Next Article in Journal
Development of a Quadruplex RT-qPCR for the Detection of Porcine Rotaviruses and the Phylogenetic Analysis of Porcine RVH in China
Next Article in Special Issue
Assessing the Occurrence of Host-Specific Faecal Indicator Markers in Water Systems as a Function of Water, Sanitation and Hygiene Practices: A Case Study in Rural Communities of Vhembe District Municipality, South Africa
Previous Article in Journal
Phylogeographic Aspects of Bat Lyssaviruses in Europe: A Review
Previous Article in Special Issue
The Fascinating Cross-Paths of Pathogenic Bacteria, Human and Animal Faecal Sources in Water-Stressed Communities of Vhembe District, South Africa
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Pathogens
Volume 12
Issue 9
10.3390/pathogens12091090
pathogens-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles thumb_up
...
Endorse textsms
...
Comment
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Evaluation of a Tetracycline-Resistant E. coli Enumeration Method for Correctly Classifying E. coli in Environmental Waters in Kentucky, USA

by
Callie Boggs
1,
Kidus Shiferawe
1,
Eckhardt Karsten
2,
Jayden Hamlet
3,
S. Travis Altheide
4 and
Jason W. Marion
1,5,*
1
Environmental Health Science and Sustainability Program, Eastern Kentucky University, Richmond, KY 40475, USA
2
Department of Microbiology, Miami University, Oxford, OH 45042, USA
3
School of Natural Sciences and Mathematics, Stockton University, Galloway, NJ 08205, USA
4
Medical Laboratory Science Program, Eastern Kentucky University, Richmond, KY 40475, USA
5
Eastern Scientific LLC, Richmond, KY 40475, USA
*
Author to whom correspondence should be addressed.
Pathogens 2023, 12(9), 1090; https://doi.org/10.3390/pathogens12091090
Submission received: 21 July 2023 / Revised: 20 August 2023 / Accepted: 25 August 2023 / Published: 28 August 2023
(This article belongs to the Special Issue Water-Borne Pathogens)
Browse Figure
Versions Notes

Abstract

:
The global concern over antimicrobial resistance (AMR) and its impact on human health is evident, with approximately 4.95 million annual deaths attributed to antibiotic resistance. Regions with inadequate water, sanitation, and hygiene face challenges in responding to AMR threats. Enteric bacteria, particularly E. coli, are common agents linked to AMR-related deaths (23% of cases). Culture-based methods for detecting tetracycline-resistant E. coli may be of practical value for AMR monitoring in limited resource environments. This study evaluated the ColiGlow™ method with tetracycline for classifying tetracycline-resistant E. coli. A total of 61 surface water samples from Kentucky, USA (2020–2022), provided 61 presumed E. coli isolates, of which 28 isolates were obtained from tetracycline-treated media. Species identification and tetracycline resistance evaluation were performed. It was found that 82% of isolates were E. coli, and 18% were other species; 97% were identified as E. coli when using the API20E identification system. The MicroScan system yielded Enterobacter cloacae false positives in 20% of isolates. Adding tetracycline to ColiGlow increased the odds of isolating tetracycline-resistant E. coli 18-fold. Tetracycline-treated samples yielded 100% tetracycline-resistant E. coli when the total E. coli densities were within the enumeration range of the method. ColiGlow with tetracycline shows promise for monitoring tetracycline-resistant E. coli in natural waters and potentially aiding AMR surveillance in resource-limited settings among other environments.

1. Introduction

Human health risks related to antimicrobial resistance (AMR) are a great global concern with a recent estimate associating 4.95 million deaths in 2019 to antibiotic resistance [1]. A recent assessment from the U.S. Centers for Disease Control and Prevention (CDC) described that the U.S. has had 35,000 deaths from 2.8 million infections from antibiotic-resistant bacteria [2]. While progress reducing the prevalence of AMR infections and mortality has been made in the U.S. in the last decade [2], many nations and regions are having trouble responding to the threat posed [3]. The AMR threat is particularly greater in regions with inadequate water, sanitation, and hygiene [1,4] as evidenced in a recent metagenomic analysis of 1589 fecal metagenomes which demonstrated significantly higher abundances of antimicrobial resistance genes (ARGs) in samples from nations lacking improved water and sanitation [5].
In estimating the most common pathogens linked to AMR-related deaths globally, the enteric bacteria E. coli and Klebsiella pneumoniae represent the top and third most common agents, respectively [5], in which E. coli have been implicated in 23% of the total deaths from antibiotic resistance [1]. The emergence and spread of resistant microbiota or genes from regions lacking improved water and sanitation to clinical environments has already happened [6]. Accordingly, increased densities of resistant microbes are greater in areas needing improvements to water, sanitation, and hygiene systems since these areas experience more waterborne and foodborne illnesses, and thus use more antimicrobial agents for patient and animal care [1,4]. While recommendations for antimicrobial stewardship are promoted, the scientific community advises environmental monitoring of AMR to inform data-driven approaches for guiding interventions [1,6,7,8] through a One Health framework [9,10,11].
The National Antimicrobial Resistance Monitoring System (NARMS) strategic plan for the U.S. recommends that surveillance should begin with surface waters since they integrate differentially affected ecosystems [8]. For a potential target organism to study for enteric bacteria resistance, E. coli may be a good starting point since existing water quality monitoring guidelines for freshwater recreation/bathing and drinking already use E. coli [12,13,14].
The focus of this study is assessing the validity of culture-based method results for tetracycline-resistant E. coli, while recognizing molecular approaches in general are preferred for future methodological monitoring standards [15]. Among these methods, qPCR is one expected to be among future gold-standard molecular methods for quantifying fecal indicator markers and pathogens [16]. Regarding qPCR, while tremendous progress has been made in reducing equipment, software, and consumable costs for fecal indicators and pathogens in water, molecular methods including qPCR present major obstacles for limited resource environments including high costs, specialized instruments, and training/personnel costs [16]. Thus, culture-based methods have some advantages in resource-limited environments with respect to the 20 characteristics identified by Bain et al. [17]. For low-resource settings, interest remains in new approaches for detecting E. coli and quantitatively assessing E. coli density even when results take 24 h [18], and this logic may hold true for AMR monitoring.
In generating a framework for antibiotic resistance monitoring in the water environment, int1, blaCTX-M, sul1, vanA, and tet(A), have been identified as valuable molecular targets [15]. Using these genes to inform culture-based methods, beta-lactams, sulfonamides, vancomycin, and tetracyclines would be potential media additives for assessing resistance. While all candidates have stability in spiked river samples after six days [19], beta-lactams have been reported to degrade in growth media [20]. Among the others, tetracycline stability in the environment has been more frequently raised as a point of concern in the literature [21] and tetracycline resistance among E. coli is common globally [22].
Based upon a presumed monitoring need for limited resource environments, the purpose of this study was to evaluate the classification ability of the ColiGlow™ E. coli enumeration method [23] to properly classify E. coli species identification and antibiotic resistance status upon the addition of an antibiotic. Tetracycline was selected for evaluation as a recent study from an adjacent U.S. state (Ohio) demonstrated that tetracycline resistance would likely be observed in natural waters in the U.S., as has been observed in 8.8% of 329 E. coli isolates from the Maumee River (Ohio, USA) and 27% of over 200 isolates from prairie, cropland, and hay pasture runoff in Texas, USA [24,25]. Findings from this study may have application to culture-based methods for practically assessing the presence/absence or density of tetracycline-resistant culturable E. coli with improved efficiency and lower costs than existing multi-step methods.

2. Materials and Methods

2.1. Study Area and Sample Collection

Surface water samples (n = 61) used for E. coli isolate analysis were collected in Kentucky, USA, during 2020, 2021, and 2022 from a diverse array of surface water types. The 61 samples are a portion of 112 natural water samples that were analyzed including 112 paired tests, whereby 112 tests were performed with tetracycline-treated media, and 112 tests were performed using media without the tetracycline antibiotic. All water samples were collected in sterile Whirl-Pak® bags and processed at Eastern Kentucky University (Richmond, Kentucky, USA) within six hours of collection to facilitate E. coli growth using an E. coli enumeration method (ColiGlow™) in a 96-well plate format for each sample [23].
A total of 61 isolates were evaluated. These 61 isolates were obtained from a random selection of the 96-well ColiGlow plates presenting at least one fluorescing well, whereby a fluorescing well was hypothesized to contain culturable E. coli. For assessing whether these fluorescing wells contained culturable E. coli, the 61 isolates were obtained for the purpose of evaluating the correct or incorrect classification of the E. coli species identification and tetracycline susceptibility.
To increase sample diversity and independence, only one isolate was obtained from each of these 61 selected ColiGlow plates. For plates with more than one fluorescing well, only one of the fluorescing wells in the 96-well plate was used to obtain an isolate. Therefore, 61 total isolates were obtained in this study which were randomly selected from 61 ColiGlow plates presenting presumed E. coli growth among the 224 total plates used. Given that ColiGlow plates containing tetracycline-treated samples had many samples with no growth (no fluorescing wells) there were fewer plates to randomly select for isolate evaluation. Ultimately, 33 isolates were obtained from 33 ColiGlow plates with no tetracycline in the media, and 28 isolates were obtained from 28 ColiGlow plates with tetracycline in the media.
Among the 61 samples used for obtaining isolates, 20 samples were collected in 2020 from central Kentucky (Madison County) including 16 samples from free-flowing streams in a cattle-producing region, two samples from a lake, and two samples from roadside ditches associated with cow pasture runoff. In 2021, eleven samples were collected in central and southeast Kentucky with five samples from free-flowing streams impacted by surface mining for coal, five samples from free-flowing streams not impacted by mining, and one sample from a roadside ditch associated with cow pasture runoff. In 2022, 30 samples were collected in central and eastern Kentucky with 15 samples being free-flowing streams in heavily forested areas of the Daniel Boone National Forest and 15 samples being from a mixture of urban and residential free-flowing streams in central Kentucky.

2.2. Obtaining Presumed E. coli and Tetracycline-Resistant E. coli Isolates

Each water sample was processed in accordance with a standard procedure for the ColiGlow test method. In brief, samples were processed in media with and without the tetracycline antibiotic. Specifically, 22.5 mL of the water sample was added to a 50 mL tube containing 2.5 mL of the liquid culture media for the selective differential growth of E. coli. Then, another 22.5 mL of the sample was added to a different tube containing the growth media plus either 320 µg (half tetracycline) or 640 µg of tetracycline (Fisher Scientific, Fair Lawn, New Jersey, USA, product code: BP912-100). The half tetracycline concentrations used were intended to increase test sensitivity while enabling the enumeration of E. coli with intermediate resistance.
The sample mixtures were then inverted at least 30 times, poured into a multi-channel pipette reservoir, and then distributed into 96-well plates with a multi-channel pipette with 200 microliters being placed in each well. The prepared 96-well plates were covered and incubated at 35 °C for 24 h. Following incubation, the plates were viewed under a longwave ultraviolet light and fluorescing wells were counted as positive wells for presumed E. coli growth due to their β-D-glucuronidase enzymatic activity cleaving the 4-methylumbelliferyl-β-D-glucuronide (MUG) included in the growth media to produce the fluorogenic compound 4-methylumbelliferone (4-MU). Using the number of positive wells in conjunction with a most probable number (MPN) table provided with the ColiGlow test, the MPN was estimated per 100 mL of sample within a range. The range of the test kit is 14–1479 MPN per 100 mL for 1 glowing well to 95 glowing wells. A negative test was reported as <14 MPN per 100 mL and a test with 96 glowing wells was reported as overrange (>1479 MPN per 100 mL). An example of the ColiGlow method with fluorescing wells in the media with and without tetracycline is presented in Figure 1.
For obtaining isolates, a sterile 10 µL calibrated loop was dipped in one fluorescing well from a positive plate and then inoculated using a three-phase streak technique onto a modified membrane-Thermotolerant E. coli (modified mTEC) agar plate. This process was performed for the 61 selected plates with fluorescing wells. Following inoculation, the modified mTEC plates were incubated at 2 h for 35 °C and then 22 h at 44.5 °C in accordance with the incubation temperatures used in U.S. EPA Method 1603 [26]. Following incubation, the magenta-colored colonies were presumed to be E. coli [26]. The streak technique enabled the growth of many isolated colonies per plate, and among these colonies, one colony was selected at random from all the colonies present on each modified mTEC agar plate to be used for subsequent species identification and tetracycline resistance evaluation. If no magenta-colored isolate appeared following incubation on modified mTEC agar, then a randomly selected isolate from the colonies on the plate was used for subsequent evaluation of tetracycline resistance and species identification to discern which species was associated with the fluorescence in the 96-well plate.
Similar research evaluating the specificity of Aquatest media with RUG™ for detecting, enumerating, and properly classifying E. coli used between one and five isolates per plate based upon colony morphology type [27]. In this study, only one isolate per sample (per mTEC plate) was used, as the laboratory resources required for subsequent antibiotic susceptibility analysis and species identification per isolate are substantial. Like similar research [27], isolates most likely to be E. coli were selected.

2.3. Species Identification

During 2020 and 2021, species identification was performed using API20E strips (bioMerieux, Marcy l’Etoile, France) in accordance with the manufacturer’s instructions using one API20E strip per isolate. In 2022, the university acquired new instrumentation allowing species identification to be performed using a Beckman Coulter MicroScan® (Brea, CA, USA) with one Gram-negative Urine Combo 85 panel per isolate. The newer technology permitted simultaneous species identification and antibiotic susceptibility testing.
Prior to inoculating the API20E strip and MicroScan panels, each isolate was inoculated from modified mTEC to blood agar and incubated for 18 h. A fresh isolate from blood agar was then inoculated into the API20E strip and MicroScan panels in accordance with the manufacturer’s instructions.
Both identification methods (API20E and MicroScan) utilize a combination of biochemical tests during and after an 18 h incubation that generate color changes which provide a practical means for the identification of Gram-negative bacteria and members of the Enterobacteriaceae when compared against each manufacturer’s reference library. The MicroScan panels are advantageous over the API20E strips due to less human error and reliance on an automated microplate reader attached to PC and software [28,29]. The API20E strips rely upon an approach developed in the 1970s [28] that has been demonstrated recently to be error prone due to having a limited library of species and strains [29].

2.4. Assessment of Tetracycline Susceptibility/Resistance

Assessment of tetracycline resistance occurred using minimum inhibitory concentration (MIC) methods. The MIC value obtained for each isolate in the presence of tetracycline was used to determine resistant isolates; whereby isolates with a tetracycline MIC > 16 µg/mL were deemed resistant, ≤4 µg/mL were susceptible, and the intermediate breakpoint was 8 µg/mL [30].
During 2020 and 2021, tetracycline MICs were obtained using tetracycline E-strip methods (bioMerieux, Marcy l’Etoile, France). Specifically, an inoculum (0.5 McFarland standard) was prepared for each isolate evaluated using the colonies from the same blood agar plates that were also used for species identification. Each inoculum was swabbed onto Mueller–Hinton agar to promote the growth of lawns. Immediately following plate inoculations, the E-strip containing a gradient of concentrations of tetracycline was placed on the media surface of each plate. The inoculated plates with the E-strips were then incubated for 24 h at 35 °C. Following incubation, MIC values were obtained by reading the printed concentration on the strip at the intersection with the zone of inhibition [31].
In 2022, tetracycline resistance was assessed using the MIC method that co-occurred with species identification procedures using a Beckman Coulter MicroScan® (Brea, CA, USA) Urine Combo 85 panel. In addition to the biochemical tests, the Urine Combo 85 panel obtained the MIC values for several antibiotics including tetracycline.
While two methods were used for tetracycline susceptibility testing, both methods presumably performed comparably. While comparisons of these methods for tetracycline susceptibility testing are not readily described in the literature, other comparisons exist. In the case of gentamicin susceptibility testing among Enterobacterales [32], Colistin resistance testing among E. coli [33], and ertapenem susceptibility among Enterobacteriaceae [34], the two methods used in this study provide similar results in those comparison studies.

2.5. Statistical Analysis

All data analyses were performed with Stata 14 [35]. Data analyses included summary statistics (mean, median, standard deviation) and cross-tabulations. Statistical tests included Chi-square and Fisher’s Exact tests for comparing frequency differences in the cross-tabulations; whereby Fisher’s Exact tests were used if any cell in the cross-tabulation had less than five observations. Logistic regression was used for obtaining odds ratios. Following logistic regression, model discrimination was assessed using the area under the ROC curve (AUC) [36].

3. Results and Discussion

3.1. Species Identification of Isolates

Among the 61 isolates investigated, 50 (82%) were identified as E. coli (Table 1). Among the eleven (18%) isolates not identified as E. coli, six were identified as Enterobacter cloacae. The other five isolates were identified as Kluyvera ascorbata, Kluyvera intermedia, Klebsiella pneumoniae, Serratia odorifera, and Citrobacter braakii. Variables potentially impacting species identification include the species identification method (API20E versus MicroScan) and the total density of E. coli and other bacteria in the sample associated with each isolate.
Among the presumed E. coli-positive (fluorescing) wells from the ColiGlow plates, one isolate was obtained per well. Since only one isolate was obtained per well, the interpretation should also consider the possibility that E. coli may also have been present, but by chance was not picked as a colony from among the many isolated colonies that appeared on the modified mTEC agar plates following inoculation from the positive wells from the ColiGlow plates. As a study limitation, since only one colony was obtained from each modified mTEC plate, there was a possibility that E. coli were among the many colonies capable of growth in the media from the ColiGlow method and on modified mTEC agar. In similar research evaluating Aquatest containing the novel substrate RUG™, upon inoculating fluorescing wells on modified mTEC agar, numerous colonies were observable following incubation, and in their study, as many as five morphologies could be present [27]. In that study, one isolate of each colony morphology was evaluated. In this study evaluating ColiGlow, the most likely candidate isolate (magenta colony) was selected given the practical limitations of testing the tremendous diversity and number of all isolated colonies that were growing on the mTEC agar.
For enhancing the likelihood of recovering E. coli isolates, using lessons learned from prior research [27], less differential media (MacConkey agar) was not used and modified mTEC agar was used, coupled with the elevated incubation temperature (44.5 °C) for obtaining thermotolerant isolates on the modified mTEC agar. This approach was consistent with studies evaluating Colilert-18™ [27,36] and the novel substrate RUG™ [27] as a means of isolating E. coli. Among all the isolates obtained in this study, only one (C. braakii) did not present magenta on the modified mTEC agar, but instead presented as a white colony with no magenta colonies present. In that case, it was presumed that the isolate was not likely E. coli as the modified mTEC agar colonies should present with magenta or red coloration to indicate β-D-glucuronidase enzymatic activity, which is the same enzyme associated with fluorescence in positive wells from the ColiGlow method as well as other E. coli identification/enumeration methodologies including Colilert-18 and AquaTest-RUG [27,37]. Specifically, the modified mTEC agar contains the chromagen 5-bromo-6-chloro-3-indolyl-β-D-glucuronide [26], which produces the magenta color in colonies exhibiting β-D-glucuronidase activity which catabolizes the compound [38,39].
It is noteworthy that while most (92% to 96%) E. coli from water have been reported to have β-D-glucuronidase activity within 24 h [40,41], there are some E. coli that do not express this enzyme and would likely not be detected by the ColiGlow method, modified mTEC plate, or other methods relying on β-D-glucuronidase to differentiate E. coli such as Colilert and AquatTest-RUG, among others [27,37]. A possible reason for isolating microorganisms that were not E. coli, listed in Table 1, is that β-D-glucuronidase has been observed in other microorganisms, including many of the Enterobacteriaceae, among others [42,43]. These organisms would be capable of growing in ColiGlow media and could present as false positives if E. coli were also not co-located from the sample. Table 1 presents six species other than E. coli that were isolated, and like evaluations of Aquatest-RUG, Colilert-18, and Aquatest, E. cloacae and K. pneumoniae were isolated [27,37,44]. Other research has observed detections of K. ascorbata [45], Citrobacter spp. [46], E. cloacae [46], and Klebsiella spp. [46] from water or wastewater samples containing Colilert media and other media using similar enzymatic differentiation [47].

3.2. Tetracycline Impact on E. coli Selection and Species Identification

In the tetracycline treated media, 28 (100%) of 28 were identified as E. coli versus 22 (67%) of 33 from the regular media (Table 1). There was a significant difference in the successful E. coli selection and identification frequency between the isolates obtained from the tetracycline-treated media versus the media without tetracycline (Fisher’s exact p = 0.001). The tetracycline in the media likely acted as an inhibitor or eliminated tetracycline susceptible bacteria (Figure 1), which would have included susceptible E. coli. Tetracycline has been used in some growth media to improve selectivity [48], and antibiotics in general are among the most used selective agents [49] possessing abilities to greatly reduce the diversity of organisms [50] limiting possible co-dependent organisms. Exploring this hypothesis, samples with fewer E. coli in the original source water were associated with a greater likelihood of an isolate selected that was correctly classified as E. coli. A plausible explanation for 100% E. coli recovery and isolate identification in the tetracycline-treated samples with growth is that by having less microbial diversity and a reduction in the total microbial load due to tetracycline, the treatment differentially imperiled the survival, growth, and/or selection of the non-target (non-E. coli) species.

3.3. Microbial Load and Likelihood of Selecting Non-Target Species as Isolates

When examining the relationship between the number of fluorescing ColiGlow wells (presumed to contain E. coli) and the likelihood of obtaining an E. coli isolate, the ColiGlow plates that had all 96 wells fluorescing were significantly more likely to have an isolate selected other than E. coli relative to the ColiGlow plates with less growth (Fisher’s Exact Test p = 0.005). Specifically, 91% of isolates were identified as E. coli when the 96-well plate did not have all 96 wells fluorescing (≤1479 MPN per 100 mL) versus 63% when all 96 wells of the ColiGlow plate were fluorescing (>1479 MPN per 100 mL).
In comparison with other studies, using the same or similar enzyme-substrate as the ColiGlow method, the likelihood of false positives increased when non-target bacteria were in greater abundance [51,52,53,54] resulting in the recommendation [51] or actual use of substantial dilutions to enhance successful E. coli recovery for reducing the number of false positives [52]. When comparing the results in Table 2 with other studies, 11 (18%) of 61 isolates were not classified as E. coli, which if treated as false positives, would be higher than recent studies using Colilert-18, Aquatest, and Aquatest-RUG in temperate and subtropical waters that also used modified mTEC agar for colony isolation [27,44]. Specifically, those studies had false positive rates below 5%. Other studies, using original (non-modified) mTEC agar as isolation media for Colilert-18 had false positive rates of 7.4% and 36%, respectively [27,54].
When comparing the potential false positive rate from the ColiGlow method when the E. coli density estimates were within the range of the method (less than 96 wells glowing), the likelihood of recovering a non-E. coli species was 9%, which was closer to the false-positive rate in the more recent studies using Colilert-18 and related methods [27,44]. It is plausible that E. coli were also co-located in the glowing wells but were not the selected colony from the modified mTEC plates used for isolate analysis.

3.4. Evaluation of Tetracycline Treatment for Screening Tetracycline-Resistant E. coli

Among 28 isolates obtained from the ColiGlow plates containing tetracycline that were evaluated for tetracycline resistance, 25 (89%) were tetracycline resistant and three (11%) were susceptible to tetracycline. When examining the three that were susceptible to tetracycline that were recovered and isolated from tetracycline-treated media, these three isolates were from ten tests that used 320 µg per 25 mL (12.8 µg/mL) versus the 640 µg per 25 mL (25.6 µg/mL). When the higher concentration was examined, 18 (100%) of 18 isolates from the higher tetracycline treatment group were tetracycline resistant versus seven (70%) of ten isolates from the treatment using half the tetracycline dose per sample, which represents a significant difference (Fisher’s Exact Test p = 0.037).
Research evaluating if these antibiotic concentrations are most appropriate remains limited when examining resistance using complex environmental matrices such as contaminated surface waters. Several studies enumerating E. coli when tetracycline and other antibiotics have been added to Colilert were carried out from 2005 to 2010 with Ohio River (USA) surface water, [55] municipal and hospital wastewater in Ireland [56], Mud Creek (Fayetteville, Arkansas, USA) surface water impacted by wastewater effluent [57], and recently using irrigation waters on a commercial farm in Maryland, USA. [58]. The Irish study supplemented Colilert with tetracycline to achieve a concentration of 4 µg/mL and reported that 40 (100%) of 40 isolates obtained from the positive Colilert tests were tetracycline resistant. The Arkansas study used concentrations of 4, 8, and 16 µg/mL. For this research using the ColiGlow method on Kentucky waters, the lower tetracycline concentration (12.8 µg/mL) had a false-positive rate of 30% for tetracycline resistance. Differences between study methods may have contributed to the discrepancy as the Irish study used significant dilutions for ensuring samples were within the test ranges of Colilert, and by using dilutions, the dilutions reduced the complexity of the aquatic matrix which may have optimized antibiotic effectiveness. Additionally, details on the tetracycline resistance breakpoint value used in that study were not apparent, limiting comparability.
A recent (2023) Maryland, USA, study was supplemented with 4 µg/mL (low dose) and 16 µg/mL (high dose) of tetracycline to be consistent with the Irish study [56], and current Clinical Laboratory Standards Institutes breakpoint values [59], that also aligns with the 2021 NARMS Interpretive Criteria for Susceptibility Testing [30]. In comparing low- and high-dose tetracycline treatment, significantly more growth occurred in the lower-dose samples. There was no discussion in that study on evaluating isolates for tetracycline resistance between the no dose, low-dose, and high-dose treatment groups; however, opportunities for future research were articulated.
In this study, due to the breakthrough of tetracycline susceptible organisms in the lower-dose group, the higher tetracycline concentration was subsequently used to decrease the false-positive rate. Earlier research examining tetracycline resistance in dairy cattle feces using agar-based methods that supplemented MacConkey agar with 32 µg/mL specifically used a “twofold-higher concentration” [60] of the MIC breakpoint [61]. When concentrations are used below the MIC, susceptible bacteria have been reported to survive and/or use bacterial SOS repair systems that promote horizontal gene transfer and genome mutation [62]. Accordingly, a susceptible microorganism that survived in a sub-MIC growth media could be selected and grown on the modified mTEC agar-lacking tetracycline, and then present tetracycline susceptibility when challenged by higher tetracycline concentrations in a MIC-based susceptibility analysis. Alternatively, susceptible E. coli bacteria surviving in sub-MIC conditions can obtain resistance genes from non-target organisms thereby having resistance induced when using sub-MIC conditions of the growth media. The sub-MIC concentrations of 1 to 15 µg/mL for tetracycline have been associated with increasing the abundance of resistance genes [63,64] which supports a perspective of attempting to maintain >15 µg/mL when evaluating tetracycline resistance among E. coli. Future guidelines or standards for tetracycline concentrations in screening media are recommended [15,54]. Levels in the range of >15 µg/mL [63,64] to 32 µg/mL [60] are likely needed, depending upon the density of bacteria and interfering substances in the sample. In this study, the results were mostly obtained from media containing a relatively high tetracycline concentration that would be 25.6 µg/mL when mixed with pure water, which would decrease the likelihood of sub-MIC conditions. Actual tetracycline concentrations were not measured after adding natural water samples. If measures were possible, greater understanding of how tetracycline concentrations vary overtime with samples containing varying levels of contamination may be useful for informing guidelines for media standards containing tetracycline or other antibiotics.
Overall, the use of tetracycline significantly enhanced the likelihood of isolating tetracycline-resistant E. coli relative to the probability of obtaining a resistant isolate from among ColiGlow plates with media that lacked tetracycline. Specifically, 25 (89%) of the 28 E. coli isolates from the tetracycline-treated ColiGlow plates exhibited tetracycline resistance versus 15 (32%) of the 22 E. coli isolates obtained from ColiGlow plates without tetracycline treatment (Chi-Square p < 0.001). In comparison of all isolates, including non-E. coli isolates, the difference in obtaining a tetracycline-resistant isolate was also significant between the tetracycline-treated and no treatment groups (Chi-Square p < 0.001).
Among the 50 E. coli isolates evaluated in this study, the odds of observing a tetracycline-resistant E. coli isolate were 17.9 times higher when tetracycline was used in the media versus when it was not used (Odds Ratio = 17.9; 95% C.I.: 4.0–79.7). The area under the ROC curve was 80.7%, which is on the low end of the excellent discrimination range (80−90%) for a test [32]. Studies attempting to enumerate culturable tetracycline-resistant E. coli or obtain isolates would benefit from the inclusion of tetracycline in selection media.

3.5. E. coli Densities and Likelihood of Obtaining Tetracycline-Resistant Isolates

Among the 61 isolates evaluated in this study, 19 were obtained from ColiGlow plates associated with a water sample that exceeded the E. coli density of the ColiGlow test range (>1479 MPN per 100 mL). Table 3 demonstrates that the prevalence of tetracycline resistance (84%) among the isolates associated with the highest densities of E. coli was significantly higher than the prevalence (55%) in isolates associated with E. coli densities within the range of the ColiGlow method (Fisher’s Exact Test p = 0.043). Among the 50 water samples that led to the successful recovery of E. coli isolates, 12 (100%) of the 12 isolates selected from samples associated with a ColiGlow test that exceeded the test range were resistant, which was significantly higher than the frequency of resistance among samples associated with in-range E. coli densities (Fisher’s Exact Test p = 0.002).
Consistent with prior research, increasing E. coli densities were associated with a greater likelihood of observing increased tetracycline resistance among culturable E. coli [57], potentially driven by precipitation and runoff. Numerous studies have established relationships between increased densities of molecular markers of tetracycline resistance following precipitation events, particularly in urban and agricultural landscapes [65,66,67,68,69]. While increases in the E. coli densities likely increase the diversity of the E. coli populations which may contain antibiotic-resistant types, an increase in the overall amount of microbial, soil, and chemical contaminants could interfere with antibiotics and reduce the tetracycline exposure concentrations at the microbe level in the media to sub-MIC concentrations which could enable susceptible organisms to survive and/or induce resistance, rather than detect existing resistance, from within the sampled natural environment [60,61].

3.6. Challenges for Interpreting and Generalizing Species Identification Results

The study results are based upon a limited number of isolates (n = 61) obtained from 61 individual 96-well plates that were selected from among all the plates with presumed E. coli growth. Future studies evaluating recovered isolates from fecal indicator bacteria detection and enumeration media should consider examining multiple recovered isolates to strengthen the study design.
Another limitation of the study pertains to study results being aggregated from two different identification methods. While a limitation, the findings also generated additional research questions pertaining to identification methods. When limiting the species identification analyses to the results obtained by the API20E method, 30 (97%) of 31 were identified as E. coli which is comparable to the results reported for Aquatest-RUG (97%), modified mTEC agar with membrane filtration (97%), and Colilert-18 (98.5%) when the API20E identification method was used in related research [27]. When using the identification results from the MicroScan panels, 20 (67%) of 30 were identified as E. coli (Table 4). The difference in these results is significant (Fisher Exact Test p = 0.003) and may reflect either improved sensitivity in the MicroScan method or a significantly greater abundance of false-positive species in those samples. MicroScan methods are modern advancements relative to the API20E identification approach, and in early editions of the MicroScan methodology, the technology outperformed the API20E method by correctly identifying E. coli in 95% of samples versus 84% in the parallel comparison study [70]. In a recent study of isolates obtained from Colilert-18, the API20E only correctly classified 50% of the isolates from the Enterobacter genus [46]. The most frequent non-E. coli species detected in our study with the MicroScan method was E. cloacae, representing six (20%) of thirty isolates evaluated. E. cloacae was not detected in any of 31 isolates assessed by the API20E identification system, and while speculative, this organism may have been incorrectly classified as E. coli in the API20E tests; however, no side-by-side comparison was completed.
Beyond MicroScan and API20E identification methods, molecular or MALDI-TOF approaches are promising. Molecular approaches targeting and/or quantifying the uidA may have a similar false-positive detection as experienced in culture-based assays as the uidA gene encodes for beta-glucuronidase [38] which is responsible for the fluorescence observed in this study and other enzyme-substrate tests such as Colilert. The uidA gene can exist in E. cloacae, K. pneumonia, and other coliforms [40,71]. For improving identification, multiplex PCRs [71], as well as 16S rRNA and/or MALDI-TOF approaches have been recommended due to their high accuracy and reliability [72].

3.7. Value of Applying Tetracycline-Treated Culture-Based E. coli Detection Methods

The use of tetracycline in this study reduced the total number of tetracycline-susceptible E. coli and other susceptible bacteria. The three-year study that led to the study of 61 isolates included 224 ColiGlow plates, which included 112 plates containing media treated with tetracycline. Among these ColiGlow plates with tetracycline in the media, 63 (56%) of the 112 samples had no growth. Conversely, when using the regular ColiGlow method, where tetracycline was not in the media, only seven (6.25%) of those one hundred and twelve samples had no detectable E. coli growth. Together, these results indicate that when tetracycline was used in the media, there was a significant reduction in the likelihood of observing any E. coli growth (Chi-Square p < 0.001) by presumably eliminating or inhibiting the growth and/or enzymatic expression among the tetracycline susceptible E. coli. Using this information, along with the confirmation of tetracycline resistance in nearly all isolates when tetracycline was used, these data support the value of the ColiGlow method with tetracycline for screening and detecting tetracycline-resistant E. coli from water samples when the samples have E. coli densities within the range of the method. When water samples are expected to exceed the range, such as following wet weather events, the use of dilutions would likely improve the classifications for true E. coli and tetracycline resistance [52].
Practical methods for enumerating tetracycline-resistant E. coli may enhance local surveillance in limited resource settings for similar reasons needed for global water quality monitoring using total E. coli. Specifically, regions experiencing the greatest threats to water safety stand to benefit from low-cost and field-ready methods that can be employed [17,18,27], and limited resource environments have substantial overlap with areas experiencing the greatest amount of pressure from waterborne illness, mortality, and emerging antibiotic resistance [1,3,4,5,7]. Additional environments with limited laboratory resources may include agricultural areas where irrigation waters and livestock-impacted waters could also be actively monitored [58].
In addition to the practicality of enumeration methods, culturable E. coli and tetracycline-resistant E. coli enable the selection of E. coli isolates that can be studied, further enabling a comprehensive and simultaneous characterization of susceptibility to multiple antibiotics beyond tetracycline, while also enabling molecular approaches to examine or identify resistance genes specific to these isolates [73]. Ultimately, approaches using molecular- and/or culture-based methods are needed and valued for environmental AMR surveillance and the value of each remains dependent on the purpose of the surveillance [73,74]. Research applications using existing methods for quantifying both total E. coli and tetracycline-resistant E. coli densities to obtain the percentages of resistant E. coli have been proposed for informing how water quality parameters or contaminants may influence the relative abundance of tetracycline-resistant E. coli versus total E. coli [58].

4. Conclusions

The odds of isolating a tetracycline-resistant E. coli increased nearly 18-fold when tetracycline was used in the ColiGlow method relative to when tetracycline was not added. The addition of tetracycline to the media at the concentrations used in this study resulted in 100% of the E. coli isolates also being observed as tetracycline-resistant when the total E. coli density in the sample water was within the range of the enumeration method (<1479 MPN per 100 mL). The successful recovery of E. coli isolates and tetracycline-resistant isolates diminished when samples had elevated densities of E. coli and other enteric bacteria that also express β-D-glucuronidase activity. The use of dilutions aimed at achieving densities within the enumeration range of the method may enhance the discrimination of the method for correctly classifying E. coli and tetracycline resistance. The species identification methods used in this study may have led to misclassification and future studies aimed at identifying bacteria isolated from media using β-D-glucuronidase enzymatic should consider MALDI-TOF or molecular identification methods. Overall, these results demonstrate that the ColiGlow method, with and without tetracycline, has acceptable discrimination for identifying E. coli and tetracycline-resistant E. coli, respectively. In tandem, these methods and related methods have value for surveillance efforts aimed at understanding how environmental factors influence the densities and spatiotemporal distributions of tetracycline-resistant culturable E. coli in natural waters.

Author Contributions

C.B. was responsible for writing—original draft preparation, investigation -field and laboratory data collection in year 1, data curation—aggregation, and formal analysis—data analysis; K.S. was responsible conceptualizing the study, investigation—field and laboratory data collection in year 1, and data curation in year 1; E.K. was responsible for investigation—field and laboratory data collection in year 2 and data curation in year 2; J.H. was responsible for investigation—field and laboratory data collection in year 3 and data curation in year 3; S.T.A. was responsible for conceptualizing the study, supervision—laboratory management for species identification and antibiotic susceptibility data collection efforts in all years, supporting the development of the methodology, and provision of resources for laboratory methods; J.W.M. supervised all studies, conceptualized all studies, was responsible for final draft preparation, and managed all studies. All authors have read and agreed to the published version of the manuscript.

Funding

This research was funded in 2020 by the Eastern Kentucky University Board of Regents’ Board Innovation Fund and the research was funded in 2021 and 2022 by the National Science Foundation (NSF) Division of Biological Infrastructure, Award Number 1950355, Research Experiences for Undergraduates Site: Disturbance Ecology in Central Appalachia, PI: David Brown, Co-PI: Kelly Watson.

Institutional Review Board Statement

No human participants were involved in this study.

Data Availability Statement

Raw data available upon request.

Acknowledgments

We are thankful to Larissa Watkins and Nickson Rotich who aided in the development of the field and laboratory protocols while at Eastern Kentucky University (EKU). We are thankful to Olivia Salazar from the University of Virginia at Wise for her support in data collection efforts with J.H. during the NSF-REU program. We are thankful to Tom Martin, the EKU Center for Economic Development, Entrepreneurship and Technology, and the 2018-19 EKU Board of Regents for enabling the project. We are grateful for the administrative support provided by Sarah Rose and Ismail El-Amouri in the EKU College of Health Sciences.

Conflicts of Interest

C.B., K.S., E.K., J.H. and T.S.A. declare no conflict of interest. J.W.M. declares being the inventor on an international PCT application and pending U.S. patent application by Eastern Kentucky University for a method for the detection of E. coli and antibiotic-resistant bacteria in water. J.W.M. declares being the principal for Eastern Scientific LLC, which is licensed by Eastern Kentucky University to commercialize the method. Eastern Scientific LLC has a pending U.S. trademark application for ColiGlow™. The funders of the research had no role in the design of the study, in the collection, analyses, interpretation of data, writing of the manuscript, or in the decision to publish the results. The content provided here represents the original findings of the authors and does not necessarily represent the views or interpretations of Eastern Kentucky University, Miami University of Ohio, Stockton University, or the National Science Foundation.

References

  1. Murray, C.J.; Ikuta, K.S.; Sharara, F.; Swetschinski, L.; Aguilar, G.R.; Gray, A.; Han, C.; Bisignano, C.; Rao, P.; Wool, E.; et al. Global burden of bacterial antimicrobial resistance in 2019: A systematic analysis. Lancet 2022, 399, 629–655. [Google Scholar] [CrossRef]
  2. CDC. Antibiotic Resistance Threats in the United States; U.S. Department of Health and Human Services, CDC: Atlanta, GA, USA, 2019. Available online: https://www.cdc.gov/drugresistance/pdf/threats-report/2019-ar-threats-report-508.pdf (accessed on 11 July 2023).
  3. Patel, J.; Harant, A.; Fernandes, G.; Mwamelo, A.J.; Hein, W.; Dekker, D.; Sridhar, D. Measuring the global response to antimicrobial resistance, 2020–2021: A systematic governance analysis of 114 countries. Lancet Infect. Dis. 2023, 23, 706–718. [Google Scholar] [CrossRef] [PubMed]
  4. Ashbolt, N.J.; Pruden, A.; Miller, J.H.; Riquelme, M.V.; Maile-Moskowitz, A. Antimicrobial resistance: Fecal sanitation strategies for combatting a global public health threat. In Global Water Pathogens Project; Part 3 Bacteria; Pruden, A., Ashbolt, N., Miller, J., Eds.; UNESCO and Michigan State University: East Lansing, MI, USA, 2018; Available online: https://www.waterpathogens.org/book/antimicrobal-resistance-fecal-sanitation-strategies-combatting-global-public-health-threat (accessed on 12 July 2023).
  5. Fuhrmeister, E.R.; Harvey, A.P.; Nadimpalli, M.L.; Gallandat, K.; Ambelu, A.; Arnold, B.F.; Brown, J.; Cumming, O.; Earl, A.M.; Kang, G.; et al. Evaluating the relationship between community water and sanitation access and the global burden of antibiotic resistance: An ecological study. Lancet Microbe 2023, in press. [Google Scholar] [CrossRef] [PubMed]
  6. Walsh, T.R.; Weeks, J.; Livermore, D.M.; Toleman, M.A. Dissemination of NDM-1 positive bacteria in the New Delhi environment and its implications for human health: An environmental point prevalence study. Lancet Infect. Dis. 2011, 11, 355–362. [Google Scholar] [CrossRef]
  7. Hendriksen, R.S.; Munk, P.; Njage, P.; Van Bunnik, B.; McNally, L.; Lukjancenko, O.; Röder, T.; Nieuwenhuijse, D.; Pedersen, S.K.; Kjeldgaard, J.; et al. Global monitoring of antimicrobial resistance based on metagenomics analyses of urban sewage. Nat. Comm. 2019, 10, 1124. [Google Scholar] [CrossRef]
  8. FDA; CDC; USDA. The National Antimicrobial Resistance Monitoring System: Strategic Plan 2021–2025; U.S. Department of Health and Human Services, FDA: Silver Spring, MD, USA, 2021. Available online: https://www.fda.gov/media/79976/download (accessed on 13 July 2023).
  9. White, A.; Hughes, J.M. Critical importance of a one health approach to antimicrobial resistance. EcoHealth 2019, 16, 404–409. [Google Scholar] [CrossRef]
  10. Singh, K.S.; Anand, S.; Dholpuria, S.; Sharma, J.K.; Blankenfeldt, W.; Shouche, Y. Antimicrobial resistance dynamics and the one-health strategy: A review. Environ. Chem. Lett. 2021, 19, 2995–3007. [Google Scholar] [CrossRef]
  11. Booton, R.D.; Meeyai, A.; Alhusein, N.; Buller, H.; Feil, E.; Lambert, H.; Mongkolsuk, S.; Pitchforth, E.; Reyher, K.K.; Sakcamduang, W.; et al. One Health drivers of antibacterial resistance: Quantifying the relative impacts of human, animal and environmental use and transmission. One Health 2021, 12, 100220. [Google Scholar] [CrossRef] [PubMed]
  12. WHO. WHO Recommendations on Scientific, Analytical and Epidemiological Developments Relevant to the Parameters for Bathing Water Quality in the Bathing Water Directive (2006/7/EC)—Final Report; World Health Organization: Geneva, Switzerland, 2018. Available online: https://cdn.who.int/media/docs/default-source/wash-documents/who-recommendations-on-ec-bwd-august-2018.pdf?sfvrsn=5c9ce1e0_6 (accessed on 13 July 2023).
  13. U.S. EPA. Recreational Water Quality Criteria; U.S. Environmental Protection Agency Office of Water: Washington, DC, USA, 2012. Available online: https://www.epa.gov/sites/default/files/2015-10/documents/rwqc2012.pdf (accessed on 13 July 2023).
  14. WHO. Guidelines for Drinking-Water Quality—Fourth Edition Incorporating the First and Second Addenda; World Health Organizaton: Geneva, Switzerland, 2022. Available online: https://www.who.int/publications/i/item/9789240045064 (accessed on 13 July 2023).
  15. Liguori, K.; Keenum, I.; Davis, B.C.; Calarco, J.; Milligan, E.; Harwood, V.J.; Pruden, A. Antimicrobial resistance monitoring of water environments: A framework for standardized methods and quality control. Environ. Sci. Technol. 2022, 56, 9149–9160. [Google Scholar] [CrossRef]
  16. Paruch, L. Molecular Diagnostic Tools Applied for Assessing Microbial Water Quality. Int. J. Environ. Res. Public Health 2022, 19, 5128. [Google Scholar] [CrossRef]
  17. Bain, R.; Bartram, J.; Elliott, M.; Matthews, R.; McMahan, L.; Tung, R.; Chuang, P.; Gundry, S. A summary catalogue of microbial drinking water tests for low and medium resource settings. Int. J. Environ. Res. Public Health 2012, 9, 1609–1625. [Google Scholar] [CrossRef] [PubMed]
  18. Brown, J.; Bir, A.; Bain, R.E. Novel methods for global water safety monitoring: Comparative analysis of low-cost, field-ready E. coli assays. Npj Clean Water 2020, 3, 9. [Google Scholar] [CrossRef]
  19. Dinh, Q.T.; Alliot, F.; Moreau-Guigon, E.; Eurin, J.; Chevreuil, M.; Labadie, P. Measurement of trace levels of antibiotics in river water using on-line enrichment and triple-quadrupole LC–MS/MS. Talanta 2011, 85, 1238–1245. [Google Scholar] [CrossRef]
  20. Brouwers, R.; Vass, H.; Dawson, A.; Squires, T.; Tavaddod, S.; Allen, R.J. Stability of β-lactam antibiotics in bacterial growth media. PLoS ONE 2020, 15, e0236198. [Google Scholar] [CrossRef]
  21. Amangelsin, Y.; Semenova, Y.; Dadar, M.; Aljofan, M.; Bjørklund, G. The Impact of Tetracycline Pollution on the Aquatic Environment and Removal Strategies. Antibiotics 2023, 12, 440. [Google Scholar] [CrossRef] [PubMed]
  22. Poirel, L.; Madec, J.Y.; Lupo, A.; Schink, A.K.; Kieffer, N.; Nordmann, P.; Schwarz, S. Antimicrobial resistance in Escherichia coli. Microbiol. Spectr. 2018, 6, 4. [Google Scholar] [CrossRef]
  23. Eastern Scientific. ColiGlow; Eastern Scientific: Richmond, KY, USA, 2023; Available online: http://www.coliglow.com (accessed on 14 July 2023).
  24. Mukherjee, M.; Gentry, T.; Mjelde, H.; Brooks, J.P.; Harmel, D.; Gregory, L.; Wagner, K. Escherichia coli antimicrobial resistance variability in water runoff and soil from a remnant native prairie, an improved pasture, and a cultivated agricultural watershed. Water 2020, 12, 1251. [Google Scholar] [CrossRef]
  25. Mukherjee, M.; Marie, L.; Liles, C.; Mustafa, N.; Bullerjahn, G.; Gentry, T.J.; Brooks, J.P. Elevated incidences of antimicrobial resistance and multidrug resistance in the Maumee River (Ohio, USA), a major tributary of Lake Erie. Microorganisms 2021, 9, 911. [Google Scholar] [CrossRef]
  26. U.S. EPA. Method 1603: Escherichia coli (E. coli) in Water by Membrane Filtration Using Modified membrane-Thermotolerant Escherichia coli Agar (Modified mTEC); U.S. Environmental Protection Agency, Office of Water: Washington, DC, USA, 2014. Available online: https://www.epa.gov/sites/default/files/2015-08/documents/method_1603_2009.pdf (accessed on 15 July 2023).
  27. Genter, F.; Marks, S.J.; Clair-Caliot, G.; Mugume, D.S.; Johnston, R.B.; Bain, R.E.; Julian, T.R. Evaluation of the Novel Substrate RUG™ for the Detection of Escherichia coli in Water from Temperate (Zurich, Switzerland) and Tropical (Bushenyi, Uganda) Field Sites. Environ. Sci. Water Res. Technol. 2019, 5, 1082–1091. [Google Scholar] [CrossRef]
  28. Altheide, S.T. Biochemical and culture-based approaches to identification in the diagnostic microbiology laboratory. Clin. Lab. Sci. 2019, 32, 166–175. [Google Scholar] [CrossRef]
  29. Topić Popović, N.; Kepec, S.; Kazazić, S.P.; Strunjak-Perović, I.; Bojanić, K.; Čož-Rakovac, R. Identification of environmental aquatic bacteria by mass spectrometry supported by biochemical differentiation. PLoS ONE 2022, 17, e0269423. [Google Scholar] [CrossRef]
  30. FDA; CDC; USDA. 2021 National Antimicrobial Resistance Monitoring System Interpretive Criteria for Susceptibility Testing; U.S. Department of Health and Human Services, FDA: Silver Spring, MD, USA, 2021. Available online: https://www.fda.gov/media/108180/download (accessed on 14 July 2023).
  31. CLSI Document M100-S20; Performance Standards for Antimicrobial Susceptibility Testing: 20th Informational Supplement. Clinical and Laboratory Standards Institute: Wayne, PA, USA, 2010.
  32. Cayci, Y.T.; Ulker, K.H.; Birinci, A. Evaluation of three different methods for susceptibility testing of gentamicin in carbapenem resistant Enterobacterales. Le Infez. Med. 2021, 29, 568–573. [Google Scholar]
  33. García-Meniño, I.; Lumbreras, P.; Valledor, P.; Díaz-Jiménez, D.; Lestón, L.; Fernández, J.; Mora, A. Comprehensive statistical evaluation of Etest®, UMIC®, MicroScan and disc diffusion versus standard broth microdilution: Workflow for an accurate detection of Colistin-Resistant and Mcr-Positive E. coli. Antibiotics 2020, 9, 861. [Google Scholar] [CrossRef]
  34. Lee, M.; Chung, H.S. Different antimicrobial susceptibility testing methods to detect ertapenem resistance in Enterobacteriaceae: VITEK2, MicroScan, Etest, disk diffusion, and broth microdilution. J. Microbiol. Meth. 2015, 112, 87–91. [Google Scholar] [CrossRef] [PubMed]
  35. StataCorp. Stata Statistical Software, Release 14.; StataCorp LP.: College Station, TX, USA, 2015. [Google Scholar]
  36. Hosmer, D.W., Jr.; Lemeshow, S.; Sturdivant, R.X. Applied Logistic Regression, 2nd ed.; John Wiley & Sons: Hoboken, NJ, USA, 2013. [Google Scholar]
  37. Chao, W.L. Evaluation of Colilert-18 for the detection of coliforms and Escherichia coli in tropical fresh water. Lett. Appl. Microbiol. 2006, 42, 115–120. [Google Scholar] [CrossRef] [PubMed]
  38. Jefferson, R.A.; Burgess, S.M.; Hirsh, D. Beta-Glucuronidase from Escherichia coli as a Gene-Fusion Marker. Proc. Natl. Acad. Sci. USA 1986, 83, 8447–8451. [Google Scholar] [CrossRef]
  39. Frampton, E.W.; Restaino, L.; Blaszko, N. Evaluation of the β-Glucuronidase Substrate 5-Bromo-4-Chloro-3-Indolyl-β-D-Glucuronide (X-GLUC) in a 24-Hour Direct Plating Method for Escherichia coli. J. Food Prot. 1988, 51, 402–404. [Google Scholar] [CrossRef]
  40. Martins, M.T.; Rivera, I.G.; Clark, D.L.; Stewart, M.H.; Wolfe, R.L.; Olson, B.H. Distribution of uidA Gene Sequences in Escherichia coli Isolates in Water Sources and Comparison with the Expression of Beta-Glucuronidase Activity in 4-Methylumbelliferyl-beta-D-Glucuronide Media. Appl. Environ. Microbiol. 1993, 59, 2271–2276. [Google Scholar] [CrossRef]
  41. Rice, E.W.; Allen, M.J.; Edberg, S.C. Efficacy of Beta-Glucuronidase Assay for Identification of Escherichia coli by the Defined-Substrate Technology. Appl. Environ. Microbiol. 1990, 56, 1203–1205. [Google Scholar] [CrossRef] [PubMed]
  42. Frampton, E.W.; Restaino, L. Methods for Escherichia coli Identification in Food, Water and Clinical Samples Based on Beta-Glucuronidase Detection. J. Appl. Bacteriol. 1993, 74, 223–233. [Google Scholar] [CrossRef]
  43. Tryland, I.; Fiksdal, L. Enzyme Characteristics of β-D-Galactosidase-and β-D-Glucuronidase-Positive Bacteria and Their Interference in Rapid Methods for Detection of Waterborne Coliforms and Escherichia coli. Appl. Environ. Microbiol. 1998, 64, 1018–1023. [Google Scholar] [CrossRef]
  44. Bain, R.E.; Woodall, C.; Elliott, J.; Arnold, B.F.; Tung, R.; Morley, R.; du Preez, M.; Bartram, J.K.; Davis, A.P.; Gundry, S.W.; et al. Evaluation of an Inexpensive Growth Medium for Direct Detection of Escherichia coli in Temperate and Sub-Tropical Waters. PLoS ONE 2015, 10, e0140997. [Google Scholar] [CrossRef] [PubMed]
  45. Pitkänen, T.; Paakkari, P.; Miettinen, I.T.; Heinonen-Tanski, H.; Paulin, L.; Hänninen, M.L. Comparison of Media for Enumeration of Coliform Bacteria and Escherichia coli in Non-Disinfected Water. J. Microbiol. Methods 2007, 68, 522–529. [Google Scholar] [CrossRef] [PubMed]
  46. Kämpfer, P.; Nienhüser, A.; Packroff, G.; Wernicke, F.; Mehling, A.; Nixdorf, K.; Fiedler, S.; Kolauch, C.; Esser, M. Molecular Identification of Coliform Bacteria Isolated from Drinking Water Reservoirs with Traditional Methods and the Colilert-18 System. Int. J. Hyg. Environ. Health 2008, 211, 374–384. [Google Scholar] [CrossRef] [PubMed]
  47. Olstadt, J.; Schauer, J.J.; Standridge, J.; Kluender, S. A Comparison of Ten USEPA Approved Total Coliform/E. coli Tests. J. Water Health 2007, 5, 267–282. [Google Scholar] [CrossRef]
  48. Power, D.A.; Johnson, J.A. Difco™ & BBL™ Manual. Manual of Microbiological Culture Media, 3rd ed.; Becton, Dickinson and Company: Sparks, MD, USA, 2009; Volume 359, p. 60. [Google Scholar]
  49. Bonnet, M.; Lagier, J.C.; Raoult, D.; Khelaifia, S. Bacterial Culture through Selective and Non-Selective Conditions: The Evolution of Culture Media in Clinical Microbiology. New Microbes New Infect. 2020, 34, 100622. [Google Scholar] [CrossRef]
  50. Martínez, J.L. Effect of Antibiotics on Bacterial Populations: A Multi-Hierarchical Selection Process. F1000Research 2017, 6, 51. [Google Scholar] [CrossRef]
  51. Pisciotta, J.M.; Rath, D.F.; Stanek, P.A.; Flanery, D.M.; Harwood, V.J. Marine Bacteria Cause False-Positive Results in the Colilert-18 Rapid Identification Test for Escherichia coli in Florida Waters. Appl. Environ. Microbiol. 2002, 68, 539–544. [Google Scholar] [CrossRef]
  52. Sercu, B.; Van De Werfhorst, L.C.; Murray, J.L.; Holden, P.A. Cultivation-Independent Analysis of Bacteria in IDEXX Quanti-Tray/2000 Fecal Indicator Assays. Appl. Environ. Microbiol. 2011, 77, 627–633. [Google Scholar] [CrossRef]
  53. Tiwari, A.; Niemelä, S.I.; Vepsäläinen, A.; Rapala, J.; Kalso, S.; Pitkänen, T. Comparison of Colilert-18 with Miniaturized Most Probable Number Method for Monitoring of Escherichia coli in Bathing Water. J. Water Health 2016, 14, 121–131. [Google Scholar] [CrossRef]
  54. Chao, K.K.; Chao, C.C.; Chao, W.L. Evaluation of Colilert-18 for Detection of Coliforms and Eschericha coli in Subtropical Freshwater. Appl. Environ. Microbiol. 2004, 70, 1242–1244. [Google Scholar] [CrossRef] [PubMed]
  55. Chou, G. Effects of Land Use in the Ohio River Basin on the Distribution of Coliform and Antibiotic Resistant Bacteria in the Ohio River. Master’s Thesis, Marshall University, Huntington, WV, USA, 2011. Available online: https://mds.marshall.edu/etd/218 (accessed on 27 August 2023).
  56. Galvin, S.; Boyle, F.; Hickey, P.; Vellinga, A.; Morris, D.; Cormican, M. Enumeration and Characterization of Antimicrobial-Resistant Escherichia coli Bacteria in Effluent from Municipal, Hospital, and Secondary Treatment Facility Sources. Appl. Environ. Microbiol. 2010, 76, 4772–4779. [Google Scholar] [CrossRef] [PubMed]
  57. Akiyama, T.; Savin, M.C. Populations of Antibiotic-Resistant Coliform Bacteria Change Rapidly in a Wastewater Effluent Dominated Stream. Sci. Total Environ. 2010, 408, 6192–6201. [Google Scholar] [CrossRef] [PubMed]
  58. Stocker, M.; Smith, J.; Pachepsky, Y. Spatial Variation of Tetracycline-Resistant E. coli and Relationships with Water Quality Variables in Irrigation Water: A Pilot Study. Appl. Microbiol. 2023, 3, 504–518. [Google Scholar] [CrossRef]
  59. CLSI. Performance Standards for Antimicrobial Susceptibility Testing, 30th ed.; CLSI Supplement M100; Clinical and Laboratory Standards Institute: Wayne, PA, USA, 2020. [Google Scholar]
  60. Sawant, A.A.; Hegde, N.V.; Straley, B.A.; Donaldson, S.C.; Love, B.C.; Knabel, S.J.; Jayarao, B.M. Antimicrobial-Resistant Enteric Bacteria from Dairy Cattle. Appl. Environ. Microbiol. 2007, 73, 156–163. [Google Scholar] [CrossRef]
  61. NCCLS Document M31-A2; Performance Standards for Antimicrobial Disk and Dilution Susceptibility Tests for Bacteria Isolated from Animals, 2nd ed. National Committee for Clinical Laboratory Standards: Wayne, PA, USA, 2002.
  62. Serwecińska, L. Antimicrobials and Antibiotic-Resistant Bacteria: A Risk to the Environment and to Public Health. Water 2020, 12, 3313. [Google Scholar] [CrossRef]
  63. Gullberg, E.; Cao, S.; Berg, O.G.; Ilbäck, C.; Sandegren, L.; Hughes, D.; Andersson, D.I. Selection of Resistant Bacteria at Very Low Antibiotic Concentrations. PLoS Pathog. 2011, 7, e1002158. [Google Scholar] [CrossRef]
  64. Lundström, S.V.; Östman, M.; Bengtsson-Palme, J.; Rutgersson, C.; Thoudal, M.; Sircar, T.; Blanck, H.; Eriksson, K.M.; Tysklind, M.; Flach, C.F.; et al. Minimal Selective Concentrations of Tetracycline in Complex Aquatic Bacterial Biofilms. Sci. Total Environ. 2016, 553, 587–595. [Google Scholar] [CrossRef]
  65. Duff, J.A.; Aslan, A.; Cohen, R.A. Land Use and Environmental Variables Influence Tetracycline-Resistant Bacteria Occurrence in Southeastern Coastal Plain Streams. J. Environ. Qual. 2019, 48, 1809–1816. [Google Scholar] [CrossRef]
  66. Di Cesare, A.; Eckert, E.M.; Rogora, M.; Corno, G. Rainfall Increases the Abundance of Antibiotic Resistance Genes within a Riverine Microbial Community. Environ. Pollut. 2017, 226, 473–478. [Google Scholar] [CrossRef]
  67. Williams, N.L.; Siboni, N.; McLellan, S.L.; Potts, J.; Scanes, P.; Johnson, C.; James, M.; McCann, V.; Seymour, J.R. Rainfall Leads to Elevated Levels of Antibiotic Resistance Genes within Seawater at an Australian Beach. Environ. Pollut. 2022, 307, 119456. [Google Scholar] [CrossRef] [PubMed]
  68. Lee, C.; Agidi, S.; Marion, J.W.; Lee, J. Arcobacter in Lake Erie Beach Waters: An Emerging Gastrointestinal Pathogen Linked with Human-Associated Fecal Contamination. Appl. Environ. Microbiol. 2012, 78, 5511–5519. [Google Scholar] [CrossRef] [PubMed]
  69. Carney, R.L.; Labbate, M.; Siboni, N.; Tagg, K.A.; Mitrovic, S.M.; Seymour, J.R. Urban Beaches Are Environmental Hotspots for Antibiotic Resistance following Rainfall. Water Res. 2019, 167, 115081. [Google Scholar] [CrossRef] [PubMed]
  70. O’Hara, C.M.; Tenover, F.C.; Miller, J.M. Parallel Comparison of Accuracy of API 20E, Vitek GNI, MicroScan Walk/Away Rapid ID, and Becton Dickinson Cobas Micro ID-E/NF for Identification of Members of the Family Enterobacteriaceae and Common Gram-Negative, Non-Glucose-Fermenting Bacilli. J. Clin. Microbiol. 1993, 31, 3165–3169. [Google Scholar] [CrossRef] [PubMed]
  71. Molina, F.; López-Acedo, E.; Tabla, R.; Roa, I.; Gómez, A.; Rebollo, J.E. Improved Detection of Escherichia coli and Coliform Bacteria by Multiplex PCR. BMC Biotechnol. 2015, 15, 48. [Google Scholar] [CrossRef]
  72. Osińska, A.; Korzeniewska, E.; Korzeniowska-Kowal, A.; Wzorek, A.; Harnisz, M.; Jachimowicz, P.; Buta-Hubeny, M.; Zieliński, W. The Challenges in the Identification of Escherichia coli from Environmental Samples and Their Genetic Characterization. Environ. Sci. Pollut. Res. 2023, 30, 11572–11583. [Google Scholar] [CrossRef]
  73. McLain, J.E.; Cytryn, E.; Durso, L.M.; Young, S. Culture-based methods for detection of antibiotic resistance in agroecosystems: Advantages, challenges, and gaps in knowledge. J. Environ. Qual. 2016, 45, 432–440. [Google Scholar] [CrossRef]
  74. Wuijts, S.; van den Berg, H.H.; Miller, J.; Abebe, L.; Sobsey, M.; Andremont, A.; Medlicott, K.O.; van Passel, M.W.; de Roda Husman, A.M. Towards a research agenda for water, sanitation and antimicrobial resistance. J. Water Health 2017, 15, 175–184. [Google Scholar] [CrossRef]
Pathogens 12 01090 g001
Figure 1. The results of one water sample evaluated for E. coli density by the ColiGlow method without tetracycline (left), and the ColiGlow method containing tetracycline (right) whereby the fluorescing wells under longwave ultraviolet light are indicative of presumed E. coli growth.
Figure 1. The results of one water sample evaluated for E. coli density by the ColiGlow method without tetracycline (left), and the ColiGlow method containing tetracycline (right) whereby the fluorescing wells under longwave ultraviolet light are indicative of presumed E. coli growth.
Pathogens 12 01090 g001
Table 1. The species and tetracycline susceptibility status of microorganisms isolated in this study from fluorescing ColiGlow wells from growth media with and without tetracycline.
Table 1. The species and tetracycline susceptibility status of microorganisms isolated in this study from fluorescing ColiGlow wells from growth media with and without tetracycline.
Isolates from ColiGlow
Wells without Tetracycline		Isolates from ColiGlow
Wells with Tetracycline
Species (no.)SusceptibleResistantSusceptibleResistant
Escherichia coli (50)15 7 3 25
Enterobacter cloacae (6)2 4 00
Kluyvera ascorbate (1)01 00
Kluyvera intermedia (1)1000
Klebsiella pneumoniae (1)0100
Serratia odorifera (1)0100
Citrobacter braakii (1)1000
Total (61)1914325
Table 2. The species and tetracycline susceptibility status of microorganisms isolated in this study from fluorescing ColiGlow wells from growth media with and without tetracycline.
Table 2. The species and tetracycline susceptibility status of microorganisms isolated in this study from fluorescing ColiGlow wells from growth media with and without tetracycline.
Species Identification
Reported by ID Method		Frequency (%)
E. coli Density
within Method Range		E. coli Density
over Method Range
E. coli38 (90.5%)12 (63.2%)
Not E. coli 4 (9.5%)7 (37.8%)
Overall42 (100%)19 (100%)
Table 3. The tetracycline susceptibility status of microorganisms isolated in this study from fluorescing ColiGlow wells when E. coli densities are in-range or over-range (>1479 MPN per 100 mL) in the water sample.
Table 3. The tetracycline susceptibility status of microorganisms isolated in this study from fluorescing ColiGlow wells when E. coli densities are in-range or over-range (>1479 MPN per 100 mL) in the water sample.
Tetracycline
Susceptibility		Frequency (%)
E. coli Density
within Method Range		E. coli Density
over Method Range
Tetracycline-Resistant23 (54.8%)16 (84.2%)
Tetracycline-Susceptible19 (45.2%)3 (15.8%)
Overall: All Isolates42 (100.0%)19 (100.0%)
Tetracycline-Resistant E. coli18 (52.6%)12 (100.0%)
Tetracycline-Susceptible E. coli20 (47.4%)0 (0%)
Overall: E. coli Isolates38 (100.0%)12 (100.0%)
Table 4. The frequency of 31 isolates reported as E. coli or not E. coli by the API20E identification method versus the frequency reported as E. coli or not E. coli for 30 different isolates using the MicroScan instrument.
Table 4. The frequency of 31 isolates reported as E. coli or not E. coli by the API20E identification method versus the frequency reported as E. coli or not E. coli for 30 different isolates using the MicroScan instrument.
Species Identification
Reported		Frequency (%)
API20E
Identification Method		MicroScan Urine Panel-85
Identification Method
E. coli30 (96.8%)20 (66.7%)
Not E. coli 1 (3.2%)10 (33.3%)
Overall31 (100%)30 (100%)
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Boggs, C.; Shiferawe, K.; Karsten, E.; Hamlet, J.; Altheide, S.T.; Marion, J.W. Evaluation of a Tetracycline-Resistant E. coli Enumeration Method for Correctly Classifying E. coli in Environmental Waters in Kentucky, USA. Pathogens 2023, 12, 1090. https://doi.org/10.3390/pathogens12091090

AMA Style

Boggs C, Shiferawe K, Karsten E, Hamlet J, Altheide ST, Marion JW. Evaluation of a Tetracycline-Resistant E. coli Enumeration Method for Correctly Classifying E. coli in Environmental Waters in Kentucky, USA. Pathogens. 2023; 12(9):1090. https://doi.org/10.3390/pathogens12091090

Chicago/Turabian Style

Boggs, Callie, Kidus Shiferawe, Eckhardt Karsten, Jayden Hamlet, S. Travis Altheide, and Jason W. Marion. 2023. "Evaluation of a Tetracycline-Resistant E. coli Enumeration Method for Correctly Classifying E. coli in Environmental Waters in Kentucky, USA" Pathogens 12, no. 9: 1090. https://doi.org/10.3390/pathogens12091090

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop