Rapid Testing of Resistance of Timber to Biodegradation by Marine Wood-Boring Crustaceans

Summary

This protocol presents a method for assessing the feeding rate of the wood-boring crustacean, Limnoria, by measuring faecal pellet production. This method is designed for use in non-specialist labs and has potential for incorporation into standard testing protocols, to evaluate enhanced wood durability under marine conditions.

Abstract

Wood-boring invertebrates rapidly destroy marine timbers and wooden coastal infrastructure, causing billions of dollars of damage around the globe every year. As treatments of wood with broad spectrum biocides, such as creosote and chromated copper arsenate (CCA), are now restricted in marine use by legislation, naturally durable timber species and novel preservation methods of wood are required. These methods undergo testing in order to meet regulatory standards, such as the European standard for testing wood preservatives against marine borers, EN 275. Initial investigation of durable timbers species or wood preservative treatments can be achieved quickly and inexpensively through laboratory testing, which offers many advantages over marine field trials that are typically costly, long-term endeavours. Many species of Limnoria (gribble) are marine wood-boring crustaceans. Limnoria are ideal for use in laboratory testing of biodegradation of wood by marine wood-borers, due to the practicality of rearing them in aquaria and the ease of measuring their feeding rates on wood. Herein, we outline a standardizable laboratory test for assessing wood biodegradation using gribble.

Introduction

Wood-borers can cause extensive damage to marine wooden structures, such as sea defences, piers, and aquaculture structures; the replacement or restoration of which costs billions of dollars per annum worldwide^1,2,3. In order to protect these structures, timber is often treated to reduce biodegradation. However, due to the restriction of use of broad-spectrum biocides in Australia, EU, UK, and USA, in the marine environment, new modification techniques and species of wood that are naturally durable to borers are sought after^4,5,6,7. Novel techniques for the preservation of wood in the marine environment require thorough testing in order to meet regulatory standards and limit environmental impacts from hazards such as leaching of any chemical preservative. For example, the European standard, EN 275, which is the current European standard from 1992, is used to evaluate wood preservation treatments against marine wood-borer damage^8,9. This standard, along with other legislations against the use of biocidal compounds, such as CCA^4,5,6,7 and creosote¹⁰, necessitates sustainable, non-toxic methods of wood protection and the use of naturally durable timber species to replace biocidal treatments^11,12. Marine trials, such as those specified in EN 275, require long exposure periods and are thus expensive and slow to yield meaningful results. Laboratory tests, however, provide a much quicker alternative to test methods of preserving timber products against marine wood-borer attack, allowing rapid evaluation of adjustments to treatment schedules¹³. Results from this rapid laboratory experiment are designed to inform novel modification processes of wood and to identify timber species with natural durability to borer damage. A low feeding rate and vitality can indicate increased resistance in potential products and this information can then be fed back to industry partners to allow them to improve designs. Our method allows a nimble and rapid response, that is desirable in industry, and once promising products have been identified, results can be supplemented with those from marine trials.

Gribbles (Limnoria) are a genus of isopod crustacean in the family Limnoriidae. There are over 60 species of Limnoria worldwide^13,14,15, with three common species found in the UK, Limnoria lignorum, Limnoria tripunctata and Limnoria quadripunctata¹⁶. They bore tunnels on the surface of wood that is submerged in seawater, often causing economically significant damage. Gribbles are highly abundant in coastal UK waters and are easy to maintain under laboratory conditions, making them ideal organisms for the study of wood biodegradation by marine wood-boring invertebrates. Evaluating the feeding rates and vitality of gribbles on different timber species and wood preservation methods can determine the efficacy of their resistance to biodegradation. The following protocol sets out a standard method for measuring gribble feeding rates, developed from that described by Borges and colleagues^12,17, in addition to streamlining the introduction of image analysis to make the process operable in non-specialist labs. Image analysis is also used to reduce the practical limitations of manually counting large number of samples. Durability in long-term marine testing, according to the British Standard EN350-1:1994, are graded in reference to Pinus sylvestris sapwood¹⁸. In the short-term laboratory testing presented here, we use Scots pine (Pinus sylvestris L) sapwood as a control to testing heartwood of the species ekki (Lophira alata Banks ex C.F Gaertn), beech (Fagus sylvatica L), sweet chestnut (Castanea sativa Mill) and turpentine (Syncarpia glomulifera (Sm.) Nied). Average faecal pellet production and vitality among eight replicates per wood species was used as an indicator of durability. We provide illustrative data collected from a typical evaluation, using the gribble species Limnoria quadripunctata and a range of naturally durable timber species. Limnoria quadripunctata, identified by the keys provided by Menzies (1951), was selected as the optimal species for biodegradation trials due to the fact that it is the most well-studied member of the family and is well-established as a model species for use in biodegradation trials. This protocol is also applicable for testing woods of different treatments although the control used should be untreated replications of the same species.

Protocol

1. Preparing Test Sticks

1. After any treatment processes are complete, cut dry wood into test sticks to size 2 mm x 4 mm x 20 mm (Figure 1). Air dry sticks to a constant weight, under laboratory conditions. Use at least 5 replicates of each wood being tested.

Figure 1: Test sticks used in short-term laboratory testing to assess gribble feeding rates.  Test wood sticks sized 2 mm x 4 mm x 20 mm. From left to right: ekki, turpentine, sweet chestnut and beech heartwood and Scots pine sapwood. Scale bar 4 mm. Please click here to view a larger version of this figure.

2. Vacuum Impregnation
   1. Post wood preparation (i.e., cutting and treatment, if applicable), place sticks under a mesh in a food-safe plastic container, inside the vacuum desiccator and replace lid ensuring there is a tight seal, facilitated by a coating of vacuum grease (Figure 2).
   2. Attach a three-way valve between the tubing connecting the desiccator and pump, with a third tube leading to open air (Figure 2). Ensure that the three-way valve is closed off to the air and run the pump to achieve a vacuum of between -0.75 to -1.0 bar within the vacuum desiccator and hold this vacuum for 45 minutes - 1 hour.
   3. Submerge the open end of the third tube into a container of seawater. Switch the pump off and close the valve leading to the pump, then slowly open the valve until seawater is drawn by the vacuum into the desiccator. Allow the water to flow until it fills the plastic container, above the level of the mesh.
   4. Then withdraw the tube from the seawater in the container, allowing air to enter, until the desiccator returns to atmospheric pressure. Keep the sticks submerged under the mesh until they sink to the bottom of the plastic container.

Figure 2: Equipment used to vacuum impregnate wood sticks with seawater, in preparation for feeding to gribbles during a laboratory feeding assay.  A) Vacuum desiccator; B) Pump; C) Pressure gauge for the vacuum desiccator; D) The three-way valve leading to the vacuum desiccator, pump and to open air or seawater (orange tube). Please click here to view a larger version of this figure.

3. Leaching Wood
   1. Submerge seawater-saturated test sticks in seawater contained in 50 mL tubes (Figure 3). Replace water regularly for a period of 20 days.
      NOTE: The leaching process applies to any experimental wood under test, including treated or natural woods.

Figure 3: Leachate from wood sticks for preparation for feeding to gribbles during a laboratory feeding assay.  Wood that was fully submerged in seawater contained in a 50 ml Falcon tube, with regular water change (1-3 days), produced distinctly coloured leachate. From left to right leachate from heartwood of; sweet chestnut, turpentine, ekki, and beech and Scots pine sapwood. Please click here to view a larger version of this figure.

2. Extracting Gribble

1. Extract individual specimens of gribble from an infested wood block. Use a pair of fine forceps and a thin (size 000/0.4 mm or smaller) paintbrush. Carefully peel back any wood that is covering the gribble burrow with the forceps
   NOTE: Burrows are found on the surface of wood and can be identified by small holes (Figure 4).
2. Once gribble have been exposed, use a paintbrush to gently pick up individuals from underneath and deposit in a petri dish filled with seawater. Check gribble under a microscope to identify species and to ensure no damage was caused while extracting.
   NOTE: Beating pleopods are a sign of vitality.
   1. Discard any females brooding eggs as gravid females have a reduced feeding capacity.

Figure 4: Image of a gribble burrow with two typical ventilation holes.  L. quadripunctata burrow on a stick of Radiata pine wood, sized 2 mm x 4 mm x 20 mm. Two smaller ventilation holes can be seen next to the burrow entrance. Scale bar 2 mm. Please click here to view a larger version of this figure.

3. Identifying Limnoria quadripunctata
   1. Identify Limnoria quadripunctata under a stereomicroscope by the four distinct tubercles, arranged in a square pattern, on the animal's pleotelson in addition to an X-shaped carina on the fifth pleonite¹⁹ (Figure 5).

Figure 5: Limnoria quadripunctata identifying features.  Image of dorsal surface Limnoria quadripunctata, taken on a stereomicroscope at x20 magnification. Identifying features shown by red arrow - indicates the X- shaped carina and blue arrow - indicates four tubercles on pleotelson. Scale bar 1 mm. Please click here to view a larger version of this figure.

3. Preparing Well Plates

1. In multi-well plates with wells of diameter 20 mm, place one test stick and 5 mL of unfiltered seawater, between 32-35 PSU, per well (Figure 6).
2. Place treatments/species of wood systematically throughout the well plate so that each type of wood is represented at least once per plate. Add one gribble per well.
   ​NOTE: Temperature should be kept stable in an incubator at 20 °C ± 2 °C for the species L. quadripunctata, other species of Limnoria can be used with adjustments to the temperature made to suit the specific species.
3. Keep plates in constant dark conditions as the photoperiod does not have an effect on gribble feeding rate¹⁵.

Figure 6: Experimental set up for gribble feeding assay.  An example of a 12 multi-well plate used in the laboratory testing of gribble feeding rate. Each well contains 5 ml seawater and one test stick (20 mm x 4 mm x 2 mm) of different wood species; Scots Pine sapwood and ekki, beech, sweet chestnut, and turpentine heartwood. Scale bar 20 mm. Please click here to view a larger version of this figure.

4. Collecting and Counting Faecal Pellets and Assessing Vitality.

1. Twice per week, remove the test stick and each gribble (one per well) from the well plate and place into a freshly pre-prepared well plate (containing 5 mL of seawater per well [32-35 PSU, 18-22 °C]).
2. Use a paintbrush to gently brush off any faecal pellets from the stick before transferring and retain the faecal pellets within the original well.
   NOTE: Prior to transferring the gribble to a fresh well plate, vitality can be assessed on a scale of 1-5; 1= dead, 2 = passive, not on the wood, 3 = actively swimming or beating pleopods, not on the wood, 4 = crawling on the surface of the wood, 5 = burrowed into the wood.
3. Image Processing
   1. Use a fine paintbrush to separate any clumps so that individual pellets are visible and brush pellets away from the very edges of the well. Take a detailed photograph under a stereo microscope, at magnification x4 and upload to a computer (Figure 7).
      ​NOTE: Ensure the pellets are in focus and the background is uniform, with no shadows or light reflections on the surface of the water.

Figure 7: Image of gribble faecal pellets.  L. quadripunctata faecal pellets (small, cylindrical, brown pellets) from feeding on Radiata pine wood in one well of a multi-well plate. Taken at x4 magnification. Images prior to manipulation for image analysis (see Figure 7). A) Example of a suitable image to be used for automated counting in ImageJ. Pellets are sufficiently spread out and away from the edges of the well. The well is centred and there are no obstructions or reflections. B) An example of an image that is unsuitable for image analysis. The well is off-centre, cutting off the bottom half. Blue (dotted) circle shows light reflection off the surface of the water. Orange (solid) circle shows pellets that are clumped too closely together and too near the edge of the well. Red (dashed) circle shows a wood chip that was not removed. Scale bar 10 mm. Please click here to view a larger version of this figure.

2. Process to Generate Faecal Pellet Count Using ImageJ.
   1. Download ImageJ (latest version as of 03/08/21, 1.8.0_172) from https://imagej.nih.gov/ij/download.html or run from the computer’s browser. 
   2. Upload a stack of images by dragging and dropping or by selecting File | Import | Image sequence | Browse. Do not change any parameters then select Okay.
   3. Next, use the circle tool to select the bottom section of the well containing the faecal pellets. Remove the well edges, select Edit | Clear outside. Make the image binary, select Process | Make binary.
   4. Calibrate by selecting Analyse | Set scale and choose the number of pixels per millimeter for the image (for example 10 pixels = 1 mm). Count the pellets, select Analyse | Analyse particles.
   5. In the box next to Size (unit²), select a lower threshold that is the same as the smallest size pellet, using the unit scale set earlier (for example, if 10 pixels = 1 mm and the smallest pellet is 0.5 mm, choose 5-infinity).
   6. In the Show drop down box, select Outlines and then tick Summarise and press Okay (Figure 8).
      NOTE: Further information can be found at https://imagej.nih.gov/ij/docs/guide/index.html

Figure 8: A flow diagram of the process used in ImageJ to count faecal pellets.  A) Importing an image sequence in the File tab of ImageJ. B) The browse button in the 'Import Image Sequence' dialog box to import a sequence of images from a local device. C) Using the circle tool to select area containing faecal pellets D) Clear outside button in the edit tab area to remove outside of selected area. E) Make binary button in the process tab. F) Set scale button in the Analyse tab. Distance in pixels is equivalent to the number of pixels to one unit of measurement (mm). G) Analyse particles button in the Analyse tab. Size (unit^2) set to the lower threshold of faecal pellet size, in pixels, to infinity. Show 'outlines' and 'summarise' are selected. Please click here to view a larger version of this figure.

4. Data Analysis
   1. Convert pellet counts to pellets per day, which gives and indirect measure of feeding rate. Discard data from any moulting individuals on days that moulting occurred (Figure 9).
      NOTE: Moulting occurs over 1-3 days and can be identified when a full moult of the exoskeleton can be seen.

Figure 9: Example of a gribble moult.  Gribble (L. quadripunctata) moulting, on a Radiata pine wood test stick sized 20 mm x 4 mm x 2 mm. Moults are indicated by red circles. Scale bar 2 mm. Please click here to view a larger version of this figure.

Representative Results

A feeding experiment of L. quadripunctata was conducted over 20 days, using five different wood types (Scots pine (Pinus sylvestris L) sapwood, and heartwood of beech (Fagus sylvatica L), ekki (Lophira alata Banks ex C. F Gaertn), sweet chestnut (Castanea sativa Mil), and turpentine (Syncarpia glomulifera (Sm.) Neid)) (See Table of Materials), in November 2020. Eight replicate sticks were used per wood species and one specimen of Limnoria quadripunctata was fed per stick. All gribble were acquired from stocks that are maintained in aquaria at the Institute of Marine Sciences, University of Portsmouth, UK. Stocks are regularly supplemented with wild collections from the south coast of England. Animals are well acclimatized to the stable and consistent culture conditions prior to the experiment. Wood sticks (20 mm x 4 mm x 2 mm) were leached in seawater for two weeks prior to the feeding trial. One gribble, one test stick and 5 mL of seawater were placed per well in a 12 multi-well plate and kept in an incubator at stable conditions of 20 °C (± 0.2 °C) and in constant dark conditions. Faecal pellets were counted and collected every 2 to 5 days, with full water changes at each collection. Eight replicates of each wood species were used, giving a total of forty sticks with one individual gribble each. Seawater used for leaching wood and used throughout the experiment was obtained directly from the aquarium used to rear specimens. Seawater conditions are stable in the aquarium and stable in the incubator. Evaporation from the small volume of water used per well is minimised by the lid design of the well plates and full water changes occurring every 2-5 days.

Pellets were counted automatically using Image J (version 1.8.0_112).

Gribble feeding on Scots pine sapwood wood as a control, produced the most faecal pellets per day consistently, apart from at Day 20 where pellet production was overtaken by beech. Ekki produced the lowest faecal pellets per day of all the wood species tested. The second highest faecal pellet production was seen on beech, followed by sweet chestnut and turpentine. There was an increase in faecal pellet production in all species from Day 5 to Day 7. Pellet production dropped in all species, other than ekki, between Day 7 and Day 12, possibly due to the increased time between water changes. After this, faecal pellet production remained fairly consistent among each of the wood species. From Day 14, Scots pine decreased in daily faecal pellet production, while beech increased (Figure 10).

Figure 10: Number of faecal pellets per day (n=40) (mean ± SE) produced by L. quadripunctata using different wood species, over 20 days. Turpentine, sweet chestnut, beech and ekki heartwood tested, with Scots pine sapwood used as a control. Please click here to view a larger version of this figure.

The highest vitality of score (5) was seen in most individuals feeding on Scots pine wood, other than the one dead individual. 5 indicates animals that have burrowed into the wood and this was seen only on Scots pine sapwood and beech heartwood. By Day 12 for Scots pine and Day 20 for beech, all living individuals had burrowed into the wood. Sweet chestnut had the highest percent mortality but did not increase over time. The remainder of living individuals stayed at a vitality of 4 (crawling on the wood surface), apart from at Day 14 where two individuals were off the wood (vitality of 3). Ekki and turpentine also had the majority of individuals at a vitality of 4 over the duration of the experiment, apart from at Day 14 and Day 5 for turpentine. Mortality did not show an increase over time across any of the wood species. Only burrowing was seen to increase on Scots pine and beech while the other three wood species mostly remained at a vitality of 4 (Figure 11).

Figure 11: Vitality of individuals over time, as a percentage of replicates, feeding on different wood species.  Turpentine, sweet chestnut, beech and ekki heartwood tested, with Scots pine sapwood used as a control. Of eight replicates per wood species, the percentage at different vitalities were plotted over the 20-day experimental period. Dark blue represents a vitality of 5 (burrowing), light blue a vitality of 4 (on wood), grey a vitality of 3 (off wood but active), purple a vitality of 2 (off wood and passive) and black shows a vitality of 1 or dead individuals. Please click here to view a larger version of this figure.

Results from this testing method can be used to identify wood types or treatments that have an increased resistance to marine wood-borer damage. Then, marine field trials, as described in the European Standard EN 275, can be conducted and durability can be graded (0= 'no attack', 1= 'slight attack', 2= 'moderate attack', 3= 'severe attack', 4='failure'²⁰) in addition to comparison to non-durable control wood.

Discussion

Prior to selecting gribble specimens to be used in the feeding experiment, individuals should be screened to assess their suitability. There can be some variation in feeding rate between individuals due to differences in size, so only fully grown adult specimens should be selected. No significant difference between feeding rate of individuals between 1.5 mm and 3 mm length was detected by Borges et al., 2009¹⁷. Female Limnoria brood their eggs, during which time have a reduced feeding rate. Therefore, any brooding females should be checked for and discarded while selecting specimens. Similarly, moulting individuals will also have a reduced feeding rate²¹. Therefore, faecal pellet counts on days when individuals are moulting should be discarded¹⁷. As moulting occurs for more than a day, moults are counted when a full exoskeleton moult can be seen on pellet collection days. Limnoria, when creating their burrows, have an increased faecal pellet production and will also produce more frass (fine wood waste that is not incorporated into the faecal pellets). The high levels of frass can interfere with identification of faecal pellets but can carefully be removed under stereomicroscope observation, using a pipette or fine paintbrush, prior to image capture for automatic counting. Alternatively, pellets can be counted manually.

The software ImageJ requires quality, in-focus images for image processing. To this end, images should be captured in which faecal pellets are not obstructed by the well walls and a paintbrush should be used to separate individual faecal pellets. The background of the image must be uniform with no areas of light or shadow, which would interfere when the image is transformed to binary for processing in ImageJ. There is no need to adjust contrast or light prior to image processing. When importing a stack of images, all photographs must be taken in the same plane so no errors occur while processing.

Vacuum impregnating wood with seawater causes the wood to sink and become readily accessible to the gribble. Leaching wood prior to exposing it to gribbles will remove any water-soluble extractives that may impact their feeding rate or cause mortality¹². Mortality due to extractives in the water is not representative of mortality to be expected in the sea, where extractives will become rapidly diluted. Well plates should be kept at a constant temperature that is the optimal for the gribble species being tested. The common southern British species, L. quadripunctata, feeds well between 15 and 25 °C and has an optimum feeding rate at 20 °C¹⁷ so well plates can be conveniently kept in an incubator at a constant 20 °C ± 0.5 °C.

Assessing the vitality of the feeding gribble detects sublethal or pre-lethal effects of wood treatments or naturally durable timbers. A high vitality of 5 indicates that the gribble is demonstrating natural behaviour by burrowing into the wood and suffers no adverse effect from contact with it. A vitality of 4 shows that while not having burrowed into the wood, the gribble is still comfortable to crawl along its surface. A score of 3 is given to gribble that are not on the wood, but instead actively swimming in the water or are stationary but with rapidly beating legs and pleopods. A low vitality of 2 means that the gribble is exposed and/or has little energy. This may come about from a prolonged period of low feeding rate or from extractives either leaching into the water or becoming accessible during feeding. If high mortality is seen after 7-8 weeks, this may be due to starvation, as starved gribbles (kept in wells with just 5 ml of seawater and no wood) can survive for this long (personal observation).

The benefits of using a short-term laboratory assay as opposed to longer-term marine field trials, is that novel treatments and wood products can be rapidly tested to identify their potential to be used commercially. Furthermore, such assays can facilitate rapid optimization of treatment processes. If a significantly lower faecal pellet production is seen compared to a control wood, then testing can be supplemented by marine trials. Slevin et al., 2015²³ and Westin et al., 2016²⁴ demonstrate a good correlation between laboratory and field assessments through testing the same wood in two different settings, indicating a capable predictive ability of the former. A short-term assay can be run for several weeks. Starved gribbles can survive for 7-8 weeks when kept in well aerated water without wood (personal observation) which may provide additional comparison if investigating the mortality response to different types of woods. However, through recent, unpublished observations, there is no significant fluctuation in faecal pellet production over a time period longer than 20 days, other than when mortality begins to occur. In addition, previous methods, such as that used by Borges et al., 2008 and 2009, runs for 15 days. Therefore, 20 days is a sufficient time for a rapid laboratory-based test to provide indication of wood durability.

While this method is suitable for short-term trials, findings should be complemented by long-term marine field experiments. Laboratory conditions cannot replicate the variety of biotic and abiotic factors that may affect wood in the marine environment. Biofouling organisms, along with other species of marine wood-borers (such as shipworms) may still be present and cause damage to the wood^25,26. In addition, abrasion from wave-thrown shingle or sand can wear wood down, which may then become accessible for gribbles²⁷. However, a standard laboratory method can provide an initial screening of new products which show promise for marine applications. By assessing the faecal pellet production and vitality, woods that are better at reducing gribble feeding rate can be identified.

Due to the regulations and restrictions of wood preservatives, such as CCA and creosote, it is important to find novel products to replace these treatments. Timber is subject to high levels of biodegradation in the marine environment but is still one of the most renewable construction materials available and retains its strength and structure well in seawater^27,28. Not only will timber that is resistant to biodegradation reduce costs but will also be more environmentally friendly than using alternative materials such as concrete or steel, which require high energy input during manufacture^29,30, or broad-spectrum biocide preservatives that may leach out and affect the surrounding ecosystem^{31,32,33,34,35,36,37}.

Materials

Name	Company	Catalog Number	Comments
12-well cell culture plates	ThermoFisher Scientific	150200
50ml Falcon tubes	Fisher Scientific	14-432-22
Adjustable volume pipette	Fisher Scientific	FBE10000	1-10 ml
Beech	G. Sawyer (consultant in timber technology)	Fagus sylvatica	Taxonomic authority: L
Ekki	G. Sawyer (consultant in timber technology)	Lophira alata	Taxonomic authority: Banks ex C. F. Gaertn.
Forceps	Fisher Scientific	10098140
Incubator	LMS LTD	INC5009
Microporous specimen capsules	Electron Microscopy Sciences	70187-20
Petri dish	Fisher Scientific	FB0875713
Scots Pine	G. Sawyer (consultant in timber technology)	Pinus sylvestris	Taxonomic authority: L.
Size 00000 paintbrush	Hobby Craft	5674331001	Size 000 or 0000 also acceptable
Sweet Chestnut	G. Sawyer (consultant in timber technology)	Castanea sativa	Taxonomic authority: Mill
Turpentine	P. Evans (Professor, Dept. Wood Science, University of British Columbia)	Syncarpia glomulifera	Taxonomic authority: (Sm.) Nied.
Vacuum desiccator	Fisher Scientific	15544635