Dave S. Bachoon

Quantification of enterococci and bifidobacteria in Georgia estuaries using conventional and molecular methods.

Fecal pollution is a serious threat to the estuarine environment along the Georgia coast. Culture-dependant and molecular methodologies were utilized to compare and evaluate the abundance of fecal indicator bacteria in four Georgia estuaries (Darien River, Frederica River, Gulley Hole Creek, and St. Marys River). The functionality of enterococci and bifidobacteria as indicator organisms in marine environments was assessed, as well as Bifidobacterium adolescentis densities. At each study site, enterococci were enumerated as colony forming units (CFU) on mEI agar. For quantitative polymerase chain reaction (qPCR), genus- and species-specific primer sets were used to quantify bifidobacteria and B. adolescentis as 16S rRNA gene copies and enterococci as tuf gene copies. A high correlation (r=0.925) was observed between CFU and qPCR enumeration of enterococci. Enterococci densities in the estuarine rivers ranged from 3-449CFU/100ml on mEI plates and 4.58-5.39Log(10) gene copies/100ml by qPCR. Bifidobacteria densities ranged from 3.62-4.14Log(10) gene copies/100ml and suggested the Frederica River as least affected by fecal bacteria and the Darien River as most affected by fecal pollution. A correlation of 0.46 was observed among qPCR densities of enterococci and bifidobacteria at all sample sites. Quantitative polymerase chain reaction detection of B. adolescentis was a rapid (i.e., less than 2h) indicator of presumptive human fecal pollution and suggested that Gulley Hole Creek, the Darien River, and the St. Marys River were affected by fecal bacteria derived from a human source. Gulley Hole Creek and the Darien River had the highest levels of fecal pollution detected in the studied estuaries. Molecular quantification of bifidobacteria may be a more accurate method of determining immediate health risks associated with fecal pollution in estuarine water than traditional and contemporary assessments of enterococci.

Effects of humic substances on fluorometric DNA quantification and DNA hybridization

DNA extracts from sediment and water samples are often contaminated with coextracted humic-like impurities. Estuarine humic substances and vascular plant extract were used to evaluate the effect of the presence of such impurities on DNA hybridization and quantification. The presence of humic substances and vascular plant extract interfered with the fluorometric measurement of DNA concentration using Hoechst dye H33258 and PicoGreen reagent. Quantification of DNA amended with humic substances (20-80 ng/microl) using the Hoechst dye assay was more reliable than with PicoGreen reagent. A simple procedure was developed to improve the accuracy for determining the DNA concentration in the presence of humic substances. In samples containing up to 80 ng/microl of humic acids, the fluorescence of the samples were measured twice: first without Hoechst dye to ascertain any fluorescence from impurities in the DNA sample, followed with Hoechst dye addition to obtain the total sample fluorescence. The fluorescence of the Hoechst dye-DNA complex was calculated by subtracting the fluorescence of the impurities from the fluorescence of the sample. Vascular plant extract and humic substances reduced the binding of DNA onto the nylon membrane. Low amounts (<2.0 microg) of humic substances derived from estuarine waters did not affect the binding of 100 ng of target DNA to nylon membranes. DNA samples containing 1.0 microg of humic substances performed well in DNA hybridizations with DIG-labeled oliogonucleotide and chromosomal probes. Therefore, we suggest that DNA samples containing low concentrations of humic substances (<20 ng/microl) could be used in quantitative membrane hybridization without further purification.

Comparison of four polymerase chain reaction methods for the rapid detection of human fecal pollution in marine and inland waters.

We compared the effectiveness of three PCR protocols for the detection of Bifidobacterium adolescentis and one PCR protocol for detecting Bacteroidales as indicators of human fecal pollution in environmental samples. Quantitative PCR indicated that a higher concentration of B. adolescentis DNA was recovered from sewage samples on the 0.2 μm filters compared to the 0.45 μm filters, and there was no evidence of qPCR inhibitors in the DNA extracts. With the Matsuki method (1999), B. adolescentis was detected only in undiluted sewage samples. The King method (2007) performed well and detected B. adolescentis in all of the sewage dilutions (from undiluted to 10−4). In contrast, the Bonjoch approach (2004) was effective at detecting B. adolescentis at lower dilutions (10−3) of sewage samples and it gave false positive results with some (3/8) pig fecal samples. Human-specific Bacteroidales (HuBacs) were detected in the lower diluents of sewage samples but was positive in pig (6/8) and cattle fecal samples. PCR detection of B. adolescentis in marine samples from Puerto Rico and freshwater samples from Georgia indicated that the PCR method of King et al. (2007) and the modified Layton method for HuBac were in agreement in detecting human fecal pollution in most sites.

Assessment of non-point sources of fecal pollution in coastal waters of Puerto Rico and Trinidad.

Traditional and molecular methods (PCR) were used to detect, quantify and identify the source of fecal pollution in coastal sites of Puerto Rico and Trinidad. Enterococci and Escherichia coli standard plate counts were used as a general indicator of fecal contamination while the PCR detection of Bifidobacteria adolescentis and human or bovine specific Bacteroidales were used to examine potential sources. Seven of 14 sites in Trinidad including Maracas Bay which is a major public beach contained significant fecal contamination based on enterococci numbers counts exceeding established thresholds for areas of direct contact. Forty six percent of the 27 stations in Puerto Rico were over the established thresholds for enterococci and 49% according to E. coli counts. About 31% of the stations examined in Puerto Rico had evidence of human derived fecal contamination. Human fecal pollution was detected in only one station from Trinidad. Bovine derived contamination was detected only once.

Detection of Helicobacter pylori in the coastal waters of Georgia, Puerto Rico and Trinidad

Fecal pollution in the coastal marine environments was assessed at eleven sampling locations along the Georgia coast and Trinidad, and nine sites from Puerto-Rico. Membrane filtration (EPA method 1604 and method 1600) was utilized for Escherichia coli and enterococci enumeration at each location. Quantitative polymerase chain reaction (qPCR) amplification of the 16S ribosomal RNA gene was used to determine the presence of the Helicobacter pylori in marine samples. There was no significant correlation between the levels of E. coli, enterococci and H. pylori in these water samples. H. pylori was detected at four of the 31 locations sampled; Oak Grove Island and Village Creek Landing in Georgia, Maracas river in Trinidad, and Ceiba Creek in Puerto Rico. The study confirms the potential public health risk to humans due to the widespread distribution of H. pylori in subtropical and tropical costal marine waters.

Detection of human-derived fecal contamination in Puerto Rico using carbamazepine, HF183 Bacteroides, and fecal indicator bacteria.

The level of fecal pollution in 17 sites in Puerto Rico was determined by Escherichia coli (E.coli) enumeration using an enzyme substrate medium and Quanti-Tray®/2000. Human fecal pollution was identified using an enzyme-linked immunosorbent assay for the detection of carbamazepine (CBZ) and quantitative polymerase chain reaction (qPCR) detection of the human Bacteroides marker, HF183. Carbamazepine was detected in 16 out of 17 sites, including Condado Lagoon, a popular recreational area. Elevated E.coli levels (>410 CFU 100 mL(-1)) were detected in 13 sites. Average CBZ concentrations ranged from 0.005 μg L(-1) to 0.482 μg L(-1) and 7 sites were positive for HF183. Higher CBZ concentrations were associated with the detection of HF183 (Mann-Whitney test; U=42.0; df=7; 1-tailed P value=0.013). This was the second study to determine surface water concentrations of CBZ in the Caribbean and the first in Puerto Rico.

Molecular detection of atrazine catabolism gene atzA in coastal waters of Georgia, Puerto Rico and Trinidad

In this study, quantitative polymerase chain reaction targeting the atrazine catabolism gene, atzA, was used to detect the presence of atrazine degrading bacteria as an indicator of atrazine contamination in 11 sites in Georgia, nine coastal sites in Puerto Rico and 11 coastal sites in Trinidad. The atzA gene was detected in five stations in Georgia (Oak Grove Island entrance, Blythe Island Recreation Park, Jekyll Island., Village Creek Landing and Dunbar Creek Sea Island Rd Bridge). In Puerto Rico gene was detected in five sites (Boquilla, Oro Creek, Fishers Association, Ceiba Creek and Sabalos Creek) while seven sites in Trinidad (Carli Bay, Las Cuevas Bay, Quinam Bay, Salybia River, Salybia Bay, Maracas River and Maracas Bay) showed the presence of atzA.

Detection of verotoxin producing Escherichia coli in marine environments of the Caribbean.

The goal of this study was to determine the potential for Enterohemorrhagic Escherichia coli O157:H7 (EHEC) contamination in tropical marine waters. Samples were collected from urban, suburban, and rural sites around the islands of Puerto Rico and The Republic of Trinidad and Tobago. Quantification of E. coli and EHEC was evaluated using MI plates and qPCR. EHEC was detected in six sites in Puerto Rico: West of La Parguera Town, Boquilla, Oro Creek, Fishers Association, Joyuda Lagoon, and Boqueron Wetland Creek and in two rural sites in Trinidad: Balandra Bay and Quinam Bay. Plate count enumeration of E. coli was not a reliable indicator for the presence of EHEC. The sites where EHEC was detected on both islands are used for recreational bathing, water sports and recreational/commercial fisheries and therefore pose a public potential health risk.

Evaluation of microbiological water quality in Point Judith Pond (Rhode Island, USA): Quantitation of fecal pollution and presence of human pathogenic bacteria

Water and sediment samples were collected in the Bluff Hill Cove area of Point Judith Pond from 12 stations along three parallel transects, and eelgrass samples were taken at two sites per transect. Fecal enterococci (FE) in the water samples exceeded USEPA regulatory limit of 100 MPN/100 mL for safe recreational and shellfish harvesting use in 24 out of 41 samples. Sediment samples had the lowest FE counts, whereas eelgrass had the highest counts. Using qPCR, H. pylori was detected in 8 out of 41 samples, but E. coli O157:H7 was not detected. Overall, higher levels of H. pylori were detected in 33% of the eelgrass samples and 33% of the sediment samples as compared to 9% in the water column. Microbial source tracking, using the Bacteroides marker HF183, indicated the presence of human fecal bacteria in 7 out of 41 samples, four of which were positive for H. pylori.

Evaluation of physical, chemical and microbiological parameters of water quality in the Harris Neck estuarine marshes along the Georgia coast

Analysis of the physical, chemical and biological parameters assessing water quality in Harris Neck estuary indicated that the average dissolved oxygen level was 8.6 mg/L, it maintained moderate levels of total dissolved nitrogen (2.7-4.6 mg/L) and total dissolved phosphorous (<0.05 mg/L), chlorophyll a was above 5.0 μg/L and it is contaminated with low levels of fecal bacteria. Bifidobacterium adolescentis, a putative marker of human fecal pollution, was detected once at stations 3 and 5. Overall the Harris Neck water quality analyses indicated a relatively pristine and a healthy functioning marine environment.