A. Overview: The sequence coding for the HTT polyglutamine and polyproline tracts is amplified from genomic DNA using MiSeq-compatible PCR primers (i.e. PCR primers including HTT locus-specific primers and the complete sequence of a barcoded Illumina adapter – "Figure 1":https://www.nature.com/protocolexchange/system/uploads/6459/original/LibraryPreparationByPCRforHTT.png?1530611667). This makes the PCR product obtained directly sequenceable. Post-PCR, a fraction of each PCR product are pooled together and cleaned-up using AMPure XP beads. This PCR clean-up step also allows removal of primer dimers. The sequencing library is then quality controlled using Qubit, Bioanalyzer and qPCR and sequenced on the MiSeq platform.
B. MiSeq-compatible PCR primer design
The MiSeq-compatible PCR primers were designed based on the structure of the TruSeq combinatorial dual (CD) index adapters (p23 "Supplementary document 2":https://www.nature.com/protocolexchange/system/uploads/6661/original/SupplementaryDocument2-illumina-adapter-sequences-Feb2018.pdf?1530635414) with the addition of spacers between the sequencing primer binding site and the locus-specific primer (Figure 1). These spacers are used to increase nucleotide diversity at the beginning of the sequencing reads. Spacers on the P5 primer enable cluster detection and validation on the MiSeq flow cell which is not possible without enough nucleotide diversity. This is required for the smooth sequencing of amplicon libraries prepared with the same locus-specific primer pair [e.g. 17]. Spacers on the P7 primer should improve sequencing quality .
The HTT locus-specific primers used are HS319F (5’-GCGACCCTGGAAAAGCTGATGA-3’) and 33935.5 (5’-AGCAGCGGCTGTGCCTGC-3’) which respectively bind 26 bp 5’ upstream of the CAG repeat and 26 bp 3’ downstream of the CCG repeat.
The protocol has been validated using the two following i5 and i7 index sets:
The TruSeq CD set of primers (Table 1) corresponds to an earlier version of the primer design that only allowed the sequencing of up to 96 samples per MiSeq run and did not include spacers on the P7 primers. The Nextera XT Kit v2 set of primers corresponds to the latest version of the primer design for which the following criteria were used in the spacer design:
• Spacers on the P5 primer (0 to 7 bases) were manually designed to maximise nucleotide diversity within the first 25 cycles of sequencing. We have limited ourselves to using seven base spacers on the i5 primer in order not to use too much read length.
• Spacers on the P7 primer (0 to 2 bases) were manually designed to stagger the sequencing within the trinucleotide part of the amplicons.
• Spacers were associated with i5 and i7 indexes to minimise differences in melting temperature (Tm) between MiSeq-compatible primer pairs (spacers with higher Tm with indexes with low Tm, spacers with average Tm with indexes with average Tm, spacers with low Tm with indexes with higher Tm). This was done to minimise differences in PCR yield between primer pairs because they are all used at the same annealing temperature during PCR (see PCR conditions below) although they have slightly different Tm.
C. Sequencing library preparation by a single PCR amplification
PCR amplifications are performed in 15 µl reactions containing 20 ng of genomic DNA (diluted in nuclease-free water), 10% (v/v) DMSO, 0.666 µM of each MiSeq-compatible PCR primer, 1X ‘Custom PCR Master Mix + βME’ and 1 U of Taq DNA polymerase. PCRs are manually prepared in 96-well plates using multichannel pipettes (8 i5 indexes x 12 i7 index = 96 index pairs for each plate. 96 index pairs in total for the TruSeq CD index set and 384 index pairs in total for the Nextera XT Kit v2 index sets) as illustrated in "figure 2":https://www.nature.com/protocolexchange/system/uploads/6667/original/Figure2_indexesInPCRplates.png?1530612191.
A PCR plate is prepared for 94 DNA samples of interest and two controls (for each plate, these two controls are: one no-template controls (NTC) and one HD positive control, i.e. a DNA sample from an individual carrying an HD-causing allele of known size)
For each 96-well plate the PCRs are set up as follows:
1) Add 2 µl of a working solution (5 µM) of each primer to each well of the PCR plate using an 8-channel p10 pipette for the P5 primers and a 12-channel p10 pipette for the P7 primers (according to the plate maps shown in figure 2)
2) Add 5 µl of template DNA (solution at 4 ng/µl) from each sample of interest and from the positive control to each well of the 96 well plate
3) Add 5 µl of nuclease-free water to the well of the NTC
4) Prepare the PCR master mix for the 96 reactions (volume of each reagent for one tube: 2.8 µl of nuclease-free water, 1.5 µl of 10X ‘Custom PCR Master Mix + βME’*, 1.5 µl of DMSO, 0.2 µl of Taq polymerase at 5 units/µl)
- The concentration of βME in the 10X ‘Custom PCR Master Mix + βME’ is 0.48% (v/v). This leads to a final βME concentration of 0.048% (v/v) in the 15 µl PCR (as indicated in section Reagents/PCR)
5) Add 6 µl of this PCR master mix to each well of the plate
6) Seal the PCR plate
7) Place the PCR plate in a PCR machine and start the PCR. PCR cycling conditions are: (96°C, 5 min)
28 cycles of (96°C, 45 s), (58.5°C, 45 s) and (72°C, 3 min)
(72°C, 10 min)
D. Control of the PCR amplification
To make sure that PCR amplification has occurred on each plate, 1 µl of each positive and NTC PCR products is collected after PCR. A 1:10 dilution of the positive control PCR products is prepared from the 1 µl collected and the DNA concentration of these dilutions is then measured using a Qubit fluorometer using the Qubit® dsDNA HS Assay Kit (Life Technologies) following the manufacturer’s instructions. The dilutions of the positive control PCR products are then adjusted to 0.5 ng/µl. That dilution of the positive controls and the neat NTCs are then checked by capillary electrophoresis on a Bioanalyzer (Figure 3).
Considering the full length of the MiSeq-compatible primers with Nextera XT Index Kit v2 indexes (Table 2), the expected size of the PCR products for alleles of structure (CAG)Q1CAACAGCCGCCA(CCG)7(CCT)2 is 305 to 314 bp if Q1 = 15, 350 to 359 bp if Q1 = 30, 395 to 404 bp if Q1 = 45 and 440 to 449 bp if Q1 = 60. The Bioanalyzer concentration of the PCR products (300 to 450 bp) for the HD control should be between 0.2 and 1.5 ng/µl.
E. AMPure XP size selection and PCR clean-up
Note: volumes in the section below correspond to 384 PCR products. These volumes would have to be adjusted if processing fewer PCR products.
After PCR amplification and the absence of contamination of the NTCs have been confirmed, 5 µl of each of the 384 products (including NTCs) are used for downstream pooling and purification, which leaves approximately 10 µl of the PCR product available in the PCR plates. These 10 µl, remaining in the PCR plates, are stored at 4°C as a back-up. Five µl of each of the PCR products are pooled and mixed together (total volume of the pool for 384 samples 1900 µl). The 1900 µl pool is distributed in six DNA LoBind 1.5 ml tubes (300 µl per tube). From that step, DNA LoBind 1.5 ml Tubes are used. Each of the 300 µl aliquots is then purified with AMPure XP beads using a Dyna Mag-2 magnetic stand as follows:
- The temperature of the beads and of the pool of HTT PCR products are allowed to stabilise at room temperature
- All the mixing is performed by gently flushing solutions up and down 10 times. No vortexing, no centrifugation
E.1. 0.6X AMPure XP clean-up to get rid of the primer dimers
This first AMPure XP purification has two aims: (i) to get rid of the primer dimers (size selection) and (ii) to concentrate the sequencing library.
At this stage, there are six aliquots of PCR products in six 1.5 ml tubes.
- E.1.1. Accurately measure the volume of each aliquot (this is better done directly by aliquoting a known volume of aliquot). The following protocol assumes each aliquot is 300 µl at this stage. Overestimation of the volume of the aliquot of PCR product can lead to a reduction of efficiency in the size selection against smaller fragments during clean-up which can, in turn, result in the carryover of primer dimers
- E.1.2. Vortex AMPure XP beads to resuspend
- E.1.3. In each of the 1.5 ml DNA LoBind tubes containing the aliquots of HTT PCR products, add 0.6 µl of resuspended AMPure XP beads (180 µl) for each µl of HTT PCR product
- E.1.4. Mix well by gently pipetting up and down 10 times
- E.1.5. Incubate 5 min at room temperature
- E.1.6. Place the tubes on the magnetic stand. Loosen the tube cap (without opening it) to allow easy opening of the cap in subsequent steps (this will avoid disturbing the beads when opening the tube on the magnetic stand)
- E.1.7. Leave the tubes on the stand for 8 minutes
- E.1.8. Carefully remove the supernatant and transfer it to another tube (as a backup in case something goes wrong). Be careful not to disturb the beads that contain the DNA targets
Leave the tubes on the magnetic stand while performing the following steps E.1.9. to E.1.13.:
- E.1.9. Add 500 µl of freshly prepared 80% ethanol to each tube without disturbing the beads (the beads must be completely covered by the ethanol. Use more than 500 µl of 80% ethanol if this is not the case)
- E.1.10. Incubate at room temperature for 30 seconds
- E.1.11. Carefully remove and discard the supernatant
- E.1.12. Repeat steps E.1.9. to E.1.11. once
- E.1.13. Leave the tubes on the magnetic stand with the lid open and air-dry the beads for 5 minutes. During drying, collect excess ethanol with a small tip and a 10 µl pipette
Caution: Do not over dry the beads. This may result in lower recovery of DNA target. The surface of overdried beads looks cracked.
- E.1.14. Remove the tubes from the magnetic stand. Elute DNA target by adding 45 µl of nuclease free water to the beads. First, use the water to detach the beads from the side of the tube by pipetting the water onto the bead pellet several times. When all the beads and water are at the bottom of the tube, mix well by gently pipetting up and down 10 times
- E.1.15. Incubate the tubes at room temperature for 2 min out of the magnetic stand
- E.1.16. Place the tubes in the magnetic stand. Loosen the tube cap (without opening it) to allow easy opening of the cap in subsequent steps (this will avoid disturbing the beads when opening the tube on the magnetic stand)
- E.1.17. Incubate 5 min at room temperature
- E.1.18. Without disturbing the bead pellet which is on the side of the tubes, carefully collect 40 to 43 µl of supernatant, i.e. eluted DNA, from each tube
- E.1.19. Collect the supernatant from each of the six tubes and pool the six supernatants in a single new 1.5 ml DNA LoBind tube
- E.1.20. Thoroughly mix the pool of these six supernatants by gently pipetting up and down
E.2. 1.4X AMPure XP clean-up to concentrate the sequencing library
At this stage, the sequencing library is contained in a single 1.5 ml tube and the total volume of the sequencing library is 240 to 260 µl. Take 50% of the eluted DNA (about 130 µl) and put aside and store at 4°C as a backup. A second round of AMPure XP clean-up (using 1.4 µl of the bead solution for each µl of sequencing library to concentrate) is performed on the other 50% (about 130 µl) to increase the DNA concentration of the MiSeq library and further clean it up.
- E.2.1. Vortex AMPure XP beads to resuspend
- E.2.2. In the 1.5 ml DNA LoBind tube containing the HTT sequencing library (130 µl), add 182 µl (1.4X) of resuspended AMPure XP beads
- E.2.3. Mix well by gently pipetting up and down 10 times
- E.2.4. Incubate 5 min at room temperature
- E.2.5. Place the tube on the magnetic stand. Loosen the tube cap (without opening it) to allow easy opening of the cap in subsequent steps (this will avoid disturbing the beads when opening the tube on the magnetic stand)
- E.2.6. Leave the tube on the stand for 8 minutes
- E.2.7. Carefully remove the supernatant and transfer it to another tube (as a backup in case something goes wrong). Be careful not to disturb the beads that contain the DNA targets
Leave the tube on the magnetic stand while performing the following steps E.2.8 to E.2.12:
- E.2.8. Add 200 µl of freshly prepared 80% ethanol to the tube without disturbing the beads (the beads must be completely covered by the ethanol. Use more than 200 µl of 80% ethanol if this is not the case)
- E.2.9. Incubate at room temperature for 30 seconds
- E.2.10. Carefully remove and discard the supernatant
- E.2.11. Repeat steps E.2.8. to E.2.10. once
- E.2.12. Leave the tube on the magnetic stand with the lid open and air-dry beads for 5 minutes. During drying, collect excess ethanol with a small tip and a 10 µl pipette
Caution: Do not over dry the beads. This may result in lower recovery of DNA target. The surface of overdried beads looks cracked.
- E.2.13. Remove the tube from the magnetic stand. Elute the target DNA by adding 90 µl of nuclease free water to the beads. First, use the water to detach the beads from the side of the tube by pipetting the water and pouring it onto the bead pellet several times. When all the beads and water are at the bottom of the tube, mix well by gently pipetting up and down 10 times
- E.2.14. Incubate the tube at room temperature for 2 min out of the magnetic stand
- E.2.15. Place the tube in the magnetic stand. Loosen the tube cap (without opening it) to allow easy opening of the cap in subsequent steps (this will avoid disturbing the beads when opening the tube on the magnetic stand)
- E.2.16. Incubate 5 min at room temperature
- E.2.17. Without disturbing the bead pellet which is on the side of the tube, carefully collect 85 µl of the clear supernatant
Note: NO beads should be collected. It is safe to transfer a volume which is 5 µl smaller than the elution volume. If beads are collected when trying to collect 85 µl, replace the supernatant in the tube, put the tube back on the magnetic stand for 3 to 5 min and then collect less than 85 µl of clear supernatant in order not to collect any beads.
- E.2.18. Transfer the clear supernatant to a new 1.5 ml DNA LoBind tube
F. Quality control of the size selection and PCR clean-up
The DNA concentration of the purified library is then measured using a Qubit fluorometer using the Qubit® dsDNA HS Assay Kit following the manufacturer’s instructions. This quantification is used to prepare a dilution of the library that can be run on a Bioanalyzer. The sequencing library, diluted to 0.5 ng/µl, is then run on a Bioanalyzer to: (i) check the library is of the expected size; (ii) check the library is free of primer dimers; and (iii) estimate the average size of the library (Figure 4). Based on Bioanalyzer estimates, the aim is to have a library of at least 30 µl at 10nM.
Note: the above QC SHOULD NOT be used as a basis for loading the library on the MiSeq. We have ONLY been able to properly load our HTT libraries on MiSeq based on library quantification by qPCR.
G. Sequencing library quantification by qPCR
The quantity of sequenceable molecules in the sequencing library is then measured by qPCR using the KAPA Library Quantification Kits (KAPABIOSYSTEMS).
- G1. Using the concentration (in ng/µl) estimated by Qubit and the average fragment size estimated by the Bioanalyzer, a rough molarity is calculated. This rough molarity estimate is used to dilute the sequencing library to approximately 8 pM within the range of the qPCR standards.
- G2. The qPCR is performed as instructed in the Kapa Quantification user manual (using high ROX that is the dye recommended for our qPCR machine). All samples and standards are run in triplicate.
- G3. The concentration of the sequencing library is then calculated and corrected for size using the average fragment size estimated by the Bioanalyzer.
H. MiSeq Sequencing
The library is then sequenced following Illumina guidelines for an amplicon MiSeq run, using MiSeq Reagent Kit v3 with a cluster density of 1000k cluster / mm2 but with 5% PhiX spike-in (the PhiX library allows increasing nucleotide diversity during the run and also serves as a sequencing control). The sequencing library is then diluted to 4 nM (based on the qPCR molarity estimate) and 5% PhiX library are added. The mix “sequencing library + PhiX” is then denatured according to the Illumina protocol (MiSeq System, Denature and Dilute Libraries guide) and loaded at 12 pM onto the MiSeq. We used the MiSeq Control software version 2.5.05. If the samples correspond to individuals that are not affected by HD then the 600 cycles of sequencing can be used to produce two reads of 300 nt per cluster. However, if some samples correspond to individuals affected by HD or suspected to be affected, the 600 cycles of sequencing should be used to produce a forward read (i.e. Read 1) of 400 nt and a reverse (i.e. Read 2) of 200 nt. After the run (65 h) the MiSeq Reporter software (version 2.5.1) is used to demultiplex the reads corresponding to the index pair used (96 TruSeq CD or 384 Nextera XT Index Kit v2 index pairs). This demultiplexing is done using the default demultiplexing parameters of the MiSeq Reporter software, i.e. “clusters are assigned to a sample when the index sequence matches exactly or there is up to a single mismatch per index” (as indicated in the MiSeq Reporter Software Guide). The MiSeq Reporter software outputs the sequencing reads in .fastq files, two (Read 1 and Read 2) for each of the 96 or 384 indexes (5 to 120 MB per file, ~10 GB per run) as well as two (Read 1 and Read 2) undetermined reads (reads corresponding to the PhiX control library, which is not indexed, and reads for which the indexes could not be identified).