The coronavirus disease 2019 pandemic has reversed essential tuberculosis (TB) treatment and reduced the TB burden. While there is a large difference between the number of people newly diagnosed with TB, which is reported to be 5.8 million, and the almost 10 million people diagnosed with TB in 2020, the incidence of TB is still high.[1] For decades, efforts have been made to develop techniques to effectively diagnose TB, but with little progress.[2] Acid-fast bacillus (AFB) smear microscopy has an application history of 135 years, and it is the most widely used test, despite the speculation that it misses more than half of all positive cases.[3],[4] It has a poor record for TB detection and diagnosis in patients with extrapulmonary TB or pediatric TB, as well as patients coinfected with human immunodeficiency virus (HIV) and Mycobacterium tuberculosis (M.tb), immunocompromised patients, and elderly individuals.[5] Despite these limitations, sputum AFB smear microscopy has been the primary method for diagnosing pulmonary TB, especially in low-and middle-income countries. Recently, diagnostic methods based on sputum AFB microscopy have been investigated to overcome the current limitations of TB diagnosis using other biological samples, such as blood, urine, and exhaled air.[6],[7],[8] Among these, urine as a sample is safe for the personnel involved in its collection, is noninvasive, and does not require special equipment or medical technicians.

In patients with active TB, cell-free DNA (cfDNA) from M.tb circulating in the bloodstream may be detected in the urine after filtration in the kidneys. However, the concentration of M.tb transrenal DNA (trDNA) in urine is low. Therefore, the use of M.tb trDNA as a prime diagnostic tool is limited.[9],[10],[11]

In this study, we determined the possibility of extracting trDNA from the urine of hospitalized patients with TB and analyzed its use as a biomarker for confirming treatment outcomes. We explored two methods to extract trDNA and found that the anchor extraction method was more effective in extracting small amounts of M.tb trDNA present in the patients' urine than the modified magnetic beads extraction method.[12],[13] We also evaluated the application of trDNA as a biomarker to predict treatment outcomes using sequential urine samples from patients with TB who were hospitalized.

Study cohort

Fifty-one routine urine samples were obtained from 40 patients diagnosed with TB from June 2018 to May 2019. The included patients were hospitalized for varying durations, diagnosed with M.tb on the basis of positive microbial cultures, and also clinically diagnosed with active TB using the TB disease code.[14]

Ethics

Public Institutional Review Board designated by the Ministry of Health and Welfare approved this study (Approval No. P01-201908-33-001).

Urine collection and processing

After performing routine investigations on the 40 admitted patients, residual urine samples were collected from the patients at 8 am every day. Sputum samples from the same patient were also collected and processed using GeneXpert. Twenty milliliters of urine samples were treated with urine preserves of cfDNA (Streck, La Vista, NE, USA) within 2 h after collection. Thereafter, trDNA was extracted from these samples using two methods on the same day.[15],[16]

GeneXpert MTB/RIF assay

The Xpert® MTB/RIF assay (Cepheid, CA, USA) results were recorded in the following report format: negative or positive, and then subdivided into four subtypes (very low, low, medium, and high), including detailed cycle threshold values. GeneXpert test was performed according to the manufacturer's instructions.[17]

Oligomer design

Control DNA

For the positive control, we used artificial double-stranded DNA of size 99 bp designed and synthesized in M.tb IS6110 (GenBank: AJ242908.1, from 2075th to 2173rd). After serial dilution of the synthesized positive control DNA, the Bio-Rad Droplet Digital polymerase chain reaction (ddPCR) QX200 System was used to confirm its presence (Bio-Rad, Hercules, CA, USA) with a locked nucleic acid (LNA) probe and primer.

Transrenal DNA extraction oligomers

The trDNA was extracted using two methods. The 51 samples were divided into two subgroups, and each group was subjected to a different extraction method. Nineteen samples (Nos. 1–19) were processed using MX-8 L DNA automatic extraction equipment (Apintech, Seoul, Korea) and an extraction kit using magnetic beads,[15] and the remaining 32 samples (Nos. 20–51) were processed using the anchoring method with extraction oligomers.[16] The extracted oligomers were located within the IS6110 sequences (forward: 2071–2105, reverse: 2144–2178) [Table 1]. All sequences are described in [Figure 1].

Click here to view

Click here to view

Transrenal DNA amplification primers and probe set

We designed primers and probes for trDNA amplification. Based on the nucleotide sequence of IS6110, forward and reverse primers were synthesized to be stable at 52.5°C–56.3°C. The probe was designed for a temperature of 61.6°C. The reporter was fluorescein phosphoramidite (FAM), and black hole quencher 1 (Integrated DNA Technology, USA) was synthesized as presented in [Table 1] and [Figure 1].

Transrenal DNA extraction

Two methods were used to extract trDNA from urine samples obtained from patients with TB.

Modified magnetic bead-based extraction method

Nineteen urine samples (Nos. 1–19) were processed and trDNA was extracted using magnetic beads and an MX-8 L automatic DNA extraction device (Apintech, Seoul, Korea). A total of 8 mL of urine samples, 8 mL of cfDNA-binding buffer, 120 μL of proteinase K, and 2 μL of antifoams (in the kit) were used, according to the protocol of the manufacturer of MX-8 L.

Anchoring extraction method

The remaining 32 samples (Nos. 20–51) were processed and trDNA was extracted using the anchoring method designed for this study. The extraction oligomer was labeled with biotin and designed within the IS6110 sequence to selectively extract TB trDNA among various trDNAs. Urine samples were incubated at 92°C for 5 min, and then at 65°C for 1 min for each sample. Subsequently, 257 μL of ×20 saline-sodium citrate buffer (SSC) (SSC buffer, Sigma, USA) and 3 μL of oligomers F and R (extraction oligomers, 10 pmol/μL) were added to the samples, which were then slowly mixed for 15 min at room temperature.[18],[19],[20] Next, using Streptavidin Streptavidin MagneSphere® Paramagnetic Particles (Promega, Madison, WA, USA), the samples were washed five times with ×0.5 SSC buffer and placed at room temperature for 20 min. After removing the supernatant using a DynaMag™-2 Magnet (Thermo Fisher Scientific, Invitrogen™, CA, USA), the samples were washed three times with ×0.1 SSC buffer. After removing the supernatant, the precipitate was vortexed, and the remaining supernatant was separated using a DynaMag™-2 Magnet and stored in an e-tube.

Droplet digital polymerase chain reaction

For ddPCR, the Bio-Rad QX200 ddPCR system (Bio-Rad, Hercules, CA, USA) was used according to the manufacturer's instructions.[21] The reaction was conducted with a final volume of 20 μL. The reaction mixture comprised 10 μL of ddPCR Supermix for probes (no deoxyuridine triphosphate) (Bio-Rad, Hercules, CA, USA), 2 μL of IS6110 trDNA primer/probe mix (TaqMan probe labeled with FAM), 1 μL of sample trDNA, and water (H2O) to make up the volume. The reaction mixture was placed in the sample well of a DG8 cartridge (Bio-Rad). Then, 70 μL of droplet generation oil for probes (Bio-Rad) was loaded into the oil well and droplets were allowed to form in the droplet generator (Bio-Rad). After processing, the droplets were transferred to a 96-well PCR plate (Bio-Rad Laboratories). PCR amplification was carried out on a C1000 Touch instrument. A thermal cycler (Bio-Rad) was used with the following thermal profile: hold at 95°C for 10 min, 40 cycles at 94°C for 30 s and 60°C for 1 min (ramp 2°C/s), and 1 cycle at 98°C for 10 min and end at 4°C. After amplification, the plate was loaded onto a droplet reader (Bio-Rad), and the droplets from each well of the plate were read automatically. QuantaSoft software was used to count the PCR-positive and PCR-negative droplets to obtain absolute quantification of the target DNA. The quantification of each target is expressed as the number per microliter of reaction.

Patient characteristics

The 40 hospitalized patients enrolled in this study were clinically diagnosed with active TB. The age groups of the patients ranged from the 30s to the 90s, and 85% of them were men. The duration of hospitalization ranged from < 1 month to more than 6 months. The results of the anti-HIV antibody test were negative in all patients. The demographic data of the patients are presented in [Table 2].

Confirmation of positive control DNA

Analysis of polymerase chain reaction assay sensitivity in targeting IS6110

Positive control DNA was used as a reference to (i) measure the extraction limit of the trDNA and (ii) obtain a comparable quantification of the amplified product. Control DNA was diluted from an initial concentration of 6 μg/μL to 6 × 10 − 5 pg/μL (ddPCR concentration: 2.42 copies/μL) to simulate the amount of detectable trDNA [Figure 2].

Click here to view

Comparison of Transrenal DNA yield between the extraction methods

Modified magnetic bead-based extraction method

TrDNA in samples from group 1 (Nos. 1–19) was extracted using an automated machine. Initially, these samples could not be quantified, and the samples were amplified using ddPCR with the LNA primer and probe [Figure 3]. This method confirmed that only 4 of the 19 samples met the minimum concentration of 2.42 copies/μL compared to the control DNA. The other 15 samples in group 1 did not yield a significant amount of amplified trDNAs.

Click here to view

Anchoring extraction method using hybridization capture

The trDNA extracted from group 2 samples also did not reach the minimum value of DNA quantification, which was measured using the NanoDrop-2000 spectrophotometer (Thermo Fisher Scientific, Waltham, MA, USA); therefore, the same method of ddPCR was used to amplify the samples. When the extracted trDNA of group 2 was compared with the synthetic control DNA, 16 of the samples met the minimal concentration of 2.42 copies/μL, whereas the remaining 16 samples had negligible amounts of trDNA.

Although one of the limitations of this study is that all samples were not subjected to extraction using the same method, the results showed that group 2 (Nos. 20–51) had a better yield than group 1 (Nos. 1–19). Compared to the baseline set by synthetic control DNA, 4 of the 19 samples in group 1 had higher copy numbers, whereas 16 of the 32 samples in group 2 had higher copy numbers.

Further comparison of the two methods in terms of the ddPCR concentration of extracted trDNA is presented in [Figure 4]: group 1 in green (n = 19) and group 2 in orange (n = 32). A significant difference (P = 0.0117) was observed between the groups, according to the t-test. The trDNA concentration (mean ± standard error of the mean) of group 2 (anchoring extraction) was 8.621 ± 2.153, which was considerably higher than that of group 1 (magnetic bead extraction), with a concentration of 1.228 ± 0.4283.

Click here to view

To study trDNA, which exists in very small amounts in the blood, there should be an accurate and objective standard to compare trDNA. Synthetic control DNA used in the present study can be used as a reference for subsequent trDNA extractions and amplifications. For the practical purpose of discerning the infinitesimal amount of M.tb trDNA extracted, we must aim to develop a procedure or method that combines the LNA probe-primer structure investigated in the present study, but is more sensitive to smaller amounts of trDNA and yields a higher concentration of trDNA during extraction.

After increasing the extraction rate of trDNA, samples from both groups were amplified using ddPCR. Among the two approaches, magnetic bead-based and anchoring extraction, the latter proved more efficient based on the results of the present study. Although the anchor extraction method is labor intensive, it has been proven to be more suitable for the selective extraction of M.tb trDNA. However, both methods require further improvements to be more efficient, standardized and automated for practical assessments.

The results of the extracted trDNA or GeneXpert sputum analysis were not affected by the duration of each patient's stay in the hospital. However, the number of TB-causing bacilli in the sputum decreased after successful treatment [Figure 5]. Some researchers have hypothesized that there are potential trends in the amount of trDNA detected after the conversion of AFB smear negativity.[14] The present study did not show any trend of an increase in extracted trDNA.[22],[23]

Click here to view

A limitation of the present study was the lack of comparison with other detection modalities and types of samples for TB detection. Another limitation was that the present study only included seven patients hospitalized and evaluated over a 5-month interval; therefore, the present study cannot be categorized as a long-term study. To accurately analyze the existence of trends, a similar study must be conducted for a longer period, such as following a patient from the beginning of their positive sputum-smear TB diagnosis through smear-negative conversion and other treatment stages at timely intervals.

Some of the negative GeneXpert test samples had interesting results; 11 of the 15 were positive for M.tb trDNA. Considering that the GeneXpert test shows remarkably high sensitivity in molecular diagnosis using sputum, it is encouraging that 73% (11 of the 15) of samples that were negative in the GeneXpert test were detected as positive in the trDNA testing. This finding suggests that tests targeting trDNA could be an alternative to sputum-based test methods.

Although there was no confirmation of the existence of the trend mentioned above, cfDNA was detected after the negative result of the GeneXpert test of the sputum [Figure 6]. The possibility of using trDNA as an alternative diagnostic tool to sputum-dependent diagnosis is still largely unexplored. This investigation using trDNA as a biomarker must be a continuous research commitment. Monitoring TB treatments using methods other than AFB smear or culture conversion is vital in determining the success of a potential treatment. Potential biomarkers, such as trDNA, have to be identified and developed to better understand TB treatment.

Click here to view

Further reading

World Health Organization. International standards for TB care. Wkly Epidemiol Rec 2006;81:43-47.

Patents

Primers, probes, and kits for TB diagnosis using urine samples: PCT/KR2020/015712 KR10-2019-0148294.

Oligomers for extraction of M.tb trDNA from urine: PCT/KR2021/000299, KR 10-2020-0004031.

Data availability statement

The processed data are available from the corresponding author on request.

Ethical policy and institutional review board statement

The Ethics Committee of Masan National TB Hospital approved the study (IRB-398837-2018-E25, Korea).

Financial support and sponsorship

This study was supported by the National Research Foundation of Korea (Grant number 2018R1A5A2021242) and the TB Clinical Research Program of the Clinical Research Center, Masan National TB Hospital (Grant number 4631-304-210-13) by the Korean Government.

Conflicts of interest

There are no conflicts of interest.