Development of a high throughput human stool specimen processing method for a molecular Helicobacter pylori clarithromycin resistance assay
Authors:
Natalie Clines aff001; Erin Beckman aff001
Authors place of work:
Meridian Bioscience, Inc., Cincinnati, Ohio, United States of America
aff001
Published in the journal:
PLoS ONE 14(12)
Category:
Research Article
doi:
https://doi.org/10.1371/journal.pone.0224356
Summary
It has become critical to detect Helicobacter pylori (H. pylori) infection due to the link to gastric cancer with some strains. These strains are also increasing in resistance to antibiotics with clarithromycin leading the way as the first line treatment. Resistance to clarithromycin has been shown to correlate with the A2142G, A2142C, and A2143G mutations on the rrl gene. In the last few decades, non-invasive specimens, such as stool, have been a reliable alternate to gastric biopsy for immunoassay tests. More recently, it has been proven feasible for stool to be used in molecular based tests. Many of the core laboratories in the United States need a high throughput sample preparation to run this test. Here, a high throughput assay is compared to a previously published manual sample prep H. pylori molecular based assay. Using the Magna Pure 96 (Roche), at least 96 stool species and 96 biopsy specimens can be tested in an 8-hour shift of a clinical lab. The high throughput sample prep had a positive percent agreement (PPA) of 87% compared to the manual sample prep using the same testing configuration. The genotype predictions from the high throughput assay matched genotype predictions from the manual sample prep with the same stool sample 92% of the time. A concordance rate of 89% was observed with genotype predictions from the high throughput assay of the same patient stool and biopsy. In stool samples from the high throughput assay, there was 100% concordance between the quantitative polymerase chain reaction (qPCR)-derived genomic prediction and DNA sequencing data. The high throughput workflow can get more patients tested faster in addition to detection of mutations associated with clarithromycin resistance.
Keywords:
Helicobacter pylori – Lysis (medicine) – antibiotics – Specimen preparation and treatment – Antibiotic resistance – Helicobacter pylori infection – Biopsy – High throughput sequencing
Introduction
Helicobacter pylori has infected one-half of the world’s population [1]. It has become increasingly important to test for H. pylori infection as there is strong evidence showing a high correlation between infection and gastric cancer [2,3]. Although gold standard testing includes an invasive endoscopic procedure where a gastric biopsy is retrieved, there are tests where other specimen types can be used [4]. One commonly performed test is the urea breath test (UBT) in which the patient ingests a capsule containing 13C-labeled urea. Once the capsule is digested, the patient breathes into a capture vessel, releasing isotopic CO2 and ammonia [5]. The amount of CO2 released is measured against the patient’s baseline. The change in CO2 is used to determine if H. pylori is present in the stomach. There are also rapid immunoassay tests and ELISA assays available for H. pylori that involve collection and use of stool specimens [6,7]. Once the patient is diagnosed with an H. pylori infection, a regimen of antibiotics with a proton pump inhibitor (PPI) is prescribed. Typically, the first line of treatment includes the antibiotic clarithromycin (CLA) [8,9], although drug resistance to this antibiotic is increasing in prevalence [10]. The prevalence of clarithromycin resistance in the pediatric population in Italy, for example, was shown to be 25% from 2002 to 2007 [10]. It is important to detect resistance to minimize the number of repeat treatments and circumvent treatment failures. Currently, there are no FDA cleared tests for detecting H. pylori and CLA resistance using non-invasively collected specimens. A recently published manuscript described a novel molecular assay using analyte specific reagents (Meridian Bioscience, Cincinnati, Ohio) to amplify and detect H. pylori DNA in stool specimens as well as detect mutations associated with CLA drug resistance [11]. A manual sample preparation method was employed in that instance. There is a need for core laboratories to process a larger number of non-invasive H. pylori specimens in an automated fashion. The objective of this study was to use a H. pylori positive population to compare an optimized high throughput, semi-automated workflow to a reference manual workflow described previously with stool and stomach biopsy specimens.
Materials and methods
Samples and processing methods
Unpreserved stool and biopsy specimens were collected from patients who exhibited symptoms of H. pylori infection. Specimens were collected as part of the patients’ standard of care (symptomatic) in a private practice in Bologna, Italy and were de-identified upon testing. Specimens were stored at -20°C prior to testing. Mostly positive stool specimens were processed with the test method which was a high throughput, semi-automated sample preparation using the Roche MagNA Pure 96 (MP96). The MagNA Pure 96 DNA and Viral NA Small Volume Kit, Version 07, “Pretreatment of Stool Samples” was used and the instructions followed (cat no. 06543588001; Roche) for both the stool and biopsy specimens. The volumes used in this protocol were scaled down to correspond with 200 μL of sample transferred to the MagNA Pure 96 Processing Cartridge for on-board processing. The lysis buffer used for stool pre-treatment was MagNA Pure Bacteria Lysis Buffer (cat no. 04659180001; Roche). The pre-treatment protocol used for biopsy specimens is specified in the instructions for the MagNA Pure Tissue Lysis Buffer (cat no. 04805160001; Roche). This protocol recommends use of lysis beads, however, historical data indicates loss of this specific target during the mechanical homogenization and was, therefore, removed from the process. The homogenizer used was the Qiagen Tissuelyser II. For both specimen types, stool and tissue biopsy, 20 μl of the exogenous control (cat no. BIO-11025; Bioline) was added. The program run on the instrument, Pathogen Universal 200 3.1, was also the same for both specimen types. The sample input volume onto the MP96 was 200 μl with a corresponding 50 μl elution volume. One 96-well plate was tested for each specimen type containing a few select negative samples to detect any possible cross-contamination events.
Reference testing
The reference manual method used was the QIAamp Fast Stool kit (Qiagen) following the manufacturer’s instructions for performing human DNA isolation. For this process, a qPCR exogenous control (cat no. BIO-11025; Bioline) was spiked into each stool sample prior to extraction which acted as a workflow performance control as well as in indicator of inhibition. Both test and reference testing were carried out at the same time. Biopsies, collected from symptomatic patients from the MP96 assay were compared to bidirectional sequencing and stool results from the same patient. Duplicate biopsy specimens were not available for reference testing with the manual method.
Detection and analysis
All testing for H. pylori and CLA resistance was carried out using TaqMan real-time PCR amplification as described previously using the Qiagen RotorGene Q instrument [5]. A qPCR master mix was created using the following components: 2X SensiPLUS master mix (catalog no. BIO-11021; Bioline); H. pylori primer set (catalog no. ASR100; Meridian Bioscience) at a final concentration of 400 nM; H. pylori probe (cat no. ASR101; Meridian Bioscience) at a final concentration of 100 nM; H. pylori clarithromycin resistance (HPCR) probe (cat no. ASR102; Meridian Bioscience) at a final concentration of 100 nM; and control primers and probe (cat no. BIO-11025; Bioline) at a final concentrations of 40 nM and 100 nM, respectively. The H. pylori TaqMan probes’ binding sites were within the same amplified target on the H. pylori rrl gene. The Hp detection probe was detected in the green channel (emission wavelength, 510 nm) and the HPCR probe was detected in the red channel (emission wavelength, 660 nm). For the eluted samples processed using the MP96 workflow, 10 mg/mL BSA stock was added to a final concentration of 0.2 mg/ml in the qPCR master mix to reduce possible invalids due to interfering substances from stool. The volume of the MP96 processed samples added to the final qPCR mix was 10 μl. For specimens processed using the reference manual method, the addition of BSA was unnecessary and the sample volume input was 20 μl. Molecular-grade water was used in all reactions to adjust the final reaction volume to 50 μl. The following qPCR thermoprofile was used: 95°C for 5 min followed by 50 cycles each of 95°C for 20 seconds and 65°C for 60 seconds. The final step of the qPCR program was a melt step of the red probe (emission wavelength, 660 nm). The melt step was conducted between 50°C and 70°C with a 10 second wait between 1°C step increments.
The simultaneous detection of H. pylori and the prediction of CLA resistance/susceptibility was analyzed as previously described [11]. If amplification was observed in the green channel with the H. pylori probe, the specimen was positive. If no amplification crossed the set threshold of 0.1 normalized fluorescent units after 50 cycles, the sample was considered negative. If a specimen was positive for H. pylori, typically a single melt peak with the HPCR probe was observed. The RotorGene Q software would assign a single melt temperature (Tm) using a heavy digital filter on the red (HPCR probe) signal. The presence or absence of a mutation associated with CLA resistance would show a different melt profile and Tm value so resistance or susceptibility can be predicted. The stool specimens for both the test and reference assays had a Tm cutoff of 60°C to indicate wild type (≥60°C; susceptible) or mutant (<60°C; resistant) (Fig 1). In the case of an indeterminate melt result, the change in Ct, also known as delta Ct, was analyzed to predict CLA resistance or susceptibility. The Ct value from the normalized red channel (HPCR probe) was subtracted from the Ct value of the normalized green channel (Hp probe). If the difference between Ct values was higher than a shift of 2 Ct values, the sample was considered CLA susceptible (Fig 1). Those samples with a Ct shift of less than 2 were considered CLA resistant.
For biopsy specimens from the MP96 high throughput assay, a similar algorithm was used to predict CLA resistance. The Tm cutoff for biopsy samples was different than that of stool. Biopsies were given a cutoff of ≥57°C for CLA sensitivity and <57°C for CLA resistance. Bidirectional sequencing was used to verify the absence or presence of the mutations associated with CLA resistance of H. pylori positive samples. Purification of the amplified DNA followed by bidirectional sequencing was performed by ACGT, Inc. (IL, USA). Sequences were aligned using Sequencher software (Gene Codes Corporation).
Results
Stool specimens
A total of 96 stool specimens were processed using the MP96 high throughput and reference manual methods. Table 1 shows the overall step-wise workflow of the high throughput method. Of the 96 specimens tested, 76 were positive and 20 were called negative by the manual reference assay. The MP96 assay had a positive percent agreement (PPA) of 86.8% with 66 positive confirmations compared to 76 by the manual reference method. The positive predictive value (PPV) of the MP96 assay was 92.7%.
H. pylori CLA resistance/susceptibility was also predicted and compared to the reference manual assay. After initial testing with the MP96, genotype predictions of 59 of the 66 positive specimens were concordant with manual reference testing. Of the 7 genotype discordant specimens, 5 were retested and were favorably resolved. Therefore, of the 66 positive specimens by the high through-put test assay, 61 were concordant on genotype prediction with the manual reference assay. Bidirectional sequencing was carried out on the 66 MP96 assay qPCR products. We observed a 100% concordance between the qPCR-derived genomic prediction and DNA sequencing data. The PPA and PPV rates had 95% confidence intervals compared to the reference assay for Hp (Fig 2). This plot includes concordance rates with a 95% confidence interval of the observed proportion for 66 H. pylori positive specimens from the MP96 assay compared to the manual reference assay qPCR predictions and the bidirectional sequencing results.
Biopsy specimens
A total of 96 biopsy specimens from H. pylori symptomatic patients were tested with the MP96 test assay. Of the 96 specimens, 87 were positive for H. pylori and were bi-directionally sequenced. The sequencing results showed a 94% concordance rate compared to the biopsy MP96 qPCR assay. Some sequencing results showed either a mixed population result or inadequate sequencing results in 4 of the samples while 1 out of 87 samples did not match the qPCR genotype prediction for CLA resistance/susceptibility.
A subset of 32 positive biopsies had patient matched stools for further analysis. 4 results were indeterminate compared to the MP96 stool assay and 1 result was indeterminate compared to the manually processed stool assay. Of the 28 positive results, the genotype predictions of the biopsies versus that of the stool specimens from the MP96 stool assay and the reference manual stool assay were 89.3% (25/28) and 93.6% (29/31), respectively. There was a total of 5 total discordant specimens from the two comparator assays. Four out of the 5 discordant results were confirmed of their assay genotype prediction by their respective sequence result. One out of 5 discordant result was shown to be a mixed population after sequencing. Table 2 shows a breakdown of genotype concordance of the analyzed results.
Discussion/Conclusion
The results from stool and biopsy specimens show that sample prep automation is a feasible workflow for H. pylori detection and genotyping. Cutoff temperatures were different with stool compared to biopsy. This may be because less target is present in a stool sample due to degradation of free DNA. The difference in the amount of target resulted in a shift of the melt curves. Cutoffs were confirmed with bi-directional sequencing. The lower sensitivity shown by the high throughput assay could be due to the mechanism of DNA isolation from each method. The MP96 uses a silica bead to bind DNA under the optimal conditions, while the manual QIAamp method employs a silica membrane.
This study demonstrates that the number of specimens tested in a typical 8-hour shift can be scaled up; this includes any potential hands-off time for other laboratory tasks. An example of an 8-hour shift where the elapsed time for each task in the workflows are compared (Fig 3). For this study, the MP96 samples were processed off-board in individual tubes and then added to a 96-well processing plate to load on-board the MP96. If the appropriate equipment is available, an off-board processing in a deep well 96-well plate and a multi-channel pipette may reduce the risk of cross-contamination. Typically, a batch of 24 specimens is an ideal number for the manual method because it corresponds to the number of samples that most table top centrifuges can simultaneously handle. The stool specimens from the MP96 assay had a PPA of 87%. The PPV of stool specimens from the MP96 assay was 93%. The PPV was a biased result since the number of positive and negative stools is not an accurate indication of prevalence. With more negative specimens, the indication is that the PPV would increase.
It is well documented that three mutations on the rrl gene, A2142G, A2142C, and A2143G, are associated with CLA resistance [12,13]. Both stool and biopsy specimens showed a high concordance with respect to the sequencing results. H. pylori positive specimens can be infected with a mixed strain population [14,15]. Here, we observed a few cases that were confirmed by sequencing where one direction exhibited a wild type sequence and the other direction was a mutant sequence. These were categorized as indeterminate.
The high throughput, semi-automated assay using the MagNA Pure 96, is a reliable sample prep method for H. pylori testing in stool and biopsy specimens. It also can be used for the simultaneous prediction of CLA resistance.
Supporting information
S1 Table [xlsx]
Raw biopsy data.
S2 Table [xlsx]
Raw stool data.
Zdroje
1. Hunt RH, Xiao SD, Megraud F, Leon-Barua R, Bazzoli F, van der Merwe S, et al. World gastroenterology organization global guideline: Helicobacter pylori in developing countries. J Gastrointestin Liver Dis. 2011; 20(3):299–304. 21961099
2. Infection and Helicobacter pylori. IARC Monogr Eval Carcinog Rosks Hum 1994; 61:177–240.
3. De Vries AC, Kuipers EJ. Helicobacter pylori eradication for the prevention of gastric cancer. Aliment Pharmacol Ther. 2007; 26(2):25–35.
4. Ricci C, Holton J, Vaira D. Diagnosis of Helicobacter pylori: invasive and non-invasive tests. Best Pract Res Clin Gastroenterol. 2007; 21(2):299–313. doi: 10.1016/j.bpg.2006.11.002 17382278
5. Luo JC. Noninvasive diagnostic methods for Helicobacter pylori infection. J of Chinese Med Assoc. 2015; 78:83–84.
6. Fallone CA, Chiba N, vanZanten SV, Fischbach L, Gisbert JP, Hunt RH, et al. The Toronto consensus for the treatment of Helicobacter pylori infection in adults. Gastroenterology. 2016; 151 (10): 51–69.
7. Chey WD, Leontiadis GI, Howden CW, Moss SF. ACG Clinical Guideline: treatment of Helicobacter pylori infection. Am J Gastroenterol. 2017; 112: 212–239. doi: 10.1038/ajg.2016.563 28071659
8. Mitui M, Patel A, Leos NK, Doern CD, Park JY. Novel Helicobacter pylori sequencing test identifies high rate of clarithromycin resistance. J Ped Gastroenterol and Nut 2014; 59(1):6–9.]
9. Toracchio S, Cellini L, Di Campli E, Cappello G, Malatesta MG, Ferri A, et al. Role of antimicrobial susceptibility testing on efficacy of triple theory in Helicobacter pylori eradication. Aliment Pharmacol Ther 2000; 14(12):1639–43. doi: 10.1046/j.1365-2036.2000.00870.x 11121913
10. Giorgio F, Principi M, De Francesco V, Zullo A, Losurdo G, Di Leo A, Et al. Primary clarithromycin resistance to Helicobacter pylori: Is this the main reason for triple therapy failure? World J Gastrointest Pathophysiol. 2013; 4(3):43–46. doi: 10.4291/wjgp.v4.i3.43 23946886
11. Beckman E, Saracino I, Fiorini G, Clark C, Slepnev V, Patel D, et al. A novel stool PCR test for helicobacter pylori may predict clarithromycin resistance and eradication of infection at a high rate. J Clin Microbiol. 2017; 55 (8): 2400–2405. doi: 10.1128/JCM.00506-17 28515219
12. Abdollahi H, Savari M, Javad Zahedi M, Darvish Moghadam S, Hayatbakhsh A. Detection of the A2142C, A2142G, A2143G Mutations in 23S rRNA gene conferring resistance to clarithromycin among Helicobacter pylori isolates in Kerman, Iran. Iran J Med Sci. 2011; 36(2): 104–110. 23359224
13. Ribeiro ML, Vitiello L, Miranda MCB, Benvengo YHB, Godoy APO, Mendonca S, et al. Mutations in the 23S rRNA gene are associated with clarithromycin resistance in Helicobacter pylori isolates in Brazil. Ann Clin Microbiol Antimicrob. 2003; 2: 11. doi: 10.1186/1476-0711-2-11 14633281
14. Wang G, Rahman MS, Humayun MZ, and Taylor DE. Multiplex sequence analysis demonstrates the competitive growth advantage of A-to)G mutants of clarithromycin resistant Helicobacter pylori. Antimicrob Agents Chemother. 1999; 43(3): 683–685. 10049289
15. Cao Q, Didelot X, Wu Z, Li Z, He L, Li Y, et al. Progressive genomic convergence of two Helicobacter pylori strains during mixed infection of a patient with chronic gastritis. Gut. 2015; 64(4):554–61. doi: 10.1136/gutjnl-2014-307345 25007814
Článek vyšel v časopise
PLOS One
2019 Číslo 12
- Řekněte mi, co jíte, a já vám řeknu, jaká je úroveň vaší kognice
- Původní český test ALBA pro odhalení mírných kognitivních deficitů získal certifikaci MZ ČR
- Duševní onemocnění se u teenagerů přenáší v rámci školní třídy
- Od kostiček k lidské kůži. Jak si postavit 3D biotiskárnu?
- Prof. Libor Grubhoffer: Nová varianta mpox je příležitostí pokročit v poznání původce varioly
Nejčtenější v tomto čísle
- Methylsulfonylmethane increases osteogenesis and regulates the mineralization of the matrix by transglutaminase 2 in SHED cells
- Oregano powder reduces Streptococcus and increases SCFA concentration in a mixed bacterial culture assay
- The characteristic of patulous eustachian tube patients diagnosed by the JOS diagnostic criteria
- Parametric CAD modeling for open source scientific hardware: Comparing OpenSCAD and FreeCAD Python scripts
Zvyšte si kvalifikaci online z pohodlí domova
Všechny kurzy