The Floating Lab: Standard Operational Procedure for Collecting and Filtering Seawater Samples from Operating Ferries for Environmental DNA Analysis

Summary

Here, we present a novel protocol to retrieve marine eDNA through the collection and filtration of seawater samples from operating ferries and other commercial vessels.

Abstract

Genetic tools to retrieve information from environmental DNA traces are well-established for both targeted and taxonomically broad biodiversity characterization. Yet, in the marine context, collecting eDNA samples from poorly accessible areas, such as in abyssal or offshore waters, may still be a limitation. The use of scheduled ferries or commercial ships crossing large expanses of open sea can constitute valuable opportunistic platforms for the collection of environmental samples. The advantages are countless, from the possibility of collecting data in any weather condition and at any time of day to the repeatability over time, being the routes constant, and with a total reduction in costs and fuel emissions. Previous work has shown that the approach is possible and successful. However, the process of acquiring samples from the bellies of these large marine vessels may not be immediate. This work illustrates in every detail how the collection and filtering of seawater from large ships can be prepared, arranged, and carried out.

Introduction

In the last decades, the analysis of environmental DNA (eDNA), namely the detection of genetic material left by organisms in the environment, has rapidly grown as a promising monitoring tool for freshwater and marine animal communities. This molecular technique offers the possibility to simultaneously identify multiple taxa within a single sample (metabarcoding approach), allowing the detection of the biological communities inhabiting or crossing the sampled area^1,2,3,4,5.

Currently, the real challenge is reaching sampling spots, as it is often economically and/or logistically infeasible to collect samples across large geographical scales, resulting in pelagic areas being usually under-sampled compared to more accessible coastal areas⁶. Using ferries or other commercial vessels allows for highly replicable sampling transects, as they follow specific shipping routes that are typically constant over the season and cover large areas^7,8. Moreover, using ferry routes as sampling means allows sample collection at any time of the day and during every season, regardless of the weather and sea conditions, besides presenting numerous other advantages (Supplementary Figure 1). Preliminary studies have proved both its feasibility and accuracy for assessing biodiversity indexes^9,10. Water samples can be collected from the ferry engine room via a derivation pipe intercepting marine cooling water upstream of the engines. LIFE-CONCEPTU MARIS (CONservation of CEtaceans and Pelagic sea TUrtles in Med: Managing Actions for their Recovery In Sustainability) is the first project in which this proof of concept has been put into practice on a large scale: it aims at the collection of 500 samples (over two years), and it involves the engagement of not one but several ferry companies and several carriers, allowing to highlight any critical issues and refine sampling strategies to better adapt to the most diverse contexts that may be encountered on board. We describe below the characteristics of the project for its component relating to the large-scale collection of marine environmental DNA samples (for the description of the project, visit https://webgate.ec.europa.eu/life/publicWebsite/project/details/5707).

The decision to undertake a sampling campaign using a scheduled ferry implies some actions that must be taken before the actual sampling, which is the aspect directly addressed in this work. Here, we briefly mention the actions that must precede onboard sampling from any commercial vessel.

This is an aspect that should not be overlooked and which determines the success of the collection campaign. Shipping companies are commercial and not research entities and, therefore, are not, and are not required to be, familiar with the aims and objectives of the scientific research intended to be carried out onboard. This implies an important and delicate initial phase in which the aims of the proposed scientific project, which can be multidisciplinary, must be transmitted and motivated by the top management of the shipping company itself. In the specific case of the CONCEPTU MARIS project, this was made possible by a decades-long relationship in which the shipping companies had already been involved in visual monitoring projects (e.g., Fixed Line Transects [FLT]⁷), which determined a relationship of mutual esteem necessary to be able to obtain the green light for access to the engine rooms indispensable for the sampling approach covered by this work.

The kind of sampling regime adopted in the CONCEPTU MARIS project is twofold: the samples are taken both in fixed sampling stations (FSS, see below) and on the occasion of the sightings of "rare" cetaceans. These represent species that are least sighted during FTL transects¹¹, namely: common dolphin (Delphinus delphis), Cuvier's beaked whale (Ziphius cavirostris), pilot whale (Globicephala melas), Risso's dolphin (Grampus griseus), sperm whale (Physeter macrocephalus) and fin whale (Balaenoptera physalus). An additional water sample is collected only when the sighting of the rare species reported by the visual census team (FLT team) occurs more than half an hour before or after one of the FSSs. The samplings carried out in conjunction with the sightings will necessarily all be diurnal.

Fixed sampling stations (FSSs), whose geo-position remains invariable over cruises, are identified beforehand and agreed upon with the entire working team. FSSs' geographic positions are selected according to 1) the presence of a site of biological interest based on previous observational/literature data; 2) priority given to points indicating habitat changes on bathymetric maps (e.g., edge of continental shelf); 3) homogeneous coverage of the designated shipping lanes, thus selecting roughly equidistant sampling sites (about 35-45 nautical miles apart), in order to both covering the whole route and foreseeing the collection of night-time samples too.

In order to ensure that adjacent samples are taken at different times of the day and that there is enough time between two consecutive samples to complete sample processing, it is advisable to take samples from adjacent FSSs, one on the outward journey and one on the return trip. This means that if the FSSs are numbered according to the chronological order in which they are sampled, they will not appear in a consecutive order on the map. For example, if 6 FSSs are selected, 3 will be sampled on the outward journey and 3 on the return journey. “Their order along the route on the map will be PortAFSS1-FSS6-FSS2-FSS5-FSS3-FSS4-PortB, with the three sampling stations in Italics (FSS4, FSS5, and FSS6) being surveyed in the return journey (see Figure 1).

Each single FSS has a unique identification number (i.e., none of the FSSs identified on different routes shares the same identification number). Numbering is assigned the first time an FSS is sampled, and consecutive numbering will follow. Therefore, at the end of the project, low-numbered FSSs will be those that have been tested for several years or for which the oldest data are available.

Each sample consists of ca. 13 L of seawater collected at each sampling station. The seawater is decanted directly from the ferry's derivation pipe into "Bag-in-the-Box" (BiB) containers, namely sterile foil laminated plastic bags, until sample processing (see below). The characteristics and advantages of the Bag-in-Box Sampling System (BiBSS) are illustrated in detail by Valsecchi et al. (2021)⁹ and are also reported here in Figure 2 for convenience.

Protocol

1. Getting ready in the laboratory

1. Gear checklist
   1. Consult the checklist for the items to be taken on board for both sample collection and sample filtration. It is shown in Supplementary File 1. Each item and its use will be explained in the sections below.
   2. Print a copy of Supplementary File 1 and fill it in before departure for the sampling campaign.
2. Preparing the Bag-in-Box sampling system
   1. Prepare the Bags-in-Box in the laboratory, under the fume hood if possible, before boarding. Seal the three openings (the one on the bag, part of the cap that closes the bag, and the outlet from the tap) with sealing film (Figure 3, Table of Materials).
   2. Roll the bags, with the cap detached and sealed inside, tightly and close it with an adhesive tape to prevent reopening during transportation (Figure 3).
   3. Perform this operation by making sure not to let air into the bag during the aforementioned sealing maneuver, both (i) to avoid possible contamination carried by air entering the bag and (ii) to prevent the folded bag from becoming bulkier (if it contains air).

2. Getting ready on board

1. Coordination with crew members
   1. As soon as on board, share the selected sampling points with the captain and his crew so that the times when the ferry will be closest to each point selected as a "Fixed Sampling Station" (FSS) will be indicated.
   2. Note down the sampling time for each FSS and the current (or average, if available) cruising speed of the ferry for that route: this allows estimating the extension (km) of the sampled stretch of sea.
   3. As the ferry never passes exactly on the same point, correct the coordinates of each seawater sample collection after the actual sampling.
      NOTE: The exact sampling interval is the most important data for a faithful identification of the sea area actually sampled; therefore, it is useful to print a map of the rough points where reporting the actual sampling coordinates (e.g., in Supplementary File 2).
   4. If the company grants access to the engine room, ask to be escorted to the room.
   5. Wear the personal protective equipment (PPE) before entering the engine room. Here, the environment is inhospitable (deafening noise and high temperatures) and full of potentially insidious points (steep stairs, large gaps in the floors). Once permission to enter the engine room has been obtained, equip with all essential safety equipment, such as safety shoes, helmet, and hearing protection headphones.
   6. If the filtering surface is provided by the shipping company (it is usually advisable to have one's own folding table), check that this area can be sanitized between samples using a 10% bleach solution.
2. Settling up in the engine room
   1. As there are various seawater inlets on ships, identify with the engine room manager, which allows a minimum path for the in-taken water through the ship's pipes to minimize contamination of the water sample with substances and organisms that may be found in the pipes.
   2. Once the most suitable point/inlet for collecting the incoming sea water has been identified, ask the staff if the sample collection tap can be kept slightly open for the entire duration of the cruise in order to continuously rinse the pipe used for sample collection with "local" water. Dispose of the seawater that flows incessantly from the tap in the bilge.

3. Water sample collection (in the engine room)

NOTE: Figure 4 shows the phases of seawater sample collection.

1. Preparation for water collection
   1. Bring all necessary gear (first part Supplementary File 1 checklist) to the engine room. Get to the engine room at least 15 min before the planned sampling time to sterilize the area where the BiBs are handled using a 10% bleach solution.
   2. If the dedicated tap has not been left open for the whole cruise, open it for the collection of seawater samples, and let the water run for at least 5 min (the water is disposed in the bilge) before the actual withdrawal (Figure 4A).
   3. Prepare the filtering station.
      NOTE: There are several measures in setting up the filtering station that can significantly improve the efficiency of the filtration (see step 4.3): 1) use high-capacity vacuum flasks (2 L, 4 L, or 5 L flask, depending on the volume to be withdrawn). This, in addition to saving the time needed to empty the flask between 1 L and the next one, minimizes the handling of the filtration cylinder and, therefore, any contamination; 2) one can use multiple filter flasks arranged in series, a measure that allows the simultaneous filtration of multiple samples (bags). This measure is recommended in tight sampling regimes, where there is otherwise the risk of not having the necessary time to process the bags before reaching the docking port.
   4. Use a permanent marker to prepare and label the BiB by writing on it: 1) the unique alphanumeric identification code of the sample on both sides of the BiB (important especially if the BiB are not processed straight away on board and are filtrated on land) 2) the date; 3) the exact time of starting filling the BiB.
2. Water collection
   1. Remove the sealing film from the opening of the BiB and start filling it until completely full (about 13 L) (Figure 4B).
   2. When the bag is nearly full, remove the sealing film from the lid, but not from the tap opening: maintain the sealing film around the tap safety secure (red plastic tab) until filtration (Figure 4C). Seal the BiB with the lid, pressing it tightly until completely closed.
   3. Take note of the exact time when the sample collection is completed: it allows to calculate the total duration (minutes) of sample collection.
3. After water collection
   1. Write with the permanent marker on the BiB the exact time when the sample collection is completed and report all data on the sample collection/filtration form (Supplementary File 2).
   2. Transfer the filled-in BiB to the storage room or to the filtration area (a dedicated space in the engine room itself or in a cabin).
   3. Transfer all data on the sample collection/filtration form (Supplementary File 2).
      NOTE: Supplementary File 2 shows the sample collection/filtration sheet to be completed on board in all its parts.
      1. For the part relating to the collection of the sample (green), enter the start and end times of the seawater sampling from the engine room.
      2. Secondarily, after consultation with crew members or GPS records, enter the geographic coordinates relative to the beginning and end of the sampling and cruising speed. This allows you to subsequently calculate the length of the sampled sea segment.
         NOTE: As regards to the part relating to filtering (blue), it will be indicated whether filtration occurs on board or later on land: filtering times will be indicated for each of the 3 filters.

4. Seawater sample filtration

1. Prepare for filtration
   NOTE: From this point on, the filtration process can be managed according to different protocols based on the specific objectives of the research project. The protocol described here allows for efficient filtration of a large amount of seawater (more than 10 liters) directly on board, avoiding the volume constraints inherent in other commonly used commercial kits or protocols. When possible, filter water samples on board: filtration can take place either in the engine room itself or in a cabin, preferably located near the engine room. It is always desirable to filter seawater samples immediately after collection, both to limit eDNA degradation and for convenience, thus avoiding disembarking with large volumes of seawater to be processed. If not possible, samples can be stored in the Bag-in-Box containers and transported to be filtered subsequently.
   1. In the provided space dedicated to filtration on board (in the engine room or a dedicated cabin), get prepared with everything listed in the second part of the checking list in Supplementary File 1 and assemble the filtering system as shown in Figure 5.
   2. Prepare the filtration cylinder (Table of Materials) as shown in Figure 6 and Figure 7. Use a new cylinder for each station, a new pair of single-use tweezers, or a pair of sterilized ones.
      NOTE: A single cylinder can be used multiple times to filter samples from the same location. However, it needs to be sterilized with 10% bleach, followed by freshwater rinsing, before being left to air-dry in the laboratory and then singularly packed to be stored until the next campaign.
   3. Before starting the filtration, fill in the sample collection/filtration data log (Supplementary File 2). This is necessary both to keep track of the number of liters processed and to measure the filtering time (indicative of the amount of particulate in the sample). As soon as the vacuum pump is activated, start the timer to measure the filtering time.
2. Filtration on board
   1. In order to minimize contamination risk (especially when several bags are processed simultaneously, see below), isolate the flow of water to be filtered from the surrounding environment by placing an insulating sleeve - plastic bag - between the tap and the filtering cylinder (see Figure 7).
      NOTE: This is an important precaution as the sensitivity of the technologies used for the analysis of environmental DNA makes the approach particularly susceptible to environmental contamination (from the operator or cross-contamination between samples).
   2. Activate the vacuum pump and filter up to 4 L of water for each filter. Keep the cylinder always full during filtration to avoid air getting into the system and slow down the filtration process.
      NOTE: As for the choice of membrane’s porosity, this depends on the purpose of the research, the target taxonomic groups and the quantity of water that is intended to be processed. This issue has been dealt with in previous works^9,12.  in which filters of different porosity were compared. For the study of vertebrates and the choice to process 4 liters of water per filter, a membrane with a porosity of 0.45 µm proved the best solution, allowing to process large volumes of water without causing cluttering of the filter due to saturation of the pores.
   3. When the level of the filtered water inside the flask reaches the 4 L level mark, turn off the vacuum pump and stop the timer, reporting the filtering time value on the data log.
      NOTE: From each BiB of 12/13 L, 3 technical replicates will be obtained, of 4 L each (named A, B and C).
   4. At this point, using the tweezers (and, if needed, a sterile toothpick to help lift the filter edge), recover the first filter (filter A) from the cylinder, as shown in Figure 8, avoiding any possible damage to the filter.
   5. Fold the filter in half (with the side that retained the biological material folded on itself) and wrap it in aluminum foil, immediately writing the sample number and the ID of filter replicate (for example, 15A) on the wrapping. Store it in the onboard freezer until disembark.
   6. Mount a new filter (filter B) inside the filtration cylinder to process another 4 L sample (Figure 9). Empty the flask and start again with filtering the second replicate until 4 L are completely filtered. Recover filter B as explained and repeat the same steps for filter C.
3. Good practices
   1. To shorten filtering times, use a large vacuum flask capable of accommodating all the filtered water processed through a single filter (therefore, in this case, at least 4 L capacity).
   2. Reduce the filtering times considerably by filtering multiple bags simultaneously by arranging the filtering apparatus in parallel (i.e., multiple vacuum flasks served by a single pump).
      NOTE: In the CONCEPTU MARIS campaign, both strategies were adopted (Figure 10). Note that maximizing filtering efficiency is very important as the number of samples collected along a route can be large. It is, therefore, necessary to process the bags in a short time in order to avoid accumulating samples with the risk of having to disembark the bags if there is not enough time to process them on board.
   3. In order to monitor the incidence of false positives (contamination), include in the filtering procedure, at least once per cruise, a "blank-bag", filtering in parallel with the samples a bag filled with drinking water taken from sealed bottles. This control sample does not require replicates and treat this as the rest of the samples and include it in the NGS run.
4. Filters' storage and transportation
   1. After filtration, store all filters between -4 °C and -20 °C until further laboratory processing.
   2. Collect the samples from the onboard freezer just prior to disembarkation and prepare them for transport at low temperatures using transportable coolers, as shown in Figure 11.
      NOTE: If the samples are not filtered on board, the seawater bags are more safely transported if kept vertically with the cap held on the top side, supported in pairs inside shopping bags inside plastic crates to facilitate transport (Figure 12).

Results

Executing the protocol as described, a number of filters equal to the number of FSS multiplied for 3 replicas will be collected by the operator. Each filter retains biological traces (environmental DNA) on one of the two sides. The analytical steps subsequent to eDNA extraction strongly depend on the research question, target, and goal. For instance, in the pilot study by Valsecchi et al.⁹, the described eDNA collection protocol allowed the design of a replicable and systematic marine megafauna su...

Discussion

As the protocol steps are simple and straightforward, there are some relevant considerations to consider when working on commercial ferries. The cooling-water derivation pipe is usually located in the engine room area, to which access may be restricted or subject to prior authorization. Therefore, making sure to have such authorization, together with the appropriate personal protective equipment, is pivotal (Figure 4). Furthermore, water collection and filtration should take place in the mos...

Materials

List of materials used in this article

Name	Company	Catalog Number	Comments
4.7 cm diameter filter membranes	Biosigma S.p.A.	https://www.biosigma.com/	Filter membranes of 0.45 μm porosity, individually packaged
4/5 L vacuum flasks	Duran	https://www.duran-bottle-system.com/	5 L glass flask with stopper
Absorbent paper pads	Advantec MFS, Inc.	https://www.advantecmfs.com/category/absorbent-pads-1	Filter paper absorbent pads, size 47 mm.
Bag-in-Box (BiB)	G.M.V. Agricenter S.r.l.	https://www.gmvagricenter.it/	10 liters bags-in-box with faucet
Filtering cylinders	Sartorius AG	https://shop.sartorius.com/in/industrial-microbiology-filtration-devices/biosart-100-monitors/p/M_Biosart_100_Monitors#	Biostart 100 Monitor cylinders, with 0.45 μm membrane included
Sealing film	Merck	https://www.sigmaaldrich.com/IN/en/product/sigma/hs234526a?utm_source=bing&utm_medium= cpc&utm_campaign=all+product_ dsa_WW_%28bing+ebizpfs%29& utm_id=626946489&utm_content= 1166583023365595&msclkid= d2b55b82440b1aef56fc3ca803c 6486f&utm_term=%2Fproduct%2F	4 inches x 125 feet roll of laboratory sealing film
Vacuum pump	KNF Group	https://knf.com/en/us	Laboport portable vacuum pump N. 96