PCR-Based Replicon Typing of Plasmids in E.Coli Producing Esbls in Patients with UTI in Iraq
Sawsan Qahtan Taha Al-Quhli 1 Safaa Abed Latef Al Maeni 2 Department of Microbiology, College of Medicine, University of Anbar, Iraq 1
Department of Biotechnology, College of Science, University of Anbar, Iraq2
Objectives:The aim of this work was to study the distribution of antibiotics resistance genes of ESBLs in clinical E. coli in patients with UTI as well as identification and classification of bacterial plasmids associated with virulence and antimicrobial resistance by using PCR-Based Replicon Typing (PBRT).
Materials and methods: ESBLs- producing E. coli were preliminary screened by cephalosporin confirmation tests. Antimicrobial susceptibility test was performed of all ESBLs- producing E.
coli. For genotypic detection, CTX-M, TEM and SHV ESBL genes were determined by PCR.
PBRT was performed by PBRT 2.0 KIT (Diatheva, Italy, 2020), as well as ESBLs resistance transfer assay was performed by conjugation method.Results: The prevalence of ESBL in this study is high. Overall, bla CTX-M was the commonest genotype (93.6%) in E. coli by PCR. 26 replicons were detected in CTX-M producing E. coli strains:" HI- 1, HI- 2, I1- α, I- 2, X- 1, X- 2,X- 3,X -4, L, I1y, N, FIA, FIB, FIC, FII, W, Y, P1,I1 A/C, T, k, U, R, B/O, HIB-M and FIB-M
"; 25 replicons were detected of TEM and SHV by PBRT. After conjugation was performed;
transconjugants could be obtained at ratio 100% in E. coli producing CTX-M; and 88% in TEM and SHV.
Conclusions: The PBRT scheme exhibited anexcellent specificity and sensitivity, detecting new epidemiologically relevant replicons: P1, N2 and X4 in ESBLs-producing E. coli. The PBRT method for the(plasmids analysis) by the(PBRT 2.0) kitretrenches the detection time and provided that simplifies the steps for identification and classification ofE. coli plasmids.
PCR- Based Replicon Typing (PBRT) is a useful technique for detecting and monitoring the interaction of replicons with specific plasmid-borne resistant genes and for detecting the mobilization potential of resistance genes among different bacterial plasmids (1).Bacterial plasmid diagnostic is very difficult because the several plasmids in same bacterial cell make it unmanageable to assess the bacterial plasmid- content of the bacterial - strain. PBRT method based on PCR amplification of replicons has therefore been used as a plasmid identification and typing technique throughout the world in recent years.(2). The (PBRT)method is a PCR- based auscultating initially certified and developed by" Dr. Alessandra Carattoli"( senior scientist) at the Italy ; industrialized by "Diatheva" subsequently.However, in Iraq generally, there are no studies about molecular identification and classification of plasmid conferring drug resistance in these clinical isolates and strain genotypes exacts a distinction to sprawls of antimicrobial resistance
ESBL genes in patients with UTI.
Materials and methods
Clinical isolates. Midstream urine samples were collected from 715 patients suffering from UTI referred to Al Anbar hospitals during (January 2019 to October 2019).All isolates were diagnosedroutine analysis,and automated diagnosis by API 20E and VITEK 2 compact system.
Antimicrobial Susceptibility Testing
E. coli's antimicrobial susceptibility test was estimated by updated' KirbyBauer disk diffusion method" in the Mueller–Hinton agar against Ampicillin, Pipracillin ,Mecillinam, Imipenem,Meropenem, Gentamicin,Amikacin,Tobramycin, Ciprofloxacin, Levofloxacin, Nalidixic acid, Chloramphenicol,Nitrofurantoin, Trimethoprim-Sulfamethoxazole, Tetracyclin,Doxycycline, Minocycline,Tigecycline, Rifampicin,Fosfomycin,Colistin and Polymyxin B, as well as screening method antibiotics (Oxoid, UK) according to CLSI (3).
Escherichia coli ATCC 25922 were used as a control strain.
Screening methods for ESBLs detection
The CLSI has proposed disc diffusion methods for screening for ESBL production by E. coli(3).
For ESBL production, these methods can be screened by noting particular zone diameters that suggest a high degree of suspicion for ESBL production. Ceftazidime, Cefotaxime, Ceftriaxone, Cefpodoxime, Cefepime and Aztreonam were used. Even then, using more than one of these screening agents increases detection sensitivity. If any ESBL diameter indicates suspicion, the condition should be checked using phenotypic confirmatory tests.
Confirmatory Tests for ESBLsdetection: Phenotypic Tests
The creation of E. coli isolates toESBLswas tested to at least onecephalosporinindicator resistant.
(4); Double Disk Synergy (DDST),Broth Microdilution, CombinedDisk Test (CDT), ESBL E Test,inhibitor-potentiated- disk- diffusion test(IPDD)Test, Culture on (ChromeID- ESBLs and Brilliance ESBLs agar) andESBL NDP Test (5). As ESBL positive or negative controls, Klebsiella pneumonia ATCC 700603 and Escherichia coli ATCC 25922 were used for all phenotypic methods ofdetection ofE.coli.
Molecular Detection of ESBLs Genes
For ESBL production, molecular analysis has been conducted for all phonotypicalESBL E. coli isolates,for detection of molecular E. coli; the traditional Polymerase chain reaction was performed by theE. coli harboring ESBL genes (PCR).Total bacterial genomic DNA was isolated according to the Promega kit (USA).
For detecting of ESBLs"Ambler: class A" of E.coli"CTX-M, TEM and SHV" encoding, standard PCR as a prototype with full bacterial genomic DNA. For PCRs with these primers, thermal cycling conditions (shv genes) consisted: 30 cycles at"94°C for 30 S", "68°C for 60S" and "72 for 60 S", and a final extension of" 72°C for 7min". A PCR with primers and conditions defined by
4270 genes (bla CTX-M) was performed (8). Amplicon is an intragenic (544)bp fragment. In 1-hour W/V agarose gel, 10 μl of the pcr product was electrophorised. 140V with an intelligent DNA ladder.
Table 1: sequences of SHV and TEM
Gene Primers' Sequences (5'→3') P. s. (bp) Refernce SHV F: CGC CGG GTT ATT CTT ATT TGT CGC
1016 R:TCT TTC CGA TGC CGCCGCCAGTCA 6
TEM F:ATA AAA TTC TTG AAG ACG AAA
1080 R:GAC AGT TAC CAA TGC TTA ATC A 7
F: TTTGCGATGTGCAGTACCAGTAA R:CGATATCGTTGGTGGTGCCATA
* F: Forward sequences, R: Reverse sequences.
The "resistance plasmids" in the ESBLs- E.coli isolates wereidentified by using "PBRT-kit":
(Diatheva, Italy, 2020). Complete bacterial DNAwas prepared according to Promega kit (USA) and used as a pcr prototype according to" the PBRT 2.0 kit protocol". The following 8 multiplex PCR tests (groups of incompatibility) with "positive –controls" were used for amplifying "30 replicons":- (HI-1, HI -2, I1 -α, I-2, X-1, X-2, X-3,X-4, L, I1y, N, FIA, FIB,FIC, FII,FIIS, FIIK, FIB KN, FIB KQ, W, Y, P1, A/C, T, K, U, R, B/O, HIB-M and FIB-M) in this kit representing the main plasmidIncompatibility- groups and plasmid genes in the "Enterobactericeae" set (2).
Transfer of ESBLs resistance assays (conjugation)
The plasmid that confers cephalosporin resistance in all ESBL-producing E. coli strains were transferred by plasmid-free, rifampicin-resistant E. coli strain K12 J53R (positive lac) was used as acceptor (9). The Luria Bertani (LB) boron crop was mixed with the respective donor strain at 1:2 and the mixture was" incubated at 37°C for 18 h", a mixture was spread (100) μl on rifampicin- containing 250 mg/l and the Ceftazidime 2 mg/l (CAZ) rifampicin-containing MacConkey agar plates for (TEM) or (SHV) donors strains styling overnight (10).For CTX-M:Producing-Donors;
they were selected on MacConkey agar- plates containing rifampicin 250 mg/l as well as 2.5 mg/l of Cefotaxime (CTX) (10). When resistance plasmid was transferred in mating experiments: DNA was extracted using Promega kit (USA) and plasmid replicates were determined using PCR-based replication typing (2).
All the data was analyzed with descriptive statistical analysis using SAS , 2012 software, .In order to compare the different classes, we used chi-squared testfor descriptive parameters.
Results and Discussion
In our study, E. coli was the most predominant species(45.1% 166 out 368 positive urine samples).High-ratio explanation for these bacteria is found in the urine, thereby triggering autoinfection. Additionally, after joining the human bladder;E. coli can bind to the bladder wall and form a "biofilm" that resists body immune response (11).Antibiotic susceptibility test showed full resistance of ESBLs- producing E. coli to Ampicillin, Pipracillin, Azetronam, third generation Cephalosporin and to Cefepime, but 100% sensitive to Carbapenems & Mecillinam. Pathogenic bacteria typically resist several antibiotics in hospitals due to increased antibiotic selection pressure.Weobserved that no resistant in the pattern to Carbapenems and Mecillinam, because they may be not used asa choice for the management of multidrug-resistant E. coli.
Our study found that E. coli producing ESBLs were 100% resistant to (Ceftazidime, Cefotaxime, Ceftriaxone, Cefpodoxime), Cefepime and Aztreonam. Due to the large amount consuming of thirdgeneration cephalosporins, cefepime and Aztreonam. Furthermore, resistance of ESBLs isolates to non-Beta-lactam antibiotics were threshed; Gentamicin, Tobramycin, Ciprofloxacin, Levofloxacin, Nalidixic acid, Chloramphenicol, Trimethoprim-Sulfamethoxazole, Tetracycline, Doxycycline and Rifampicin with resistant rates of respectively 44%, 50.4%, 71.2%, 70.4%, 88.8%, 33.6%, 83.2%, 76%, 55.2% and 64% against E .coli.Multi-drug resistance is a big concern in treating uropathogens ESBLs. This (MDR) may be attributable to plasmids with multiple resistance genes transferred from one uropathogenic bacterium to another. ESBL is usually encoded with resistant genes (aminoglycosides, chloramphenicol, sulfonamides, trimethoprene and other antigens) that significantly restrict alternative therapies (including the use of multiple resistant plasmids). Amikacin, Nitrofurantoin, Minocyclin, Tigecycline, Fosfomycin,Colstin and polymyxin- B. Showed best performance(high sensitivity ratio) percent against E.coliindicating these medications can still be used to treat UTIs. The antibiotic resistance of ESBL-producing strains to Beta-lactam and non-Beta lactam antibiotics which are usually recommended for initial treatment of UTI; was significantly higher (p < 0.01) than that of ESBL-non-producing strains in our study. Similarobservations have been published by other studies (12).
In the present study, we found 125(75.30%) ESBLs producers' from166 E. coli isolates based on MIC ESBLs screening breakpoints and confirmatory tests bydifferent phenotypic methods; high Sensitivites and Specificities of thesetestsforidentification of ESBL-production E. coli. The overall high sensitivities (100%) of the E. coli producing ESBL by broth micro dilution method, IPDD and the sensitivity of the combined disk method, with and without clavulanic acid as screening substrates for clavulanic acid synergy.
Among all ESBLs phenotypic detection methods, (DDS)test was the least sensitive phenotypic ones. The Combined disc test tends to have a higher Sensitivity 100% with ESBLs-positive isolates (11 of E.coli) that were missed by DDST. As a phenotypic method, CDM was shown to be more sensitive than the DDS test; thus, CDM is better than DDST for detection of ESBLs (13), as well as we found that The IPDD test seemed to be a better method than the DDST, showing 100
4272 percent sensitivity and specificity in confirming ESBL production. DDST failed to detect 11 isolates ofE.coli therefore; 91.2 percent sensitivity E.coli ESBL producers by this test. The sensitivity of the DDST as a phénotypic tool depends on the exact location of the disks (14).
In our research, the E-strip sensitivity is 100% in E. coli with 100 percent specificity; thus, the (E- strip) phenotypic testing method is responsive, accurate and comfortable. In addition, our analysis showed that cefpodoxime and ceftazide have a larger zone (increase) than cefotaxime; and that even the IPDD testing is more sensitive when ceftazidime disk is used than when the cefotaxime disk is used, as has the cephalosporins used in the IPDD test (14(
The advantage of the IPDD test is that isolates produced in ESBLs may be differentiated from non-produced ESBL isolates by (almost) mm cut-off for area increase and three Oxyimino cephalosporins may also be measured on one agar plate. Ceftazidime is an excellent platform for most ESBL enzymes. The new Brilliance ESBLs, by comparison, proved worthwhile and worked just as well as the chrom ID ESBLs to detect ESBLs. Having an excellent Negative Predictive Value (NPV) of (24) h,E.coli was a promising phenotype that could easily remove patients not transporting producers of ESBL. The ESBL NDP Test was recently developed for rapid identification of ESBLs-Enterobactericeae. This test is simple, sensitive and precise, based on hydrolysis detection in the beta-lactam ring of Cefotaxime (15).
Results of the PCR showed that the most common ESBL chanced gene in E.coli urinary isolates was CTX-M. The gene CTX-M was discovered in 93.6 percent E. coli. In this analysis, the frequency of the TEM-type and SHV-type enzymes were 90.4% and 60% respectively. Devi et al.
have found rising prevalence of E.coli in recent studies. CTX-M-type ESBL development ofE. coli and Klebsiella pneumoniae (16). High prevalence of ESBL-induced UTI in many hospitals worldwide has been identified (17; 18(.The
Replicons could be determined in ESBLs-producing E. coli strains strains in (Table -1), PCR- based Replicon Type 2.0 Kit; 26 repliconswere detected in CTX-M producingE.coli. HI-1, HI-2, I1-α, I-2, X-1, X-2,X-3,X-4, L, I1y, N, FIA, FIB, FIC, FII, W, Y, P1,I1 A/C, T, K, U, R, B/ O, HIB-M, FIB-M Strengths: (fig. 1(
The TEM replicas in ESBL-produced E. coli strains (Table-1); 25 replica: HI-1, I1-α, I-2, X-2, X- 3, X-4, L, I1y, N, FIA, FIB, FIC FII, W, Y, P1, A/C, T, K, U, R, B/O, HIB-M and FIB-M were detected (fig.3). 25 replicas were found on SHV generating E in the same table (Table-1). HI-1, HI-2, I1-α, I-2, X-1, X-2,X-3,X-4,I1y, N, FIA, FIB, FIC, FII, W, Y, P1, A/C, T, K, U, R, B/O, HIB-M and FIB-M .(fig. 5).
PBRT in E. coli producing CTX-M, TEM and SHV after conjugation was performed (Table-1);
transconjugants could be obtained at ratio 100% in E. coli producing CTX-M (Fig. 2). A mating experiment (conjugation) was successfulinour studywithCTX as a selective agent. In vitro,large number of the strains capable of transferring their resistance was CTX-M producers, which involved transconjugation in gene dissemination (10). The replicons rep B/O, FIB and FII which were found in the E. coli donors producing TEM (Fig.3) as well as rep I-2, B/O and FII in the E.
coli donors producing SHV (Fig 5) were absent in the transconjugants of E .coli donors producing SHV (Fig 5); therefore the ratio is 88% in E. coli producing TEM and SHV (Fig. 4 & 6). Inc I plasmids were more likely to be transferred by conjugation than Inc F plasmids (19).
In the present study the replicon FIIS was not recognized by PBRT, because this replicon PCR recognizes a divergent FII replicon identified on Salmonella virulence plasmids; as well as in our
results the replicons FIIK, FIB KN and FIB KQ was not recognized by this method. The FIIK PCR recognizes a divergent FII replicon identified on plasmids from Klebsiella spp but not in E.
coli (20). In this study X Replicons (X1, X2, X3 and new replicon X4) were associated with the three ESBL types studied CTX-M,TEM and SHV(Fig. 1,3,5) of E. coli strains by PBRT 2.0 kit;
but Incx family: (X1, X2, X3) types of IncX plasmids are detected by PBRT, as defined the previous studies. Typing plasmid is an important parameter for evolution research, epidemiology and antibacterial resistance spread (2; 21). The replicate content well-defined by the "PBRT"
method which used for"Microbiological analyzing" of the outbreaks bacterial clones and for monitoringspreading of particular groups of epidemiologically related resistant determinants; but genetically unassociated bacterial isolates (22). In 2017, Carloni et al. concluded that "PBRT-kit"
detected replicates in" 100% "of the analyzed strains; pretended to increase commercial sensitivity and specificity (2). It has been surmised that larger than "10,000 strain" of Enterobacteriaceae tested by using this technique to classify epidemic plasmids that have spread away species confines in very remote geographer locations in Enterobacteriaceae (22;23).
Table (1). Number of Replicons according to CTX-M, TEM and SHV identified in E. coli before and after conjugation by PCR-based Replicon typing.
ESBL – genes of E. coli
CTX-M TEM SHV
Presence of replicon types before
Replicons transfer after conjugatio n
Presence of replicon types before conjugatio n
Replicons transfer after conjugatio n
Presence of replicon types before
Replicons transfer after conjugation
HI1 + HI1 + HI1 +
HI2 + I1-α + HI2 +
I-1α + M + I-1α +
M + N + M +
N + I2 + N +
I-2 + B\O - I2 -
B\O + FIB - B\O -
FIB + FIA + FIB +
B1 + P1 + FIA +
w + w + P1 +
L + L + W +
X- 3 + X3 + X3 +
I1Y + I1Y + I1Y +
T + T + T +
A\C + A\C + A\C +
N2 + N2 + N2 +
U + U + U +
X -1 + X1 + X-1 +
R + R + R +
X-2 + X -2 + X-2 +
K + K + K +
HIB-M + HIB - M + HIB-M +
FIB-M + FIB – M + FIB – M +
F-I I + F-II - F-II -
X4 + X4 + X4 +
25 Replicon 22 Replicon
Fig. 1.Agarose gel electrophoresis analysis of multiple amplicons in CTX-M of E.coli obtained via amplification of positive controls using all PCR mixes (M1 to M8).
Fig. 2. Agarose gel electrophoresis analysis of multiple amplicons in CTX-M of E.coli after conjugation obtained via amplification of positive controls using all PCR mixes (M1 to M8).
Fig. 3. Agarose gel electrophoresis analysis of multiple amplicons in TEM of E.coli obtained via amplification of positive controls using all PCR mixes M1 to M8).
Fig. 4. Agarose gel electrophoresis analysis of multiple amplicons in TEM of E. coli after conjugation obtained via amplification of positive controls using all PCR mixes ( M1 to M8).
Fig. 5. Agarose gel electrophoresis analysis of multiple amplicons in SHV of E.coli obtained
4276 via amplification of positive controls using all PCR mixes (M1 to M8).
Fig. 6. Agarose gel electrophoresis analysis of multiple amplicons in SHV of E.coli after conjugation obtained via amplifications of positive- controls using all pcr mixes (M-1 to M- 8).
1. Sukmawinata, E.; Uemura,R; Sato, W.;Mitoma, S.; Kanda, T.;Sueyosh,M.; (2020). Inc/1 Pasmid Associatted with blaCTX-M-2 Transmission in ESBL-Producing Escherichia coli Isolated
from Healthy Thoroughbred Racehorse, Japan.J.Antibiotics 9020070.vol.9.Issue2;10.3390.
2. Carloni, E.; Andreoni, F.; Omiccioli, E.; Villa, C.; Magnani,M.; Carattoli, A.(2017).
Comparative analysis of the standard PCR-Based Replicon Typing (PBRT) with the commercial PBRT- KIT.Elsever.Plasmid. 90:10-14.
3. Pandit R. , Awal P., Shrestha S., Joshi G., Rijal B P. , and Narayan Prasad Parajuli N P.(2020).Extended-Spectrum β-Lactamase (ESBL) Genotypes among Multidrug-Resistant Uropathogenic Escherichia coli Clinical Isolates from a Teaching Hospital of Nepal. Article ID6525826/ https://doi.org/10.1155/2020/6525826.
4. Abdelmoktader A and El Far A. (2019) Beta-Lactamases and ESBL. Virol Immunol J, 3(4):
5. EUCAST (2020). The European Committee on Antimicrobial Susceptibility Testing.
6. Nüesch-Inderbinen T, Kayser H F, Hächler H (1996). Detection of genes coding for extended- spectrum SHV beta-lactamases in clinical isolates by a molecular genetic method, and comparison with the E test Eur J Clin. Micro. Infect Dis15 (5):398-402.
7. Wu, L.Thonpson, D.K.;Li,G, Hurt R.A.;Tidge, J.M. and Zhou, J.(2001). Development and evaluation of functional gene arrays for detection of selected genes in the environment.Appl.
8. Edelstein, M. Pimkin, M. Palagin, I. Edelstein, I.and Stratchounski,L.(2003).Prevalence and molecular epidemiology of CTX-M extended-spectrum beta-lactamase-producing Escherichia coli and Klebsiella pneumoniae in Russian hospitals. Antimicrob Agents
9. Marcadé G, Deschamps C, Boyd A and Arlet G (2009). Replicon typing of plasmids in Escherichia coli producing Extended-Spectrum-B-lactmases. J. of Antim. Chemoth. 63(1):67- 71.
1o. Schmitt J, Jacobs E and Schmidt H (2007). Molecular characterization of extended-spectrum beta-lactamases in Enterobacteriaceae from patients of two hospitals in Saxony, Germany.
Journal of Medical. Microbiology 56 (Pt 2):241-9.
11. Chaudhary M.; Payasi A. (2015). Molecular characterization and in vitro suscebtibilities of B- lactamase producing Escherichia coli, Klebsiella species,Acinetobacter baumannii, Pseudomonas aeruginosa and Staphylococcus aureus to CSE 1034 and other B-lactams.
Asian Pac J Top Med; 7(Suppl1): 217-23.
12. Tayh, G.; Al Laham, N.; Yahia, H.P.; Sallem, R.P. Elottol, A.E.; and Slama,K.B.(2019).
Extended-Spectrum β-Lactamases among Enterobacteriaceae Isolated from Urinary Tract Infections in Gaza Strip, Palestine. Res. ID 4041801,11 p
13. Tantry, A.; Bilal, A. Rahiman, S. Shaik, M. Nabi, M. (2018). Detection of Extended-Spectrum ß-lactamases Production by Escherichia coli: A Phenotypic Comparative Study. Journal of Pure and Applied Microbiology 12(4):2245-2252.
14. Harwalkar A, Sataraddi J, Gupta S, Srinivasa H (2013). The detection of ESBL-producing Escherichia coli in patients with symptomatic urinary tract infections using different diffusion methods in a rural setting. J Infect Public Health 6(2):108-14.
15. Nordmann P, Dortet L, Poirel L. (2012). Rapid detection of extended-spectrum-β-lactamase- producing Enterobacteriaceae. J. Clin. Microbiol. 50:3016–3022. 10.1128/JCM.00859-12 [PMC free article] [PubMed] [CrossRef] [Google Scholar].
16. Devi LS, Broor S, Rautela, RS (2020). Increasing Prevalence of Escherichia coli and Klebsiella pneumoniae Producing CTX-M-Type Extended-Spectrum Beta-Lactamase, Carbapenemase, and NDM-1 in Patients from a Rural Community with Community Acquired
Infections: A 3-Year Study. Intern. J. of applied and Basic MedicalResarch.
17. Shakya, P.; Shrestha, D.; Maharjan, E., Sharma, V.K. and Paudyal, R. (2020). ESBL Production among E. coli and Klebsiella spp. Causing Urinary Tract Infection: A Hospital Based Study. The open Microbiology Journal.Vol.14.
18. Shash, R.Y., Elshimy, A.A., Soliman , M.Y. and Mosharafa, A. A.(2019). Molecular Characterization of Extended Spectrum β-Lactamase EnterobacteriaceaeIsolatedfrom Egyptian Patients with Community- and Hospital-Acquired UrinaryTract Infection Am. J.
Trop. Med. Hyg., 100(3), pp. 522–528.
19. Timofte D, Maciuca I E, Williams N J, Wattret A and Schmidt V (2016).Veterinary Hospital Dissemination of CTX-M-15. Extended-Spectrum B-Lactamase–Producing E. coli ST410 in
the United Kingdom.Microb. Drug Resist. 1; 22(7): 609–615.
20. Villa L et al. (2010). Replicon sequence typing of IncF plasmids carrying virulence and resistance determinants.J Antimicrob Chemother.65:2518-29.
21. Villa, L. and Carattoli (2020). Plasmid Typing and Classification. Springer Nature Switzerland and AG.Vol.2075.pp 309-321.
22.Carattoli, A. (2013). Plasmids and the spread of resistance. Int. J. Med. Microbiol. 303: 298–
23. Liakopoulos, A., Mevius, D.J., Olsen, B., Bonnedahl, J., (2016). Thecolistin resistance mcr-1 gene is going wild. J. Antimic.Chemo.71:2335 .