The requirements and challenges of a mobile laboratory for onsite water microbiology assessment

Despite significant efforts deployed to increase the ecological awareness of Earth's inhabitants, the chemical and microbial safety of the global supply of water used for drinking is constantly threatened by the spoilage and spillage of this highly demanded essential resource. The growing human population exerts increased anthropogenic and agricultural activities, leading to pristine areas urbanization, pollution generated by the abovementioned factors and industries, as well as climate changes. It is suspected that these factors have contributed directly or indirectly to the emergence or re-emergence of waterborne infectious diseases (Szewzyk et al. 2000; Epstein 2001; Theron & Cloete 2002; Brettar & Höfle 2008).

Where it is economically and technologically feasible, the assessment of the microbial risk posed by the water supply is generally performed by accredited environmental microbiology laboratories which use classical microbiology methods designed to monitor for the presence of faecal contamination indicators (FCI) such as E. coli and enterococci (Rompré et al. 2002). However, growing knowledge indicates that the indicators seldom fail to provide an adequate evaluation of the risk posed by several groups of waterborne pathogens such as viruses (e.g. Norovirus), bacteria (e.g. Vibrio cholerae), and parasites (Cryptosporidium spp., Giardia intestinalis, etc.) (OECD 1999; Costán-Longares et al. 2008; USEPA 2008; Payment & Locas 2011; Wu et al. 2011).

In light of the increasingly reasonable affordability of molecular biology technologies, our working hypothesis is that the multiparametric (multiple targets) assessment of a putatively contaminated water sample can provide a better evaluation of the microbial risk posed by waterborne pathogens susceptible to be present in a single water sample. Accordingly, molecular microbiology tools should be prioritized to enable a faster and more specific detection of human pathogenic microorganisms (OECD 1999; Weintraub 2006; Brettar & Höfle 2008; USEPA 2008; Figueras & Borrego 2010; Maheux et al. 2011a, 2011b, 2012). However, it must be acknowledged that the detection of pathogens more specifically in drinking water is complicated by the necessity of recovering and detecting the very few microbial targets present in a significant sample size, for example as low as one (1) colony forming unit (CFU) of E. coli per 100 mL or one (1) Cryptosporidium oocyst in one (1) litre of water (LT2ESWR; USEPA 2006). While most water analysis are performed in specialized laboratories, in less privileged areas of the world, where these infrastructures do not exist, a mobile laboratory developed to provide an appropriate (compact) analytical capacity in microbiology would be beneficial for the development and health of communities.

This article describes the water microbiology program of Atlantis, the requirements that have driven the design of the Atlantis classical and molecular microbiology (CMM) laboratory module and the selection of its instrumentation, the challenges and lessons learned from performing water microbiology in the field, and the main deliverables of this applied research program.

In 2002, a group of researchers of Université Laval led by Dr Éric Dewailly from the Unité de recherche en santé publique obtained a grant from the Canada Foundation for Innovation (CFI; www.innovation.ca) for the design, construction, and implementation of a mobile laboratory complex, Atlantis (www.atlantis.ulaval.ca). The CFI was created in 1997 by the Government of Canada to reinforce or increase the capacity of Canadian scientists to perform world-class research and technology development, and for recruiting researchers of talent.

At the time of its conception and to the best of our knowledge, the Atlantis mobile laboratory (AML) complex was a unique multidisciplinary research infrastructure designed to enable field measures of environmental and health impacts of global environmental changes in coastal regions, in the disciplines of analytical chemistry, ecotoxicology, and microbiology. A recent survey of the literature shows that there are at least three types of mobile laboratory configurations applicable to microbiology and/or molecular biology: 1 ° mobile autonomous or semi-autonomous infrastructures (Higgins et al. 2003; Ouedraogo et al. 2008; Grcevic 2010; Canier et al. 2013; Njanpop-Lafourcade et al. 2013), the configuration chosen for Atlantis, 2 ° portable equipment installed in permanent or temporary locations (Grolla et al. 2005, 2011, 2012; Inglis et al. 2011), and 3 ° laboratory without walls (Inglis 2013).

In addition to basic research objectives in the abovementioned disciplines, the planning and development of field missions of Atlantis enabled the inclusion of public outreach activities elaborated to 1 ° facilitate interactions with and training of local professionals, researchers, and science students, and 2 ° promote education and technology transfer to children, the general public, and/or local public health authorities or academic institutions.

Each laboratory module was designed to accommodate a selection of scientific instruments and equipment compatible with the range of bioanalytical tests each research team proposed and was expected to perform within the physical, environmental, and operational constraints of a standard maritime container. Thus, whenever possible in the 2001–2003 period, instruments were selected with respect to size and known (conventional laboratory) reliability but, in many instances, instruments did not travel very well (breakage, calibration problems, etc.) nor operated efficiently in extreme conditions (heat, cold, humidity, etc.)…

The support modules were designed to provide a housing capacity for four persons, as well as electrical support (conversion of local current to 115–120 V AC, power distribution, and capacity for diesel-powered electricity generation) and mechanical (water distribution and repair shop) for the whole complex, field transportation (4WD pickup truck, hydraulic wheeled axles for moving modules, and inflatable boat), and underwater exploration (scuba diving equipment and air tank compressor).

In order to provide enough flexibility for microbiology and molecular biology, the CMM module was equipped with a reverse-osmosis water purification system, a tabletop autoclave, an undercounter freezer (−25 °C) for molecular biology reagents and refrigerators. Initial requirements of the module included an air conditioning system for more comfortable working conditions and a ventilation system for providing a controlled air flow to prevent aerosol contamination of the molecular microbiology sector.

For membrane filtration of water samples, the module was equipped with a tri-headed filtration manifold, a vacuum pump, and microbiology incubators. For other microbiology procedures including performing water analysis using defined substrate liquid methods (Colilert^® and Enterolert™), the module also carried a portable UV light, a fluorescence microscope, and a bioMérieux VITEK^® microbiological identification system.

For the preparation of nucleic acid extracts from concentrated water samples and preparation of PCR reagents and reactions, the module was equipped with a Class II biological hood, two (2) thermal cyclers, a vortex mixer, a Thermomixer, a refrigerated tabletop centrifuge, microcentrifuge, and a selection of micropipettors. For post-PCR analysis, a Bioanalyzer 2100 (Agilent) for chip-based capillary electrophoresis was installed in the classical microbiology sector.

Owing to its expertise in CMM applied to the rapid diagnostics of infectious diseases (Bergeron et al. 2000; Bergeron 2008), the research group of Michel G. Bergeron, from the Centre de recherche en infectiologie of Université Laval, was recruited to develop a research program for onsite assessment of the microbial quality of water within what would become the CMM module of the AML complex. In line with the grand objectives of the CFI and in an effort to address the wishes expressed by international experts (OECD 1999), the Bergeron group foresaw the opportunity of developing rapid and innovative PCR-based bioanalytical tools for detecting FCI and waterborne pathogens and render these tools amenable to field testing, as well as providing education activities in microbiology.

The Health Policy Brief published in 1999 by the Organisation for Economic Co-operation and Development (OECD), following the Interlaken workshop on molecular technologies for safe drinking water (OECD 1999) is a cornerstone document on which the research program of the CMM module was based. The following elements have been considered to develop the program:

1. the recognition that FCI seldom fail to provide an adequate evaluation of the risk posed by waterborne pathogens, many being fastidious or non culturable or capable to enter into a viable but non culturable (VBNC) state (Li et al. 2014);

2. the fact that current methods approved for the microbial assessment of a water sample prevent the simultaneous or retrospective analysis or of the same sample for another FCI or for a pathogen;

3. the (increasing) capacity of molecular microbiology methods such as multiplex PCR and rtPCR, microarrays, and biosensors to enable multiparametric detection;

4. the necessity of demonstrating that molecular methods can be useful methods to assess the microbial quality of water, while providing faster turnaround time and delivery of results than classical microbiology, and;

5. the limited human resources and budget of the AML, and hence of the CMM.

In the context of the AML, the initial plan was to determine the contamination of potable drinking water, recreational water, and seawater for the presence of FCI and selected pathogens. Taking into account the intricacies of analysing these water types (enumeration, presence of inhibitors, adequate selection of target[s], etc.) and since classical (end-point) PCR is essentially qualitative, we opted to concentrate the program on the detection of common FCI and pathogens from treated and untreated water used for drinking. On the basis that some tests approved for testing drinking water operate on the presence/absence principle i.e. that detection must be demonstrated at one (1) or more CFU or microbial particle (MP) (in the case of a fastidious, non culturable, or VBNC microorganism) per volume tested (generally 100 mL), this decision was highly instrumental to the scientific success of the program (Maheux et al. 2011a, 2011b, 2012, 2013). For comparison against standard culture-based methods and to validate new PCR-based assays, DNA extracts from collected samples were conserved and processed in the main laboratory in Québec City.

Membrane filtration is a method of choice for the concentration of MPs from a water sample prior to incubation of the filter on a selective and/or chromogenic solid medium used to grow and enumerate bacterial FCI or pathogens. As putatively contaminated drinking water may contain as low as one CFU or MP per volume tested, this strategic choice oriented the research towards the development of an alternate sample preparation procedure which enabled the efficient recovery of very low numbers of MPs immobilised at the surface of or within the fibres of a cellulose ester filtration membrane. This was accomplished by dissolving the membrane in organic solvents before the nucleic acids extracted from the MP(s) are enriched by whole genome amplification (WGA), in a manner similar to microbiological enrichment very much in usage in food microbiology. This highly innovative filtration-based sample preparation procedure dubbed Concentration and Recovery [of MPs], Extraction of Nucleic Acids, and Molecular Enrichment (CRENAME) is a compact method enabling the multiparametric microbial assessment (i.e. renders the detection of multiple MPs feasible) of a single drinking water sample (Maheux et al. 2011a).

The CMM module has served during several field missions where its capacities have been tested on ground and on water. For the researchers, their graduate students, and staff, these field missions have revealed a number of challenges before and during said scientific endeavours.

Shortly after Hurricane Fabian struck Bermuda (5 September 2003), the AML complex was shipped by cargo to Bermuda and installed at the Bermuda Biological Station for Research (BBSR; Figure 1(c)). This two-month mission, funded in part by the XL Foundation, constituted the first deployment of Atlantis and served to test the functionality of the complex and the resistance of the equipment and infrastructure. We also had the opportunity of performing a small-scale assessment of the contamination of household tank water by coliform bacteria (Lévesque et al. 2006).

In the Fall of 2004 and after railroad transportation to Churchill (Manitoba, Canada), the analytical chemistry and CMM modules of the AML complex were embarked on the CCGS Amundsen, a scientific icebreaker which was renovated through a CFI grant awarded to Université Laval (Figure 1(d) and 1(e); http://www.amundsen.ulaval.ca). During the Arctic Net study, fourteen (14) Inuit communities of Nunavik (Canada) have been visited and, among other activities, the CMM module served for the microbial assessment of treated and raw water used for drinking by culture-based methods for the detection of E. coli and enterococci (Martin et al. 2007). For the development of the CRENAME method, the CMM module was also used to prepare nucleic acid samples amplified by WGA, prior to the PCR detection of FCI and parasites (Cryptosporidium and Giardia) in Québec City.

This large multidisciplinary study, an initiative of the Cree Board of Health (www.creehealth.org), aimed to address issues related to environmental contamination and impacts on human health (Boissinot et al. 2007; Huppé et al. 2011; Nieboer et al. 2013). In the first arm of this study done in Mistissini (Québec, Canada; June-August 2005), we evaluated the microbiological quality of raw source water used for drinking by testing for total coliforms, E. coli, and Enterococcus sp. using membrane filtration and colorimetric standard methods. In particular, we longitudinally tested water from twelve (12) environmental sites traditionally used for domestic water collection (Bernier et al. 2009). During the first and then the second arm of the study undertaken in the James Bay communities of Wemindji (June–July 2007) and Eastmain (July–August 2007), the CMM module also served to prepare WGA- and PCR-amplified nucleic acid samples which could be later tested in Québec City for the presence of FCI, parasites, and other waterborne pathogens.

The Caribbean EcoHealth Programme (CEHP) was a Canadian-funded initiative focused on the integration of environment and health research in the Caribbean islands (Forde et al. 2011). The CEHP agenda was defined in close collaboration with professors from regional universities (University of West Indies, St. George's University, and Ross University) as well as with scientists from the Ministries of Health of the participating countries. It included large regional studies such as the Burden of Gastrointestinal Illness studies, a study of exposure to zoonoses, heavy metals and persistent organic pollutants in maternal blood, the microbial contamination of rainwater cisterns, and microbial contamination of seawater. The AML has been based in Grenada (Summer 2008 to Spring 2009), Dominica (Spring 2009 to Summer 2010), and Barbados (Fall 2010 to Fall 2012). A number of research projects developed in the AML also served for human resources capacity building and training of technicians and graduate students from the region. For example, in 2008 on location in Grenada, the CMM has been used to train local staff in doing classical (VITEK, standard colorimetric and membrane filtration methods for the detection of total coliforms, E. coli, Enterococcus sp., etc.) and molecular detection of microbial agents. In 2010, during an interdisciplinary field course on Oceans and Human Health, the AML hosted technology demonstrations and laboratory exercises.

Since the end of the CEHP in 2012, parts of AML are still active in the Caribbean islands, key instruments are now operated in regional universities, and laboratory modules have been based in Bermuda where the CMM serves for microbiological water testing. The expertise in place in local academic and governmental institutions and, more importantly, the collaboration between Caribbean Community (CARICOM) countries helped to develop a multidisciplinary and multi-institutional network of indigenous, regional, and international professionals working on finding sustainable workable initiatives to environmental and priority public health problems.

The decision of designing a field mission of the AML required significant scientific, political, and financial support since such an endeavour necessitated the mobilization of highly qualified personnel (scientists, graduate students, or research associates) and support staff, to operate a laboratory complex evaluated to 3 million CAD₂₀₀₃.

The transportation of several maritime containers to an abroad country (e.g. Bermuda) or remote area (Nunavik or Nunavut) required a careful planning to ensure an efficient timing of human and material resources. Interactions with local officials to obtain credentials, permits, and required authorizations for the customs clearance and performing sample collection, using a motor vehicle (pick-up truck), organizing public outreach activities, training of the local resources, or preparing scientific reports, for examples, had to be planned with the guidance of collaborators of the project to facilitate information exchange and minimize drawbacks attributed to cultural differences or political impacts.

Prior to installation of the laboratory complex on the field site, the infrastructure requirements related to electrical power, water distribution, and communication (phone, Internet, Skype, etc.) were discussed and adequately planned. The power supply was especially important since a majority of analyses require reagents which must be stored in the cold (4 °, −20 °, −80 °C, or −152 °C) in the modules and voltage/cycle converter could be necessary in some areas. The availability of cold storage capacity in local installations was determined as part of a contingency plan.

Field monitoring is not frequent in microbiology mainly due to materials and equipment requirements. Performing culture-based methods in tropical areas where temperatures in excess of 40 °C are typical may require adequate air conditioning such that incubation is done at an adequate temperature (e.g. 35–37 °C), thereby conferring some advantages to PCR in these conditions. By contrast, the cost, complexity, and relative material frailty of several molecular biology instruments and thermal susceptibility of expensive protein-based reagents have hampered the introduction of PCR-based methods for onsite monitoring. The planning of a field mission requires that enough microbiological materials and temperature-stable reagents are loaded in the module and that temperature-sensitive reagents are safely shipped to and/or stored at the site where the laboratory complex will be installed before the module is made operational.

The development of methods for field monitoring requires an integration of physical, methodological, and temporal factors which may affect the robustness and performance of required analyses and we have uncovered that it was necessary to validate said methods both in the main laboratory and in the Atlantis infrastructure. For example, the implementation of conventional PCR amplification, for onsite monitoring could have led to a recurrent molecular contamination problem caused by the release of amplification products (amplicons) in a severely limited environment. Since then, PCR assays have been converted to the real-time PCR (rtPCR) format where a reaction tube does not need to be opened after the amplification process. This closed tube assay format also opens the possibility of performing multiplex amplification for the simultaneous detection of a microbial panel consisting of up to 4 or 5 FCI and waterborne pathogens from a single water sample.

Most instruments used in bioanalytical life sciences are relatively fragile and not necessarily designed to withstand frequent handling and packing, as well as harsh shipping conditions. Therefore, precautions must be taken to safely stow equipment in modules potentially exposed to intense vibrations and extreme temperature variations during short-to-long distance transportation on the haul of a truck or of a railroad wagon, on the deck of a cargo ship or icebreaker, in addition to moving on local roads (of variable roughness). Indeed, we have witnessed damage to the containers, to internal structures, and to instruments containing mobile internal parts. A capacity for infrastructure repair or equipment/instrumentation troubleshooting, verification procedures, as well as a contingency plan are therefore required.

The robustness of scientific instruments was not only strained during storage and transportation but also severely challenged during operation of the module. Indeed in the original CMM module configuration, deficient air conditioning and ventilation have contributed to extreme temperature variations that have caused instrument failures (VITEK system and portable computers for example), discomfort of students and professionals working in the CMM and paradoxical situations such as incubating bacteria at 35 °C when the laboratory temperature exceeded 40 °C. These situations could be circumvented by upgrading the air conditioning system of the module. Instrument repair is also problematic since accessing remote areas or foreign countries is often prohibitively costly to allow onsite servicing by a specialized technician. The local research team has proven to be resourceful and made a clever use of web-based photos and videos to fix problems.

In addition to basic research activities performed onsite by graduate students or research professionals, the personnel of Atlantis was given the opportunity of using the infrastructure as an educational tool. Local students interested in learning the basics of microbiology and water analysis were offered training opportunities with team members during field missions. In the regions visited by Atlantis, conferences or open door activities have been organized to display the laboratories of the complex, describe the research programs run concomitantly by the research teams, and explain the role of environmental assessment in public health. Exchanges with local authorities contributed to a better understanding of local issues and traditional practices such as the preferred selection of raw water collection sites which were later shown to be less contaminated by FCI than other tested sites.

In itself, Atlantis can be envisioned as a technological showcase which provided to graduate students an opportunity of performing science in remote environments but also to expose, initiate, and train local students to science basics, for example in Mistissini, Wemindji, Northern Québec as well as in the Caribbean islands.

The Atlantis laboratory complex was a unique infrastructure that has proven its usefulness for field monitoring of treated and untreated water used for drinking in remote communities. For the CMM module, budgetary and methodological limitations have hampered the development of an infrastructure which could be efficiently used for performing molecular microbiology in the field although we demonstrated that PCR-based methods can be used to assess the microbial quality of drinking water (Maheux et al. 2011a). This latter application constitutes an extreme analytical problem we circumvented by demonstrating that CRENAME and rtPCR can provide a presence/absence method for the rapid detection of E. coli/Shigella in drinking water (Maheux et al. 2011b, 2012). Notwithstanding the fact that the reagents used for CRENAME-rtPCR do not currently exist in a stable (dried) state, we believe that the technology is now mature enough to be tested in the field and that cross-contamination problems known to occur with PCR-based methods shall be alleviated by performing closed-tube rtPCR tests.

Regulations of environmental authorities such as the United States Environmental Protection Agency have begun to support molecular microbiology for assessing the microbial quality of water, in the form of Method A, a quantitative rtPCR (qrtPCR) assay for detecting and enumerating enterococci in recreational water (USEPA 2010). We believe that our research might have contributed to open this avenue for molecular microbiology.

Andrée F. Maheux held a scholarship from the Nasivvik Center for Inuit Health and Changing Environment (Canadian Institutes of Health Research). The construction of the Atlantis complex was supported by grant FCI-5251 from CFI. The authors acknowledge the precious support from the XL Foundation (Bermuda 2003), Arctic Net (Nunavik 2004), and Niskamoon Corporation (Mistissini 2005 and Wemindji 2006). Canadian Global Health Research Initiative's Teasdale-Corti programme which is managed by the International Development Research Centre. Additional funding and support has been provided by the Canada-World Bank POPs Fund, the Pan American Health Organization, the Public Health Agency of Canada, the Global Health Research Capacity Strengthening Program (GHR-CAPS), the Institut national de santé publique du Québec (INSPQ), the Centre de recherche du Centre hospitalier universitaire de Québec (CRCHUQ), the Lepercq Foundation, the University of the West Indies, Ross University Medical School, Ross University School of Veterinary Medicine, the Caribbean Epidemiology Centre, and various national governments in the CARICOM region (Caribbean islands 2008–2012) for the field missions.

Andrée F. Maheux and Luc Bissonnette contributed equally to this work.

Éric Dewailly died accidentally on 17 June, 2014. This article is dedicated to this passionate researcher and to his revolutionary vision of how sciences could be applied in partnership with local communities in the field, to improve public health globally, including remote areas and isolated populations.