Drinking water quality from sources other than tap water is increasingly becoming a source of concern in many communities. Communities in the Eastern Coachella Valley (ECV), Riverside County, California, USA have raised concerns regarding bulk drinking water from water vending machines (WVMs) found in public vendors. To address concerns, we conducted microbiological contamination assessments of drinking water from WVMs in the ECV using heterotrophic plate counts (HPC), the presence of total coliforms using IDEXX technology, and real-time PCR (qPCR). We also measured temperature, pH, electrical conductivity, and free chlorine concentration. Twenty-five WVMs were sampled by using positively charged NanoCeram® filters in the field. Results indicated 32% of WVMs had total coliforms, and 21% had HPC above Environmental Protection Agency (EPA) requirements. Through qPCR, we found 81% of WVMs had Salmonella spp., Listeria monocytogenes, and Campylobacter jejuni, 76% had Enterococcus faecalis, and 90% had Pseudomonas aeruginosa. Results indicated most WVM samples we collected contained genetic material of pathogenic microorganisms and therefore, did not meet EPA drinking water standards. There is an urgency to enforce WVM maintenance through drain flushing, spigot cleaning, rust removal, filter replacement, and limits to physico-chemical parameters.

  • Drinking water quality from vending machines was assessed in a disadvantaged rural community in Southern California.

  • Real-time qPCR was used to determine water quality via a select panel of indicator bacteria.

  • Drinking water vending machines were found to be contaminated by using cultivatable IDEXX and qPCR methods.

  • We emphasize following current regulations for maintaining water vending machines.

Access to potable and safe drinking water is paramount to human health, and is a basic human right recognized by the United Nations General Assembly in their Human Right to Water and Sanitation Program (United Nations 2010). One in three people worldwide do not have access to safe drinking water (WHO 2014). While substantial progress has been made in developed countries to ensure safe drinking water to communities guided by stringent regulations, disease outbreaks due to contaminated drinking water remain a serious public health problem. A collaborative national Waterborne Disease and Outbreak Surveillance System (WBDOSS) formed by the Centers for Disease Control and Prevention (CDC), United States Environmental Protection Agency (US EPA), and the Council of State and Territorial Epidemiologists (CSTE) indicated that from 1971 to 2006, 780 disease outbreaks in the USA were related to drinking water contamination (Craun et al. 2010). Recently, there has been a striking increase in drinking water consumption from sources other than tap water around the world (Hu et al. 2011; Patel et al. 2016; Praveena et al. 2018). Alternative sources of drinking water include water vending machines (WVM) and bottled water. As of 2002, there were over 7,000 Glacier branded WVMs throughout the USA (Sharp & Walker 2002).

Water crises, such as that during 2014 in Flint, Michigan where drinking water was contaminated with lead and microorganisms (Denchak 2018) indicate that even in the USA, the threat of drinking water contamination remains an important and unequivocal concern. Moreover, the inability of some states to meet the EPA standards for safe drinking water (Mayer & Goldman 2016; Philip et al. 2017; Bird et al. 2019) is a serious threat to the human right to safe water.

In January 2018, more than five thousand residents of the state of California were supplied with water that did not meet the Safe Drinking Water Act (SDWA) requirements, and two-thirds of tribal communities had inadequate home plumbing (Feinstein 2018). In order to address the problem, the state of California enacted the Human Right to Water Law in 2012, stating that all people in the state have the right to clean and safe drinking water. Furthermore, the California State Water Resources Control Board acknowledged the importance of addressing the water problem for disadvantaged communities (Feinstein 2018).

The largest WVM provider in California is the Primo Water Company which owns the Glacier vending machine network. The non-profit organization, Environmental Working Group, released a report in 2001 in which it verified that the Glacier Water company reported over 60% of its sales went to Latino or Asian customers (Sharp & Walker 2002). The targeting of water sales to immigrant customers has been recognized as a concern as far back as 1999 by the Metropolitan Water District of Southern California (MWDSC), which initiated a community water awareness campaign to help water utilities reclaim tap-water-drinking customers. The campaign used programs targeted to immigrant populations that informed households about the safety of tap water (The Metropolitan Water District of Southern California 1999). In California, WVMs are regulated by the Health and Safety Code (HSC) (California Legislative Information 1995), the federal Food and Drug Administration (FDA), and the United States Environmental Protection Agency (USEPA). California requires that vending machine owners inspect, service, and sanitize each unit, and perform bacteriological analyses at least every 6 months. The HSC, section 111115, states that the California Department of Environmental Health should annually inspect at least 20% of licensed WVMs in both rural and urban counties to check for microbial or chemical contamination and ensure they meet the USEPA requirements for safe drinking water (CDPH 2014).

In the Eastern Coachella Valley (ECV) many households are reliant on WVMs because of contamination of some household water originating from the groundwater supply that feeds their private wells. Most households who use private wells are in mobile home parks and purchase drinking water from WVMs (Sharp & Walker 2002). Although there are limited studies reporting the microbial contamination unique to WVM water, there are no known published data that report this problem in the ECV. A Los Angeles-based study by Schillinger & Du Vall Knorr (2004) found statistically significant associations between operator visits and poor machine conditions, current permits, servicing intervals, presence of fungi, and presence of Pseudomonas aeruginosa. A study in Tucson, AZ found heterotrophic plate count (HPC) bacteria in 73% of the WVMs studied and Pseudomonas aeruginosa in 23% of WVMs examined despite the use of advanced disinfection systems that included carbon filters, ultraviolet treatment, and reverse osmosis (RO) filter components (Cardaci et al. 2016). However, these studies did not use large volumes of water, and did not couple cultivation techniques with real-time qPCR. The aim of the current study was to analyze physico-chemical parameters and investigate the microbial contamination of drinking water from commercial, self-standing WVMs of the ECV.

Study area

The present study was carried out in the ECV, a desert area situated in southern California (337°N, 116.2°W). It extends southeast into Riverside County for approximately 72.4 km from the San Bernardino Mountains to the northern shore of the Salton Sea. It is limited on the West by the San Jacinto Mountains and the Santa Rosa Mountains, and on the North and East by the Little San Bernardino Mountains (Figure 1). The ECV is made up of four unincorporated rural communities; Thermal, Oasis, Mecca, and North Shore. These communities are populated by Latino service industry and agricultural-worker families that contribute approximately 430 million dollars per annum to the Gross Domestic Product of the USA (PUCDC 2020).

Figure 1

Map featuring California State, San Bernardino county, and the Eastern Coachella Valley.

Figure 1

Map featuring California State, San Bernardino county, and the Eastern Coachella Valley.

Close modal

Filtration manifold set-up

We designed a field filter manifold (FFM) with a pump, flow meter, and filter housing to pump water from multiple WVM bottles in one sampling day (Figure 2). We fabricated a plastic briefcase with dimensions 48 L × 5 W × 35D cm, with two pieces of plywood attached to both sides and fitted with screws. A cartridge housing module was attached to the top side of the briefcase and attached to the wood with screw bolt clamps. We attached a barb adapter (0.95 cm) on the input of the cartridge housing casket. A Pacific Hydrostat utility 12-volt DC/0.95 cm HP self-priming pump (Pacific Hydrostar, Washington, USA) was connected to the cartridge housing by a stainless brass pipe nipple (0.95 cm) on the pump input and cartridge output. The pump was attached to the upper side of the briefcase by four screws. We attached a hose barb adapter (0.95 cm) to the pump outlet, and a DigiFlow 600R flow meter (Kao Chen Enterprises, Taichung City, Taiwan) was attached vertically to a vinyl hose on the lower inner part of the briefcase with a hose adapter (0.95 cm) on both the pump input and output. A small hole was cut in the lower left side of the case to allow the hose adapter of the flow meter to protrude out the side of the briefcase. Another approximately 3 m braided vinyl hose of 0.95 cm internal diameter was connected to the output of the flow meter to allow the filtered water to flow out as wastewater. A mating pair of XT 60 connectors (Hobbyking, Seattle, USA) were soldered to the pump wire on one side with the battery on the other and covered with a heat shrink wire connector. A low voltage buzzer alarm was attached to the 5,000 mAH 11.1 V Lithium Polymer battery.

Figure 2

Close up labelled image of the field filtration membrane (FFM) apparatus we designed equipped with a positively charged NanoCeram® filter for the filtration of 113 L of drinking water from each vending machine of the Eastern Coachella Valley. Part of the FFM are labelled as: a. 5-gallon bottle water; b. Cartridge housing module containing a positively charged NanoCeram® filter; c. Self-priming pump utility 12-Volt DC/0.95 cm; d. 5,000 mAH 11.1 V lithium polymer battery; e. Low voltage buzzer alarm; f. Flow meter. Only (a), (b) and the vinyl hose connecting them where sterilized.

Figure 2

Close up labelled image of the field filtration membrane (FFM) apparatus we designed equipped with a positively charged NanoCeram® filter for the filtration of 113 L of drinking water from each vending machine of the Eastern Coachella Valley. Part of the FFM are labelled as: a. 5-gallon bottle water; b. Cartridge housing module containing a positively charged NanoCeram® filter; c. Self-priming pump utility 12-Volt DC/0.95 cm; d. 5,000 mAH 11.1 V lithium polymer battery; e. Low voltage buzzer alarm; f. Flow meter. Only (a), (b) and the vinyl hose connecting them where sterilized.

Close modal

Sample collection and filtrations

Prior to sampling, we sterilized the American Maid 5-gallon Polyethylene terphalate water bottles (Walmart, Little Rock, Arkansas, USA) we used for samples with household bleach, then ultraviolet (UV) light using a 40 W Bioshield UVC light (Pentair, St. Paul, Minneapolis, USA). Polycarbonate clear tubes were also sterilized by plunging them into a container half full of water and autoclaving them for 15 minutes at 121 °C. During sterilization, we added 0.5 mL of 6% household bleach and 5 L of deionized (DI) water to the American Maid 5-gallon Water Bottle. We shook the bottle vigorously for 2–5 min and ensured the entire internal surface of the bottle was covered. After a rest of approximately 10 min, the procedure was repeated three times. Afterward, we rinsed the bottle with DI water and then added 5 mg of sodium thiosulfate to neutralize the chlorine during sampling. An ultraviolet (UV) lamp was introduced into the bottle and turned on for 5 min. We remotely operated the light using a switch to avoid human exposure to UV.

A total of 113.7 L of WVM water was filtered through our FFM from each WVM. For this study, we chose 25 sites (Figure 3) to represent water sources in the ECV. Drinking water samples were collected from free standing WVMs in malls and stores of the ECV. All samples were collected in sterile American Maid 5-gallon Water Bottles containing sodium thiosulfate to neutralize any residual disinfectant from the WVMs as described above. A 500 mL sample of drinking water was collected on site in a beaker for physico-chemical analyses. During sampling, we carefully opened and placed bottles in the chest of the coin-operated WVMs. Immediately after sampling, water bottles were transported to the filtration area. Wearing clean nitrile gloves, we introduced autoclaved polycarbonate tubes into the 5-gallon bottles one at a time, and the filtration procedure started by connecting the battery to the pump. From each WVM, a total of approximately 113.7 L of water was passed through a NanoCeram® filter. After filtration, the cartridge housing containing the NanoCeram® filter was immediately dismantled from the filtration apparatus and placed in an insulated storage and transport cooler with ice, to ensure microorganisms on the filters remained at between 1–10 °C during transportation. In the laboratory, a visual infrared thermometer was used to measure the temperature of the cartridge housing.

Figure 3

Map of water vending machines sampled in the Eastern Coachella Valley for microbiological analysis. Most vending machines were located in Coachella where the communities converge for shopping and buying of water. The longest distance from point to point was about 18 kilometers.

Figure 3

Map of water vending machines sampled in the Eastern Coachella Valley for microbiological analysis. Most vending machines were located in Coachella where the communities converge for shopping and buying of water. The longest distance from point to point was about 18 kilometers.

Close modal

In the laboratory, we prepared a dispersant solution made of Tween 80 and sodium pyrophosphate (0.01 μL of NaPP) (Hill et al. 2005), used to remove bacteria and viruses from the positively charged NanoCeram® filters. Following the primary concentration, approximately 300 mL of the elution solution were added to the cartridge housing containing the filter. The unit was then resealed and shaken vigorously for 5 min, followed by a 10 min pause. This procedure was repeated three times.

Centricon plus 70 ultrafilters (30-KDA cutoff; Millipore, Billerica, MA) were used to further concentrate the eluted solution. According to the manufacturer's recommendation, the Centricon plus 70 ultrafilters were pre-wetted by adding 70 mL of Nanopure water, followed by centrifugation (1,900 × g for 8 min). The Centricon plus 70 ultrafilters were inverted and centrifuged (800 × g for 2 min) to collect the remaining water, which was then discarded. We added 70 mL of the NanoCeram® filter eluate to the Centricon filter and concentrated it via centrifugation (1,900 × g for 8 min). We collected the bacterial concentrate via inversion of the filter and further centrifugation for 2 min at 800 × g. The procedure was repeated three times. The total volume of 210 mL filtered sample using the Centricon plus 70 ultrafilters yielded 2 mL of concentrated sample. We further mixed the 1 mL volumes from the secondary filtration with 1 mL of 50% diluted glycerol, into 2 mL cryonic vials, which we then labelled and store at −80 °C until needed.

We calibrated the OAKTON pH meter (Cole-Parmer, Illinois, USA) using a three-point calibration process (pH 4.01, pH 7.00, pH 10.00).

Laboratory analysis

Samples from WVMs were streaked on R2A agar for HPC and colonies counted in duplicates. Following incubation (at 35 °C for 48 h), colonies on plates were counted and results expressed as colony-forming units per milliliter (CFU·100 mL−1). We averaged all counts for each vending machine. We used the Colilert reagent test (IDEXX, Maine, USA) to detect and count total coliform and E. coli cells in drinking water samples. The Colilert reagent was added, and the bottle sealed, mixed, and poured into a Quanti-Tray. We sealed the Quanti-Trays using IDEXX Quanti-Tray Sealer and incubated them at 35 °C for 24 h. The presence of total coliform was confirmed by a yellow color in Quanti-Tray wells. We determined the presence of E coli by fluorescence emitted when Quanti-Trays were exposed to UV light after 24 h incubation.

DNA extraction

To extract DNA from environmental microorganisms, we used GenEluteTM Bacterial Genomic DNA KIT (Sigma-Aldrich, Missouri, USA), stored extracted DNA in 2 mL cryovial tubes, and stored them at −80 °C until needed. The subunit ribosomal ribonucleic acid (SSU) rRNA genes were amplified by PCR and qPCR from the extracted DNA samples. After extraction, we quantified the concentration of nucleic acid using a NanoDrop spectrophotometer (ND-1000, Thermo Fisher Scientific, Delaware, USA).

Extraction of DNA from bacterial cultures

We enriched Salmonella typhimurium (ENVH Carolina 155351A), Listeria monocytogenes (ATCC® 7644), Pseudomonas aeruginosa (ATCC® 27853), Enterococcus faecalis (ATCC® 29212), and Campylobacter jejuni (ATCC® 33291) overnight in TSB 1:10 and extracted DNA using GenEluteTM Bacterial Genomic DNA KIT (Sigma-Aldrich, Missouri, USA) as described above. We used the gDNA from these microorganisms as positive controls for the presence of targeted microorganisms in qPCR experiments.

Primer design

Primers used in this study are shown in Table 1 and were designed using SILVA databases (Quast et al. 2012). Primers were selected based on annealing temperature, small amplicon size, and specificity to the selected microorganisms. Specificity of the primer was tested using Primer-Blast-NCBI (Boratyn et al. 2019). Standard curves were prepared (data not shown) to confirm the specificity of primers.

Table 1

Primers list of bacteria chosen for a WVM panel with justification and reference

Target microorganismsPrimer sets (5′-3′)SizeJustificationReferences
Salmonella spp. FGGAAACGGTGGCTAATACC 103 Local Liu et al. (2018)  
 CCTCACCAACAAGCTAATCC   contaminant  
Listeria GATGATCAGGTAGATAGGTTTGG 119 Local Gião & Keevil (2014)  
monocytogenes CCTAACTGAGCCCTTTCTTC   contaminant  
Campylobacter CCCTATCAAACTCCGAATACC 93 Local Whiley et al. (2013)  
jejuni GGTAGTCTGGGTTGTTTCC   contaminant  
Pseudomonas GAGCAGGTTGAAGGTTAGG 111 Similar Liguori et al. (2010)  
aeruginosa GCTAATCAAGCTCGGAGATAG   study  
Enterococcus TTGTGTTATGAACCCTCTAACC 124 Indicator Girolamini et al. (2019)  
faecalis GGTTCCCTCAGAATGGTTG   bacteria  
E. coli (all strains) CTATGTGTTGTTGGGTAGGG 423 Indicator Luby et al. (2015)  
 GATGTTACCTGATGCTTAGAGG   bacteria  
Target microorganismsPrimer sets (5′-3′)SizeJustificationReferences
Salmonella spp. FGGAAACGGTGGCTAATACC 103 Local Liu et al. (2018)  
 CCTCACCAACAAGCTAATCC   contaminant  
Listeria GATGATCAGGTAGATAGGTTTGG 119 Local Gião & Keevil (2014)  
monocytogenes CCTAACTGAGCCCTTTCTTC   contaminant  
Campylobacter CCCTATCAAACTCCGAATACC 93 Local Whiley et al. (2013)  
jejuni GGTAGTCTGGGTTGTTTCC   contaminant  
Pseudomonas GAGCAGGTTGAAGGTTAGG 111 Similar Liguori et al. (2010)  
aeruginosa GCTAATCAAGCTCGGAGATAG   study  
Enterococcus TTGTGTTATGAACCCTCTAACC 124 Indicator Girolamini et al. (2019)  
faecalis GGTTCCCTCAGAATGGTTG   bacteria  
E. coli (all strains) CTATGTGTTGTTGGGTAGGG 423 Indicator Luby et al. (2015)  
 GATGTTACCTGATGCTTAGAGG   bacteria  

Real-time PCR

Real-time PCR (qPCR) was performed using a C1000 Touch Thermal Cycler CFX 96 (Bio-Rad, Hercules, USA). Each 96-well plate reaction mixture (25 µL) contained 12.5 μL iTaq Universal SYBR Green Supermix (Bio-Rad, California), 7.5 μL Nanopure distilled water, 1 μL of forward primers, 1 μL of reverse primers, and 1 μL of gDNA. In the 96 -well plate set-up, we had positive control wells with gDNA from targeted microorganisms obtained from our laboratory. Negative controls wells had no gDNA added. The thermocycling program was 40 cycles at 95 °C for 3 min for the initial cycle, 95 °C for 10 s, and 55 °C for 30 s. All amplifications were done on a CFX 96 Real-Time System (Bio-Rad, California). The targeted microorganisms (Table 1) were selected from previous studies on WVMs (Pseudomonas aeruginosa, HPC), common water quality indicators (E. coli, Coliform, and Enterococcus faecalis), and pathogens (Listeria monocytogenes, Salmonella spp., Campylobacter jejuni) that we detected in a previous wastewater study in the same community.

Physico-chemical parameters

In this study, we narrowed the physico-chemical parameters to temperature, EC, TDS, pH, and free chlorine. The physico-chemical results are presented in Table 2. The pH values measured in this study ranged from 5.1–9.5. Samples from WVM 1 had the highest pH of 9.5, while WVM 9 had the lowest pH of 5.1. Electrical conductivity ranged from 0.00 to 5.7 μS· cm−1. Results from the TDS were all below 300 mg·L−1. A total of 40% of water samples from WVMs had no residual free chlorine. In this study, free chlorine concentrations ranged from 0.04 mg·L−1 to 0.2 mg·L−1 Cl2 (Table 2). Water temperatures measured immediately after sampling from WVMs ranged from 18.2 °C to 32.1 °C (Table 2).

Table 2

Physico-chemical parameters of WVMs in the ECV including TDS

WVMTemp. °CConductivity μS·cm−1pHFree chlorine mg·L−1TDS mg·L−1
WVM1 19.5 0.24 9.15 0.2 0.12 
WVM2 30.5 0.01 6.5 0.00 0.00 
WVM3 26 0.48 7.48 0.26 0.47 
WVM4 29.2 0.03 0.00 0.01 
WVM5 28.1 0.1 6.23 0.03 0.03 
WVM6 32 0.00 6.9 0.02 0.00 
WVM7 26.8 3.91 5.3 0.1 0.16 
WVM8 30.9 6.83 7.5 0.00 3.34 
WVM9 25.1 2.67 5.19 0.05 0.12 
WVM10 22.5 0.3 7.1 0.00 0.1 
WVM11 18.2 0.1 5.6 0.00 0.09 
WVM12 19.4 0.5 6.4 0.05 0.2 
WVM13 32.8 0.1 6.3 0.12 65 
WVM14 32.1 0.2 6.7 0.08 81 
WVM15 32.1 0.3 0.07 3.12 
WVM16 28.5 5.7 – 0.00 8.36 
WVM17 29.5 0.2 6.29 0.00 98 
WVM18 20.1 1.6 8.7 0.08 8.01 
WVM19 26 7.2 0.03 
WVM20 29 3.2 6.1 0.00 
WVM21 22 5.9 0.01 16 
WVM22 29 5.5 0.00 10 
WVM23 27 0.3 6.4 0.00 7.9 
WVM24 31 4.1 5.9 0.03 19 
WVM25 32 2.8 – 0.01 13 
WVMTemp. °CConductivity μS·cm−1pHFree chlorine mg·L−1TDS mg·L−1
WVM1 19.5 0.24 9.15 0.2 0.12 
WVM2 30.5 0.01 6.5 0.00 0.00 
WVM3 26 0.48 7.48 0.26 0.47 
WVM4 29.2 0.03 0.00 0.01 
WVM5 28.1 0.1 6.23 0.03 0.03 
WVM6 32 0.00 6.9 0.02 0.00 
WVM7 26.8 3.91 5.3 0.1 0.16 
WVM8 30.9 6.83 7.5 0.00 3.34 
WVM9 25.1 2.67 5.19 0.05 0.12 
WVM10 22.5 0.3 7.1 0.00 0.1 
WVM11 18.2 0.1 5.6 0.00 0.09 
WVM12 19.4 0.5 6.4 0.05 0.2 
WVM13 32.8 0.1 6.3 0.12 65 
WVM14 32.1 0.2 6.7 0.08 81 
WVM15 32.1 0.3 0.07 3.12 
WVM16 28.5 5.7 – 0.00 8.36 
WVM17 29.5 0.2 6.29 0.00 98 
WVM18 20.1 1.6 8.7 0.08 8.01 
WVM19 26 7.2 0.03 
WVM20 29 3.2 6.1 0.00 
WVM21 22 5.9 0.01 16 
WVM22 29 5.5 0.00 10 
WVM23 27 0.3 6.4 0.00 7.9 
WVM24 31 4.1 5.9 0.03 19 
WVM25 32 2.8 – 0.01 13 

: missing data due to equipment failure.

We present results from microbial analyses in Table 3. Results for HPC demonstrated that 21% of WVMs had HPC above 500 CFU· mL−1. HPC ranged from 6.9 × 103 CFU· mL−1 for WVM 5 to 3.20 × 103 CFU· mL−1 for WVM 8. The presence of total coliforms (Table 3) was identified in 32% of WVMs. The highest coliform count was found in WVM 19.

Table 3

Biological characteristics of WVMs in the ECV including HPCs

SamplesHPCs (CFU· mL−1)Total coliforms (MPN·100 mL−1)
WVM1 <1 <1 
WVM2 <1 <1 
WVM3 <1 2.1 
WVM4 <1 <1 
WVM5 6.9 ×103 <1 
WVM6 <1 <1 
WVM7 3.5 ×103 12.2 
WVM8 3.2 ×106 <1 
WVM9 1.13 ×104 <1 
WVM10 <1 <1 
WVM11 <1 <1 
WVM12 <1 <1 
WVM13 <1 <1 
WVM14 8.3 ×104 <1 
WVM15 <1 <1 
WVM16 <1 12.1 
WVM17 <1 601.5 
WVM18 <1 <1 
WVM19 <1 1,011.2 
WVM20 <1 17.1 
WVM21 <1 185 
WVM22 <1 260.3 
WVM23 <1 <1 
WVM24 <1 <1 
WVM25 <1 <1 
SamplesHPCs (CFU· mL−1)Total coliforms (MPN·100 mL−1)
WVM1 <1 <1 
WVM2 <1 <1 
WVM3 <1 2.1 
WVM4 <1 <1 
WVM5 6.9 ×103 <1 
WVM6 <1 <1 
WVM7 3.5 ×103 12.2 
WVM8 3.2 ×106 <1 
WVM9 1.13 ×104 <1 
WVM10 <1 <1 
WVM11 <1 <1 
WVM12 <1 <1 
WVM13 <1 <1 
WVM14 8.3 ×104 <1 
WVM15 <1 <1 
WVM16 <1 12.1 
WVM17 <1 601.5 
WVM18 <1 <1 
WVM19 <1 1,011.2 
WVM20 <1 17.1 
WVM21 <1 185 
WVM22 <1 260.3 
WVM23 <1 <1 
WVM24 <1 <1 
WVM25 <1 <1 

Our qPCR results showed that 81% of WVMs sampled had Salmonella spp., Listeria monocytogenes and Campylobacter jejuni, 76% had Enterococcus faecalis, and 90% Pseudomonas aeruginosa. Results from the qPCR from the 25 WVMs are presented in Figure 4.

Figure 4

qPCR results of Percent Positives of elected Bacteria in WVMs.

Figure 4

qPCR results of Percent Positives of elected Bacteria in WVMs.

Close modal

The pH measures the acid–base equilibrium achieved by compounds dissolved in water, as well as extent of coagulation and flocculation processes of water-borne chemicals (Masood et al. 2015). Although the World Health Organization (WHO) guidelines recommend pH ranges should be maintained between 6.5 and 8.5, we found that 52% of water samples from WVMs in the ECV had pH below this recommended value. In WVM 1 we recorded a pH of 9.5, which can lead to metal leaching in water (Pehkonen et al. 2002). Electrical conductivity measures the ability of water to transmit an electrical current and is measured on a scale from 0 to 50,000 μS·cm−1 and is used to indicate taste and scaling problems. There are no guidelines given by the WHO concerning EC. In this study, the highest EC (9 μS·cm−1) was measured in water from WVM 22. Since high conductivity can lower water quality by giving it a mineral taste, based on our results, we suggest that water from the WVMs we tested had less ionic material with no odor. These results indicate that drinking water samples had low ionic concentrations and therefore low dissolved solids. Variation in conductivity among samples may be caused by end-consumer application filtration methods and changes in water temperatures, among other factors. Total dissolved solids are the inorganic and organic matter present as solutes in water. In general, the quantity of TDS is proportional to the degree of pollution in a water sample (Butler & Ford 2018). Although there are no guidelines established for TDS, WHO recommends 600 mg·L−1 or less for aesthetic reasons (WHO 2017). The values found from drinking water samples we investigated were all within the aesthetic range proposed by the WHO. These results indicate that filtration methods used in WVMs of the ECV were successful in removing inorganic minerals and salts, such as potassium, calcium, sodium, bicarbonates, chlorides, magnesium, and sulfates, among others. However, the presence of residual chlorine in drinking water from vending machines indicates filtration failure or complete absence of filtration for WVMs directly connected to the water district's supply tap. Generally, filtration methods used for WVMs remove chlorine from drinking water samples to improve the taste of water. We found that 40% of WVMs had no residual chlorine. However, the removal of chlorine in drinking water could cause the reactivation of viable, although not culturable microorganisms, as well as poor water quality (Zhang & DiGiano 2002; Li et al. 2018).

We suggest the considerable variance in concentrations of HPC across samples may be explained by bacterial regrowth in drinking water. The presence of heterotrophic bacteria in our water samples suggested that these WVMs have the potential for pathogenic microorganisms and biofouling. Furthermore, it signifies that there may be a problem with filtration systems or maintenance at WVMs, and that regrowth occurs. In Tucson, Arizona, Chaidez et al. (1999) found 73% of samples had HPC in drinking water from WVMs greater than 500 CFU·mL−1. Microbiological quality control of drinking water from WVMs is essential to ensure compliance with standard regulations. Results of the current study showed that 21% of WVM samples had HPC above standard regulations set by the EPA and are likely indicators of poor water treatment. These HPC results are similar to those of Lévesque et al. (1994), who found a significant count of HPC in bottled water coolers in Quebec City, Canada. When connected to a water district's main, WVMs require a minimum of maintenance, with the carbon filter requiring servicing at 3–6 month intervals, the UV lamp at 1 year, and the RO membrane replaced every 2–3 years. The carbon filter and RO membrane are two components that can quickly become fouled with biofilm if not changed according to the recommended schedule. HPC bacteria can quickly colonize carbon filters producing more bacteria in the product water than the source (Raymond et al. 1979). Tubing, spigots, nozzles, and drains between treatment steps may also support biofilm development and are not often maintained or sanitized when the carbon, RO, and UV components are exchanged.

Lévesque et al. (1994) found that 44% of the nozzles on home water coolers in Quebec City, Canada, were colonized by coliforms, while Kneller et al. (1990) found the drains of automatic WVMs in the USA were contaminated with HPC by up to 106 CFU per sampled area, and had 23% of the drains contaminated with E. coli, which was noted to be a potential problem for transmission of infectious microorganisms.

Our qPCR results showed that 81% of WVMs studied had Salmonella spp., Listeria monocytogenes and Campylobacter jejuni, 76% had Enterococcus faecalis, and 90% had Pseudomonas aeruginosa. These results can be explained by the fact that qPCR analyses identify any DNA fragments (free DNA, inactivated, or DNA fragments from viable microorganisms) in very low concentrations. We identified several species of disease-causing pathogens that may threaten the general health of the ECV community. According to the EPA National Primary Drinking Water Regulations (EPA 2020), These microorganisms should not be present in drinking water at any time. Overall, water from WVMs failed to meet criteria of water quality safe for consumption. Most WVMs either had HPC values that were higher than acceptable (500 CFU.mL−1). As recommendations, we propose the FDA implement a requirement for vendors of WVMs to flush their drains and pipes, and cleanse nozzles at least once per month. This will remove biofilms formed inside hoses and reduce microbial concentrations. Flushing and nozzle cleaning could potentially be integrated into current machine maintenance schedules. Hoses coated with nanoparticles able to control microbial growth may be a preferred approach in controlling biofilm formation in nozzles and drains. In public locations, WVMs should be avoided if recent, dated inspection certificates are absent that indicate the device has been inspected and maintained within the last year. These certificates and electronic validations assist the consumer in understanding that the machine's filters were changed at least within the preceding year.

The current study addresses a serious problem regarding the microbiological quality of drinking water from WVMs in the ECV and stresses the importance of enforcing governmental monitoring and population awareness of the risks of microbial contamination from WVMs. There is a need for future studies to address the presence of biofilms and rust in relation to WVMs contamination, to quantify the pathogenic microorganisms present in WVMs, and to determine the health risks these microorganisms pose for populations in both local and global contexts.

We thank the California Institute for Rural Studies (CIRS), the Metropolitan Water District World Water Forum funding, and the Loma Linda University Department of Earth and Biological Sciences for funding this project. We would like to thank Brian McMinn of the USEPA National Exposure Research Laboratory for providing NanoCeram® filters. Our appreciation is extended to the Kerby Oberg laboratory for use of their instruments. We wish to thank Brittney Springer, Erika Altamirano, Anderson Garrett, Kate Ball, and the summer students that participated to this project, as well as Dustin Baumbach for his cartographic skills in producing the ECV map. We are grateful to three anonymous reviewers for their suggestions for improving the clarity of the manuscript. This research is part of the Loma Linda University Environmental Microbiology Research Laboratory.

Data cannot be made publicly available; readers should contact the corresponding author for details.

Bird
K.
Boopathy
R.
Nathaniel
R.
LaFleur
G.
2019
Water pollution and observation of acquired antibiotic resistance in Bayou Lafourche, a major drinking water source in Southeast Louisiana, USA
.
Environmental Science and Pollution Research
26
(
33
),
34220
34232
.
Boratyn
M.
Thierry-Mieg
J.
Thierry-Mieg
D.
Busby
B.
Madden
L.
2019
, Jul 25
Magic-Blast, an accurate Rna-Seq aligner for long and short reads
.
BMC Bioinformatics
20
(
1
),
405
.
https://doi.org/10.1186/s12859-019-2996-x
.
Butler
A.
Ford
G.
2018
Evaluating relationships between total dissolved solids (Tds) and total suspended solids (Tss) in a mining-influenced watershed
.
Mine Water and the Environment
37
(
1
),
18
30
.
https://doi.org/10.1007/s10230-017-0484-y
.
California Department of Public Health
2014
Water Vending Machine, Inspection Procedure & Operating Requirements
.
California Legislative Information
1995
(11/22/2020). Bottled, Vended, Hauled, and Processed Water. California Legislative Information. Retrieved 11/25/20, available from https://leginfo.legislature.ca.gov/faces/codes_displayText.xhtml?lawCode=HSC&division=104.&title=&part=5.&chapter=5.&article=12
Cardaci
R.
Burgassi
S.
Golinelli
D.
Nante
N.
Battaglia
M.
Bezzini
D.
Messina
G.
2016
Automatic vending-machines contamination: a pilot study
.
Global Journal of Health Science
9
(
2
),
63
.
Chaidez
C.
Rusin
P.
Naranjo
J.
Gerba
C. P.
1999
Microbiological quality of water vending machines
.
International Journal of Environmental Health Research
9
(
3
),
197
206
.
Craun
F.
Brunkard
M.
Yoder
S.
Roberts
A.
Carpenter
J.
Wade
T.
Calderon
L.
Roberts
M.
Beach
J.
Roy
L.
2010
Causes of outbreaks associated with drinking water in the United States from 1971 to 2006
.
Clinical Microbiology Reviews
23
(
3
),
507
528
.
https://doi.org/10.1128/CMR.00077-09
.
Denchak
M.
2018
(November 08, 2018). Flint Water Crisis: Everything You Need to Know. The Natural Resources Defense Council (NRDC). Retrieved 11/25/20, available from https://www.nrdc.org/stories/flint-water-crisis-everything-you-need-know
Feinstein
L.
2018
Measuring progress toward universal access to water and sanitation I California defining goals, indicators, and performance measures
.
Pacific Institute
57
,
Girolamini
L.
Lizzadro
J.
Mazzotta
M.
Iervolino
M.
Dormi
A.
Cristino
S.
2019
Different trends in microbial contamination between two types of microfiltered water dispensers: from risk analysis to consumer health preservation
.
International Journal of Environmental Research and Public Health
16
(
2
),
272
.
https://doi.org/10.3390/ijerph16020272
.
Hill
R.
Polaczyk
L.
Hahn
D.
Narayanan
J.
Cromeans
T.
Roberts
J.
Amburgey
J.
2005
Development of a rapid method for simultaneous recovery of diverse microbes in drinking water by ultrafiltration with sodium polyphosphate and surfactants
.
Applied and Environmental Microbiology
71
(
11
),
6878
6884
.
https://doi.org/10.1128/AEM.71.11.6878-6884.2005
.
Hu
Z.
Morton
L.
Mahler
R.
2011
Bottled water: United States consumers and their perceptions of water quality
.
International Journal of Environmental Research and Public Health
8
(
2
),
565
578
.
https://doi.org/10.3390/ijerph8020565
.
Kneller
P.
Jayaswal
R.
Eils
L.
1990
Sanitation controls for cold Cup soft drink vending machines
.
Dairy, Food and Environmental Sanitation
10
(
8
),
499
502
.
Lévesque
B.
Simard
P.
Gauvin
D.
Gingras
S.
Dewailly
E.
Letarte
R.
1994
Comparison of the microbiological quality of water coolers and that of municipal water systems
.
Applied and Environmental Microbiology
60
(
4
),
1174
1178
.
Liguori
G.
Cavallotti
I.
Arnese
A.
Amiranda
C.
Anastasi
D.
Angelillo
I.
2010
Microbiological quality of drinking water from dispensers in Italy
.
BMC Microbiology
10
,
19
19
.
https://doi.org/10.1186/1471-2180-10-19
.
Liu
H.
Whitehouse
C.
Li
B.
2018
Presence and persistence of Salmonella in water: the impact on microbial quality of water and food safety
.
Frontiers in Public Health
6
,
159
159
.
https://doi.org/10.3389/fpubh.2018.00159
.
Luby
S.
Halder
A.
Huda
T.
Unicomb
L.
Islam
M.
Arnold
B.
Johnston
R.
2015
Microbiological contamination of drinking water associated with subsequent child diarrhea
.
The American Journal of Tropical Medicine and Hygiene
93
(
5
),
904
911
.
https://doi.org/10.4269/ajtmh.15-0274
.
Masood
Z.
Rehman
H.
Baloch
A.
Akbar
N.
Zakir
M.
Gul
I.
Gul
N.
Jamil
N.
Din
N.
Ambreen
B.
2015
Analysis of physicochemical parameters of water and sediments collected from rawal dam Islamabad
.
American-Eurasian Journal of Toxicological Sciences
7
(
3
),
123
128
.
Patel
A. I.
Grummon
A. H.
Hampton
K. E.
Oliva
A.
McCulloch
C. E.
Brindis
C. D.
2016
A trial of the efficacy and cost of water delivery systems in San Francisco bay area middle schools, 2013
.
Preventing Chronic Disease
13
,
E88
E88
.
https://doi.org/10.5888/pcd13.160108
.
Pehkonen
S.
Palit
A.
Zhang
X.
2002
Effect of specific water quality parameters on copper corrosion
.
Corrosion
58
(
2
),
156
165
.
Philip
A.
Sims
E.
Houston
J.
Konieczny
R.
2017
63 Million Americans Exposed to Unsafe Drinking Water. USA Today (August 15, 2017)
.
Pueblo Unido Community Development Corporation
2020
Building Sustainable Communities in the Eastern Coachella Valley
.
Retrieved 11/25/20, available from: https://pucdc.org/
Quast
C.
Pruesse
E.
Yilmaz
P.
Gerken
J.
Schweer
T.
Yarza
P.
Peplies
J.
Glöckner
F. O.
2012
The silva ribosomal Rna gene database project: improved data processing and web-based tools
.
Nucleic Acids Research
41
(
D1
),
D590
D596
.
https://doi.org/10.1093/nar/gks1219
.
Raymond
T.
Martin
A.
Edwin
G.
1979
Testing of home use carbon filters
.
Journal of the American Water Works Association
71
(
10
),
577
579
.
Schillinger
J.
Du Vall Knorr
S.
2004
Drinking-water quality and issues associated with water vending machines in the city of Los Angeles
.
Journal of Environmental Health
66
,
25
31
.
43; quiz 45–46
.
Sharp
R.
Walker
B.
2002
Is Water From Vending Machines Really ‘Chemical-Free’?
Environmental Working Group
.
The Metropolitan Water District of Southern California
1999
Community Outreach
.
The Metropolitan Water District of Southern California
.
United Nations
2010
International Decade for Action ‘Water for Life’ 2005–2015
.
United Nations
.
United States Environmental Protection Agency
2020
(February 14, 2020) National Primary Drinking Water Regulations. United States Environmental Protection Agency. Available from: https://www.epa.gov/ground-water-and-drinking-water/national-primary-drinking-water-regulations
Whiley
H.
van den Akker
B.
Giglio
S.
Bentham
R.
2013
The role of environmental reservoirs in human campylobacteriosis
.
International Journal of Environmental Research and Public Health
10
(
11
),
5886
5907
.
https://doi.org/10.3390/ijerph10115886
.
World Health Organization
2017
(24 April 2017)
Guidelines for Drinking-Water Quality: Fourth Edition Incorporating the First Addendum
.
World Health Organization
.
Retrieved 11/25/20, available from: https://www.who.int/publications/i/item/9789241549950
World Health Organization & UNICEF
2014
Progress on Drinking Water and Sanitation: 2014 Update
. In:
Progress on Drinking Water and Sanitation: 2014 Update
.
World Health Organization
, p.
80
.
This is an Open Access article distributed under the terms of the Creative Commons Attribution Licence (CC BY-NC-ND 4.0), which permits copying and redistribution for non-commercial purposes with no derivatives, provided the original work is properly cited (http://creativecommons.org/licenses/by-nc-nd/4.0/).