Described here is the use of a methylation-specific probe amplification method to analyze methylation levels of LINE-1 elements in mesenchymal stem cells treated with osteosarcoma-derived extracellular vesicles. Ultracentrifugation, a popular procedure for separating extracellular vesicles from fetal bovine serum, is also demonstrated.
Methylation-specific probe amplification (MSPA) is a simple and robust technique that can be used to detect relative differences in methylation levels of DNA samples. It is resourceful, requires small amounts of DNA, and takes around 4–5 h of hands-on work. In the presented technique, DNA samples are first denatured then hybridized to probes that target DNA at either methylated or reference sites as a control. Hybridized DNA is separated into parallel reactions, one undergoing only ligation and the other undergoing ligation followed by HhaI-mediated digestion at unmethylated GCGC sequences. The resultant DNA fragments are amplified by PCR and separated by capillary electrophoresis. Methylated GCGC sites are not digested by HhaI and produce peak signals, while unmethylated GCGC sites are digested and no peak signals are generated. Comparing the control-normalized peaks of digested and undigested versions of each sample provides the methylation dosage ratio of a DNA sample. Here, MSPA is used to detect the effects of osteosarcoma-derived extracellular vesicles (EVs) on the methylation status of long interspersed nuclear element-1 (LINE-1) in mesenchymal stem cells. LINE-1s are repetitive DNA elements that typically undergo hypomethylation in cancer and, in this capacity, may serve as a biomarker. Ultracentrifugation is also used as a cost-effective method to separate extracellular vesicles from biological fluids (i.e., when preparing EV-depleted fetal bovine serum [FBS] and isolating EVs from osteosarcoma conditioned media [differential centrifugation]). For methylation analysis, custom LINE-1 probes are designed to target three methylation sites in the LINE-1 promoter sequence and seven control sites. This protocol demonstrates the use of MSPA for LINE-1 methylation analysis and describes the preparation of EV-depleted FBS by ultracentrifugation.
DNA methylation is a major epigenetic modification occurring in human cells. DNA methylation refers to the linkage of methyl groups to cytosine residues in CpG dinucleotides. Such dinucleotides are usually found in clusters (CpG islands) at the 5' region of genes1. In normal cells, most of these dinucleotides exist in an unmethylated state, which allows DNA transcription. Incidentally, many cancers are associated with hypermethylated CpG islands and transcriptomic silencing2, especially in tumor suppressor genes, which in turn contribute to various hallmarks of cancer3.
On the other hand, long interspersed nuclear elements-1 (LINE-1s or L1s) are repetitive, transposable DNA elements that normally have high levels of methylation at CpG islands. Methylation of LINE-1 prevents translocation and helps maintain genome integrity. In several types of cancer, LINE-1 is hypomethylated, resulting in activation and subsequent retrotransposition-mediated chromosomal instability4. LINE-1 accounts for nearly 17% of the human genome5, and its methylation status may serve as an indicator of global genomic methylation levels6. Global LINE-1 hypomethylation is considered to precede the transition of cells to a tumor phenotype7; therefore, it holds promise as a potential marker for early cancer onset.
Currently, there are several methods for methylation analysis, including pyrosequencing, methylation-specific PCR, microarrays, and chromatin immunoprecipitation1. The use of next-generation sequencing has also made it possible to incorporate genome-wide approaches to detection of DNA methylation. Many of these methods rely on bisulfite-treated DNA, in which unmethylated cytosines are converted to uracil and methylated cytosines remain unchanged. However, working with bisulfite-treated DNA has several pitfalls, such as incomplete conversions of unmethylated cytosines to uracil, biased amplification of sequences, and sequencing errors8.
In methylation-specific probe amplification (MSPA), probes composed of two oligonucleotides target DNA sequences containing a restriction site (GCGC) for the methylation-sensitive restriction enzyme HhaI9. After the probes hybridize to DNA, each sample is divided into two sets. Probes in the first set undergo ligation, while probes in the second set undergo ligation followed by HhaI-mediated digestion at unmethylated CGCG sites. Both sets of samples are then amplified by PCR, and the products are separated by capillary electrophoresis. Probes at unmethylated sites are digested by HhaI and are not amplified during PCR, resulting in no peak signals. By contrast, probes at methylated sites are protected from digestion and are therefore amplified during PCR, subsequently generating peak signals10.
MSPA has several advantages over alternative methods. First, it requires a low amount of DNA (50–100 ng) and is well-suited for analysis of DNA from formalin-fixed paraffin embedded samples10. It does not require bisulfite-treated DNA; in fact, it is unsuitable for DNA that is modified in this way. Many samples can be analyzed at the same time, and MSPA probes can be designed such that they target multiple genes or sequences simultaneously. Additionally, the probes are specific and sensitive for methylated DNA as the HhaI restriction site corresponds to a sequence that is typical of CpG islands10.
This study investigated the effects of osteosarcoma (OS)-derived extracellular vesicles (EVs) on LINE-1 methylation in adipose tissue-derived mesenchymal stem cells (AT-MSCs; Figure 1). EVs are nanoscale, membrane-bound vesicles secreted by most cell types. They carry proteins, lipids, mRNA, microRNA, and additional molecules from parent cells11,12. EVs mediate intercellular communication and play important roles in several pathophysiological conditions13,14. A recent study showed that cancer-derived EVs may transfer active LINE-1 to recipient cells15. It has been reported earlier that EVs from the HOS-143B cell line can alter the methylation status of LINE-1 in MSCs, in addition to other genetic effects16.
When growing cells for EV isolation, it is important to use EV-depleted fetal bovine serum [FBS] in the growth medium, since FBS-derived EVs may interfere with EVs from other sources and hamper the results17,18. Ultracentrifugation is one of the most common methods for depleting EVs from FBS. It is a relatively simple and cost-effective procedure compared to alternatives such as ultrafiltration and commercial EV-depleted FBS19. Here, the protocol also demonstrates how to prepare EV-depleted FBS by ultracentrifugation.
This article presents a detailed protocol for the aforementioned techniques, from isolation of EVs from an OS cell line to the methylation analysis of LINE-1 in OS-EV treated MSCs (Figure 1).
This study was approved by the Ethics Committee of Helsinki and Uusimaa Hospital District (ethical approval D. No. 217/13/03/02/2015).
1. Preparation of EV-depleted FBS by ultracentrifugation
2. Isolation of osteosarcoma-derived EVs
3. Characterization of OS-EVs
NOTE: Purified EVs can be characterized by western blotting (WB), nanoparticle tracking analysis (NTA), and transmission electron microscopy (TEM)16.
4. AT-MSC culture
NOTE: Human adipose tissue for mesenchymal stem cell isolation was provided as liposuction aspirate (Department of Plastic Surgery, Laser Tilkka Ltd., Finland). Written informed consent was taken from the lipoaspirate donors, who were undergoing elective liposuction procedures.
5. Treatment of MSCs with OS-EVs
6. LINE-1 methylation assay
7. Fragment data analysis
The main goal of this study was to evaluate the epigenetic effects of OS-EVs on MSCs. OS-EVs were isolated from HOS-143B cells using the standard differential centrifugation method. Expression of the typical EV markers CD63, Hsp70, and TSG101 by western blotting confirmed the presence of OS-EVs. (Figure 2A). Absence of calnexin signal indicated purity of the OS-EV isolate. Additional indication of purity was observed with TEM, with intact vesicles of various sizes being present (Figure 2B). The average OS-EV particle concentration was 7.63×1011/mL (Figure 2C). The size distribution of EVs ranged from 50–500 nm, with around 80% of particles falling within 50–200 nm range (Figure 2D).
AT-MSCs were treated with OS-EVs and DNA was extracted from the MSCs at different timepoints. LINE-1 methylation was analyzed by MS-MLPA and calculated in terms of methylation dosage ratio. Results from TP 0 were used to determine the baseline methylation levels (dashed line) (Figure 3). At TP 3 (green columns), a decrease in average LINE-1 methylation dosage ratio was observed in MSCs when treated with OS-EVs, falling under baseline methylation levels. This hypomethylating phenomenon was more subtle at TP 7 (blue columns), and the average methylation dosage ratio was slightly higher in both untreated and EV-treated MSCs compared to the baseline.
Figure 1: Experimental workflow. EVs were isolated from HOS-143B cells by differential centrifugation and characterized by western blotting, TEM and NTA. AT-MSCs were treated with OS-EVs at different time points (TP 0, 3, an 7). DNA was extracted from MSCs and LINE-1 methylation was analyzed by MSPA. Please click here to view a larger version of this figure.
Figure 2: Characterization of EVs. (A) The presence of EV markers CD63, Hsp70 and TSG101 confirmed the presence of EVs by western blotting, while no bands were observed for calnexin, indicating the purity of the EV isolate. 10 µg of protein was loaded for both HOS-143B protein lysate and OS-EVs. (B) TEM confirmed the presence of intact EVs of various sizes in the isolate. (C) Particle concentration and (D) size distribution of OS-EVs were determined by NTA measurements. Please click here to view a larger version of this figure.
Figure 3: Methylation dosage ratios of LINE-1 from MSCs either untreated or treated with OS-EVs. The dashed grey line represents baseline methylation value, corresponding to TP 0 values. Green columns represent average methylation dosage ratios without and with OS-EV treatment for TP 3 samples, while blue columns represent the same for TP 7 samples. Both TP 3 and TP 7 samples showed a decrease in methylation levels after treatment with OS-EVs. However, the difference was greater in TP 3 samples, where the methylation level after EV treatment was also lower than the baseline value. Please click here to view a larger version of this figure.
MSPA reagent mix recipes | Thermocycler program | |||||
No. | Mix | Reagent | Volume (1x, µl) |
Temperature (°C) |
Time | |
1 | Probe hybridization mix |
Reference probe-mix (black cap) | 1.0 | 95 | 60 s | |
Custom LINE-1 probe-mix | 0.5 | 60 | 18 h | |||
Hybridization buffer (yellow cap) | 1.5 | 8 | hold | |||
Total | 3.0 | |||||
2 | Post- hybridization mix |
Ligase buffer A (black-ringed cap) | 3.0 | 48 | hold | |
H2O | 10.0 | |||||
Total | 13.0 | |||||
3 | Ligation mix | Ligase buffer B (white cap) | 1.50 | |||
H2O | 8.25 | |||||
Ligase enzyme (green cap) | 0.25 | 48 | 50 min | |||
Total | 10.00 | 98 | 5 min | |||
4 | Ligation-Digestion mix |
Ligase buffer B (white cap) | 1.50 | 20 | hold | |
H2O | 7.75 | |||||
Ligase enzyme (green cap) | 0.25 | |||||
HhaI enzyme | 0.50 | |||||
Total | 10.00 | |||||
5 | Polymerase mix |
PCR primer mix (brown cap) | 1.00 | 72 | 2 min | |
H2O | 3.75 | 95 | 30 s | |||
Polymerase enzyme (orange cap) | 0.25 | 60 | 30 s | |||
Total | 5.00 | 72 | 60 s | |||
72 | 20 min | |||||
8 | hold |
Table 1: MSPA reagent mix recipes and thermocycler/PCR programs. The volumes mentioned represent one DNA sample. It should be noted that the polymerase mix should be prepared for 2x as many samples as previous mixes, since there are two sets of tubes at this stage. Details of the subsequent thermocycling or PCR program are provided to the right.
This study illustrates how MSPA can be used to detect and quantify the methylation status of a specific genetic element. LINE-1 was the focus here, but the probes can be designed to target a range of genes and sequences. Moreover, there is a growing list of probe mixes available for different applications. MSPA is a simple and robust technique for DNA methylation analysis that does not require bisulfite conversion10. The complete procedure from sample preparation to data analysis takes around 2 days but only involves 4–5 h of actual hands-on work. It is applicable for small amounts of DNA (as low as 70 ng), as demonstrated here.
The most important part of the methylation analysis protocol is the preparation of custom probe-mixes, such as the LINE-1 probe-mix in this study. Due to their different lengths, LINE-1 probe oligonucleotides were synthesized in the range of 4–40 nM, so the dissolution step and subsequent dilutions had to be performed carefully, as per the manufacturer's instructions25. The PCR and several thermocycling programs of the protocol as used here differ from those in the original manufacturer's version.
There are a few precautions relating to the HhaI enzyme. First, the volume of enzyme used in the ligation-digestion mix depends on the manufacturer, and versions of the enzyme that are resistant to heat-inactivation are not suitable for MSPA. With some enzymes, there may be instances of incomplete digestion, which would result in the formation of faulty PCR products and aberrant peak signals. Lastly, the extent to which the final PCR products are diluted for capillary electrophoresis depend on the probes. Initial runs are likely to involve optimizations, for which it is recommended to test a range of dilutions.
There are certain limitations of MSPA, such as the selective nature of HhaI-mediated DNA digestion. HhaI cleaves DNA only at unmethylated GCGC sequences and disregards other instances of CpG dinucleotides that are not enclosed by a G and C, even if they are located within CpG islands. As a result, the methylation status of such dinucleotides (and those that are located outside the target sequence of the probes) cannot be determined. The LINE-1 probe-mix can include more probes containing additional GCGC sequences from the promoter region24, thereby representing a greater number of methylation sites.
Alternatively, other methylation-specific restriction enzymes, such as SacII and MluI, that have a different restriction site than that of HhaI can be additionally used in MSPA, which may provide a broader representation of global methylation levels26. Secondly, as observed with TP 7 samples, the differences in methylation dosage ratio can be quite subtle. This may be due to the high copy number of LINE-1s in the genome5, which have high levels of global methylation under normal conditions and will largely be unaffected by transient hypomethylation events in only a few CpG sites. Moreover, the MSCs are heterogenous in terms of the stage of differentiation and of the cell cycle, which can affect the baseline methylation value. Lastly, MSPA can only detect relative differences in DNA methylation and requires reference samples for this reason. In such instances, methods that directly detect local methylation may be more appropriate, though they may require the bisulfite conversion step27.
Since this study involved the use of extracellular vesicles, we also demonstrated the preparation of EV-depleted FBS, which is an essential component of cell culture media for EV isolation. Ultracentrifugation is an inexpensive process that can effectively separate EVs from the rest of the FBS. However, the 19 h of centrifugation makes it a more time-consuming method than alternatives, such as ultrafiltration and commercial versions19. Ultracentrifugation also requires careful handling of the sample, such as when balancing the tubes before centrifugation and collecting the supernatant afterwards. Despite these challenges, it is a popular method for separating EVs from biological fluids.
The authors have nothing to disclose.
This work was funded by University of Helsinki project funding (WBS490302, WBS73714112) Helsinki University Hospital State funding for university-level health research (Y1014SUL05, TYH2016130), Finnish-Norwegian Medical Foundation, and the Selma and Maja-Lisa Selander Fund (Minerva Foundation). We thank Walter Pavicic for providing the modified MSPA protocol and for related technical support. We are grateful to Teemu Masalin (University of Helsinki) for helping us with video production.
1 mL syringe | Terumo | SS+01T1 | for NTA |
24-well plate | Corning | 3524 | MSC cell culture |
3730xl DNA Analyzer | Applied Biosystems, ThermoFisher Scientific | 3730XL | |
50 mL centrifuge tube | Corning | 430829 | |
Beckman Optima LE-80K Ultracentrifuge | Beckman | ||
BlueStar Prestained Protein Marker | Nippon Genetics | MWP03 | WB: protein marker |
Calnexin (clone C5C9) | Cell Signaling Technology | 2679 | WB, dilution 1:800 |
CD63 (clone H5C6) | BD Biosciences | 556019 | WB, dilution 1:1000 |
Centrifuge 5702 R | Eppendorf | 5703000010 | For conditioned media and cells |
Centrifuge 5810 | Eppendorf | 5810000010 | For spinning down 96-well plate |
Centrifuge tube (polyallomer, 14×95 mm) | Beckman | 331374 | Ultracentrifugation |
DMEM/F-12 + GlutaMAX medium | Gibco, Life Technologies | 31331-028 | For AT-MSC culture |
Fetal bovine serum | Gibco, Life Technologies | 10270-106 | |
GeneScan 500 LIZ size standard | Applied Biosystems, Life Technologies | 4322682 | for capillary electrophoresis |
GenomePlex Complete Whole Genome Amplification (WGA) Kit | Sigma | WGA2-10RXN | for MSPA negative control |
Hi-Di formamide | Applied Biosystems, Life Technologies | 4311320 | for capillary electrophoresis |
HOS-143B cell line | ATCC | CRL-8303 | |
Hsp70 (clone 5G10) | BD Biosciences | 554243 | WB, dilution 1:1000 |
IRDye 800CW Goat anti-mouse | Li-Cor | 926-32210 | WB: secondary |
IRDye 800CW Goat anti-rabbit | Li-Cor | 926-32211 | WB: secondary |
LINE-1 probe-mix primers | IDT | Sequences in Table 1 | |
MicroAmp Optical 96-well reaction plate with barcode | Applied Biosystems, Life Technologies | 4306737 | also requires sealing film |
Micro BCA Protein Assay kit | ThermoFisher Scientific | 23235 | measure protein concentration |
MiniProtean TGX 10% gels | Bio-Rad | 456-1034 | WB: gel electrophoresis |
NanoSight LM14C | Malvern Instruments | for NTA | |
Nitrocellulose membrane 0.2 µm | Bio-Rad | 1620112 | WB: protein transfer |
NucleoSpin Tissue XS | Macherey-Nagel | 740901.50 | for DNA extraction |
Odyssey Blocking Buffer | Li-Cor | 927-40000 | WB: blocking, antibodies |
PBS, 1X | Corning | 21-040-CVR | |
Penicillin-streptomycin | Gibco, Life Technologies | DE17-602E | Antibiotics for culture media |
Protein LoBind tube, 0.5 mL | Eppendorf | 22431064 | For storing Evs |
REVERT Total Protein Stain and Wash Solution Kit | Li-Cor | 926-11015 | WB: total protein staining |
RKO cell line | ATCC | CRL-2577 | for MSPA positive control |
RPMI medium 1640 + GlutaMAX | Gibco, Life Technologies | 61870-010 | For HOS-143B cell culture |
SALSA MLPA HhaI enzyme | MRC-Holland | SMR50 | |
SALSA MLPA reagent kit | MRC-Holland | EK1-FAM | |
SALSA MLPA P300 probe-mix | MRC-Holland | P300-100R | |
Swinging rotor SW-28 | Beckman Coulter | 342207 | Ultracentrifugation |
Syringe filter, 0.22 µm | Jet Biofil | FPE-204-030 | sterile filtering FBS |
Tecnai 12 | FEI Company | equipped with Gatan Orius SC 1000B CCD-camera (Gatan Inc., USA); for TEM |
|
TBS, 1X tablets | Medicago | 09-7500-100 | WB: buffer |
Trans-Blot Turbo | Bio-Rad | WB: transfer | |
Thermal cycler | ThermoFisher Scientific | TCA0096 | |
TrypLE Express | Gibco | 12604-021 | for trypsinization of cells |
TSG101 (clone 4A10) | Sigma | SAB2702167 | WB, dilution 1:500 |