High-throughput Preparation of DNA, RNA, and Protein from Cryopreserved Human iPSCs for Multi-omics Analysis

Abstract

We describe the procedure to isolate genomic DNA, RNA, and protein directly from cryopreserved iPSC vials using commercially available solid-phase extraction kits, and report the relationship between the macromolecules yields and experimental and storage factors. Sufficient quantities of DNA, RNA, and protein were recoverable from as low as 1 million cryopreserved cells across 728 distinct iPSC lines suitable for whole-genome sequencing, RNA sequencing, and mass spectrometry experiments. Nucleic acids extracted from iPSC stocks cryopreserved up to 4 years maintained sufficient quantity and integrity for downstream analysis with minimal genomic DNA fragmentation. An expected positive correlation exists between cell count and DNA/RNA yield, with comparable yields recovered between cells across different cryostorage timespans. This protocol provides an effective way to simultaneously isolate iPSC biomolecules for multi-omics investigations.

INTRODUCTION:

Here we describe an experimental protocol to simultaneously extract the biological macromolecules in DNA, RNA, and protein from cryopreserved induced pluripotent stem cells (iPSCs) such as commonly received from biorepository service requests for various downstream quality assurance and research applications. We show that DNA, RNA, and protein can be effectively extracted directly from cryopreserved stock using accessible protocols. There was no correlation between storage duration and yield, as sufficient yields of high-quality molecules were extracted from iPSCs cryopreserved for approximately 4 years.

In this article, two protocols are listed. The first protocol details the method of biomolecule extraction used and how to quantify and qualify the resulting DNA isolated, and provides the storage instructions for isolated RNA and protein pellets. The second protocol describes the method of quantifying and determining quality of the extracted RNA.

BASIC PROTOCOL 1

QIAShredder and AllPrep^® DNA/RNA/Protein MiniKit Extraction and subsequent DNA quantification and quality analysis

The following steps use the commercially available Qiagen AllPrep^® DNA/RNA/Protein Mini Handbook, but other similar solid-phase extraction (SPE) columns may be used (AllPrep® DNA/RNA/Protein, 2014). All centrifugation steps are performed at room temperature (22-24 °C). The experimenter will isolate one or more samples in a streamlined, simultaneous manner using various buffers and several types of spin columns to homogenize, separate, and purify distinct biomolecule products. Observe chemical safety precautions in handling β-mercaptoethanol is advised including the use of a fume hood. In our hands, the protocol reliably yielded isolated DNA and RNA eluted samples, alongside a denatured protein pellet of gross content from original cell sample. DNA quality was checked by running 1 μL aliquots in a 2% agarose gel run at 110 V for 35 min, alongside a 1 kb DNA ladder. The gel was stained with MidoriGreen DNA dye. After a successful run, the DNA was imaged on a MultiDoc-It™ Imaging System. Good quality DNA will visualize as a bright band closest to the negative end of the gel with minimal smearing. DNA yield may be quantified using various spectrophotometers, such as ThermoFisher NanoDrop and QuBit^® and Agilent Bioanalyzer. The following steps assume the use of the QuBit^® dsDNA BR Assay Kit User Manual from Life Technologies (Qubit® dsDNA BR Assay Kits, 2015).

Materials:

(unless otherwise noted, please store at room temperature)

Materials

AllPrep DNA/RNA/Protein Mini Kit (50) (QIAGEN, cat. no. 80004)

QIAshredder Kit (250) (QIAGEN, cat. no. 79656)

QIAshredder Mini Spin Column (QIAGEN (in AllPrep kit))

Allprep DNA Mini Spin Column (QIAGEN (in AllPrep kit))

RNeasy Mini Spin Column (QIAGEN (in AllPrep kit))

Buffer RW1 (QIAGEN (in AllPrep kit))

Buffer RLT (QIAGEN (in AllPrep kit))

Buffer RPE (QIAGEN (in AllPrep kit))

Buffer AW1 (QIAGEN (in AllPrep kit))

Buffer AW2 (QIAGEN (in AllPrep kit))

Buffer EB (QIAGEN (in AllPrep kit))

RNaseZap (Invitrogen, cat no. AM9780)

Absolute Ethanol (100% ethanol, 200 proof) (Fisher BioReagents, cat. no. BP2818500)

2-Mercaptoethanol, ≥99.0% (Sigma-Aldrich, cat. no. M6250)

UltraPure Distilled Water (Invitrogen, cat. no. 10977015)

Dulbecco’s Modified Eagle Medium Nutrient Mixture F-12 (Ham) (Life Technologies, cat. no. 11330032) (store at 4 °C)

Trypan Blue Stain 0.4% (ThermoFisher Scientific, cat. no. T10282)

MicroAmp Optical 96-Well Reaction Plate (Life Technologies, cat. no. N8010560)

Qubit dsDNA BR Assay Kit (ThermoFisher Scientific, cat. no. Q32853) (see individual components for storage conditions)

Gel Loading Dye, Blue (6X) (New England Biolabs, cat. no. B7021S)

UltraPure™ TBE Buffer, 10X (Invitrogen™, cat. no. 15581044)

1 kb DNA Ladder (New England Biolabs, cat. no. N3232S) (store at −20 °C)

UltraPure™ Agarose (Invitrogen™, cat. no. 16500100)

MIDORI^Green Advance (NN Genetics, cat. no. MG04) (store at 4 °C)

Equipment

Research® Plus Single Channel Pipette, Adjustable 0.1-2.5 μL micropipette (Eppendorf, cat. no. 3123000012)

Research® Plus Single Channel Pipette, Adjustable 20-200 μL micropipette (Eppendorf, cat. no. 3123000055)

Research® Plus Single Channel Pipette, Adjustable 100-1000 μL micropipette (Eppendorf, cat. no. 3123000063)

RPT Filter Tips, 10 μL XL (TipOne, cat. no. 1120-3810)

RPT Filter Tips, 200 μL, graduated (TipOne, cat. no. 1120-8810)

RPT Filter Tips, 1000 μL, graduated (TipOne, cat. no.1111-2830)

15 mL High-Clarity Polypropylene Conical Tubes (Corning, cat. no. 352096)

Safe-Lock Tubes 1.5mL, natural (Eppendorf, cat. no. 022363204)

Protein LoBind Tubes 2.0 mL (Eppendorf, cat. no. 022431102)

Isotemp Digital Control Water Bath Model 205 (ThermoFisher Scientific, cat. no. 154625Q)

Compact Drybath D 100-120V (ThermoFisher Scientific, cat. no. 8887 1002)

Sorvall Legend Micro 21 Microcentrifuge (ThermoFisher Scientific, cat no.75002436)

Sorvall Legend X1 Centrifuge (ThermoFisher Scientific, cat. no. 75004221)

Countess II FL Automated Cell Counter (ThermoFisher Scientific, cat. no. AMQAF1000)

Countess Cell Counting Chamber Slides (ThermoFisher Scientific, cat. no. C10283)

Mini Sub-Cell GT Gel Caster (Bio-Rad, cat. no. 1704422)

Mini-Sub Cell GT UV-Transparent Gel Tray, 7 x 10 cm (Bio-Rad, cat. no. 1704435)

15-Well Comb for use with Mini-Sub cell GT systems (Bio-Rad, cat. no. 1704465)

Wide Mini-Sub Cell GT UV-Transparent Gel Tray, 15 x 7 cm (Bio-Rad, cat. no. 1704426)

30-Well Comb for use with wide Mini-Sub cell GT systems (Bio-Rad, cat. no. 1704449)

Mini-Sub Cell GT Cell (Bio-Rad, cat. no. 1704406)

Wide Mini-Sub Cell GT Cell (Bio-Rad, cat. no. 1704405)

PowerPac™ Basic Power Supply (Bio-Rad, cat. no. 1645050)

QuBit 4 Fluorometer (ThermoFisher Scientific, cat. no. Q33238)

GelDoc-ItTM Imaging System 2UV Transilluminator (UVP, cat. no. 361809)

Protocol steps — Step annotations:

Preparation

1. [QIAGEN instructions] Prepare RLT buffer by adding 10 μL 2-Mercaptoethanol, ≥99.0% to every 1 mL of RLT used.

2. [QIAGEN instructions] Add absolute ethanol to QIAGEN buffers RPE, AW1, and AW2 as per the manufacturer’s instructions.

3. Prepare 70% ethanol by mixing absolute ethanol to UltraPure Distilled Water in a 70:30 ratio.

4. [QIAGEN instructions] Warm EB buffer to 70 °C on Compact Drybath D 100-120V.

5. Spray down work area with RNaseZap to avoid RNA isolate deterioration.

Cell Homogenization and Lysis

1. Retrieve vials of cryopreserved iPSCs each containing ~1-2 million live cells or the equivalent of one 6-well of 80% confluent cells from liquid nitrogen storage.

Samples with sufficient DNA have been extracted from one vial of cryopreserved cells, but it is recommended to process at least 2 vials per sample at the same time to ensure sufficient yield.

• 2.Thaw cells in an Isotemp Digital Control Water Bath Model 205 at 37 °C for 2 min, then transfer to a 15 mL High-Clarity Polypropylene Conical Tube.

Samples should be placed in a multi microtube holder to quicken process. This thaw step is more liberal than thawing for cell culture, though care should be taken to not leave cells thawed for too long.

• 3.Add 9 mL growth media (Dulbecco’s Modified Eagle Medium Nutrient Mixture F-12 (Ham), for a total of 11 mL) and spin down at 300 g for 3-7 min (3 for 10 or less samples), 5 min for 12 samples, and 7 min for 20-24 samples).

Other formulations of DMEM, such as GlutaMAX™, are also acceptable.

• 4.Aspirate/decant media and resuspend pellet in 1 mL growth media.
• 5.Count cells (mix 10 μL cells, 10 μL Trypan Blue Stain 0.4%) using a Countess II FL Automated Cell Counter (10 μL per side on counting slide).
• 6.Spin down cells in a Sorvall Legend X1 Centrifuge at 300 g for 5 min and aspirate media.

Though the specific centrifuge mentioned was used in our experiments, any centrifuge with tube-appropriate holders may suffice.

• 7.[QIAGEN instructions] Using a 1,000 μL micropipette add 350 μL of QIAGEN Buffer RLT to each sample and pipette/vortex until cells are disrupted.
• 8.[QIAGEN instructions] Pipet the lysate directly into a QIAGEN QIAshredder spin column placed in a 2 mL collection tube, and centrifuge at 16,100 g in a Sorvall Legend Micro 21 Microcentrifuge for 2 min to homogenize the cells.

Alternative cell disruption and homogenization methods are also supported in the manufacturer’s protocol (AllPrep® DNA/RNA/Protein, 2014). Acceptable homogenization methods include the TissueRuptor homogenizer and syringe-and-needle.

Use of a blade-homogenizer such as TissueRuptor may result in fragmented genomic DNA while increasing yield. Use of gentler methods such as QIAshredder or syringe/needle will increase DNA integrity at the cost of yields (AllPrep® DNA/RNA/Protein, 2014). As such, the exact disruption/homogenization method must consider the final use-case of the product.

• 9.[QIAGEN instructions] Transfer the homogenized lysate to an QIAGEN AllPrep DNA spin column placed in a 2 mL collection tube. Close the tube lid and centrifuge at 8,000 g for 30 s.
• 10.[QIAGEN instructions] Transfer the AllPrep DNA spin column to new 2 mL collection tube and place aside. Take the flow-through of the previous step for RNA purification.

RNA purification/extraction

• 11.[QIAGEN instructions] To the eluent from step 7, add 250 μL of absolute ethanol, and pipet to mix.
• 12.[QIAGEN instructions] Transfer the sample to a QIAGEN RNeasy spin column and centrifuge at 8,000 g for 30 s. Transfer the flowthrough of this step to a separate 2 mL Protein Lo-Bind tube for protein purification.
• 13.[QIAGEN instructions] Using a 1,000 μL micropipette add 700 μL QIAGEN Buffer RW1, and 500 μL QIAGEN Buffer RPE two times. After each addition of buffer, centrifuge at 8,000 g for 30 s (2 min during final elution).
• 14.[QIAGEN instructions] Dry the RNeasy spin by centrifuging at 16,100 g for 1 min, and transfer to new 1.5 mL collection tube. Using a 200 μL micropipette add 55 μL of RNase-free water directly to the spin column membrane and elute RNA via centrifugation at 8,000 g for 1 min. Aliquot 5 μL of RNA into a separate tube with the same label as the original. Freeze both on dry ice.

It is recommended to freeze quickly, as RNA is not stable at room temperature for long. If extracting multiple samples, it is recommended to aliquot into a MicroAmp Optical 96-Well Reaction Plate and seal with adhesive cover.

DNA purification/extraction

• 15.[QIAGEN instructions] Sequentially, add 500 μL QIAGEN Buffer AW1 and 500 μL QIAGEN Buffer AW2 to the AllPrep DNA spin column. After each addition of buffer, centrifuge at 8,000 g at RT for 30 s.

Users of the DNeasy^® Blood and Tissue Kit may see a discrepancy in that the spin column is not changed after each elution. Changing tubes is not necessary.

• 16.[QIAGEN instructions] Transfer the Allprep DNA spin column to a new 1.5 mL collection tube. Add pre-heated 55 μL QIAGEN Buffer EB directly to the spin column membrane and incubate at room temperature for 2 min, then centrifuge at 8000 g for 1 min to elute DNA. Using a 10 μL micropipette, aliquot 5 μL of DNA into a separate tube with the same label as the original.

Protein purification/reprecipitation

• 17.[QIAGEN instructions] Add 600 μL of Buffer APP to the saved flow-through from step 10. Vortex vigorously for 30 s and then incubate at room temperature for 10 min to precipitate protein.
• 18.Centrifuge at 16,100 g for 10 min to bring down the protein pellet. Decant or aspirate the supernatant and add 500 μL of 70% ethanol to the protein pellet. Centrifuge at 16,100 g for 1 min. Aspirate all supernatant and air dry the protein pellet in chemical hood for 5 min. Freeze down at −80 °C.

This protein pellet is denatured aggregate (AllPrep® DNA/RNA/Protein, 2014). Steps for further protein solubilization and analysis are at the discretion of the experimenter.

Quantification of DNA yield

• 19.Reset the QuBit 4 Fluorometer by unplugging it for 1-2 min and then replugging.

Not resetting the machine for long periods has led to inconsistent/bizarre measurements. Please leave the spectrometer unplugged when not in use.

• 20.[Thermo Fisher Scientific instructions] Make a necessary amount of buffer using the QuBit DNA BR buffer solution mixed with the BR dye to a 200:1 ratio; this solution is stable for 1 hour. This solution must be prepared in a plastic container.

Make 1 additional sample’s worth for each run to account for pipetting loss.

• 21.[Thermo Fisher Scientific instructions] Make standard dsDNA BR solutions by mixing 10 μL of BR standards 1 and 2 with 190 μL DNA BR buffer solution. Mix thoroughly and analyze using QuBit until a standard curve is generated.

Be sure to use clear 700 μL PCR tubes. The QuBit 4 Flurometer can be bought with tubes which are specifically labeled as compatible, though ordinary clear 700 μL PCR tubes are also acceptable (Qubit® dsDNA BR Assay Kits, 2015).

• 22.[Thermo Fisher Scientific instructions] Make sample solutions by mixing 1-2 μL of aliquoted DNA with 199-198 μL buffer solution in a QuBit tube. Mix thoroughly and analyze using QuBit.

It is recommended to run 3 trials to ensure reproducible, accurate measurement was made.

• 23.Unplug QuBit and clean up. Discard solutions in non-hazardous waste beakers.

Storage

Store all macromolecules in a −80 °C freezer. Avoid excessive freeze-thawing by keeping completed samples and associated aliquots on dry ice. RNA is particularly sensitive to degradation (Yu et al., 2017).

DNA Quality Control

• 24.From the 5 μL aliquot of DNA, mix 2 μL with 13 μL UltraPure Distilled Water and 3 μL Gel Loading Dye, Blue (6X) in a small PCR tube or MicroAmp Optical 96-Well Reaction Plate chamber.
• 25.Make 2% agarose gel by mixing 2 g (weighed on balance) of UltraPure™ Agarose with 100 mL of 1X TBE buffer (diluted from UltraPure™ TBE Buffer, 10X) and microwaving for 30 second intervals and swirling until agarose is fully dissolved.

Water after microwaving is extremely hot; wear heat-resistant gloves when handling container. Be sure not to cap the container during microwave step, as this risks an explosion.

• 26.Add 5 μL MIDORI^Green Advance gel staining dye to the agarose buffer mixture for 20,000X dilution. Cast gel in Mini Sub-Cell GT Gel Caster and the appropriately sized gel tray and comb. Let cool until the gel is solid and comb difficult to remove (45 min – 1 hr at room temperature, 15 – 30 min at 4 °C).

Depending on the number of samples run, use size-appropriate gel tray and comb.

• 27.Place gel and gel tray in the appropriately sized Mini-Sub Cell GT Cell and fill with 1X TBE buffer. Load gel with 10 μL 1kb ladder and 10 μL of sample per well.
• 28.Run using PowerPac™ Basic Power Supply set to 110 V and 400 mA for 40 minutes.
• 29.Visualize gel in MultiDoc-It™ Imaging System, with exposure time of 0.8 – 1.2 s. Save image in TIF or JPEG formats.

Sample Data

DNA concentrations and quality

We have applied this protocol to at least 953 lines of patient-specific iPSCs. Of the 953 recorded lines, the cell counts of 728 lines were recorded. DNA amounts in μg and μg/million cells were derived from cell count and DNA concentrations quantified by the QuBit 4 Fluorometer, and graphed (Figure 1A, 2A). Live cell count was also recorded for these samples, and similarly graphed to DNA amounts (Figure 1B, 2B). Overall, moderately positive correlations were observed for DNA amount and cell count, both total and alive only (R² values of 0.3725 and 0.3665), and no appreciable correlations between cell count and amount of DNA per million cells, whether live cells or total cells (R² values of 0.012 (0.11) and 0.036 (0.19) respectively), were observed.

Figure 1: Correlation of cell count and DNA yield.

(A) Linear regression plot relating DNA yield to total number of cells. Standard errors were found using LINEST function in Excel, and are as follows: slope 3.6*10⁻⁸, y-intercept 0.12, and y-value 1.3. (B) Linear regression plot relating DNA yield to total number of living cells. Standard errors were found using LINEST function in Excel, and are as follows: slope 5.0*10⁻⁸, y-intercept 0.11, and y-value 1.3.

Figure 2: Correlation of cell count and DNA yield per million cells.

(A) Linear regression plot relating DNA yield per million cells to total number of cells. Standard errors were found using LINEST function in Excel, and are as follows: slope 1.2*10⁻⁸, y-intercept 0.04, and y-value 0.44. (B) Linear regression plot relating DNA yield per million cells to total number of living cells. Standard errors were found using LINEST function in Excel, and are as follows: slope 1.7*10⁻⁸, y-intercept 0.04, and y-value 0.44.

DNA quality was measured using agarose gel electrophoresis on 268 samples, and all but one sample showed adequate DNA quality (minor to negligible streaking, brightest band clearly distinguishable at the top) (Figure 3). Two example gels have been provided in this manuscript for reference.

Figure 3: Gel images of selected sample runs.

(A) UV image of DNA gel of iPSC lines listed. Lines 479 and 481 are not adequate to pass QC; the others are. Line numbers are from the Stanford CVI iPSC Biobank internal nomenclature scheme. Exposure time: 1.2 s. (B) UV image of DNA gel of iPSC lines listed. All lines shown pass QC. Line numbers are from the Stanford CVI iPSC Biobank. Exposure time: 0.8 s.

iPSCs extracted using this protocol were frozen and banked at various dates in the past from 2014 to current year. However, we found no appreciable correlation between total or per million DNA yield and frozen cell age for the 713 lines with these data (R² values of 0.0072 (0.085) and 0.0024 (0.049) respectively) (Figure 4).

Figure 4: Relation between DNA yield and cryostorage time.

(A) Linear regression plot relating DNA yield to cell age in cryo-storage. Standard errors were found using LINEST function in Excel and are as follows: slope 0.0002, y-intercept 0.11, and y-value 1.65. (B) Linear regression plot relating DNA yield per million cells to cell age in cryostorage. Standard errors were found using LINEST function in Excel and are as follows: slope 5.3*10⁻⁵, y-intercept 0.029, and y-value 0.45.

Prediction of DNA yield based on cell count

From the data and correlations obtained, an equation can be proposed to approximate the recommended amount of starting material (e.g. iPSCs) needed for a successful extraction that can be used towards WGS. A linear correlation may be used to generate a predictive equation for cell counts below the maximum recommended by manufacturer (10 million total) (AllPrep® DNA/RNA/Protein, 2014). In addition, there is no significant correlation between cell viability and DNA yield, suggesting that viability alone is not a significant dependent variable (Figure S1). From this, the linear regression in Figure 1A was analyzed using LINEST in Excel, and the equation is presented here with standard error correction (Equation 1).

$$\text{DNA Yield}(\mu \mathrm{g})=(7.54\mathrm{E}-07\pm 3.63\mathrm{E}-08)\ast \text{total cell count}+0.65$$ (Equation 1:)

Estimate of DNA yield by total iPSC count in sample

Disregarding standard error, the number of cells needed for an expected yield of 1.5 μg is 1.1 million cells.

Protein Sample Storage

Protein samples were collected as denatured pellets and stored at −80 °C. The quality and quantity of the pellets were not estimated here. Subsequent re-solubilization of pellets for protein quantification may be performed using common methods (Pullara et al., 2013).

BASIC PROTOCOL 2

Broad-range RNA quantification and quality assay using QuBit 4 Fluorometer and associated kits

This step involves quantitation and quality assay of extracted RNA using the Life Technologies QuBit 4 Fluorometer. If done correctly, RNA concentration and quality number IQ (defined by manufacturer as % RNA in solution which is large and structured/10) for each sample should be obtained. The following was adapted from the QuBit RNA BR and IQ Assay Kit User Manuals from Life Technologies (Qubit® RNA BR Assay Kits, 2015; Qubit™ RNA IQ Assay Kits, 2017).

Materials:

Materials

Extracted RNA from Protocol #1

Qubit™ RNA BR Assay Kit (ThermoFisher Scientific, cat. no. Q10210) (see individual components for storage conditions)

Qubit™ RNA IQ Assay Kit (ThermoFisher Scientific, cat. no. Q33221) (see individual components for storage conditions)

Equipment

QuBit 4 Fluorometer (ThermoFisher Scientific, cat. no. Q33238)Research® Plus Single Channel Pipette, Adjustable 0.1-2.5 μL micropipette (Eppendorf, cat. no. 3123000012)

Research® Plus Single Channel Pipette, Adjustable 20-200 μL micropipette (Eppendorf, cat. no. 3123000055)

Research® Plus Single Channel Pipette, Adjustable 100-1000 μL micropipette (Eppendorf, cat. no. 3123000063)

RPT Filter Tips, 10 μL XL (TipOne, cat. no. 1120-3810)

RPT Filter Tips, 200 μL, graduated (TipOne, cat. no. 1120-8810)

RPT Filter Tips, 1000 μL, graduated (TipOne, cat. no.1111-2830)

Protocol steps—Step annotations:

1. [Thermo Fisher Scientific instructions] Make the necessary amount of buffer by mixing the QuBit RNA BR buffer solution with the RNA BR dye in a 200:1 ratio; this solution is stable for 1 hour. This solution must be prepared in a plastic container.

2. [Thermo Fisher Scientific instructions] Make standard RNA BR solutions by mixing 10 μL of RNA BR standards 1 and 2 with 190 μL RNA BR buffer solution. Mix thoroughly and analyze using QuBit until a standard curve is generated.

Use clear 700 μL PCR tubes. The QuBit 4 Flurometer can run with clear 700 μL tubes (Qubit® RNA BR Assay Kits, 2015; Qubit™ RNA IQ Assay Kits, 2017).

• 3.[Thermo Fisher Scientific instructions] Make sample solutions by mixing 1 μL of aliquoted RNA with 199 μL buffer solution in PCR tube. Mix thoroughly and analyze using QuBit.

It is recommended to run 3 trials to ensure reproducible, accurate measurement.

• 4.[Thermo Fisher Scientific instructions] After RNA samples are quantitated, discard current solutions and buffer. Make QuBit RNA IQ buffer solution mixed with RNA IQ dye to a 200:1 ratio; this solution is stable for 1 hour. This solution must be prepared in a plastic container.
• 5.[Thermo Fisher Scientific instructions] Make standard RNA IQ solutions by mixing 10 μL of RNA IQ standards 1, 2, and 3 with 190 μL RNA IQ buffer solution. Mix thoroughly and analyze using QuBit until tests are passed. Use clear 700-μL PCR tubes.
• 6.[Thermo Fisher Scientific instructions] Make sample solutions by mixing aliquoted RNA into RNA IQ solution, for a total of 200 μL. The amount of RNA aliquot added is contingent on the concentrations of RNA found during the RNA BR assay (Table 1) (Qubit™ RNA IQ Assay Kits, 2017).
• 7.Mix thoroughly and analyze using QuBit.
• 8.Unplug QuBit and clean up. Discard solutions in a non-hazardous waste beaker.

Table 1:

QIAGEN buffers used and general purpose and composition. Equivalent formulation implies matching function, not necessarily matching ingredients (Ralf Himmelreich & Sabrine Werner, 2011; QIAGEN FAQ ID - 199, 2020)

Buffer	General Purpose	Equivalent formulation
RW1	Alcohol-containing buffer with guanidinium salt	20% by vol 1,3-butanediol, 900 mM guanidinium thiocyanate, 10 mM Tris-Cl pH 7.5
RLT	Buffer contaainingthiocyanate	N/A
RPE	Aqueous buffer	80% by vol 1,3-butanediol, 100 mM NaCl, 10 mM Tris-Cl pH 7.5
AW1	Wash buffer containing guanidinium hydrochloride	49% by vol 1,3-butanediol, 2.5 M guanidium chloride
AW2	Wash buffer containing sodium azide	80% by vol 1,3-butanediol, 100 mM NaCl, 10 mM Tris-Cl pH 7.5
EB	Aqueous elution buffer	10 mM Tris-Cl, pH 8.5

Sample Data

RNA concentrations and quality

Simultaneous RNA extraction was performed for the same 953 cell lines. Most RNA was stored in −80 °C after extraction, though 35 samples were tested for concentration and quality. 48 RNA samples were measured for total RNA yield and RNA quality using the QuBit RNA BR and RNA IQ kits. RNA yield and RNA yield per million were graphed to total and live cell counts (Figures 5 & 6). The average RNA IQ value for these samples was 8.9 with a sample standard deviation of 0.6. This translates to an 89:11 ratio of large, structured RNA to small RNA on average, with little overall deviation from the mean. This quality is more than sufficient for accurate bulk RNA sequencing (Qubit RNA IQ Assay: a fast and easy fluorometric RNA quality assessment, 2018).

Figure 5: Correlation between cell count and RNA yield.

(A) Linear regression plot relating RNA yield to total cell count. Standard errors were found using LINEST function in Excel and are as follows: slope 5.2*10⁻⁸, y-intercept 2.2, y-value 6.3. (B) Linear regression plot relating RNA yield to live cell count. Standard errors were found using LINEST function in Excel and are as follows: slope 7.2*10⁻⁶, y-intercept 2.0, y-value 6.3.

Figure 6: Correlation between cell count and RNA yield per million cells.

(A) Linear regression plot relating RNA yield per million cells to total cell count. Standard errors were found using LINEST function in Excel, and are as follows: slope 2.2*10⁻⁷, y-intercept 0.94, y-value 2.6. (B) Linear regression plot relating RNA yield per million cells to live cell count. Standard errors were found using LINEST function in Excel, and are as follows: slope 3.1*10⁻⁷, y-intercept 0.87, y-value 2.7.

A robust linear correlation was observed between RNA yield and either total or live cell (R² values of 0.756 (0.869) and 0.748 (0.865), respectively). The correlation coefficients are considerably higher than the those between total and live cell counts and DNA yields. This observation holds true for both the correlations of all samples (Figure 1) and the samples for which RNA yield and quality were measured (Figure S2, S3). Interestingly, there is a moderately negative correlation of cell count and RNA yield per million, which was also observed for DNA yields per million cells (Figure S4, S5). There is no correlation between age of the cells and RNA yield (Figure 7A). DNA and RNA yields from the same samples are moderately correlated (R² values of 0.5706 (0.7554) for DNA and RNA yield, and 0.4739 (0.6884) for DNA and RNA yield per million cells) (Figure 7B-C).

Figure 7: Comparisons between RNA and DNA yield from extraction batches.

(A) Linear regression plot relating RNA yield per million cells to cell age in cryostorage. Standard errors were found using LINEST function in Excel, and are as follows: slope 0.0016, y-intercept 0.88 and y-value 3.1. (B) Linear regression plot relating DNA yield to RNA yield of identical lines extracted. Standard errors were found using LINEST function in Excel, and are as follows: slope 0.012, y-intercept 0.41, y-value 0.94. (C) Linear regression plot relating DNA yield per million cells to RNA yield per million of identical lines extracted. Standard errors were found using LINEST function in Excel, and are as follows: slope 0.018, y-intercept 0.16, y-value 0.

REAGENTS AND SOLUTIONS:

There are no unique reagents or solutions used in this protocol. The QIAGEN and QuBit DNA/RNA kits contain some proprietary ingredients whose composition are not publicly disclosed by the manufacturers. However, various equivalent solutions and tentative recipes may be gleaned from existing patents and other online resources. Based on available information, we provide here an estimated list of compositions for these buffers (Table 1). Please note that these formulations do not necessarily represent exact manufacturer recipes, although they have been referred to be equivalent replacements (Ralf Himmelreich & Sabrine Werner, 2011; QIAGEN FAQ ID - 199, 2020).

COMMENTARY

BACKGROUND INFORMATION:

Human iPSCs are widely utilized in drug discovery and modeling of genetic and epigenetic factors of disease phenotypes (Chen, Matsa & Wu, 2016; Shi et al., 2017; Paik, Chandy & Wu, 2020; Sharma et al., 2020). Recent advances in organoid generation and gene editing have further propelled the utility of iPSCs in precision medicine applications (Lau, Paik & Wu, 2019). With decreasing cost and increasing scale and throughput of sequencing, patient-specific iPSCs are now also used in large-scale association studies to identify factors conferring risks of complex diseases (Lau, Paik & Wu, 2019; Strober et al., 2019). Diseases with subtle dependencies on patient phenotype and familial history can and have been precisely modeled using iPSCs and downstream differentiated cells. Examples include the use of iPSC-derived cardiomyocytes (iPSC-CMs) in studies on doxorubicin-induced cardiomyopathy and familial Brugada syndrome, as well as iPSC-forebrain-type cortical neurons replicating similar deficiencies derived from patients with gene-based frontotemporal dementia (Burridge et al., 2016; Belbachir et al., 2019; Zhang et al., 2017).

With increasing prevalence of iPSC studies there is now a need for collection of standardized iPSC lines to promote reproducible research. Several consortia have been established to create biorepositories of characterized iPSC lines (Huang et al., 2019; Coriell Institute Stem Cell Biobank, 2020; CIRM iPSC Repository, 2020; European Bank For iPSCs (EBiSC), 2020). The Stanford Cardiovascular Institute (CVI) is one of the largest existing cardiovascular research centers focused iPSC biorepositories, having to date reprogrammed peripheral blood mononuclear cells (PBMCs) collected from over 1,200 patients of diverse ethnicities and health statuses by using high-efficiency and non-integrating Sendai virus programming methods (Wu et al., 2019; Churko et al., 2017). This sample collection therefore supports inclusive studies on multiple ethnic groups that account for and examine genetic and epigenetic variation across samples. The Stanford CVI iPSC Biobank has provided cryopreserved cellular materials to local, national and international labs alike for iPSC studies. Cryopreservation of iPSC vials provides a standardized method to store cellular materials over long periods and allows live cells and genetic materials to be revived by investigators at remote sites. Biomolecules extracted from iPSCs using this protocol may be used in downstream multi-omics analysis; indeed, simultaneous extraction should mitigate potential batch effects. Examples of multiomics analysis possible are combined bulk analysis of DNA, RNA, and protein (WGS/ATAC-seq; bulk transcriptomic analysis; ESI/MALDI protein mass spectrometry) to map interesting phenomena of transcriptome and proteome to genome, such as patient phenotype (Lau, Paik & Wu, 2019). Such techniques require sufficient quantity and quality of product, with tolerances given for each facility: ~1.5 μg isolated DNA with no/insignificant smearing on gel electrophoresis for PCR-free WGS (Human Whole Genome Sequencing, 2020), 2.5 μg isolated RNA with RIN > 7.0 for bulk RNA sequencing (Cui et al., 2011), and proteins with relatively homogenous distributions and limited variance for successful mass spectrometry (Protein Sample Preparation for Mass Spectrometry, 2020). Though this protocol is unable to facilitate single-cell techniques such as single-cell sequencing and proteomics, we are hopeful that continuing developments in this novel field will allow the revision of these protocols to match the desired resolution (Paik et al., 2020).

CRITICAL PARAMETERS:

Cell samples were acquired from the Stanford CVI iPSC Biobank (Wu et al., 2019). Cell lines were created from patient blood samples containing peripheral blood mononuclear cells (PBMCs) and subsequently re-programmed into iPSCs. These cell lines were passaged on multiple 6-well plates, grown to about 80% confluency, and frozen down in 1 mL ThermoFisher BAMBANKER freezing media at −80°C overnight in a ThermoFisher CoolCell® Cell Freezing Container. Afterwards, the cells were categorized, labeled and banked, and placed into cryo-storage. While this protocol may be used for iPSCs outside of our own, the interpretation of our results may vary depending on previous handling protocols.

TROUBLESHOOTING:

Contaminated Samples

Contamination of product can result from placing eluents in the incorrect spin column such as by adding the ethanol-mixed eluent from the AllPrep columns back to the AllPrep columns instead of RNeasy. The general method to solve this error depends on whether the column has already been spun down.

Should the column be new and empty, immediately pipet out the incorrectly placed eluent into the correct column. The remaining eluent trapped in the column can either be spun down using the appropriate elution buffer (EB or UltraPure H₂O for DNA, Ultrapure H₂O for RNA and RLT for protein), or simply discarded, sacrificing a relatively insignificant percentage-of-yield for product in the process. No contamination is risked in this case, so additional measures to test for purity are not required. However, if the column has already been spun down with the previous eluent, steps will need to be taken to avoid contamination. To that end, in addition to the removal of the solution above the spin column, the column must be flushed with the elution buffer appropriate to the contaminating solution. This will remove most, if not all, of the contaminant. Afterwards, the resulting product must be analyzed for contamination using the appropriate methods (Table 3, Figure S6).

Table 3:

260/280 ratios of samples 701-724. Samples 702-711 were incorrectly pipetted into AllPrep columns rather than RNeasy columns after addition of 250 μL 100% ethanol.

DNA Sample Number	BioBank Line Number	260/280 ratio
JZ_DNA_0701	JZBank879	1.89
JZ_DNA_0702	JZBank880	1.86
JZ_DNA_0703	JZBank881	1.92
JZ_DNA_0704	JZBank882	1.84
JZ_DNA_0705	JZBank883	1.83
JZ_DNA_0706	JZBank884	1.85
JZ_DNA_0707	JZBank885	1.86
JZ_DNA_0708	JZBank886	1.85
JZ_DNA_0709	JZBank887	1.84
JZ_DNA_0710	JZBank888	1.94
JZ_DNA_0711	JZBank889	1.9
JZ_DNA_0712	JZBank890	1.83
JZ_DNA_0713	JZBank891	1.88
JZ_DNA_0714	JZBank892	1.87
JZ_DNA_0715	JZBank893	1.85
JZ_DNA_0716	JZBank894	1.84
JZ_DNA_0717	JZBank895	1.84
JZ_DNA_0718	JZBank896	1.84
JZ_DNA_0719	JZBank897	1.86
JZ_DNA_0720	JZBank898	1.87
JZ_DNA_0721	JZBank899	1.96
JZ_DNA_0722	JZBank900	1.9
JZ_DNA_0723	JZBank901	1.85
JZ_DNA_0724	JZBank902	1.85

QuBit reading error

Samples measured by the QuBit 4 Fluorometer may not always show an accurate reading. Reading errors occur when the DNA concentration is grossly underestimated or when subsequent readings of the same sample are very different (>20%) from each other. When this occurs, first remove the sample and invert to mix, taking care to avoid bubbles. Should bubbles form, centrifuge the solution until they disappear, and invert again for a final mix. If the problem persists, hard reset the QuBit by unplugging the machine, waiting 30 s, then re-plugging.

Low product yield

A sufficient DNA yield for whole genome sequencing is typically 1.5 μg or below, whereas a sufficient RNA yield for bulk RNA sequencing is 2.5 μg or below. None of the samples we analyzed showed insufficient RNA quantity, although a few rare samples resulted in insufficient DNA yield. Should two vials of cryopreserved iPSCs be insufficient, it is recommended to thaw and re-expand the failed line using any iPSC expansion protocol, rather than to re-attempt extraction from additional frozen stocks (iPSC Protocols, 2019). In our hands, 46 lines out of 728 analyzed in this protocol which had previously shown insufficient DNA yield were expanded and re-extracted. All these lines subsequently showed substantial increases in yield that were sufficient for whole-genome-sequencing (Table 4).

Table 4:

Lines re-expanded and re-extracted from low-yield original extractions.

iPSC line	1st extraction (low yield) DNA (μg)	2nd extraction (higher yield) DNA (μg)	Percent Increase (%)
JZ_DNA_0954	1.01	3.492	245.7%
JZ_DNA_0955	0.863	3.065	255.2%
JZ_DNA_0956	0.565	3.873	585.5%
JZ_DNA_0957	0.819	6.983	752.6%
JZ_DNA_0958	0.364	2.722	647.8%
JZ_DNA_0959	1.587	8.917	461.9%
JZ_DNA_0960	1.05	3.337	217.8%
JZ_DNA_0961	0.62	3.31	433.9%
JZ_DNA_0962	0.903	3.535	291.5%
JZ_DNA_0963	0.892	2.438	173.3%
JZ_DNA_0964	1.12	3.793	238.7%
JZ_DNA_0965	1.207	3.713	207.6%
JZ_DNA_0966	0.995	4.933	395.8%
JZ_DNA_0967	1.05	2.3	119.0%
JZ_DNA_0968	0.382	8.567	2142.7%
JZ_DNA_0969	0.596	3.748	528.9%
JZ_DNA_0970	0.417	6.45	1446.8%
JZ_DNA_0971	1.21	4.043	234.1%
JZ_DNA_0972	1.213	2.157	77.8%
JZ_DNA_0973	1.053	3.807	261.5%
JZ_DNA_0974	0.922	4.118	346.6%
JZ_DNA_0975	1.112	6.5	484.5%
JZ_DNA_0976	1.06	3.468	227.2%
JZ_DNA_0977	1.047	4.728	351.6%
JZ_DNA_0978	1.01	3.582	254.7%
JZ_DNA_0980	1.012	3.365	232.5%
JZ_DNA_0981	0.842	3.947	368.8%
JZ_DNA_0982	1.067	2.275	113.2%
JZ_DNA_0983	0.76	2.915	283.6%
JZ_DNA_0984	1.213	7.883	549.9%
JZ_DNA_0985	1.08	4.943	357.7%
JZ_DNA_0986	0.897	2.13	137.5%
JZ_DNA_0987	1.132	2.37	109.4%
JZ_DNA_0988	0.673	3.315	392.6%
JZ_DNA_0989	1.127	1.75	55.3%
JZ_DNA_0990	1.012	6.067	499.5%
JZ_DNA_0991	1.033	8.35	708.3%
JZ_DNA_0993	0.727	3.507	382.4%
JZ_DNA_0994	0.67	4.793	615.4%
JZ_DNA_0995	1.107	4.033	264.3%
JZ_DNA_0996	1.168	4.895	319.1%
JZ_DNA_0997	0.943	6.7	610.5%
JZ_DNA_0998	0.658	6.65	910.6%
JZ_DNA_0999	1.228	9.15	645.1%
JZ_DNA_1000	1.147	4.118	259.0%

Very high/inaccurate cell counts

Cell counts may be inaccurate if poor-quality or old trypan-blue dye was used. Expected iPSC counts from two frozen vials are between 2-6 million cells, assuming an 80% confluency on 2 wells of a 6-well plate. Counts that average above 10 million cells per vial should be re-measured using new trypan blue dye as they are likely inaccurate. This is especially likely if large clumps of stained cellular aggregates are observed in the cell counter. Inaccurate cell counts may not diminish extraction quality except in the rare case where the true cell count exceeds the capacity of the solid-phase extraction columns, but may negatively affect observed correlations between cell count and extraction yield.

UNDERSTANDING RESULTS:

Discrepancy between DNA and RNA correlations to cell count

We observed a robust linear relationship between RNA yield and total cell count in both overall (Figure 5A) and live cell (Figure 5B) yield, but the correlations between cell counts and DNA yield were much weaker both across all extracted samples (Figure 1) as well as across the subset of common samples whose RNA yield was quantified (Figure S2 and S3). Indeed, the DNA yield to cell count R² values for both the RNA-quantified subset and the total samples are similar, which is unlikely to be due to experimental or sampling error, but may instead suggest differences in DNA and RNA yield using the commercial solid-phase extraction kit employed here. In theory, a strong positive correlation might be expected for both RNA and DNA yields to cell count. However, this was not the case, and additional data suggest it is not a result of cryo-storage age, as there is no correlation between DNA/RNA yield and time in liquid nitrogen (Figure 4A, 7A). Moreover, only a moderate correlation in yield was found between DNA and RNA extracted simultaneously from the same lines (Figure 7B, 7C), suggesting the commercial solid-phase columns for DNA and RNA extraction have different efficiencies.

It is possible that the necessity to preserve both RNA and protein while collecting DNA without protease or RNase treatment may lead to a decrease DNA collection efficiency and consistency. We observed a negative correlation between cell counts and both RNA yield per million cells (Figure 6) as well as DNA yield per million to cell count (Figure S4, S5), suggesting the solid-phase extraction kit has a lower efficiency as the input cell count approaches column capacity.

The effect of cryostorage time on DNA/RNA yield

We found no correlation between cryostorage time under liquid nitrogen and RNA or DNA yield (Figure 4B, 7A). In addition, there is no correlation in cryo-storage length and cell viability (Figure 8), corroborating that standard iPSC culturing and cryopreservation protocols are efficient in preserving cellular and molecular integrity. Future work may continue to optimize culturing/freezing protocols to reduce variations in cell viability (Figure 8).

Figure 8: Correlation between cryostorage time and cell viability.

Linear regression plot of cryostorage time to cell viability. Standard errors were found using LINEST function in Excel, and are as follows: slope 1.5*10⁻⁵, y-intercept 0.0083, y-value 0.13.

TIME CONSIDERATIONS:

Estimated time of completion depends on number of samples extracted. One sample requires 1 hour, 12 samples need 6 hours, and 24 samples take around 9 hours. Around 1-3 hours in addition should be expected for completion of DNA and RNA quantitation and quality control, depending on number of samples extracted.

Table 2:

Volume of RNA to be added as recommended by RNA concentration. Values based on those recommended in the RNA IQ Assay Kit handbook (Qubit™ RNA IQ Assay Kits, 2017).

RNA concentration range (ng/μL)	RNA Aliquot to be added
500 – 1500	1.0 μL
300 – 500	2.0 μL
<300	4.0 μL