2Oasis Diagnostics Corporation, Vancouver, WA
3Department of Biostatistics and Medicine, UCLA School of Public Health, Los Angeles, CA
The stabilization and processing of salivary transcriptome and proteome biomarkers is a critical challenge due to the ubiquitous nature of nucleases and proteases as well as the inherent instability of these biomarkers. Furthermore, extension of salivary transcriptome and proteome analysis to point-of-care and remote sites requires the availability of self-administered ambient temperature collection and storage tools. To address these challenges, a self-contained whole saliva collection and extraction system, RNAPro•SAL, has been developed that provides rapid ambient temperature collection along with concurrent processing and stabilization of extracellular RNA (exRNA) and proteins. The system was compared to the University of California, Los Angeles (UCLA) standard clinical collection process (standard operating procedure, SOP). Both systems measured total RNA and protein, and exRNA IL-8, glyceraldehyde 3-phosphate dehydrogenase (GAPDH), β-actin and ribosomal protein S9 (RPS9) by qPCR. Proteome analysis was measured by EIA analysis of interleukin-8 (IL-8), and β-actin, as well as total protein. Over 97% of viable cells were removed by both methods. The system compared favorably to the labor-intensive clinical SOP, which requires low-temperature collection and isolation, yielding samples with similar protein and exRNA recovery and stability.
Each year, more than 1 million cancer cases are diagnosed in the United States, and lung cancer is the leading cause of cancer death worldwide. Early cancer detection is critical to successful treatment and survival. Saliva is a noninvasive, accessible medium containing an array of analytes (RNA, DNA, and proteins) for diagnostic and clinical applications, and saliva transcriptome and proteome analysis is an emerging diagnostic technology with discriminatory power for cancer and disease detection. Salivary biomarkers for early stage cancer detection, including both extracellular RNAs (exRNAs) and proteins, have been discovered for oral, breast, ovarian, pancreatic, and lung diseases (1-6). Preservation and stabilization of these diagnostic biomarkers is a critical challenge. In unprocessed whole saliva (WS), exRNA degrades rapidly at ambient temperature. With a half-life of 12 min for β-actin mRNA in WS, 94% of the initial β-actin mRNA content degrades within 1 h at ambient temperature (7,8). The limited stability of exRNAs and proteins and the ubiquitous nature of nucleases and proteases in WS have limited WS transcriptome and proteome analysis to settings requiring low-temperature collection and processing equipment, highly trained personnel, low-temperature storage facilities, and lengthy mRNA and protein isolation procedures.
A novel saliva collection and isolation technology is presented for salivary proteome and transcriptome analysis. RNA and biomarker stabilization (>14 days) is comparable to UCLA's standard clinical collection process (SOP), and the simple, low cost absorption and filtering system provides similar yields of extracellular RNA and proteins at ambient temperatures.
Saliva supernatant (SS) containing freely soluble nucleic acids and proteins as well as exosome compar tmentalized RNA and proteins, along with a reduced viable cell content, is expected to increase short-term sample stability by minimizing active ongoing secretion of nucleases and proteases. As an example, SS has been demonstrated to be a suitable vehicle for direct saliva transcriptome analysis (DSTA), eliminating the need for mRNA isolation for reverse transcription quantitative polymerase chain reaction (RT-qPCR) (9). In addition, SS prepared by the standard operating procedure (SOP) developed at UCLA, which requires (i) collection of WS on ice, (ii) on site centrifugation at 4°C to obtain SS, and (iii) the addition of an RNase inhibitor or protein inhibitor cocktail to the SS, exhibits improved exRNA and protein stability compared to unprocessed WS (see Supplementary Figure S1). Although this procedure is a significant improvement over traditional collection and isolation processes, the SOP is labor intensive, requiring careful removal of SS by pipetting without disturbing the sediment mucin and cellular debris layers, and also requires low temperature collection, processing, and -80°C long-term storage.
For point-of-care and remote site collection, a self-administered ambient temperature–compatible collection and processing system that ensures biomarker stability and does not require specialized personnel and equipment is critical. RNAPro•SAL addresses this unmet need. The system provides salivary exRNA and protein collection and recovery, cellular removal, and biomarker stability at ambient temperature, eliminating the requirement for specialized personnel and equipment. The self-contained system produces ≥97% cell-free supernatant by means of selective absorption, release, and filtering.
To demonstrate the utility of the RNAPro•SAL system, the relative performance of RNAPro•SAL and SOP specimens with respect to ambient temperature exRNA and protein analyte recovery and stability was evaluated. The UCLA SOP was selected as a quantitative reference over WS because of the rapid analyte degradation observed in WS, which renders quantitative scaling difficult. Materials and methods Saliva collection
Saliva samples were collected between 8 and 10 AM from 6 individuals between the ages 25 to 56 years with informed consent as prescribed by institutional review board policies.
Saliva samples were processed using a parallel experimental design (Figure 1B). The design uses a 10 mL passive drool WS common stock collected on ice from each participant for the modified UCLA SOP stored at room temperature and RNAPro•SAL-derived saliva processing. The 10 mL sample is mixed with a glass pipette to ensure homogeneity and split into 2 5 mL aliquots, 1 for SOP and 1 for RNAPro•SAL processing. The corresponding SSs obtained are then divided into individual subsample aliquots for storage and analysis to avoid the hazards associated with repetitive subsampling from a single stock at each time point.
In this study, a WS aliquot obtained by passive drool functions as a surrogate for passive WS absorption by the collection pad. Although the WS aliquot is not an exact mimic, the split aliquot design was chosen to ensure direct correspondence between the two processing methodologies by using a common sample pool. In the design illustrated in Figure 1B, the absorbent collection pad is placed in a 5 mL WS stock and allowed to soak for 5 min at room temperature. SS (1–1.5 mL) is obtained by compressing the saturated collection pad, and WS flows through the bifurcated filter assembly. For transcriptome analysis, RNAPro•SAL specimens were stabilized with +/- SUPERase-IN RNase Inhibitor (0.02 U/mL). For proteome analysis, ethanol was added to a final concentration of 20% v/v. In this study, ethanol was evaluated as a proteome preservative alternative to the more commonly used highly reactive protease inhibitors. RNAPro•SAL system and processing
The RNAPro•SAL system (Figure 1A) consists of an absorbent collection pad, sample volume adequacy indicator, compression tube, attachable bifurcated filter unit, and collection tube(s). WS is passively collected by placing the collection pad along the side of the tongue or against the cheek of the participant. WS released by compressing the collection pad flows though the bifurcated filter assembly to yield SS. The filter assembly is configured to remove the bulk of mucin and cellular components. SOP processing
For each participant, SS was prepared from a 5 mL WS aliquot by centrifugation at 2600 × g for 15 min at 4°C. Salivary supernatant was carefully removed from the cellular layer and transferred to a 15 mL Falcon tube. For exRNA stabilization, SUPERase- IN RNase Inhibitor stock (AM2694; Ambion, Austin, TX) was added to SS to a final concentration of 0.02 U/mL (1 L of 20 U/L stock per mL of SS). For proteomic stabilization, a stock consisting of 1 L of 10 mg/mL aprotinin (A1153; Sigma-Aldrich, St. Louis, MO), 3 L of 400 mM sodium orthovanadate (6508; Sigma-Aldrich), and 10 L of 10 mg/mL phenylmethylsulfonyl fluoride (PMSF) (P7626; Sigma- Aldrich) was prepared by dissolving the appropriate mass of each reagent in double-distilled water or isopropanol and then combining the solutions. The stock was then added to the SS to achieve a final concentration of 10 g/mL aprotinin, 1.2 mM sodium orthovanadate, and 100 g/mL PMSF. Total RNA and total protein recovery
Total RNA was quantified using Quant-it RiboGreen (R11940; Invitrogen, Carlsbad, CA) after RNA purification using the RNeasy Mini Kit (74106; Qiagen, Valencia, CA) per the manufacturer's instructions. Purified RNA was isolated from 200 L of SS and dissolved in 50 L RNase-free water. Excess DNA was removed by treatment with RNase-free DNase (AM2222 Lot #1008037; Ambion). RNA content was measured on a NanoDrop 3300 Fluorospectrometer (Thermo Scientific, Waltham, MA), with 490 nm excitation and 520 nm emission wavelengths, using a standard curve constructed from the rRNA kit calibration standards. RNA content is reported as the RNA concentration (ng/mL) of the 50 L purified eluent.
Total protein was measured from equal volumes of SS with the Pierce BCA Protein Assay Kit (Thermo Scientific) as described by the manufacturer using a SYNERGY H1 microplate reader (BioTek, Winooski, VT). Cell content
Viable cell counts were determined on a Vi-CELL XR Cell Viability Analyzer (Beckman Coulter, Brea, CA) using the Vi-CELL reagent pack (3822600; Beckman Coulter) as described by the manufacturer. Saliva analyte preservation and stability analysis
Saliva specimen stability was assessed by retention of RNA and protein analytes as a function of temperature and time. RNAPro•SAL and SOP SS samples stored at ambient temperature were used to model ambient temperature collection and shipment. In addition, RNAPro•SAL SS stability was evaluated at 4°C. For both transcriptome and proteome studies, aliquots were incubated for 0, 3, 5, and 14 days. After designated time periods at the exposure temperatures, individual subsample aliquots were transferred to -80°C for storage pending analysis.
For transcriptome analysis, saliva samples (n = 8) were collected and prepared via the SOP and RNAPro•SAL processes described earlier. Analyte stability was assessed by determination of mRNA expression levels for four endogenous salivary exRNAs— glyceraldehyde 3-phosphate dehydrogenase (GAPDH), β-actin, ribosomal protein S9 (RPS9), and interleukin-8 (IL-8)—via RT-qPCR as described previously (9). The two-step RT-qPCR assay, reverse transcription-PCR (RT-PCR) followed by qPCR measured on PTC-200 (Peltier Thermal Cycler; MJ Research, Waltham, MA) and Light-Cycler 480 II (Roche, Indianapolis, IN), was performed in duplicate for each sample at each time point. Primers were obtained from Sigma-Aldrich (St. Louis, MO).
For proteome analysis, saliva samples (n = 9) were collected and processed via the SOP and RNAPro•SAL methods described earlier. Proteome preservation was assessed by examination of the levels of total protein and the specific salivary protein markers IL-8 and β-actin as a function of storage condition and collection method. Specific protein levels were measured in duplicate per the manufacturer's instructions using the Human IL-8 ELISA Kit (EH2IL8; Thermo Scientific) and Human β-actin ELISA Kit (E91340Hu; Biomatik, Wilmington, DE). RNAPro•SAL protein data were adjusted for the ethanol volume addition (20% v/v). All P values reported were determined using the Mann-Whitney test. Results and discussion RNA and protein recovery
This study demonstrated the clinical functionality of RNAPro•SAL as an effective saliva collection and processing system with similar performance to the SOP with respect to analyte yield and stability. Total RNA yield (mean ± SD) was not significantly different (P > 0.05): 1164.56 ± 127.04 ng/mL and 1183.02 ± 167.47 ng/mL for SOP and RNAPro•SAL, respectively (Figure 2A). Total protein recovery (mean ± SD) was also not significantly different (P > 0.05): 966.98 ± 126.14 g/mL and 996.25 ± 128.41 g/mL for SOP and RNAPro•SAL, respectively (Figure 2B).
Removal of viable cells
The percentage of viable cells remaining in the SS obtained via both methods was similar: 2.73% and 2.62% for SOP and RNAPro•SAL, respectively (Figure 2C). Both methods are expected to yield samples with reduced viable cell content but via different mechanisms, sedimentation (for SOP) versus entrapment (for RNAPro•SAL). In the SOP, cell content is critically dependent on careful removal of the supernatant without disturbing the sediment layers to avoid mixing and sample contamination. Because the RNAPro•SAL system uses a self-contained entrapment and filtering mechanism, the process is expected to be less sensitive to variations in handling by different individuals. Moreover, the majority of the salivary mucin adheres to the RNAPro•SAL collection pad, which significantly reduces the characteristically high viscosity of WS samples, simplifying downstream isolation. Both methods achieve significant improvement in analyte stability over untreated WS by reducing the quantity of viable cells retained in the sample while retaining freely soluble nucleic acids and proteins as well as exosome compartmentalized RNA and proteins. The reduction in viable cell content increases short-term sample stability by minimizing active secretion of nucleases and proteases. Furthermore, the RNAPro•SAL system overcomes the requirement for low-temperature centrifugation, allowing for remote site collection while achieving the same efficiency in removing intact cells and debris and reducing associated nuclease and proteinase activity. Saliva analyte preservation and stability
Despite the mechanistic differences, both RNAPro•SAL and SOP yield samples suitable for direct transcriptomic and proteomic analysis even after 14 days at ambient temperature. Comparable mRNA expression levels were observed for SOP- and RNAPro•SAL-processed saliva specimens after 14 days at room temperature. Mean Cq values for β-actin, GAPDH, RPS9, and IL-8 mRNAs for RNAPro•SAL and SOP specimens were within experimental error (± 1 SD) and statistically indistinguishable (P > 0.05). For both methods, the mean Cq values of all four exRNAs gradually increase over the time period examined in Figure 3, A–D and E–H. At 4°C, RNAPro•SAL specimens exhibit lower mean Cq values consistent with increased stability at lower temperatures. Similar to reports for SOP samples, addition of SUPERase-IN to RNAPro•SAL samples did not enhance RNA stability at room temperature compared to untreated specimens. Spearman correlations within and between methodologies for both RNAPro•SAL and SOP were generally high at day 0 and attenuated at later time points. Aggregating across the 4 genes for RNAPro•SAL, the median correlations were 0.89 at day 0 and 0.28 at day 14, with a similar attenuation observed for SOP. The attenuation may be due to a combination of factors such as sample variation within the relatively small subject set, participant-linked differences in stability or degradation rates, and experimental uncertainty.
Proteomic recovery for RNAPro•SAL with 20% ethanol was comparable to SOP with a protease inhibitor cocktail. Mean total protein levels and IL-8 levels appear to remain constant and consistent for both samples from day 0 to day 14 (Figure 3I) with no significant difference observed in average total protein between the two methods (P > 0.05). Indices of the log ratio of RNAPro•SAL/SOP for IL-8 and β-actin were relatively constant as a function of time (Figure 3J). That the β-actin RNAPro•SAL/SOP log ratio is >0 suggests β-actin content in RNAPro•SAL versus SOP is potentially under-represented because of sedimentation and entrapment of β-actin in the lower mucin layers. The IL-8 RNAPro•SAL/SOP log ratio is <0, which may reflect selective retention of IL-8 in the RNAPro•SAL process due to entrapment within the filter. Total protein and specific protein (IL-8) Spearman correlations for the collection techniques were fairly consistent and moderately strong over the time span, that is, 0.60 on day 0 and 0.73 on day 14 with a median of 0.71 for total protein.
The RNAPro•SAL-integrated system achieves suitable enrichment of RNA and protein over cells and other cellular components for direct analysis and short-term ambient stabilization by selective absorption and filtering. Similarly, our findings coincide with a previous study demonstrating filtration as a robust alternative saliva processing technique compared to centrifugation for protein analysis (10). Moreover, the RNAPro•SAL system provides significant technical advancements over traditional blood and saliva collection methodologies for RNA analysis. Standard EDTA venous blood collections are inadequate for RNA analysis, with mRNA exhibiting differential gene dependent degradation (11) and limited short-term ambient temperature exposures (∼7 days). The RNAPro•SAL system provides RNA stabilization (>14 days) via a simple, low-cost absorption and filtering system. Although mRNA degradation is not completely eliminated, degradation is sufficiently retarded to allow for transcriptome analysis without the need for additional specialized stabilization and isolation processes within the 14-day window examined here.
This study focused on evaluating the clinical utility of the RNAPro•SAL system in salivary biomarker recovery and stability at ambient temperature after collection and processing. The effect of salivary flow rate on biomarker stability is not in the scope of this publication or of the original experimental design. Future directions may include further characterization of RNAPro•SAL through identifying the range of small molecule recovery, low abundant protein or RNA recovery, and the effect of salivary flow rate on biomarker stability to gain a better comparison with other collection methods (12,13). The next improvement of RNAPro•SAL will include cellular and DNA extraction for epidemiologic studies (14,15).
The RNAPro•SAL system provides a number of advantages over the SOP system: (i) it is compatible with self-collection at ambient temperature; (ii) it eliminates specialized accessory equipment and personnel requirements; and (iii) rapid collection (∼1 mL of SS in 5 min) is possible. The ability to simultaneously collect homogeneous samples suitable for concurrent transcriptome and proteome biomarker analysis at ambient temperature allows the RNAPro•SAL system to extend salivary diagnostics to populations lacking access to laboratory facilities and is a significant addition to the arsenal needed to fully realize saliva-based point-of-care diagnostics and treatment. Author contributions
S.H.C. conceived of and designed the study, acquired the data, analyzed and interpreted the data, drafted the article, and provided critical revision. G.A.T. and A.N.H. analyzed and interpreted the data and provided critical revision. W.L. conceived the study. T.G. analyzed and interpreted the data. R.L.B. and M.J.L. analyzed and interpreted the data and provided critical revision. L.F., M.Y., and C.S. acquired the data. F.W., D.E., P.D.S., and D.T.W.W. conceived of the study and interpreted the data and provided critical revision and general supervision.
This work is supported by a research grant to D.T.W.W.: National Institutes of Health (NIH) grants UH2 TR000923, U01 DE017593, R01 DE017170, U01 DE016275, and R21 CA0126733; GSK/ IADR Innovation Award; TRDRP grants 20PT-0032 and 21RT-0112; Department of Defense (DoD) grant LC110207; Fanconi Anemia Research Fund; Barnes Fund for Head & Neck Cancer Research; and the O'Keefe Foundation. This paper is subject to the NIH Public Access Policy.
D.T.W.W. is co-founder of RNAmeTRIX Inc., a molecular diagnostic company. He holds equity in RNAmeTRIX and serves as a company director and scientific advisor. The University of California also holds equity in RNAmeTRIX. Intellectual property that D.T.W.W invented and that was patented by the University of California and has been licensed to RNAmeTRIX. Additionally, he is a consultant to PeriRx. M.J.L. and R.L.B. are employees of Oasis Diagnostics who may benefit from this publication. G.A.T. is an independent Consultant for Oasis Diagnostics with no direct financial interest. P.D.S. is the CEO of Oasis Diagnostics. Oasis Diagnostics is the developer of RNAPro•SAL. The other authors declare no competing interests.
Address correspondence to David T.W. Wong, UCLA School of Dentistry, Los Angeles, CA. E-mail: [email protected]
1.) Elashoff, D., H. Zhou, J. Reiss, J. Wang, H. Xiao, B. Henson, S. Hu, M. Arellano. 2012. Prevalidation of salivary biomarkers for oral cancer detection. Cancer Epidemiol Biomarkers Prev. 21:664-672. 2.) Lau, C.S., and D.T. Wong. 2012. Breast cancer exosome-like microvesicles and salivary gland cells interplay alters salivary gland cell-derived exosome-like microvesicles in vitro. PLoS ONE 7:e33037. 3.) Lee, Y.H., J.H. Kim, H. Zhou, B.W. Kim, and D.T. Wong. 2012. Salivary transcriptomic biomarkers for detection of ovarian cancer: for serous papillary adenocarcinoma. J. Mol. Med. 90:427-434. 4.) Zhang, L., J.J. Farrell, H. Zhou, D. Elashoff, D. Akin, N.H. Park, D. Chia, and D.T. Wong. 2010. Salivary transcriptomic biomarkers for detection of resectable pancreatic cancer. Gastroenterology. 138:949-57 e1-7. 5.) Xiao, H., L. Zhang, H. Zhou, J.M. Lee, E.B. Garon, and D.T. Wong. 2012. Proteomic analysis of human saliva from lung cancer patients using two-dimensional difference gel electrophoresis and mass spectrometry. Mol Cell Proteomics. 11:M111.012112. 6.) Hu, S., Y. Li, J. Wang, Y. Xie, K. Tjon, L. Wolinsky, R.R. Loo, J.A. Loo, and D.T. Wong. 2006. Human saliva proteome and transcriptome. J. Dent. Res. 85:1129-1133. 7.) Jiang, J., N.J. Park, S. Hu, and D.T. Wong. 2009. A universal pre-analytic solution for concurrent stabilization of salivary proteins, RNA and DNA at ambient temperature. Arch. Oral Biol. 54:268-273. 8.) Park, N.J., Y. Li, T. Yu, B.M. Brinkman, and D.T. Wong. 2006. Characterization of RNA in saliva. Clin. Chem. 52:988-994. 9.) Lee, Y.H., H. Zhou, J.K. Reiss, X. Yan, L. Zhang, D. Chia, and D.T. Wong. 2011. Direct saliva transcriptome analysis. Clin. Chem. 57:1295-1302. 10.) Thanakun, S., H. Watanabe, S. Thaweboon, and Y. Izumi. 2013. An effective technique for the processing of saliva for the analysis of leptin and adiponectin. Peptides 47:60-65. 11.) Rainen, L., U. Oelmueller, S. Jurgensen, R. Wyrich, C. Ballas, J. Schram, C. Herdman, D. Bankaitis-Davis. 2002. Stabilization of mRNA expression in whole blood samples. Clin. Chem. 48:1883-1890. 12.) Topkas, E., P. Keith, G. Dimeski, J. Cooper- White, and C. Punyadeera. 2012. Evaluation of saliva collection devices for the analysis of proteins. Clin Chim Acta. 413:1066-1070. 13.) Mohamed, R., J.L. Campbell, J. Cooper- White, G. Dimeski, and C. Punyadeera. 2012. The impact of saliva collection and processing methods on CRP, IgE, and Myoglobin immunoassays. Clin Transl Med. 1:19. 14.) Rogers, N.L., S.A. Cole, H.C. Lan, A. Crossa, and E.W. Demerath. 2007. New saliva DNA collection method compared to buccal cell collection techniques for epidemiological studies. Am J Hum Biol. 19:319-326. 15.) Nemoda, Z., M. Horvat-Gordon, C.K. Fortunato, E.K. Beltzer, J.L. Scholl, and D.A. Granger. 2011. Assessing genetic polymorphisms using DNA extracted from cells present in saliva samples. BMC Med. Res. Methodol. 11:170.