A novel approach reveals genomic landscapes of single-strand DNA breaks with nucleotide resolution in human cells

Abstract Single strand breaks (SSBs) represent the major form of DNA damage, yet no technique exists to map these lesions genome-wide with nucleotide-level precision. Herein, we present a method, termed SSiNGLe, and demonstrate its utility to explore the distribution and dynamic changes of genome-wide SSBs in response to different biological and environmental stimuli. We validate SSiNGLe using two very distinct sequencing techniques and apply it to derive global pro�les of SSBs in different biological states. Strikingly, we show that patterns of SSBs in the genome are non-random, speci�c to different biological states, enriched in regulatory elements, exons, introns, speci�c types of repeats and exhibit differential preference for the template strand between exons and introns. Furthermore, we show that breaks likely contribute to naturally occurring sequence variants. Finally, we demonstrate strong links between SSB patterns and age. Overall, SSiNGLe provides access to unexplored realm of cellular biology, not obtainable with current approaches.


Introduction
Introduction DNA damage is now widely recognized as a major reason behind cancer and many other agingassociated diseases and as such represents a very important issue for human health 1,2 .While multiple types of DNA lesions exist, SSBs are considered the most common type of DNA damage 3 .These lesions can represent sites of oxygen radical DNA damage, intermediates in excision DNA repair pathway and products of unresolved intermediates of enzymes such as topoisomerases 3 .SSBs can further deteriorate into highly toxic double-strand breaks (DSBs) by stalling or collapsing replication fork 4 .However, by themselves SSBs can also represent a major issue for cell as they can inhibit progression of RNA polymerase 5 and in some cases cause apoptosis 6,7 .The importance of this type of lesion is underscored by existence of dedicated cellular pathways that deal with every step of xing SSBs from detection to processing to repair 3 .Defects in these pathways can lead to cellular sensitivity to genotoxic stress, embryonic lethality and a number of neurodegenerative diseases 8 .
The remarkable progress in appreciation of the ne details of the SSB repair machinery however stands in stark contrast with total absence of methods to map endogenous SSBs in a global, unbiased and genome-wide fashion with nucleotide precision.This gap in available methodologies also contrasts with a suite of comprehensive approaches developed for mapping DSBs with nucleotide-level resolution, such as BLESS 9 , BLISS 10 and others (reviewed in Ref 11 ).To our knowledge, only one SSB genome-wide mapping method can provide comprehensive and unbiased data 12 .The procedure relies on the 3'OH group of an SSB to prime a DNA polymerase I nick-translation reaction that labels downstream DNA with a biotinylated nucleotide 12 .The labeled DNA is then puri ed and subjected to next-generation sequencing 12 .However, this approach maps a region of DNA, quite possibly extending thousands of bases from the original SSB, thus precluding identi cation of the lesion with nucleotide precision.On the other hand, a nucleotide-level method to map sites of excision repair has been developed 13 , however, it cannot provide information on breaks generated by other mechanisms.Thus, all this leads to a total dearth of knowledge of nucleotide-level genome-wide patterns for this critical type of DNA lesion.Here, we develop and validate an approach, SSiNGLe (single-strand break mapping at nucleotide genome level, Fig. 1) that can provide nucleotide level maps of native SSBs genome-wide.We implement this approach to work with two next-generation sequencing (NGS) platforms: (1) a 3 rd generation single molecule sequencing (SMS) platform (Helicos/SeqLL) whose unique capabilities obviate the need for lengthy sample preparation and PCR ampli cation, and (2) the more commonly used Illumina platform.We show highly consistent and striking patterns of SSBs using both sequencing platforms.Furthermore, the results show that the genomic pattern of breaks -the SSB "breakome" -has strong potential to represent a novel dimension describing state of a biological system and a novel source of blood-based biomarkers, with potentially yet undiscovered connections to aging.Note: this procedure has been used for cultured adherent and suspension cells as well as for Peripheral Blood Mononuclear Cells (PBMCs).The cultured cells can be grown using standard conditions appropriate for particular cells lines being assayed and processed as described below starting with the "Preparation of crosslinked nuclei" section.The procedure for preparation of PBMCs is given below.
Preparation of PBMCs (optional), 1 hour: Note: here and below the times in hours are approximate and given for 1-10 samples processed in parallel.
1. Collect 2-4 ml of peripheral blood in EDTA anti-coagulation tubes and mix well by gently inverting the tube several times.Keep on ice and process as soon as possible.Do not freeze.
2. Dilute the samples with equal volume of 1× PBS at room temperature.
3. Take several of the Solarbio Lymphocyte Separation Medium into a 15 ml centrifuge tube (add 3 ml when the volume of the diluted blood is less than 3 ml, add equal volume of Solarbio Lymphocyte Separation Medium when the volume of the diluted blood is more than 3 ml).
4. Gently layer the diluted blood on the top of Solarbio Lymphocyte Separation Medium.The layering should be done very slowly that diluted blood and Solarbio Lymphocyte Separation Medium should stay as two different layers.
Note: the total volume of diluted blood and Solarbio Lymphocyte Separation Medium shouldn't more than 10 ml, otherwise the separation will be affected.Make sure that the samples are at room temperature before proceeding to the next step.
5. Centrifuge the samples for 30 min at 1000g at 25°C in the 5804R centrifuge with acceleration and brake settings of 6 and 0 respectively.6. Carefully aspirate the PBMCs formed in the interphase into a new 15 ml tube.The cells in the interphase need to be aspirated without delay.
7. Add 10 ml of sterile 1× PBS into the tube with the aspirate, gently resuspend the aspirate and then centrifuge at 250g at 25°C for 10 min.
8. Remove the supernatant and repeat the last step.9.After the centrifugation, resuspend the aspirate in 1-2 ml of sterile 1× PBS, count the PBMCs while keeping cells on ice and proceed to the crosslinking immediately as described in the section below at step 8.

Preparation of crosslinked nuclei, 2.5 hours:
Keep samples on ice unless speci ed otherwise.Samples processed with high-speed refrigerated centrifuge below.
1.For non-adherent cells, proceed directly to step 5 of this procedure.
2. For adherent cells, remove the growth medium carefully and wash the cells twice with1× PBS at room temperature.
3. Clearly remove the 1× PBS and digest with trypsin until the cells detach.
4. Inactivate trypsin with the growth medium with serum.
5. Resuspend and count the cells.
6. Collect the cells, centrifuge to remove the growth medium with serum at 100 g for 5 min at 4°C. 7. Remove the supernatant, completely resuspend the cells in the fresh growth medium with serum.
8. Start with 1-3 million non-adherent or adherent cells or 4-6 million PBMCs resuspended in 1-2 ml of either the growth medium with serum or 1× PBS (PBMCs) in a new 2 ml Eppendorf tube.9. Add 27 µl of 37% formaldehyde per 1 ml of cell culture medium or PBS while slowly shaking the cells for 10 min at room temperature.The nal concentration of formaldehyde is 1%.10.Add 100 µl of 1.375 M glycine ( nal concentration 0.125 M) per milliliter to quench formaldehyde for 5 min at room temperature.Gently pipette with tip and invert several times with the tube to mix them well.
11. Centrifuge the cells at 220 g for 10 min at 4°C. 12. Remove the supernatant and wash the cells twice with 2 ml of cold 1× PBS at 220 g for 10 min at 4°C. 13.Carefully remove the supernatant and resuspend the cell pellet in 2 ml of cold cell lysis buffer.
Note: the pelleted cells would stick to each other after adding the cell lysis buffer, therefore pipette with up and down to separate them completely.Setting the volume to one third of the pipette range can effectively avoid the pellet loss.
15.After the lysis, centrifuge the nuclei at 300 g in a benchtop centrifuge for 10 min at 4°C. 16.Aspirate the supernatant carefully.17.Wash the pellet with 500 µl cold 1× MNase buffer, centrifuge for 10 min at 4°C at 300 g. 18. Repeat steps 16-17 and remove the supernatant.19.At this step, the nuclei can be stored at -80°C for up to one month.If this is desirable, resuspend the pellet by adding 100 µl 1× MNase buffer and move the samples directly into -80°C.When ready to proceed, thaw them on the ice around 5 min until completely thawed, centrifuge for 10 min at 4°C at 300 g, remove the supernatant and proceed to the sections below.If storage is not necessary, proceed with the nuclear pellet directly to the sections below.

SDS permeabilization (optional), 2 hours:
Note: this step is only necessary if nuclei are to be treated with enzymes that cannot easily diffuse into the nuclei (having molecular weight >30 kDa) as shown in the example below for the treatment with shrimp alkaline phosphatase (SAP) that can make breaks with 3'-phosphate termini detectable in the SSiNGLe procedure.If no such treatments are necessary then proceed directly to the section "MNase fragmentation and DNA puri cation".
1. Resuspend the pellet completely in 100 µl of 1× MNase buffer and transfer the sample to a 1.5 ml Eppendorf tube.
2. Add 3 µl of 10% SDS into the tube, gently pipette up and down while moving around the tip to mix the SDS well, the nal concentration of SDS is 0.3%.Slowly shake (300 rpm) at 37°C for 1h in the thermostatic metal bath.
Note: the nuclei would stick to the pipette tip at the following step easily, therefore use ultra-low attachment tip and pipette gently to avoid loss.
3. Slowly add 1.8 µl of Triton X-100 into the solution, gently pipette up and down while moving around the tip to mix Triton X-100 well, the nal concentration of Triton X-100 is 1.8%.Incubate for 5 min at room temperature to sequester SDS. 4. Centrifuge 15 min at 1500 g with the temperature setting of 27°C in a high-speed refrigerated centrifuge.
5. Aspirate the supernatant carefully and completely.Wash the pellet with 500 µl of 1× MNase buffer, centrifuge 15 min at 1500 g with the temperature setting of 27°C in a high-speed refrigerated centrifuge.Note: prolong the centrifugation time if complete precipitation is not achieved initially.6. Repeat step 5.
7. Aspirate the supernatant carefully and use the nuclei for the following step.
Treatment with SAP (optional), 3.5 hours: 1. Wash the permeabilized nuclei with 500 µl of 1× NEB buffer 2.1, gently resuspend the permeabilized nuclei and centrifuge at 1500 g with the temperature setting of 27°C in a high-speed refrigerated centrifuge for 10 min.
2. Carefully remove the supernatant and repeat step 1.
4. Add 1 µl SAP into the solution, treat the nuclei in 50 µl of 1× NEB buffer 2.1 with 1U of SAP for 2h at 37°C in thermostatic metal bath. 5. Inactivate enzyme by incubation at 70°C for 20 min in thermostatic metal bath.
6. Collect the nuclei by centrifugation at 1500 g with the temperature setting of 27°C in a high-speed refrigerated centrifuge for 10 min.
7. Aspirate the supernatant carefully and completely.
8. Wash the pellet twice with 500 µl of 1× MNase buffer at 1500 g with the temperature setting of 27°C in a high-speed refrigerated centrifuge for 10 min.9. Remove the supernatant, use the nuclei pellet for the following step.
MNase fragmentation and DNA puri cation, 4 hours: 1. Completely resuspend the nuclei in 50 µl of cold 1× MNase buffer supplemented with 200 μg/ml BSA (1 µl of BSA) per 1 million nuclei (based on the original number of cells) in a 1.5 ml Eppendorf tube.
Note: the nuclei would stick to the pipette tip from this step easily, therefore use Ultraultra-low attachment tip and pipette gently to avoid loss.

Troubleshooting Troubleshooting
We noticed variation in MNase activity from lot to lot and also occurring with time of storage.Therefore, pre-testing to ensure fragmentation in the desired range can be helpful when using new lot or if they enzyme has not been used for some time.

Time Taken Timing
From the time that samples are ready to be processed, the whole wet lab procedure takes around 2 days for 1-10 samples and 3 days for 11-24 samples.Analysis from raw reads to the coordinates of breaks takes around 1 and 2 days respectively for such numbers of samples with 1 Gb sequencing scale per sample and using the server provided in this protocol.However, the processing times would vary depending on the amount of sequencing being processed and the computer hardware used.

Anticipated Results
QC metrics of a library ready for sequencing are shown in the Fig. 2.

3 . 6 .
Add 1200-3000 units of MNase and 25 units of RNase If per aliquot.4. Mix them well by pipetting up and down.Incubate for 30 minutes on ice. 5. Add 5.6 µl of 0.5 M EDTA into the solution and slowly mix by pipetting up and down to stop the digestion.Add 150 µl of nuclei lysis buffer to the tube and incubate for 5 min at room temperature.7. Add 1 µl of 20 mg/ml proteinase K into the mixture and incubate for 45 min at 55°C and 45 min at 65°C in the PCR thermocycler.8. Transfer the samples to 1.5 ml Eppendorf tubes.fastq_quality_trimmer -v -t 20 -l 30 -i read_1.clean.fq-o read_1.clean.fq.1 -Q33; fastq_quality_trimmer -v -t 20 -l 30 -i read_2.clean.fq-o read_2.clean.fq.1 -Q33; 2. Select read pairs where reads 1 start with 10 G's or AGTTGCGGATGGGGGGGGGG in the NovaSeq protocol and reads 2 start with 12T's.

Figure 2 .
Figure 2. QC metrics of a SSiNGLe-ILM library.The fragment size (base pairs, bp, X-axis) distribution was analyzed by LabChip GX Touch.LM and UM -lower and upper size standards.The QC was performed by the Novogene Corporation.