Skip to main content

Genomewide bisulfite sequencing reveals the origin and time-dependent fragmentation of urinary cfDNA.

Cheng THT, Jiang P, Tam JCW, Sun X, Lee WS, Yu SCY, Teoh JYC, Chiu PKF, Ng CF, Chow KM, Szeto CC, Chan KCA, Chiu RWK, Lo YMD

Publication: Clin Biochem (2017) • Vol 50 • Page 496-501PMID: 28238813View on PubMed

Purpose

This paper investigated if methylation patterns could be used to determine the cellular origin of cell-free DNA (cfDNA) in urine and investigated the effects of 37˚C storage on cfDNA size distribution and concentration. The effect of renal and hematopoietic stem cell transplant on the cellular origin of cfDNA and cfDNA fragmentation was also investigated.

Conclusion

Based on the methylation patterns, the authors were able to determine the cellular origins of cfDNA in urine which differed among specimens but were highly similar among those obtained from the same patient at different times of the same day. For transplant recipients, the percentage of cfDNA originating from the donor cell-type as determined by the methylation patterns was very strongly correlated with the percentage of cfDNA from the donor as determined by Illumina OMNI 2M SNP arrays (R2=0.97).

cfDNA concentrations decreased with increasing storage duration when urine was stored at 37˚C. Voided urine had a higher percentage of small cfDNA than that from the matched renal pelvis specimens, but the proportion of cfDNA greater than 100 bp decreased and the percentage of fragments that were 50 bp or less increased when urine from the renal pelvis was incubated at 37˚C. The size distribution of cfDNA was comparable in renal and hematopoietic stem cell transplant patients and there was no obvious difference between donor and recipient-derived cfDNA. Although there were small changes in the proportion of cfDNA from different tissues during storage, the ranking by origin was unaffected.

Studies (1)

Purpose

This study investigated if methylation patterns could be used to determine the source of cell-free DNA (cfDNA) in urine and investigated the effects of 37˚C storage on cfDNA concentration and size distribution. The effect of renal and hematopoietic stem cell transplant on the tissue source of cfDNA and cfDNA fragmentation was also examined. A total of 26 urine specimens were collected from 11 renal transplant patients and 2 hematopoietic stem cell transplant patients. Kidney biopsy or donor buffy coat specimens from nine clinically stable renal transplant patients were used to determine cfDNA origin and for SNP analysis. Additionally, paired voided and kidney pelvis urine was collected by percutaneous nephrostomy from a patient with kidney stones.  Renal and urothelial tissue was obtained during a cadaveric renal transplant. Urine was stored at 4˚C, cells and debris were removed by centrifugation and 0.45 µm filtration within 1 h, and DNA was extracted immediately from the cell-free urine  or after incubation at 37˚C for up to 12 h. cfDNA was extracted using the Wizard Plus Minipreps DNA Purification System. cfDNA was quantified by real-time PCR amplification of LEP. DNA was extracted from blood and tissue specimens using the Qiagen DNA Mini Kit. SNP data was obtained Illumina Omni 2.5 SNP array. cfDNA was bisulfite-converted and sequenced using the KAPA HTP Library Preparation Kit on an Illumina HiSeq 2500 sequencer. Data was processed by the Methy-Pipe analysis pipeline.

Summary of Findings

A total of 19,418 differentially methylated regions (DMR) were found in B-cells, urothelium, kidney, neutrophils, and T-cells. As expected, hierarchical clustering showed that the methylation patterns were very different in the blood cells from that in kidney and urothelium. Using the DMRs as a reference, the authors determined the source of the cfDNA in urine. More than 97% of the cfDNA was from the kidney for 9 of the 28 specimens, more than 50% was from neutrophils for 15 specimens, and more than 50% was from urothelium for one specimen. For the remaining three specimens, 30-46.8%, 16.2-35.8%, and 12.6-49% of the cfDNA was from kidney, urothelium and neutrophils, respectively. The percentage of cfDNA from B-cells was less than 10% and from T-cells was less than 5%. Importantly, cfDNA cellular origins were highly similar in specimens obtained from the same patient at different times of the same day. For transplant recipients, the percentage of cfDNA originating in the donor cell-type as determined by the methylation patterns was very strongly correlated with the percentage of cfDNA from the donor as determined by Illumina OMNI 2M SNP arrays (R2=0.97). During post-transplant monitoring, the fraction of cfDNA coming from the donor kidney decreased but the absolute amount of kidney cfDNA remained unchanged.

cfDNA concentrations decreased with storage duration when urine was stored at 37˚C, mimicking the conditions inside the bladder and the calculated half-life of the cfDNA ranged from 2.6 to 5.1 h. Voided urine had a higher percentage of small cfDNA than that from the matched renal pelvis specimens, but the proportion of cfDNA fragments greater than 100 bp decreased and the percentage of fragments that were 50 bp or less increased when urine from the renal pelvis was incubated at 37˚C. Although there were small changes in the proportion of cfDNA from different tissues during storage, the ranking  by cellular origin was unaffected. The size distribution of cfDNA was comparable in renal and hematopoietic stem cell transplant patients and there was no obvious difference between donor and recipient-derived cfDNA.  

  • Biospecimens

    FluidUrine
  • Preservation Types

    None (Fresh)Frozen
  • Diagnoses

    Other diagnoses
  • Platforms

    DNASNP assay
    DNABisulfite conversion assay
    DNANext generation sequencing
  • Pre-analytical Factors

    ClassificationPre-Analytical FactorValue(s)
    Biospecimen AcquisitionMethod of fluid acquisition
    Percutaneous nephrostomy
    Voided urine (spot collection)
    Biospecimen AcquisitionTime of biospecimen collection
    Two collections during same day
    StorageStorage duration
    0 h
    1 h
    10 h
    12 h
    2 h
    24 h
    3 h
    4 h
    6 h
    7 h
    9 h