE renal biopsy tissues. In fact, the low RINs we have seen here are typical of those reported from FFPE tissues in other samples.7–9 In concordance with the study of Ribeiro-Silva, 10 we designed two different primers (amplification length 86 and 132 bp) of 18S and compared the ratio between them to evaluate the degree of RNA degradation. In addition, we could amplify glomerular- and tubular-specific genes (SYNPO and CTNNB1) using primers that generate amplification lengths shorter than 100 bp, as described in other studies.10,11 Although our samples returned high Ct values for SYPNO and CTNNB1, we believe that this is a consequence of the small size of the biospecimens (needle biopsies) and low quantities of RNA that were applied to the reverse transcription reaction (50 ng per sample).
Our study also indicated that FFPE biopsies stored for 10 or 5 years yielded RNA that was of similar quality (the fixative, fixation times, tissue processing protocols, and storage conditions of the blocks were the same). However, RNA yields were lower in 10-year-old blocks than in the 5-year-old blocks. This difference suggests that prolonged storage of the embedded samples can result in increased RNA fragmentation,11,12 even after the biospecimens are dehydrated and embedded in paraffin. Another factor that could be salient is that in the 5-year versus 10-year comparison, the 5-year sections were cut from deeper in the block than the 10-year sections (so less exposed to oxidation) because the 5-year blocks had already and exclusively been used for comparison between the two kits. However, it is important to note that the integrity of the RNA was not lower in the blocks stored for longer storage time.
It is logical that samples that are fixed in formalin for longer periods of time will contain more formaldehyde cross-links and, therefore, be more resilient to time-induced RNA degradation. Although the fixation time on our biospecimens was relatively short (4–6 hours), we again make the point that these are small needle biopsies and, consequently, the fixation time is in our view adequate. However, we would not recommend such short fixation times for larger tissue blocks.
There are few reports about FFPE renal biopsy biospecimens used for gene expression analysis. 2 So, this study reassures us that FFPE renal biopsy tissue resources can be used for gene expression studies in renal diseases even after long-term storage.
Footnotes
Acknowledgments
Our research was supported by the National Natural Science Foundation of China (81670699) and the Innovation Capability Development Project of Jiangsu Province (BM2015004).
Author Disclosure Statement
No conflicting financial interests exist.