Drug-sensitivity test and analysis of drug-resistant mutations in Mycobacterium tuberculosis isolates from Kashgar,China

Abstract

Introduction

In recent years, drug-resistant Mycobacterium tuberculosis strains have gradually become widespread. Most drug resistance is related to specific mutations. We investigated M. tuberculosis drug resistance in the Kashgar area, China.

Methods

The drug-susceptibility test was conducted to clinical isolates of M. tuberculosis. Genomic-sequencing technology was used for the drug-resistant strains and the significance of DNA sequencing as a rapid aid for drug-resistance detection and the diagnosis method was evaluated.

Results

The resistance rates of clinical isolates to rifampicin (RFP), isoniazid (INH), streptomycin (SM), ethambutol (EMB), and ofloxacin (OFX) were, respectively, 4.4%, 12.3%, 8.8%, 2.6%, and 3.5%. The single- and multi-drug resistance rates were, respectively, 80.0% and 20.0%. The resistance genes RopB, katG, InhA, RpsL, rrs, gyrA, and embB displayed codon mutations, while InhA was mutated in its promoter region. Kappa scores, evaluating the consistency between DNA sequencing and the resistance ratio methods for the detection of isolates’ resistance to RFP, INH, SM, OFX, and EMB, were 1, 0.955, 0.721, 0.796, and 1, respectively.

Conclusion

The resistance rate of INH and SM is relatively high in the Kashgar area. Detection of mutations in RopB, katG, InhA, RpsL, rrs, gyrA, and embB by DNA sequencing can predict drug resistance of M. tuberculosis strains with high sensitivity and specificity, and can be used for diagnosis.

Keywords

drug allergy testing DNA sequencing drug resistance gene mutation site

Introduction

Tuberculosis (TB), together with AIDS and malaria, represents the world’s three most serious infectious diseases. TB has become one of the most serious chronic infectious diseases in the world. According to data released by the Global Tuberculosis Report 2020, the number of new infections with M. tuberculosis is close to 10 million; adults accounted for 88% and children accounted for 12% in TB patients. The numbers of dead and drug-resistant TB patients in 2019 were 1.4 and 0.465 million, respectively.¹ The number of TB patients and M. tuberculosis carriers in China exceeds 500 million, and the annual incidence rate exceeds 1.2 million, of which the number of TB accounts for about 77% of the total incidence.^2,3 Xinjiang is a province with high TB incidence. Amongst the regions of Xinjiang, Kashgar has the highest incidence, with an annual average number of patients with active TB accounting for more than one-third of the total cases in the province. Therefore, Kashgar is a key area for TB prevention and control in the Xinjiang province.⁴

Vaccination is crucial for the prevention and control of TB. Bacillus Calmette–Guérin is the only preventive vaccine approved worldwide. This vaccine confers good protection for infants and young children, but is not satisfactory for adults.⁵ Currently, most common treatments rely on anti-TB drugs to control the course of disease, but the proportion of drug-resistant TB has been increasing year after year.⁶ For the first time, anti-TB treatment reached a resistance rate of about 3.3%, and this rate can be as high as 20% in cases involving re-treatments.^6,7

The WHO distinguishes first- and second-line drugs based on their efficacy and side-effect severity. Common first-line anti-TB drugs include rifampicin (RFP), pyrazinamide, isoniazid (INH), and ethambutol (EMB).⁸ These drugs have a strong antibacterial activity, fewer side effects, and long half-life and are the first choice of treatment against TB. A large number of studies have shown that the main cause of the drug resistance is mutations in drug target genes, that is, RopB in case of RFP resistance,^9-11 katG and InhA for INH resistance,^12,13 gyrA for ofloxacin (OFX) resistance,^14,15 RpsL and rrs for streptomycin (SM) resistance,^10,16-18 and EmbB for EMB resistance.^17,18

Although the resistance rate to anti-TB drugs in Xinjiang is lower than the average rate in China,⁴ the baseline number of patients is large and conventional drug-sensitivity tests take a long time. These peculiarities are not conducive to early diagnoses and treatment of drug-resistant TB. Kashgar is located at the western border of China and is the hometown of the Uyghur people.¹⁹ Whether the nature and frequency of the mutations conferring drug resistance in clinical isolates of M. tuberculosis in this area are similar to those from other regions of China is unknown. Therefore, to investigate and analyze anti-TB drug resistance from Kashgar and to identify whether the underlying mutations are of great clinical significance, we investigated the genetic basis of the resistance of M. tuberculosis to commonly used anti-tuberculosis drugs in the Kashgar region. We evaluated the use of genomic-sequencing technology for the detection of drug-resistant strains and its significance as a rapid aid for drug-resistance detection and diagnosis methods.

Materials and methods

Culture of M. tuberculosis

One hundred and thirteen M. tuberculosis isolates were collected from TB patients at the Kashgar Tuberculosis Prevention and Treatment Center (Xinjiang Uygur Autonomous Region, China). The cultured strains were diluted at 1 mg/ml in a 7H9 medium containing 0.5% of Tween 80 (BD, USA). The strains were further diluted to 10 μg/mL and 100 ng/mL and aseptically inoculated to a modified Roche medium supplemented with fresh eggs and 2% malachite green solution (Baso, China). A medium containing nitrobenzoic acid (PNB, Sigma, USA)/thiophene-2-carboxylic acid hydrazine (TCH, Sigma, USA) was used to confirm the M. tuberculosis identity of the subcultures. The tube covers were slightly loosened and placed in an incubator at 37°C, with 5% CO₂. After 7 days of internal culture, the tube cover was tightened and placed vertically for further culture. The growth of the strains was regularly monitored and recorded. The results were reported 4 weeks later. The standard M. tuberculosis strain H37Rv, provided by the Chinese Center for Disease Control and Prevention, was used as control.

Drug-sensitivity test of the clinical M. tuberculosis isolates

The drug sensitivity tests were performed with the 1% ratio method recommended by the WHO/IUATLD’s “Guidelines for Tuberculosis Drug Resistance Surveillance (Fifth Edition).”²⁰ Bacterial suspensions diluted at 10 μg/mL and 100 ng/mL were prepared and inoculated into the drug medium containing RFP, INH, SM, OFX, or EMB and the control medium. After 28 days, the cultured M. tuberculosis strains were analyzed for the growth status and colony count, and the drug-resistance percentages were calculated. Isolates with a resistance percentage greater than 1% were considered resistant (R); otherwise, they were classified as sensitive (S). Highly diluted bacterial suspensions (10 ng/mL), which grew less than 20 colonies on the control medium, were sub-cultured from the control tube and re-submitted to the drug-sensitivity test.

Amplification and analysis of drug-resistant genes in drug-resistant strains

From the literature, we selected a screening panel of seven TB-resistance genes: RpoB, katG1/katG2/inhA, RpsL/rss, gyrA, and embB genes known to confer resistance to RFP, INH, SM, OFX, and EMB. We retrieved the gene sequences from the M. tuberculosis genome database TB Database (http://genome.tbdb.org) and used Primer 5.0 software to design primers to amplify the genes of the screening panel (Table 1). The genomes of M. tuberculosis strains classified as “drug-resistant” according to the susceptibility tests to the different drugs were extracted using the Mycobacterium Column DNA Out Extraction Kit (Tiandz, China). The genomes of the different isolates were used as templates to amplify the genes RpoB, katG1/katG2/inhA, RpsL/rss, gyrA, or embB corresponding to their respective drug resistance. The PCR products were purified using the MiniBEST DNA Fragment Purification Kit Ver 4.0 (TaKaRa, Japan) to remove unwanted reagents, including proteases, primers, dNTPs, and metal ions. The PCR products were sequenced and subsequently analyzed using DNAMAN software. The gene sequences corresponding to the H37Rv standard strain were retrieved from GenBank. The location and type of mutations present in each drug-resistant strain were characterized and quantified. The efficiency of the direct-sequencing method for the detection of drug resistance of M. tuberculosis was evaluated by taking the drug-susceptibility test as reference.

Table 1.

Primers used to determine the resistance mutations by PCR amplification and DNA sequencing.

Gene name	Primer sequence (5^′ to 3^′)	Annealing temperature, °C	Product length
RopB-F	GCGAGCTGATCCAAAACCA
RopB-R	GGTACACGATCTCGTCGCTAAC	59	447bp
katG₁-F	CGATGAGGTCTATTGGGGCA
katG₁-R	AGAGGTCAGTGGCCAGCATC	58.5	568bp
katG₂-F	CTGGCGCTTGGCAATACACC
katG₂-R	GGATCTCTTCCAGGGTGCGA	59.5	531bp
InhA-F	ACATACCTGCTGCGCAATTC
InhA-R	CCGATCCCCCGGTTTCCTC	59.7	300bp
RpsL-F	CCAACCATCCAGCAGCTGGT
RpsL-R	ATCCAGCGAACCGCGGATGA	59.5	306bp
rss-F	CGGGTTCTCTCGGATTGA
rss-R	CCACTGGCTTCGGGTGTTAC	59	994bp
embB-F	ATATTCGGCTTCCTGCTCT
embB-R	TAACGCAGGTTCTCGGTAT	53	725bp
gyrA-F	CAGCTACATCGACTATGCGA
gyrA-R	ATGAGGTACACCGAAGCCC	55	320bp

Data analysis and statistics

The results were collected from three repeated independent tests involving analyses of mutations in drug-resistant genes of the drug-resistant isolates and comparisons by chi-square tests of the consistency between the DNA-sequencing technology and the conventional proportion method. The consistency between the two techniques was calculated with SPSS Statistic 23 software and was expressed as a Kappa value: 1 ≥ Kappa ≥ 0.75 corresponded to high consistency; 0.75 ≥ Kappa ≥ 0.4 corresponded to medium consistency; and Kappa < 0.4 corresponded to low consistency. Graphical outputs were prepared with GraphPad Prism.

Results

Identification and drug sensitivity of the clinical isolates

The drug-sensitivity tests discovered 30 drug-resistant strains out of the isolates, including 2 RFP-resistant, 11 INH-resistant, 6 SM-resistant, 1 EMB-resistant, 4 OFX-resistant, 2 RFP/EMB double-resistant, 1 RFP/SM double-resistant, and 3 INH/SM double-resistant strains; no strains were resistant to all antibiotics. The single-drug resistance rate was 80.0% (24/30), and the multi-drug resistance rate was 20.0% (6/30) (Table 2).

Table 2.

Analysis of drug resistance in clinical isolates.

	RFP	INH	SM	EMB	OFX	Single resistance	Multi-drug resistance
Drug-resistant strain (cases)	5	14	10	3	4	24	6
Resistance rate (%)	4.4%	12.3%	8.8%	2.6%	3.5%	80.0%	20.0%

Abbreviations: EMB, ethambutol; INH, isoniazid; RFP, rifampicin; SM, streptomycin; OFX, ofloxacin.

Mutation analysis in genes related to the identified drug resistance by DNA-sequencing analysis

The sizes of the PCR products resulting from the amplification of fragments of the seven drug-resistance–related genes, that is, RopB, katG, InhA, RpsL, rrs, embB, and gyrA, were consistent with the expected values. The sequences of the target fragments were compared with the sequence of the standard H37Rv M. tuberculosis strain published in Genbank, and the mutation sites, codon, and amino acid changes were recorded.

Among the RFP-resistant strains, we found 5 mutation sites in the resistance-determining region of the RopB gene (Table 3). None of these mutations were in drug-sensitive strains. These mutations were located in codons 511, 516, 526, and 535, accounting for 14.28%, and in codon 531 (42.85%). Thus, these five mutations in the RopB gene may be highly indicative of a strain’s resistance to RFP (Figure 1(a)).

Table 3.

Drug-resistance–related gene mutations.

Phenotype	Gene name	Mutation site	Codon change	Amino acid change	Strain number	Count
		511	CTG→CGG	Leu→Arg	988
		531	TCG→TTG	Ser→Leu	988	1
		526	CAC→CTC	His→Leu	1033	1
		516	GAC→GTC	Asp→Val	1058
		531	TCG→TTG	Ser→Leu	1058
		535	CCC→CAC	Pro→His	1058	1
					1000
RFP resistance	RopB	531	TCC→TTG	Ser→Leu	1023	2
RFP sensitive	RopB	—	—	—	—	108
		234	GCG→GGG	Ala→Gly
		315	AGC→ACC	Ser→Thr	990
	katG	463	CGG→CTG	Arg→Leu	1007
	InhA	−15	C→T	—	1024	3
		234	GCG→GGG	Ala→Gly
		315	AGC→ACC	Ser→Thr
	katG	463	CGG→CTG	Arg→Leu	1058
	InhA	−8	T→C	—	1157	2
		232	CCG→TCG	Pro→Ser	992
	katG	315	AGC→ACC	Ser→Thr	1013	2
	katG	315	AGC→ACC	Ser→Thr	998
	InhA	−152	G→A	—	1047	2
		315	AGC→AAC	Ser→Asp
	katG	463	CGG→CTG	Arg→Leu	1067	1
		234	GCG→GGG	Ala→Gly
	katG	315	AGC→ACC	Ser→Thr
	InhA	−8	T→C	—	996	1
	katG	315	AGC→AAC	Ser→Asp	1056	1
INH resistance	katG	—	—	—	1066	1
INH sensitive	katG	234	GCG→GGG	Ala→Gly	—	69
		463	CGG→CTG	Arg→Leu
		43	AAG→AGG	Arg→Lys
	RpsL	88	AAG→AGG	Arg→Lys
	rrs	491	C→T	—	4509	1
					1015
					1021
	RpsL	43	AAG→AGG	Arg→Lys	1065	3
					987
	RpsL	88	AAG→AGG	Arg→Lys	1032	2
	rrs	491	C→T	—	1000	1
		512	C→T	—
	rrs	523	A deletion	—	1066	1
	rrs	905	A→G	—	1054	1
	RpsL	—	—	—
SM resistance	rrs	—	—	—	1058	1
	rrs	598	A→G	—	339	2
	rrs	612	A deletion	—	1003	1
	rrs	523	A deletion	—	1167	2
SM sensitive	RpsL	—	—	—	—	98
	rrs	—	—	—	—
		74	GCC→GTC	Ala→Val	107
OFX resistance	gyrA	95	AGC→ACC	Ser→Thr	163	2

	gyrA	56	CTC→CTG	—	92	1
OFX sensitive	gyrA	95	AGC→ACC	Ser→Thr	—	110
					1000
	embB	306	ATG→GTG	Met→Val	1023	2
EMB resistance	embB	306	ATG→ATA	Met→Ile	1011	1

Abbreviations: EMB, ethambutol; INH, isoniazid; RFP, rifampicin; SM, streptomycin; OFX, ofloxacin.

Figure 1.

Mutation rates in gene regions associated with drug resistance of M. tuberculosis isolates. A. Proportion of the different mutations affecting RpoB in rifampicin-resistant strains. B and C. Proportion of the different mutations affecting katG and InhA in isoniazid-resistant strains. D and E. Proportion of the different mutations affecting RpsL and rrs in SM-resistant strains.

Mutation in katG and combined mutations in katG and InhA led to INH resistance. The two mutations discovered in InhA were accompanied by mutations in katG. Mutations in InhA alone could not be associated with resistance to INH. In addition, katG was mutated in codons 232, 234, 315, and 463, while InhA had mutations at positions −8, −15, and −152 of its promoter region (Table 3). About 69% of INH-sensitive strains also had mutations in codons 234 and 463 of katG, suggesting that these mutations did not cause resistance to INH, but corresponded to single-nucleotide polymorphisms (SNPs). According to their frequency amongst the resistant isolates (Figure 1(b)), mutations in codon 315 of katG closely correlated with the resistance to INH, whereas mutations in codons 234 and 463 were not relevant since they were also present in 69% of the drug-sensitive strains. Codon mutations at position 232 accounted for 15.38% of the total number of mutated sites and were not found in the sensitive strains. Regarding InhA, the mutations at positions −8 and −15 of the promoter region accounted for 23% of the resistance mutations, and mutations at position −152 accounted for 15.38% (Figure 1(c)). These mutations were only present in the resistant strains, suggesting that they are closely linked to the resistance phenotype.

Amongst the SM-resistant strains, RpsL was mutated at 2 sites, and rrs had mutations or deletions at 6 sites. There were 3 configurations of mutations. First, mutations affecting both RpsL and rrs accounted for 10% of the strain resistance. Second, mutations in RpsL alone caused 50% of the strain resistance (Figure 1(d)). Third, mutations in rrs alone accounted for 30% of the resistance. Apart from these frequent mutations, a drug-resistant phenotype was found that involved neither of the two genes (Table 3 and Figure 1(e)). In addition, half of the mutations found at site 523 were in drug-sensitive strains. Mutations at site 523 also existed in drug-resistant strains but were combined with a second mutation. Therefore, the resistance in double-mutants may be caused by the second point mutation at position 512, rather than the mutation at position 523. The mutations at positions 598 and 612 of the rrs gene occurred in strains sensitive to SM. Therefore, a direct implication of these mutations in the acquisition of the resistance phenotype can be ruled out (Figure 1(e)).

The OFX-resistance–related gene gyrA displayed 3 mutation sites associated with the resistance of the isolates (Table 3). However, synonymous mutations were found in drug-sensitive strains, indicating that these changes cannot be responsible for OFX resistance. Given the considerable number of sensitive strains mutated at this site, this mutation corresponded presumably to SNP as was observed for katG at codons 234 and 463. In OFX-resistant strains, because the mutation at codon 95 was attributed to SNP, only the mutation at codon 74, uniquely present in all resistant isolates (100%), was considered to be related to the resistance phenotype.

All three EMB-resistant clinical isolates had mutations at codon 306 (Table 3). However, due to the small sample size, this result only shows that EMB resistance in Kashgar may be mainly caused by mutations in codon 306 and that it is necessary to collect and analyze more EMB-resistant strains in this area.

Analysis of the correlation between the results obtained by drug-sensitivity tests and DNA sequencing for the detection of drug-resistant M. tuberculosis strains

Drug-sensitivity test based on the proportion method is the gold standard to assess drug resistance of M. tuberculosis strains. We used this well-established technique to evaluate if the DNA-sequencing technology, used to detect mutations in seven gene loci associated with resistance to five different drugs, could represent a basis to assess and predict drug resistance of M. tuberculosis. Resistance to RFP, INH, SM, OFX, and EMB was determined by detecting mutations in the resistance genes RopB, katG/InhA, RpsL/rrs, gyrA, and embB. To assess the value of each gene for the diagnosis of drug resistance, we compared the two methods by calculating Kappa values, which were, respectively, 1, 0.955, 0.721, 0.796, and 1 (Table 4). These Kappa values were close to 1, indicating that DNA sequencing gave results highly consistent with those obtained with the susceptibility test based on the proportion method. Therefore, this technology can be used for rapid diagnosis well before the results of the susceptibility tests can be delivered. This method could considerably accelerate the implementation of appropriate therapy for TB patients.

Table 4.

The efficiency of DNA sequencing for the detection of drug-resistant clinical isolates.

Drug/Gene	DNA sequence	Phenotype		Count	Sensitivity (%)	Specificity (%)	Positive prediction (%)	Negative prediction (%)	Kappa value
Drug/Gene	DNA sequence	resistance	sensitive	Count	Sensitivity (%)	Specificity (%)	Positive prediction (%)	Negative prediction (%)	Kappa value
	mutation	5	0	5
	unmutated	0	108	108
RFP/RopB	total	5	108	113	100.0%	100.0%	100.0%	100.0%	1
RFP/RopB	mutation	12	0	12
INH/katG and InhA	unmutated	1	100	101	92.3%	100.0%	100.00%	99%	0.955
	total	13	100	113
	mutation	9	5	14
SM/RpsL and rrs	unmutated	1	98	101	90.0%	95.1%	64.00%	97.02%	0.721
	total	10	103	113
	mutation	2	1	3
	unmutated	0	110	110	100%	99.09%	66.66%	100%	0.796
OFX/gyrA	total	2	111	113
	mutation	3	0	3
	unmutated	0	110	110	100.0%	100.0%	100%%	100.00%	1
EMB/embB	total	3	110	113

Abbreviations: EMB, ethambutol; INH, isoniazid; RFP, rifampicin; SM, streptomycin; OFX, ofloxacin.

Discussion

The emergence and spread of drug-resistant M. tuberculosis strains, combined with a large and increasingly mobile population, pose greater challenges to the prevention and control of TB. The results of the fourth national TB epidemiological sampling survey in 2000 showed that the total drug-resistance rate of M. tuberculosis strains was 27.8%,²¹ while in 2010, this rate was 42.1%. This represents an increase of nearly 70% in 10 years. Of this increase, the resistance rate to first- and second-line anti-TB drugs represents 36.8% and 24.6%, respectively.²² According to the epidemiological survey of TB in Xinjiang in 2005 and 2010, the incidence rate of TB was much higher in the southern area than in the northern and eastern areas. The total resistance rates of the isolates were 26.4% and 36%, and the trend of drug resistance increased significantly.⁴ The drug-sensitivity tests in our study showed that the total drug-resistance rate of M. tuberculosis clinical isolates from the Kashgar area was 26.54%, which was lower than the national average. The single-drug resistance rate was 17.69%, higher than the 14.7% found in the Shihezi area, north of Xinjiang, and the 14% found in east China. The multi-drug resistance rate was 8.84%. The reason for the high single-drug resistance rate may be that the economic development of Kashgar is relatively delayed, and the residents’ economic income is low. During the treatment process, patients have repeatedly interrupted their anti-TB treatment due to economic reasons, resulting in an increased proportion of single-drug resistant strains.

The RopB gene is associated with RFP resistance. Studies showed that mutations in RopB account for more than 95% of the total RFP resistance.²³ These mutations lead to the decreased binding capacity of the drug to its target, which reduces or inhibits its effect. The genes katG and InhA are related to INH drug resistance.^24-26 Their products are involved in the biosynthesis of NAD coenzyme I, the inhibition of which stops mycobacterial acid synthesis. Mutations in these genes, among which mutations in codon 315 of katG are the most frequent, can lead to drug resistance.²⁷ In addition, the product encoded by sigI can bind the promoter region of katG, which in turn affects the expression level of katG. When sigI is mutated, the transcription level of the regulatory factor is increased, which suppresses the expression of the katG gene and reduces the bacterial drug sensitivity.

The genes gyrA and gyrB are associated with OFX drug resistance.^28-30 When these genes are mutated, the affinity of the drug for the target DNA helicase decreases, resulting in increased resistance of the strain. Studies have shown that about 85%–100% of OFX-resistant clinical isolates have mutations in the gyrA gene.^31-33 The genes RpsL and rrs are associated with SM resistance^34,35 and encode the S12 ribosome and the 16S rRNA, respectively. When RpsL and rrs are mutated, the drug’s ability to attach to the ribosome decreases, which in turn leads to drug resistance of the strain. The embB gene is associated with resistance to EMB. It encodes an arabinosyltransferase, and mutations in this gene cause resistance to EMB.^36,37

In this study, the mutation sites identified in the RFP-resistant strains are similar to those found by other researchers. Approximately 60% of the INH-resistant strains have mutations in katG, mostly at codon 315, which is consistent with the results from previous studies.³⁸ In addition, we found that mutations in the 234th codon of katG, present in both INH-resistant and -sensitive strains, may represent genetic polymorphism and are unlikely related to drug resistance. Mutations in InhA mainly affect the promoter region at positions −8, −15, and −152. Studies have shown that in addition to the above mutations, alteration at positions 8 and 16 and substitution at position 24 lead to drug resistance.³⁹ The reason why these mutations were not detected in our study might be the small sample size for INH-resistant strains and the concentration of clinical isolate collection areas. Otherwise, one drug-resistant strain could not be linked to mutations in known INH-resistant genes. This suggests that the relevant mutation site has yet to be found or that other mechanisms of drug resistance are at play. The main mutation sites in SM-resistant strains are at codons 43 and 88 of the RpsL gene and codons 426, 491, and 512 of the rrs gene, in keeping with previous research results.⁴⁰ The gene rrs also presented mutations in SM-sensitive strains, corresponding to A deletion in codons 523 and 612, and to an A598 G substitution. Since these mutations did not cause changes in the drug-resistance phenotype, they were not directly involved. The mutation sites in OFX-resistant isolates were hitting mainly the 74th codon. We discovered for the first time an Ala74Val mutation in the gyrA gene, associated with OFX resistance. Previously, an Ala74Ser mutation in this gene had been reported to cause drug resistance,³⁸ but no relevant reports have mentioned an Ala74Val mutation. The emergence of a new mutation may be due to the distance between Kashgar and the mainland, geographically separated from the northern Xinjiang region by the Tianshan Mountains. The genetic background of endemic M. tuberculosis in China presents some divergence.³⁹ The mutation sites found in EMB-resistant strains were all located at codon 306 of embB, scoring a mutation rate of 100%, which is far from the 70% rate reported for the mutations affecting the gene embABC.⁴⁰ The reason for this difference may be that the number of EMB-resistant strains isolated in this study was too low, and the source of the strains was concentrated, resulting in a high consistency for a single mutation site.

We used the resistance ratio method test as a reference to assess drug sensitivity and evaluated the sensitivity and specificity of the DNA-sequencing technology for the detection of the clinical isolates resistant to RFP, INH, SM, OFX, and EMB. A possible drawback affecting the sensitivity of the DNA-sequencing detection method is that some drug resistance stems in mutation sites are situated outside of known resistance-associated genes. This implies that the mechanisms involved in drug resistance need further elucidation. Therefore, as it stands, using the detection of mutations in the drug-resistance–associated loci of M. tuberculosis for rapid testing does not allow to rule out drug resistance. Even in the occurrence of a negative result, it remains necessary to confirm the strain’s sensitivity with the conventional proportion method. In conclusion, our study lays the foundation for the discovery of new sites and types of drug-resistant mutations, the elucidation of the drug-resistant mechanisms of M. tuberculosis, and the development of rapid drug-susceptibility tests. However, the results are based on a limited number of experimental samples, and the sample size should be enlarged to confirm these preliminary conclusions.

Conclusions

In Kashgar, the resistance rate to INH and SM is relatively high. DNA sequencing allowed the detection of resistance to RFP, INH, SM, OFX, and EMB with high consistency and to SM with moderate consistency. Detection of mutations in RopB, katG, InhA, RpsL, rrs, gyrA, and embB by DNA sequencing can predict drug resistance of M. tuberculosis strains with high sensitivity and specificity, and can be used for diagnosis.

Footnotes

Acknowledgments

We thank the Kashgar Tuberculosis Prevention and Treatment Center for their assistance in this study. We thank all participants for sample contribution.

Author contributions

ZM, ZW, and CC designed the study. ZM was responsible for the provision, integration, and writing of the article. ZW and CC reviewed the article, and the other authors provided help during the experimental process. All authors contributed to the article and approved the submitted version.

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This research was supported by the Collaborative Innovation Project for Prevention and Control of High-incidence Zoonotic Infectious Diseases in Western China (2011 Plan).

Data availability statement

The DNA sequences are not publicly available now; if necessary, please request the corresponding author.

ORCID iD

Chuangfu Chen

References

Chakaya

Khan

Ntoumi

, et al. Global tuberculosis report 2020 - reflections on the global TB burden, treatment and prevention efforts. Int J Infect Di 2021; S1201(21): 00193–00194.

Odone

Tillmann

Sandgren

, et al. Tuberculosis among migrant populations in the European union and the European economic area. Eur J Public Health 2014; 25: 506–512.

Wubuli

Xue

Jiang

, et al. Socio-demographic predictors and distribution of pulmonary tuberculosis (TB) in Xinjiang, China: a spatial analysis. PloS one 2015; 10: e0144010.

Yang

Simahule

Tai

, et al. Analysis of tuberculosis epidemiological survey conducted in 2010–2011 in Xinjiang Uygur autonomous region. Chin J Antituberc 2013; 35: 960–964.

Narayanan

. Influence of sex, age & nontuberculous infection at intake on the efficacy of BCG: re-analysis of 15-year data from a double-blind randomized control trial in South India. Indian J Med Res 2006; 123: 119–124.

, et al. Sputum bacteriology conversion and treatment outcome of patients with multidrug-resistant tuberculosis. Infect Drug Resist 2018; 11: 147–154.

Organization WH . Global Tuberculosis Report, 2015. Australas Med J, 2016.

Treu

Kokesch-Himmelreich

Walter

, et al. Integrating high-resolution MALDI imaging into the development pipeline of anti-tuberculosis drugs. J Am Soc Mass Spectrom 2020; 11: 2277–2286.

Sinha

Srivastava

Tripathi

, et al. Detection of mutations in the rpoB gene of rifampicin-resistant Mycobacterium tuberculosis strains inhibiting wild type probe hybridization in the MTBDR plus assay by DNA sequencing directly from clinical specimens. BMC Microbiol 2020; 20: 284.

10.

Zhou

Zhu

Yan

, et al. Deciphering extracellular antibiotic resistance genes (eARGs) in activated sludge by metagenome. Water Res 2019; 161: 610–620.

11.

Sun

Zhou

, et al. A rapid fluorescence polarization-based method for genotypic detection of drug resistance in Mycobacterium tuberculosis. Appl Microbiol Biotechnol 2014; 98: 4095–4105.

12.

Arora

Bothra

Prosser

, et al. Role of post-translational modifications in the acquisition of drug resistance in Mycobacterium tuberculosis. The FEBS J 2020; 288(11): 3375-3393.

13.

Armstrong

Lamont

Lanne

, et al. Inhibition of Mycobacterium tuberculosis inhA: design, synthesis and evaluation of new di-triclosan derivatives. Bioorg Med Chem 2020; 28: 115744.

14.

Safari

Moghim

Salehi

, et al. Sequence-based detection of first-line and second-line drugs resistance-associated mutations in Mycobacterium tuberculosis isolates in Isfahan, Iran. Infect, Genet Evol: J Mol Epidemiol Evol Genet Infect Dis 2020; 85: 104468.

15.

Al-Gallas

Khadraoui

Hotzel

, et al. Quinolone resistance among salmonella kentucky and typhimurium isolates in tunisia: first report of salmonella typhimurium ST34 in africa and qnrB19 in Tunisia. J Appl Microbiol 2020; 130(3): 807–818.

16.

Shrestha

Maharjan

Thida Oo

, et al. Molecular analysis of streptomycin-resistance associating genes in Mycobacterium tuberculosis isolates from Nepal. Tuberculosis (Edinburgh, Scotland) 2020; 125: 101985.

17.

Ghosh

ANS

Saha

. Survey of drug resistance associated gene mutations in Mycobacterium tuberculosis, ESKAPE and other bacterial species. Scientific Rep 2020; 10: 8957.

18.

Mokrousov

Sinkov

Vyazovaya

, et al. Genomic signatures of drug resistance in highly resistant Mycobacterium tuberculosis strains of the early ancient sublineage of Beijing genotype in Russia. Int J Antimicrob Agents 2020; 56: 106036.

19.

Wang

Han

, et al. ASAP1 gene polymorphisms are associated with susceptibility to tuberculosis in a Chinese Xinjiang muslim population. Exp Ther Med 2018; 15: 3392–3398.

20.

Organization WH . Guidelines for Surveillance of Drug Resistance in Tuberculosis. 5th edition. World Health Organization, 2015.

21.

Report on Nationwide Random Survey for the Epidemiology of Tuberculosis in 2000 . Buccetin of Chinese Antituberculosis Assoclation, 2002.

22.

Wang

Cheng

Chen

, et al. The fifth national tuberculosis epidemiological survey in 2010. Chin J Antituberculosis 2012; 34: 485–508.

23.

Zhang

Yew

. Mechanisms of drug resistance in Mycobacterium tuberculosis: update 2015. Int J Tuberc Lung Dis 2015; 19: 1276–1289.

24.

Cardoso

Cooksey

Morlock

, et al. Screening and characterization of mutations in isoniazid-resistant Mycobacterium tuberculosis isolates obtained in Brazil. Antimicrob Agents Chemother 2004; 48: 3373–3381.

25.

Bardou

Raynaud

Ramos

, et al. Mechanism of isoniazid uptake in Mycobacterium tuberculosis. Microbiol 1998; 144: 2539–2544.

26.

Hazbón

Brimacombe

Del Valle

, et al. Population genetics study of isoniazid resistance mutations and evolution of multidrug-resistant Mycobacterium tuberculosis. Antimicrob Agents Chemother 2006; 50: 2640–2649.

27.

Hazbón

Brimacombe

Bobadilla

dVM

, et al. Population genetics study of isoniazid resistance mutations and evolution of multidrug-resistant Mycobacterium tuberculosis. Antimicrob Agents Chemother 2006; 50: 2640.

28.

Wang

J-Y

Lee

L-N

Lai

H-C

, et al. Fluoroquinolone resistance in Mycobacterium tuberculosis isolates: associated genetic mutations and relationship to antimicrobial exposure. J Antimicrob Chemother 2007; 59: 860–865.

29.

Umubyeyi

Rigouts

Shamputa

, et al. Limited fluoroquinolone resistance among Mycobacterium tuberculosis isolates from rwanda: results of a national survey. J Antimicrob Chemother 2007; 59: 1031–1033.

30.

Guillemin

Jarlier

Cambau

. Correlation between quinolone susceptibility patterns and sequences in the A and B subunits of DNA gyrase in mycobacteria. Antimicrob Agents Chemother 1998; 42: 2084–2088.

31.

Takiff

Salazar

Guerrero

, et al. Cloning and nucleotide sequence of Mycobacterium tuberculosis gyrA and gyrB genes and detection of quinolone resistance mutations. Antimicrob Agents Chemother 1994; 38: 773–780.

32.

Dobner

Bretzel

Rüsch-Gerdes

, et al. Geographic variation of the predictive values of genomic mutations associated with streptomycin resistance in Mycobacterium tuberculosis. Mol Cell Probes 1997; 11: 123–126.

33.

Tracevska

Jansone

Nodieva

, et al. Characterisation of rpsL, rrs and embB mutations associated with streptomycin and ethambutol resistance in Mycobacterium tuberculosis. Res Microbiol 2004; 155: 830–834.

34.

Ramaswamy

Amin

Göksel

, et al. Molecular genetic analysis of nucleotide polymorphisms associated with ethambutol resistance in human isolates of Mycobacterium tuberculosis. Antimicrob Agents Chemother 2000; 44: 326–336.

35.

Mokrousov

Otten

Vyshnevskiy

, et al. Detection of embB306 mutations in ethambutol-susceptible clinical isolates of Mycobacterium tuberculosis from Northwestern Russia: implications for genotypic resistance testing. J Clin Microbiol 2002; 40: 3810–3813.

36.

Morlock

Metchock

Sikes

, et al. EthA, inhA, and katG loci of ethionamide-resistant clinical Mycobacterium tuberculosis isolates. Antimicrob Agents Chemother 2003; 47: 3799.

37.

Bifani

Mathema

Campo

, et al. Molecular identification of streptomycin monoresistant Mycobacterium tuberculosis related to multidrug-resistant W strain. Emerg Infect Dis 2001; 7: 842–848.

38.

Zhao

. Mutation of gyrA gene in 98 strains ofloxacin-resistant Mycobacteria tuberculosis clinical isolates in henan province. Chin J Zoonoses 2012; 28: 503–505.

39.

Yuan

, et al. Genotypic characteristics of Mycobacterium tuberculosis circulating in Xinjiang, China. Infect Dis 2016; 48: 108–115.

40.

Johnson

Jordaan

Pretorius

, et al. Ethambutol resistance testing by mutation detection. Int J Tuberc Lung Dis Official J Int Union Against Tuberc Lung Dis 2006; 10: 68–73.