Final Report: Effectiveness of UV Irradiation for Pathogen Inactivation in Surface Waters

EPA Grant Number: R829012
Title: Effectiveness of UV Irradiation for Pathogen Inactivation in Surface Waters
Investigators: Linden, Karl G. , Sobsey, Mark D.
Institution: Duke University , University of North Carolina at Chapel Hill
EPA Project Officer: Page, Angela
Project Period: August 20, 2001 through August 19, 2004 (Extended to August 19, 2005)
Project Amount: $524,848
RFA: Drinking Water (2000) RFA Text |  Recipients Lists
Research Category: Drinking Water , Water

Objective:

Ultraviolet (UV) radiation is recognized to be an inexpensive and relatively easy means to achieve disinfection of Cryptosporidium parvum and does not appear to produce disinfection byproducts at practical doses. The germicidal effects of UV against emerging pathogens and challenges related to application of UV disinfection for filtered and unfiltered surface waters needs to be assessed. The objective of this research project was to evaluate the susceptibility, repair potential, and resistance of select Contaminant Candidate List (CCL) and other pathogens and indicators to UV disinfection from low and medium pressure UV sources. The extent to which microbes are associated with water treatment particles typical in unfiltered systems and the effects of this particle association and other water quality parameters on UV disinfection also was investigated.

Our research approach integrated the expertise and experience of the co-investigators in environmental microbiology and UV dosimetry and disinfection process design to address these important questions and others. Specifically, numerous organisms, some of which are on the U.S. Environmental Protection Agency CCL, including bacteria (Mycobacterium spp.) and viruses (adenoviruses, coxsackieviruses, Echovirus, feline calicivirus, murine norovirus, poliovirus), in addition to protozoa (Giardia lamblia, Toxoplasma gondii) and various indicator organisms were evaluated for UV dose response to monochromatic and/or polychromatic UV irradiation. The relative wavelength effectiveness of UV irradiation for a virus and spore and the potential for repair and reactivation under light and dark conditions for a bacteria and cyst also were studied. The extent to which native and spiked microorganisms are associated with particles from water treatment and the effects of particle association and clumping of microorganisms on UV inactivation also was investigated, including the effect of upstream coagulation/flocculation protocols on microbe-particle association.

Summary/Accomplishments (Outputs/Outcomes):

UV Dose-Response for Pathogens and Indicators

Viruses and Bacteriophage. To elucidate the roles of physical and chemical properties of viruses and their sensitivity to UV radiation, the kinetics and extent of inactivation of some waterborne pathogenic viruses and bacteriophages with different virion sizes and genomic composition by monochromatic, low pressure UV was determined in phosphate buffered saline or a filtered drinking water. The inactivation rates of the small RNA viruses, poliovirus 1, and coxsackievirus B4, by low pressure UV were very rapid and reached approximately 4 log and greater than 5 log, respectively, within a UV dose of 30 mJ/cm2. In contrast, the inactivation of the small RNA bacteriophage, MS2, was much slower and only 2-log inactivation was achieved at a UV dose of 30 mJ/cm2. The inactivation of the large DNA virus, adenovirus 2, was relatively slow and once again, only a 2-log inactivation was achieved with a UV dose of 60 mJ/cm2. In contrast, the inactivation rates of the three large DNA bacteriophages were very rapid and reached greater than5 log with a UV dose of 10 mJ/cm2. Therefore, the inactivation of human enteric viruses and bacteriophages by UV irradiation is proven not predictable by the type and size of the virus or its nucleic acid genome.

The UV inactivation of murine norovirus, feline calicivirus, and echovirus12 was studied in phosphate buffer at room temperature using a low pressure UV source. All three viruses exhibited 2-log reduction following a UV dose of 15 mJ/cm2, whereas 4-log reductions were achieved at less than 30 mJ/cm2. Bacteriophage MS2 was more UV resistant than any of these viruses tested. Additionally, UV disinfection experiments were carried out using cell-associated echovirus 12. Compared to monodispersed echovirus 12, the cell associated virus required two to three times greater UV dose for a given level of log reduction. For instance, a dose of 85 mJ/cm2 was required for 4-log reduction of cell associated echovirus 12, whereas the non-cell associated virus required only a dose of 30 mJ/cm2.

Because adenovirus is the most UV resistant virus known, it was selected for more extensive study. As part of the adenovirus work, we developed a rapid and reliable cell culture-mRNA reverse transcription-polymerase chain reaction (RT-PCR) assay to detect and quantify adenovirus infectivity. This method was used to assess adenovirus type 41 (Ad41) infectivity. Treatment of approximately 104 Ad41 with different doses of 254 nm germicidal UV radiation resulted in a dose-dependent loss of infectivity. As UV doses were increased from 75 to 200 mJ/cm2, virus survival decreased and no virus infectivity (measured by detectable mRNA) was found at a dose of 225 mJ/cm2 or higher. These results using the cell culture mRNA RT-PCR assay indicate that Ad41 is more resistant to UV radiation than suggested in a previous study using conventional cell culture infectivity assay. The comparative effects of low and medium pressure UV irradiation on adenovirus 5 (Ad5) and Ad41 were studied in model waters. UV doses of 40, 80, and 120 mJ/cm2 resulted in 1.8-, 2.9-, and 3.5-log inactivation with low pressure lamps and 2.9-, 3.8-, and 4.7-log inactivation with medium pressure lamps, respectively. Low and medium pressure UV applied to dispersed Ad41 in buffered laboratory water resulted in log10 reductions of 0.35, 1.1, and 2.4 with low pressure lamps and 0.8, 1.6, and 3.4 with medium pressure lamps at UV doses of 50, 100, and 150 mJ/cm2, respectively. Thus, at the same UV doses the polychromatic medium pressure lamps inactivated adenoviruses more effectively than did the low pressure lamps.

Overall, bacteriophage MS2 was a good conservative surrogate for representing the UV inactivation of many viruses, with the exception of adenovirus. These data will be helpful for utilities and regulatory officials to evaluate the efficacy of UV irradiation for viruses and plan for levels of disinfection that will help ensure public health in drinking water.

Bacteria. UV inactivation and subsequent dark and photo-repair was examined for Mycobacterium terrae, a surrogate for Mycobacterium avium complex (MAC) and M. tuburculosis. Low pressure (monochromatic, 254 nm) and medium pressure (polychromatic UV output) Hg lamps were used for UV irradiation resulting in inactivation, and was followed by dark or photo-repair experiments. M. terrae was more resistant to UV disinfection than many other bacteria, with approximately 2-log reduction at a UV fluence of 10 mJ/cm2, similar to UV inactivation of M. tuberculosis. There was no difference in inactivation between monochromatic or polychromatic UV lamps. M. terrae did not undergo detectable dark repair. Photorepair resulted in recovery from inactivation by approximately 0.5 log in less than 30 minutes for both UV lamp systems. There was no practical difference between UV sources for disinfection or prevention of DNA repair for M. terrae.

UV disinfection tests were also performed on an environmental isolate of MAC, provided by Dr. Cangelosi at the University of Washington. The inactivation kinetics ofMAC by several different doses of both low pressure and medium pressure UV irradiation in phosphate-buffered saline at room temperature were evaluated. The inactivation rates of MAC by both UV irradiation types was rapid and reached greater than 4 log10 within a UV dose of 10 to 15 mJ/cm2. M. terrae appears to be a conservative surrogate for UV inactivation of MAC.

Protozoa. The inactivation of G. lamblia cysts by low pressure UV irradiation was determined using a gerbil infectivity test. Reduction of G. lamblia infectivity was very rapid, reaching a detection limit of greater than 4 log within a dose of 1 mJ cm-2. The ability of UV-irradiated G. lamblia cysts to repair UV-induced damage following UV doses of 16 and 40 mJ cm-2 also was investigated, and there was no phenotypic evidence of either light or dark repair of DNA damage caused by low pressure UV irradiation of cysts at the UV doses tested.

The UV inactivation of T. gondii was carried out with the assistance of Dr. Dubey at the U.S. Department of Agriculture and Dr. Simmons at the University of North Carolina at Chapel Hill. T. gondii oocysts were thought to be resistant to UV disinfection although extensive studies had not yet been performed. Using a mouse infectivity test protocol carried out by Dr. Dubey, the inactivation of T. gondii with UV irradiation was found to be very effective. Using low pressure UV a 4-log reduction in infectivity was achieved at a UV dose of approximately 8 mJ/cm2 in reagent water, with similar findings in natural water tests. Polychromatic UV irradiation from a medium pressure UV system provided a similar level of inactivation. Thus in addition to the known effectiveness of UV toward Cryptosporidium oocysts and Giardia cysts, UV is very effective for inactivation of Toxoplasma oocysts at low UV doses.

Indigenous Spores. Indigenous aerobic spores naturally occurring in raw/unfiltered water supplies were very resistant to UV irradiation compared to a range of different microbes in the literature (i.e., adenovirus, MS2 coliphage, and C. parvum). The inactivation of indigenous natural aerobic spores followed first-order kinetics with an inactivation coefficient ranging between 0.013 and 0.022 cm2/mJ. These data indicate a dose of between 45-77 mJ/cm2 is required for 1-log reduction, and a dose of 136-231 mJ/cm2 is required for 3-log UV inactivation. Thus, naturally occurring Bacillus spores are more UV resistant than adenovirus. These spores may be useful as surrogates for adenovirus or as biodosimetry test organisms for UV reactor validation.

Wavelength Effectiveness of UV Irradiation

The microbicidal UV fluence under polychromatic radiation from UV lamps is typically measured using the DNA absorbance spectrum as a weighting factor for the relative wavelength effectiveness. However, this DNA-based weighting does not necessarily match the spectral sensitivity of the microorganism being tested. Bacillus subtilis spores are often used for UV reactor validation in Europe, whereas MS2 coliphage is typically used for validation testing in the United States. These organisms were exposed to quasimonochromatic UV irradiation across the microbicidal spectrum at wavelengths of 214, 230, 240, 254, 265, 280, and 293 nm. MS2 was three times more sensitive to wavelengths near 214 nm compared to the 254 nm output of low pressure lamps, whereas B. subtilis spores were most sensitive to wavelengths around 265 nm. Use of these action spectra, compared to the DNA-based weighting, resulted in differences in the calculated polychromatic UV fluence. Consequently, the action spectrum, which is specific for each microorganism, has implications on the uncertainty of UV fluence determination during validation of reactors with polychromatic UV lamps. Furthermore, these data indicate that for some microorganisms, a polychromatic medium pressure UV source may be more germicidally effective than a low pressure source, but these effects need to be tested in a microorganisms specific basis.

Water Quality Impacts on UV Disinfection

Based on studies of the UV inactivation kinetics of indigenous aerobic spores in surface water compared to their laboratory-cultured spore isolates a relationship between physicochemical characteristics and UV inactivation kinetics of spore isolates was developed. Cultured isolated spores exhibited a three-stage inactivation curve consisting of shoulder, first order, and tailing regions, whereas indigenous spores exhibited only one stage of linear kinetics. Hydrophobicity of the Bacillus spore isolates was inversely related to the extent of UV inactivation before tailing occurred. Therefore, tailing in the UV inactivation curves results from aggregation of a portion of the spore population because of hydrophobic interactions, supporting the link between aggregation of spores, hydrophobicity, and UV inactivation.

Aggregation of microbes with particles can also reduce the effectiveness of UV disinfection. Aggregates that were induced by flocculation with alum were protected from UV irradiation compared to nonaggregated spores, and the difference between these systems was found to be statistically significant throughout the UV dose range tested. Electron microscopy analysis suggested that aggregate composition was nonhomogeneous with respect to the ratio of spores and clay particles among aggregates. It was estimated that 30–50 percent of the spores in the aggregates tested were protected from UV irradiation.

The impact of light scattering of particle aggregated microbes on UV disinfection was evaluated by comparing standard spectrophotometer and integrating sphere absorbance measurements for UV fluence determination. Coagulated systems significantly decreased the UV inactivation effectiveness compared to the noncoagulated system with the effects more pronounced for raw natural water. Absorbance measurement of suspensions and aggregates using standard spectrophotometry in the calculations of fluence resulted in overdosing, whereas the use of integrating sphere spectroscopy did not. The results demonstrated that aggregation protected spores from UV disinfection, and that use of proper absorbance measurement techniques, accounting for particle scattering, is essential for correct interpretation of the results.


Journal Articles on this Report : 11 Displayed | Download in RIS Format

Other project views: All 25 publications 12 publications in selected types All 12 journal articles
Type Citation Project Document Sources
Journal Article Bohrerova Z, Linden KG. Assessments of DNA damage and repair in Mycobacterium terrae after exposure to UV irradiation. Journal of Applied Microbiology 2006;101(5):995-1001. R829012 (Final)
  • Abstract from PubMed
  • Journal Article Ko G, Cromeans TL, Sobsey MD. Detection of infectious adenovirus in cell culture by mRNA reverse transcription-PCR. Applied and Environmental Microbiology 2003;69(12):7377-7384. R829012 (2004)
    R829012 (Final)
  • Abstract from PubMed
  • Full-text:
    Exit
  • Other: Applied and Environmental Microbiology PDF
    Exit
  • Journal Article Ko G, Cromeans TL, Sobsey MD. UV inactivation of adenovirus type 41 measured by cell culture mRNA RT-PCR. Water Research 2005;39(15):3643-3649. R829012 (2003)
    R829012 (Final)
  • Abstract from PubMed
  • Full-text: Science Direct Full Text
    Exit
  • Other: Science Direct PDF
    Exit
  • Journal Article Linden KG, Shin G-A, Faubert G, Cairns W, Sobsey MD. UV disinfection of Giardia lamblia cysts in water. Environmental Science & Technology 2002;36(11):2519-2522. R829012 (Final)
  • Abstract from PubMed
  • Full-text: ACS Full Text
    Exit
  • Other: ACS PDF
    Exit
  • Journal Article Mamane H, Ducoste JJ, Linden KG. Effect of particles on ultraviolet light penetration in natural and engineered systems. Applied Optics 2006;45(8):1844-1856. R829012 (Final)
  • Abstract from PubMed
  • Journal Article Mamane H, Linden KG. Impact of particle aggregated microbes on UV disinfection. II: Proper absorbance measurement for UV fluence. Journal of Environmental Engineering 2006;132(6):607-615. R829012 (Final)
  • Abstract: ASCE Abstract
    Exit
  • Journal Article Mamane H, Linden KG. Impact of particle aggregated microbes on UV disinfection. I: Evaluation of spore-clay aggregates and suspended spores. Journal of Environmental Engineering 2006;132(6):596-606. R829012 (Final)
  • Abstract: ASCE Abstract
    Exit
  • Journal Article Mamane-Gravetz H, Linden KG. UV disinfection of indigenous aerobic spores: implications for UV reactor validation in unfiltered waters. Water Research 2004;38(12):2898-2906. R829012 (2003)
    R829012 (2004)
    R829012 (Final)
  • Abstract from PubMed
  • Full-text: Science Direct Full Text
    Exit
  • Other: Science Direct PDF
    Exit
  • Journal Article Mamane-Gravetz H, Linden KG. Relationship between physiochemical properties, aggregation and U.V. inactivation of isolated indigenous spores in water. Journal of Applied Microbiology 2005;98(2):351-363. R829012 (2004)
    R829012 (Final)
  • Abstract from PubMed
  • Journal Article Mamane-Gravetz H, Linden KG, Cabaj A, Sommer R. Spectral sensitivity of Bacillus subtilis spores and MS2 Coliphage for validation testing of ultraviolet reactors for water disinfection. Environmental Science & Technology 2005;39(20):7845-7852. R829012 (Final)
  • Abstract from PubMed
  • Other: Environmental Science & Technology PDF
    Exit
  • Journal Article Shin G-A, Linden KG, Sobsey MD. Low pressure ultraviolet inactivation of pathogenic enteric viruses and bacteriophages. Journal of Environmental Engineering and Science 2005;4(S1):S7-S11. R829012 (2003)
    R829012 (Final)
  • Abstract: NRC Research Press Abstract
    Exit
  • Supplemental Keywords:

    environmental engineering, UV inactivation, action spectra, aggregation, CCL microbes, disinfection, microbiology, drinking water, health risk assessment,, RFA, Scientific Discipline, Water, Environmental Chemistry, Health Risk Assessment, Ecological Risk Assessment, Ecology and Ecosystems, Drinking Water, Environmental Engineering, cryptosporidium parvum oocysts, pathogens, other - exposure, monitoring, CCL, chlorination, microbiological organisms, exposure and effects, disinfection byproducts (DPBs), exposure, UV treatment, Other - risk management, chlorine-based disinfection, cryptosporidium , public health, treatment, microbial risk management, water quality, DBP risk management, drinking water contaminants, drinking water treatment, Giardia, water treatment, contaminant candidate list, drinking water system, dosimetry

    Relevant Websites:

    http://www.cee.duke.edu/faculty/linden/index.php Exit
    http://www.unc.edu/~sobsey/ Exit

    Progress and Final Reports:

    Original Abstract
    2002 Progress Report
    2003 Progress Report
    2004 Progress Report