Human health risks associated with harvested rainwater: implementation of biocontrol strategies

Date
2020-04
Journal Title
Journal ISSN
Volume Title
Publisher
Stellenbosch : Stellenbosch University
Abstract
ENGLISH ABSTRACT: Rainwater harvesting has been earmarked as an additional fresh water source, which could be utilised to supplement municipal water supplies, especially in water scarce regions. However, various studies have indicated that the microbial quality of this water source is substandard. These microbial contaminants may pose a significant health risk to end-users and it is recommended that treatment systems are implemented to reduce the level of contamination in rainwater. Solar disinfection (SODIS) has been identified as an easy-to-use and cost-effective strategy that could be used to disinfect water. A minimum of 6 hours solar exposure is generally required for effective disinfection of water and photocatalytic nanomaterials such as titanium dioxide (TiO2) have subsequently been employed to improve SODIS efficiency by decreasing the treatment time. Research has however, indicated that while SODIS is effective in significantly reducing the concentration of microbial contaminants in water sources, various pathogens and opportunistic pathogens employ survival strategies and persist after treatment. A combination of physical, chemical and biological treatments, which target these persistent organisms directly, should therefore be investigated. For the purpose of this dissertation, the use of Bdellovibrio bacteriovorus (B. bacteriovorus), a Gram-negative predatory bacterium, was investigated. The primary aim of Chapter 2 (published in Microbiological Research, 2019) was thus to isolate B. bacteriovorus from wastewater and investigate the interaction of this predator with Gram-negative and Gram-positive prey using culture-based (spread plating and double-layer agar overlays) and molecular methods [ethidium monoazide quantitative polymerase chain reaction (EMA-qPCR)]. The predation activity of B. bacteriovorus on the different prey cells was assessed and compared in a nutrient poor [diluted nutrient broth (DNB)] and nutrient deficient medium (HEPES buffer). A B. bacteriovorus isolate (PF13) was subsequently co-cultured with Pseudomonas fluorescens (P. fluorescens), Pseudomonas aeruginosa (P. aeruginosa), Klebsiella pneumoniae (K. pneumoniae), Staphylococcus aureus (S. aureus) and Enterococcus faecium (E. faecium). Results indicated that P. fluorescens (maximum log reduction of 4.21) and K. pneumoniae (maximum log reduction of 5.13) were sensitive to predation in DNB and HEPES buffer, while E. faecium (maximum log reduction of 2.71) was sensitive to predation in DNB and S. aureus (maximum log reduction of 1.80) was sensitive to predation in HEPES buffer. Predation of Gram-positive prey by B. bacteriovorus was thus dependent on the specific prey cells used and the media employed to assess these interactions. In contrast, for P. aeruginosa, while the culture-based analysis indicated that the cell counts were reduced, the EMA-qPCR analysis indicated that the concentration of P. aeruginosa was not significantly reduced in DNB or HEPES buffer. The use of EMA-qPCR can thus aid in accurately monitoring and quantifying both predator and prey cells during co-culture experiments in a time-effective manner. The aim of Chapter 3 (published in Water Research, 2020) was to subsequently apply B.bacteriovorus PF13 as a pre-treatment to SODIS and solar photocatalytic disinfection. Thephotocatalyst used was immobilised titanium-dioxide reduced graphene oxide (TiO2-rGO). Synthetic rainwater was seeded with K. pneumoniae and E. faecium, with results indicating that the use of B. bacteriovorus pre-treatment in combination with solar photocatalysis resulted in the greatest reduction in K. pneumoniae concentrations in the shortest treatment time, with the cell counts reduced by 9.30 logs to below the detection limit (BDL) within 120 min. In contrast, for E. faecium the most effective treatment was solar photocatalysis or SODIS without the B. bacteriovorus pre-treatment, as the viable counts of E. faecium were reduced by 8.00 logs to BDL (within 210 min) and the gene copies were reduced by ~3.39 logs after 240 min. It was thus evident that the application of B. bacteriovorusmay specifically enhance the disinfection of Gram-negative bacteria. Additionally, the use of the photocatalyst further enhanced the disinfection of the Gram-negative bacteria, while the same trend was not observed for E. faecium. Recirculating the water in solar photocatalytic reactors may, however, enhance disinfection of Gram-positive bacteria, by exerting mechano-osmotic stress on the cells and should be investigated in future research. As conflicting results regarding the interaction between B. bacteriovorus and Gram-positive bacteria have been reported, the aim of Chapter 4 (published in Microbiological Research, 2020) was to monitor and compare the expression of attack phase (AP) and growth phase (GP) genes of B. bacteriovorus in co-culture with Gram-positive and Gram-negative prey. Bdellovibrio bacteriovorus PF13 was thus co-cultured with Escherichia coli (E. coli; control), K. pneumoniae and E. faecium. Relative qPCR analysis indicated that the AP genes bd0108 (type IVa pili retraction/extrusion) and merRNA (massively expressed riboswitch RNA) were highly expressed in the B. bacteriovorus AP cells, whereafter expression in co-culture with all the prey strains was reduced. The fliC1 gene (flagellar filament) was also expressed at a high level in the AP cells, however, after 240 min of co-culture with E. faecium the expression of fliC1 remained low (at 0.759-fold), while in the presence of the Gram-negative prey, fliC1 expression increased (in comparison to the expression recorded after 30 min) to 4.62 (E. coli) and 2.69-fold (K. pneumoniae). In addition, bd0816 (peptidoglycan-modifying enzyme) and groES1 (chaperone protein) were not induced in the presence of E. faecium, however, after exposure to the Gram-negative prey, bd0816 expression increased during the early GP, while groES1 expression gradually increased during the early GP and GP. It was thus concluded that B. bacteriovorus senses the presence of potential prey when exposed to Gram-positive and Gram-negative prey however, the GP genes were not induced when B. bacteriovorus was co-cultured with E. faecium. This indicates that B. bacteriovorus may not actively grow in the presence of E. faecium and the second predatory cue (which induces active growth of B. bacteriovorus) may be lacking under the conditions employed in this study. Limited information on the expression of predatory-specific genes of B. bacteriovorus in co-culture with Gram-positive prey cells is available. Recent studies have however, indicated that B. bacteriovorus can prey on Gram-positive bacteria and investigating the expression of these predatory-specific genes may elucidate the genetic mechanisms this predator employs to survive in the presence of these atypical prey.
AFRIKAANSE OPSOMMING: Geoeste reënwater is geïdentifiseer as ʼn addisionele varswater bron wat gebruik kan word om munisipale water gebruik aan te vul, veral in areas waar water skaars is. Verskeie studies het egter gewys dat die mikrobiese kwaliteit van hierdie water nie op standaard is nie. Hierdie mikrobes kan ʼn beduidende gesondheidsrisiko vir verbruikers inhou en daarom moet water behandeling sisteme geïmplementeer word om die vlakke van hierdie mikroörganismes te verlaag. Sonkrag ontsmetting is aangewys as ʼn maklike en goedkoop strategie om water te suiwer. Vir effektiewe suiwering, moet die water vir 6 ure aan sonlig blootgestel word en daarom word fotokatalitiese nanomaterial soos titaandioksied (TiO2) dikwels gebruik om die proses te versnel en sodoende die effektiwiteit van sonkrag ontsmetting te verbeter. Navorsing dui egter daarop dat alhoewel sonkrag ontsmetting mikroörganismes in water verminder, baie patogene en opportunistiese patogene oorlewingsmeganismes gebruik om hierdie tipe behandeling te oorleef. ʼn Kombinasie van fisiese, chemiese en biologiese behandelings moet dus ondersoek word om hierdie oorlewende patogene te teiken. Die gebruik van Bdellovibrio bacteriovorus (B. bacteriovorus), ʼn Gram-negatiwe roofbakterium, is dus vir hierdie dissertasie ondersoek. Die oorhoofse doel van Hoofstuk 2 (gepubliseer in “Microbiological Research”, 2019) was dus om B. bacteriovorus uit riool te isoleer en die interaksie tussen hierdie roofbakterium en Gram-negatiewe en Gram-positiewe prooi te ondersoek deur kultuur- (spreiplate en dubbellaag-oorlegsels) en molekulêre metodes [ethidium monoasied kwantitatiewe polimerase ketting reaksie (EMA-kPKR)] te gebruik. Hierdie interaksies is ook in ʼn voedingstof-arm medium [verdunde voedingstof boeljon (VVB)] en ʼn medium sonder voedingstowwe (HEPES buffer) waargeneem en vergelyk. ʼn B. bacteriovorus isolaat (PF13) is dus saam met Pseudomonas fluorescens (P. fluorescens), Pseudomonas aeruginosa (P. aeruginosa), Klebsiella pneumoniae (K. pneumoniae), Staphylococcus aureus (S. aureus) en Enterococcus faecium (E. faecium) geïnokuleer en toegelaat om te groei. Die resultate het aangedui dat P. fluorescens (maksimum log vermindering van 4.21) en K. pneumoniae (maksimum log vermindering van 5.13) sensitief was vir predasie in VVB en HEPES buffer, terwyl E. faecium (maksimum log vermindering van 2.71) sensitief was vir predasie in VVB en S. aureus (maksimum log vermindering van 1.80) sensitief was vir predasie in HEPES buffer. Predasie op die Gram-positiewe bakterieë was dus afhanklik van die spesifieke prooi selle en die medium wat gebruik is om die interaksies te ondersoek. In teenstelling, vir P. aeruginosa het die resultate gewys dat die seltellings beduidende verminder is in VVB en HEPES buffer, maar die EMA-kPKR analises het gewys dat die konsentrasie van hierdie organisme nie beduidend afgeneem het nie. Daarom is die gebruik van EMA-kPKR voordelig omdat dit die konsentrasie en lewensvatbaarheid van beide die prooi en roofbakterium kan monitor in tweeledige kulture, op ‘n relatiewe vinnige manier. In Hoofstuk 3 (gepubliseer in “Water Research”, 2020) was die doel om B. bacteriovorus PF13 dan te gebruik as ʼn voorbehandeling vir sonkrag ontsmetting en fotokatalitiese ontsmetting. Titaandioksied gereduseerde grafeen oksied (TiO2-rGO) is as die fotokatalis gebruik. Sintetiese reënwater is met K. pneumoniae en E. faecium geϊnokuleer. Die resultate het aangedui dat die beste kombinasie vir K. pneumoniae ontsmetting die fotokatalise met B. bacteriovorus voorbehandeling was, aangesien die plaattellings met 9.30 log verminder is tot onder die opsporingslimiet binne 120 min. In teenstelling was die beste behandeling vir E. faecium sonkrag ontsmetting of fotokatalitiese ontsmetting sonder die B. bacteriovorus voorbehandeling, aangesien die plaattellings verminder is met 8.00 log tot onder die opsporingslimiet (binne 210 min) en die geen kopieë met ~3.39 log verminder is binne 240 min van behandeling. Dit was dus duidelik dat die gebruik van B. bacteriovorus die ontsmetting van Gram-negatiewe bakterieë kan verbeter. Die gebruik van die fotokatalis het ook die ontsmetting van die Gram-negatiewe bakterieë verbeter, terwyl dieselfde nie waargeneem is vir E. faecium nie. Om die water in die fotokatalitiese behandeling sisteem te sirkuleer mag die ontsmetting van Gram-positiewe bakterieë verbeter deur megano-osmotiese stres op die selle te plaas. Hierdie aspek moet in toekomstige studies ondersoek word. Teenstrydige resultate aangaande die interaksie tussen B. bacteriovorus en Gram-positiewe bakterieë is in die verlede weergegee en daarom was die doel van Hoofstuk 4 (gepubliseer in “Microbiological Research”, 2020) om die gene wat in die aanvallingsfase (AF) en die groeifase (GF) van B. bacteriovorus uitgedruk word, te monitor en te vergelyk terwyl B. bacteriovorus aan Gram-positiewe prooi en Gram-negatiewe prooi blootgestel word. Bdellovibrio bacteriovorus PF13 is daaropvolgens saam met Escherichia coli (E. coli; kontrole), K. pneumoniae en E. faecium in tweeledige kulture geϊnokuleer. Relatiewe kPKR analise het daarna aangedui dat die AF gene bd0108 (tipe IVa pili retraksie) en merRNA (massief uitgedrukte riboskakelaar RNA) teen hoë vlakke uitgedruk word in die AF selle, en dat die uitdrukking van hierdie gene daarna afneem in die teenwoordigheid van al die prooi selle. Die fliC1 geen (filament van die flagella) is ook teen hoë vlakke in die AF selle uitgedruk, maar na 240 min se groei saam met E. faecium was die vlak van fliC1 uitdrukking laag (0.759-voud), terwyl die fliC1 uitdrukking saam met die Gram-negatiewe bakterieë gestyg het (in vergelykking met die vlakke by 30 min) na 4.62- (E. coli) en 2.69-voud (K. pneumoniae). Verder is die bd0816 (peptidoglikaan modifiserende ensiem) en groES1 (chaperone proteϊen) gene nie geïnduseer terwyl B. bacteriovorus aan E. faecium blootgestel is nie, maar na die roofbakterium aan die Gram-negatiewe bakterieë blootgestel is, het die vlakke van bd0816 en groES1 beduidend toegeneem in onderskeidelik die vroeë GF en GF. Hierdie resultate het dus aangedui dat B. bacteriovorus waarneem dat moontlike prooi selle naby is wanneer dit blootgestel word aan beide Gram-negatiewe en Gram-positiewe prooi, maar dat die GF gene nie geïnduseer word in die teenwoordigheid van E. faecium nie. Dit kan daarop dui dat B. bacteriovorus nie kan groei met E. faecium as prooi nie en dat die tweede sein (wat aktiewe groei van B. bacteriovorus bewerkstellig) afwesig is in hierdie toestande. Beperkte inligting is beskikbaar vir die interaksies tussen B. bacteriovorus en Gram-positiewe bakterieë en daarom moet hierdie interaksies op ʼn genetiese vlak bestudeer word om vas te stel hoe hierdie roofbakterium oorleef in die teenwoordigheid van hierdie atipiese prooi.
Description
Thesis (PhD)--Stellenbosch University, 2020.
Keywords
Rainwater harvesting -- Health aspects -- South Africa, Predatory bacteria, Solar disinfection, Gene expression, Bdellovibrio bacteriovorus -- Genetics, Solar photocatalysis, Human health risk
Citation