Mycobacterium avium subsp. paratuberculosis in lake catchments, in river water abstracted for domestic use, and in effluent from domestic sewage treatment works: diverse opportunities for environmental cycling and human exposure

Abstract

Mycobacterium avium subsp. paratuberculosis from infected animals enters surface waters and rivers in runoff from contaminated pastures. We studied the River Tywi in South Wales, United Kingdom, whose catchment comprises 1,100 km2 containing more than a million dairy and beef cattle and more than 1.3 million sheep. The River Tywi is abstracted for the domestic water supply. Between August 2002 and April 2003, 48 of 70 (68.8%) twice-weekly river water samples tested positive by IS900 PCR. In river water, the organisms were associated with a suspended solid which was depleted by the water treatment process. Disposal of contaminated slurry back onto the land established a cycle of environmental persistence. A concentrate from 100 liters of finished water tested negative, but 1 of 54 domestic cold water tanks tested positive, indicating the potential for these pathogens to access domestic outlets. In the separate English Lake District region, with hills up to 980 m, tests for M. avium subsp. paratuberculosis in the high hill lakes and sediments were usually negative, but streams and sediments became positive lower down the catchment. Sediments from 9 of 10 major lakes receiving inflow from these catchments were positive, with sediment cores indicating deposition over at least 40 to 50 years. Two of 12 monthly 1-liter samples of effluent and a single 100-liter sample from the Ambleside sewage treatment works were positive for M. avium subsp. paratuberculosis. Since Lake Ambleside discharges into Lake Windermere, which is available for domestic supply, there is a potential for these organisms to cycle within human populations.

FIG. 1.

(a) Map of the study region in South Wales, United Kingdom, showing the River Tywi and its main tributary, the River Cothi, with the regional towns of Llandovery and Carmarthen and the capital of the principality, Cardiff. Also shown (dotted line) is the 26-km abstraction pipeline taking raw river water from the main channel of the River Tywi to the FWTW. Bar, 15 km. Samples and sites: 1, sediment from Llyn Brianne reservoir in the Cambrian mountains at an elevation of 488 m (UK Ordnance Survey grid reference SN796489); 2, epilithon from Gallt y bere on the upper Tywi (SN773459); 3, discharge effluent from human sewage treatment plant (SN766380); 4, epilithon from site at Llandovery bridge (SN761347); 5, discharge effluent from human sewage treatment plant at Llandielo (SN610213); 6, 10-liter water sample from River Cothi (SN525231); 7, 10-liter water sample from River Cothi (SN523229); 8, 10-liter water sample from River Tywi (SN508201); 9, 10-liter water samples twice weekly from 8 December 2002 to 4 October 2003 from Nantgaredig Bridge (SN 494204), which is just downstream of the junction of the River Cothi with the River Tywi; 10, sediment from the River Tywi at the domestic supply abstraction site (SN488205). The dotted area represents the potable water distribution region from the FWTW where domestic properties were sampled. (b) Diagram showing the River Tywi abstraction pumping station (large square), the 26-km pipeline (dotted line [not to scale]), and details of the FWTW (SN649032). Samples and sites: 11, epilithon and sediment from the abstraction pump house screening grill (SN488204); 12, sediment from the Lower Lliw holding reservoir (SN648036); 13, epilithon from the inflow channel leading to the FWTW (SN648036); 14, suspended solids separated by the COCODAFF treatment process in the FWTW; 15, 100-liter sample of finished water product; 16, accumulated suspended solids from sludge lagoon; 17, sediments accumulating in water inflow tanks in domestic properties (n = 54).

FIG. 2.

(a) Map of the study region in the Lake District, United Kingdom, showing the principal lakes and the regional towns of Cockermouth (A), Keswick (B), and Kendal (C). Bar, 10 km. Sediment samples (except where indicated) were taken from the following sites: 1, Bassenthwaite (NY204319); 2, Loweswater (NY122223); 3, River Cocker outlet from Crummock Water (NY150211); 4, Crummock Water (NY149214); 5, epilithon formed due to road runoff from fellside at Hause Point, Crummock Water (NY162183); 6, Derwentwater (NY259219); 7, Ullswater (NY460238); 8, Ennerdale (NY167063); 9, Buttermere (NY188154); 10, Thirlmere (NY306172); 11, Blea Water (NY452108); 12, Small Water (NY457101); 13 and 14, Coniston catchment, comprising Blind Tarn (SD263967) and Goat's Water (SD266978); 15, Coniston (SD310971). (b) Map of the Windermere catchment showing the lake (stippled) and its associated tarns, with the regional towns of Grasmere (D), Ambleside (E), Hawkshead (F), and Windermere (G) indicated. Streams and rivers are indicated by solid lines. Inverted triangles indicate domestic wastewater treatment works. Bar, 1 km. Sediment samples were taken from the following sites: 1, Codale Tarn (NY297088); 2, Easedale Tarn (NY309088); 3, Grasmere (NY334069); 4, Angle Tarn (NY357063); 5, Stickle Tarn (NY288076); 6, Elterwater (NY333042); 7, River Rothay (NY372037); 8, River Brathay (NY372032); 11, Esthwaite (SD359972); 12, Cunsey Beck (SD381936); 15, High Dam (SD363887). Sediment cores were taken from the bottom of Lake Windermere at the following sites: 10, north basin (NY379020); and 13, south basin (SD380897). Samples (1 liter) of treated clean water effluent were also obtained from the human domestic waste treatment works at Ambleside (NY372041) (site 9) and Windermere (SD385973) (site 14).

FIG. 3.

Relationship between presence or absence of M. avium subsp. paratuberculosis in river water and the flow and height of the River Tywi and rainfall in the catchment between August 2002 and April 2003. Plotted rainfall data were recorded at 9:00 a.m. each day. The solid line represents the river height data, while the dotted line represents rainfall. Open diamonds are negative results, and black diamonds are positive results for the presence of M. avium subsp. paratuberculosis.

FIG. 4.

Model of sources and sinks of M. avium subsp. paratuberculosis in the environment with respect to human exposure. This model incorporates the established route of transmission of this pathogen to humans via dairy products, especially milk, as well as hypothetical routes of transmission. It also incorporates the catchment cycle, which receives inputs from domestic livestock and wildlife reservoirs as well as agricultural practices such as slurry spreading. M. avium subsp. paratuberculosis enters water through runoff from the catchment, with aerosols from contaminated rivers and abstraction for domestic supply as potential sources of human exposure. Disposal of waste from water treatment onto catchments and the entry of effluent from sewage treatment works into sources of domestic supply constitute environmental cycles with opportunities for microbial evolution and enhanced pathogenicity.