Non-invasive comparative study of HLA genotyping between urinary and blood DNA using sequencing-based typing and third-generation sequencing

Introduction

Hematopoietic stem cell transplantation (HSCT) remains a cornerstone therapy for patients with hematologic malignancies or bone marrow failure syndromes, relying on precise human leukocyte antigen (HLA) compatibility to mitigate graft vs. host disease (GVHD) and ensure engraftment success [1], 2]. The HLA system, a highly polymorphic set of cell-surface proteins, dictates immune recognition and donor-recipient matching, where even minor allelic disparities significantly elevate risks of graft rejection and mortality [2], [3], [4]. While sequencing-based typing (SBT) remains the clinical gold standard for high-resolution HLA profiling, third-generation sequencing (TGS) has emerged as a transformative tool, resolving haplotype phasing and structural variations that challenge conventional methods [5], 6]. Despite these advancements, reliance on invasive blood sampling persists as a logistical barrier for large-scale donor registries, pediatric cohorts, and patients with contraindications to venipuncture.

Urine-derived DNA presents a compelling noninvasive alternative, yet its clinical adoption for HLA genotyping remains limited beyond niche applications in renal transplantation [7]. Early studies established the feasibility of amplifying urinary DNA via PCR, though inconsistent yields and inhibitor interference hindered robust genotyping [7], 8]. Recent advances in extraction protocols have enabled high-resolution HLA typing across multiple loci (e.g., HLA-A, -B, -C, -DRB1, -DQB1), yet critical gaps persist: (1) lack of standardization across platforms, (2) unresolved technical variability in low-concentration DNA handling, and (3) insufficient validation against gold-standard blood-based workflows [3], 9]. Notably, prior studies prioritized single-platform analyses (e.g., SBT or TGS), leaving untapped potential in leveraging their complementary strengths—SBT for high-throughput allele discrimination and TGS for full-length haplotype resolution [10].

The purpose of this study was to assess the feasibility of using urinary DNA as a noninvasive alternative to blood for high-resolution HLA genotyping by SBT and TGS, particularly in situations where blood sampling is difficult or inappropriate, and to validate its consistency and reliability compared with conventional blood-based methods.

Materials and methods

DNA extraction

The optimized DNA extraction protocol for urine samples is illustrated in Figure 1. The workflow comprises the following steps.

1. Sample preprocessing: (1) transfer 10 mL of fresh urine into a 15 mL polypropylene conical tube (Corning, USA). (2) Centrifuge at 2,500g for 10 min at room temperature (KUBOTA 4000, rotor RA-400) to pellet cellular debris. (3) Carefully aspirate and discard the supernatant, retaining the pellet. (4) Wash the pellet twice with PBS and centrifuging at 2,500 g for 5 min and then transfer to a 1.5 mL microcentrifuge tube (Eppendorf, Germany).

2. Cell lysis and protein digestion: (1) Resuspend the pellet in 200 μL of lysis buffer (LB). (2) Add 20 μL of proteinase K solution (20 mg/mL, Thermo Fisher Scientific) and vortex for 10 s.

3. DNA binding to silica membrane: (1) Add 200 μL of cell binding buffer (CBB) to the lysate and mix by pipetting 10 times. (2). Incubate the mixture at 70 °C for 10 min in a thermomixer (WEALTECH Corp. HB-2). (3). Add 100 μL of isopropanol (≥99.5 %, Sigma-Aldrich) and mix by pipetting 10 times. (4). Transfer the mixture to a silica-based spin column (Qiagen MinElute) and centrifuge at 12,000 g for 1 min at 4 °C. Remove the supernatant.

4. Inhibitor removal and washing: (1) Load 500 μL of inhibitor removal solution (IRS) onto the column. Centrifuge at 12,000 g for 1 min (4 °C) and remove the supernatant. (2) Wash the column twice with 700 μL of wash buffer (WB), and then centrifuging at 12,000 g for 1 min per wash. Ensure complete ethanol evaporation by air-drying the column for 5 min.

5. DNA elution: (1) Transfer the column to a fresh 1.5 mL microcentrifuge tube. (2) Apply 50 μL elution buffer (EB) directly onto the membrane. (3) Incubate at room temperature for 2 min, followed by centrifugation at 12,000 g for 1 min (4 °C). (4) Repeat elution with an additional 50 μL of EB and 30 μL WB to maximize yield. Store DNA at −80 °C (Table 1).

Figure 1:

Optimized urine DNA extraction workflow for HLA typing. LB, lysis buffer; PK, proteinase K; CBB, cell binding buffer; IRS, inhibitor removal solution; WB, wash buffer.

The Hamilton AutoLys STAR system is used to demonstrate an automated protocol for extracting DNA from 3 to 5 mL of blood specimens according to automated program [11]. Finally, DNA was eluted with 100 μL of EB, and stored at 2 °C–8 °C or −80 °C for long-term storage.

Results

Discussion

This study systematically evaluated the feasibility of urine-derived DNA as a noninvasive alternative for high-resolution HLA genotyping using SBT and TGS technologies. By comparing DNA quality, concentration, and HLA allele concordance between matched urine and blood samples from 11 healthy volunteers, we demonstrated that urinary DNA achieves 100 % genotype concordance across HLA-A, -B, -C, -DRB1, and -DQB1 loci compared to blood-derived DNA. These findings validate the technical robustness of urinary DNA for clinical HLA typing, particularly in populations where blood sampling is impractical or contraindicated.

Consistent with prior research in renal transplantation, our findings further demonstrate the reliability of urinary DNA. For instance, previous studies have reported HLA typing using urine-derived DNA [9], 10], 12]. For instance, one study reported >95 % concordance in HLA-DRB1 alleles among kidney transplant recipients, though the research prioritized donor-specific antibody (DSA) monitoring over comprehensive HLA analysis via RSSO typing technology [3]. Similarly, Li et al. validated urinary DNA for HLA genotyping but emphasized the need for optimized extraction protocols to enable ex vivo cell culture expansion [10]. In contrast, our findings demonstrate full concordance across five classical HLA loci through a combined SBT and TGS approach, establishing urinary DNA as a robust alternative for multi-locus HLA profiling without the need for ex vivo cell culture expansion.

A key innovation of this work is the development of a standardized and optimized workflow for urinary DNA-based HLA typing. While earlier studies highlighted technical challenges such as DNA degradation and variable purity, we addressed these limitations through a series of protocol optimizations [3], 13]. Firstly, we employed strong denaturants, such as guanidine hydrochloride, to effectively lyse urothelial cells and inactivate nucleases (e.g., DNase I) in urine, thereby preventing DNA degradation [14]. Secondly, our inhibitor removal system, utilizing Cell Binding Buffer (CBB, 40 % ethanol) and Inhibitor Removal Solution (IRS, 1 M NaCl), efficiently removes small-molecule inhibitors like urea and creatinine through phase transition precipitation. This method demonstrates a 30 % higher efficiency in inhibitor removal compared to traditional phenol-chloroform extraction [15], 16]. Additionally, the inclusion of bovine serum albumin (BSA) neutralizes residual PCR inhibitors (e.g., heme derivatives), significantly improving downstream amplification success rates [15], 16]. Thirdly, the use of 20 mg/mL proteinase K ensures complete degradation of DNA-binding proteins, enabling direct use of low-concentration DNA (<10 ng/μL) for high-throughput library preparation without the biases associated with in vitro amplification [16]. Finally, all buffers (except proteinase K) contain high concentrations of guanidine salts and ethanol as preservatives, ensuring room temperature stability for up to 12 months. These optimizations collectively guarantee the reliability and reproducibility of urinary DNA extraction.

Our findings demonstrate that urinary DNA, when extracted using our optimized protocol, serves as a highly stable and reliable source for HLA genotyping, achieving 100 % concordance between SBT and TGS platforms. The robustness of our extraction method ensures consistent performance across both techniques, enabling accurate and reproducible HLA typing results. SBT provides high-throughput routine analysis, while TGS resolves complex polymorphisms and rare alleles through full-length sequencing (5′UTR to 3′UTR) with ≥30× coverage [17], 18]. This dual-platform approach, validated by complete concordance, highlights the adaptability of urinary DNA for diverse clinical applications, including resource-limited settings, without compromising accuracy or reliability.

Despite the methodological rigor of our optimized workflow, several inherent limitations restrict the broader extrapolation of these findings. First, the cohort was small, precluding robust estimates of failure rates in sub-optimal specimens. Second, the immediate post-void processing of mid-stream urine under controlled laboratory conditions may not reflect clinical practice, where variable hydration status, ambient temperature fluctuations, and frequent hematuria or proteinuria in transplant recipients can diminish DNA recovery or introduce uncharacterized PCR inhibitors. Third, although all samples yielded concordant HLA calls, three urines contained <5 ng/μL DNA, a level that could compromise assays with higher input requirements or long-range amplification. Fourth, long-term stability at −80 °C and the ability to detect minor alleles or chimerism at the ∼30× TGS depth used remain untested. Multi-center studies incorporating pediatric, renal-compromised, and oncologic populations, as well as standardized home-collection workflows, are therefore warranted before clinical deployment.

In conclusion, this study establishes urine-derived DNA as a viable, noninvasive substrate for clinical HLA genotyping. Our approach offers a simpler and more accessible alternative, particularly suitable for large-scale screening and settings with limited resources. Further validation through larger, multicentric studies is warranted to confirm its broader applicability. Standardization of urine collection, DNA extraction, and HLA typing protocols will be critical for scaling this approach, ultimately enhancing accessibility for patients with contraindications to blood sampling and advancing precision in transplant medicine.