The effectiveness of disinfectant and steam exposure treatments to prevent the spread of the highly invasive killer shrimp, Dikerogammarus villosus

Biosecurity protocols designed to prevent the spread of invasive alien species (IAS) are now an essential aspect of IAS management. However, the effectiveness of various biosecurity treatments requires further exploration. Killer shrimp, Dikerogammarus villosus, a notoriously high impact and ecosystem destabilising invader, has rapidly spread across Europe, and is of concern to invade Northern America. In this study, we examine the effectiveness of three commonly used, broad-spectrum disinfectants to cause mortality of D. villosus: Virasure Aquatic, Virkon Aquatic, and Virkon S. Immersion and spray treatments of 1%, 2% and 4% disinfectant solutions were examined for applications of up to 300 secs immersion and for up to ten consecutive sprays. Furthermore, we assessed the effectiveness of steam (≥100 °C) treatments for up to 120 secs. For all disinfectants, immersion in 1% solutions caused 100% mortality at ≥120 secs. At higher concentrations, shorter immersion times caused complete mortality: 60 and 15 secs for 2% and 4% solutions, respectively. Five sprays of 2% and 4% solutions resulted in 100% mortality, for all disinfectants. Direct steam exposure was highly effective, with complete D. villosus mortality occurring at ≥10 secs. Overall, brief exposure to broad-spectrum disinfectants and direct steam could be used to limit D. villosus spread.

www.nature.com/scientificreports www.nature.com/scientificreports/ for actions taken to decontaminate equipment and thus aims to prevent the spread of IAS and has become a key aspect of IAS management strategies [12][13][14] . Accordingly, there is an urgent need to enhance biosecurity by identifying simple prevention protocols that minimise risk of spread yet remain user and environmentally friendly 13,[15][16][17][18] .
Aquatic disinfectants such as Virasure Aquatic, Virkon S and Virkon Aquatic are used by recreational water users and responsible authorities, including government agencies, for the decontamination of equipment. These disinfectants are available in powder or tablet form and can be applied through spray applications or immersion of the equipment into disinfectant solutions. Although broad-spectrum aquatic disinfectants have been demonstrated to kill harmful pathogenic microbes and various invasive Mollusca species 19,20 , the effectiveness of these oxidising agents in killing free-living aquatic IAS requires further study. Recent studies have shown partial effectiveness of disinfectants in killing invasive aquatic plants and invertebrates 14,17,18 . However, identification of optimal disinfectant treatment to achieve complete mortality of a range of IAS, whilst minimising time and expense, including testing several types of disinfectants is still required. Steam exposure has also been proposed as a treatment to decontaminate equipment that may have been exposed to IAS. Short applications of steam have been effective in killing both invasive macrophytes 21 and invertebrates 17,22,23 , however, further testing is still required to determine efficacy for species-specific treatments. Further, the identification of practical and efficacious biosecurity protocols is essential for the reliable uptake of biosecurity practices by environmental stakeholders and prevent behavioural barriers 15 .
The killer shrimp, Dikerogammarus villosus (Sowinsky 1894), is a highly invasive euryoecious amphipod crustacean native to the Ponto-Caspian region. It has spread and successfully colonised most of the major European inland waterways 1 having been spread by many anthropogenic vectors 5,6 and is of concern to invade Northern America. Capable of destabilising ecosystems, D. villosus is an especially damaging invader that causes profound declines of native macroinvertebrate populations 1,24 , and has been found to even prey upon fish eggs and larvae 25 . The propagule pressure associated with D. villosus is considered to be high, as one gravid female can to hold up to 190 eggs 26 , therefore introductions of even one organism may result in establishment, as seen in other amphipod species 27 . Accordingly, the need to prevent further potential spread, and therefore reduce propagule pressure of this prolific invader is clear.
In this study, we examined the efficacy of immersion and spraying of selected broad-spectrum disinfectants and direct steam exposure to cause mortality of D. villosus. To achieve this we examine the effectiveness of three commonly used disinfectants, including two previously untested disinfectants, Virasure Aquatic and Virkon Aquatic. In addition, we assess the effectiveness of direct steam spray for a range of application durations, including a very short and previously untested duration of 5 secs. Examined exposure times were designed to reflect realistic application times achievable by users of such biosecurity protocols. We hypothesise that greater concentrations and longer exposure times will cause substantial, if not complete mortality of D. villosus specimens, reducing potential propagule pressure. Equally, we expect that longer exposures of steam that induce thermal shock, rapidly killing D. villosus.
Steam spray. Total D. villosus mortality was caused by direct steam exposures of ≥10 secs, whilst exposure for 5 secs resulted in mean 70% mortality (Table 1). All control groups displayed 0% mortality. Steam treatments had a significant effect on mortality of D. villosus (χ 2 = 148.13, df = 5, P < 0.001). There were no significant differences in mortality among steam application durations (all P > 0.05).

Discussion
Immersion in disinfectant solutions was shown to be a suitable potential biosecurity treatment leading to complete D. villosus mortality. Mortality was greater at higher concentrations of disinfectant and for longer immersion durations. For all three disinfectants tested, total mortality of D. villosus was achieved following immersion times of ≥120 secs, 60 secs and 15 secs for 1%, 2% and 4% solutions, respectively. Disinfectant spray treatments were also effective. Total D. villosus mortality was observed for all three disinfectants at 2% and 4% solutions following 5 spray treatments. High mortality (>85%) was recorded following 10 spray treatments of 1% solutions. Overall, for shorter immersion times and reduced spray exposure, Virasure Aquatic solutions appeared to be marginally more effective than Virkon S and Virkon Aquatic. Steam exposure was highly efficacious, with complete mortality occurring at exposure durations of ≥10 secs and high mortality (70%) at 5 secs exposure.
Dikerogammarus villosus can adhere and remain attached to water users' equipment 28 , upon which they are capable of surviving for up to 16 days in damp conditions 12 . To inhibit the further overland spread of this highly invasive amphipod, biosecurity practices utilising disinfectant treatments would be especially beneficial for decontamination of small items of PPE and equipment. For instance, wetsuits, waders and nets could be www.nature.com/scientificreports www.nature.com/scientificreports/ completely immersed in disinfection baths, while spray applications may be more suitable for the decontamination of larger equipment, e.g. boats, outboard motors, and vehicles 21,29 . Furthermore, water intake-systems, designed to aid cooling of outboard motors, or large pipes such as those used in flood management and raw water movement could also be flushed with disinfectant solutions.
Whilst disinfectants have been shown to be effective against microbes and certain IAS [18][19][20] , evidence presented in the literature does indicate limited effectiveness against other IAS (e.g. macrophytes 16,17 18 . Previous research has identified treatment time as a barrier to good biosecurity practice 15 . We demonstrate that complete mortality can be achieved with shorter treatment durations (≥120 secs) using 1% disinfectant solutions, making this treatment potentially a useful addition to field biosecurity measures. Furthermore, we demonstrate that other disinfectants are also equally as effective as Virkon S, with equal application times of 1% solutions needed to achieve complete mortality of D. villosus.
We also found that short duration exposure to steam was effective against D. villosus, identifying 10 second exposure of steam resulting in complete mortality, in line with findings by Stebbing and Rimmer (2013) 22 . Direct application of pressurised jets of steam may prove to be highly beneficial, particularly when combined with additional cleaning methods such as hand removal, brushing or scraping 21,30 . Equally, steam treatments may aid decontamination of equipment items that are problematic to manually clean, such as niche areas or large complex structures like chain lockers, intake grates, pipework, trailers and vehicles. Whilst the effectiveness of steam to kill a number of IAS has been tested 21,23,31 , the application of steam needs to be tested against a wider range of IAS. Furthermore, considering the apparent effectiveness of thermal shock, pressurised hot water sprays should also be assessed in future studies against a range of IAS, including previously untested macrophytes 32,33 .
The provision of in-field biosecurity stations for a range of stakeholders could act as a suitable mechanism to limit IAS spread 15,21,31,34 . Installation of decontamination apparatus and facilities, with clear guidance, may facilitate utilisation of these simple but highly efficacious biosecurity techniques 13,34 . The provisioning of these biosecurity stations will also create a platform for raising awareness of IAS and biosecurity with all users. These biosecurity facilities could be placed at points of exit and entry (e.g. angling stations and boat ramps) to ensure ease of access to the steam cleaners or large soaking stations containing disinfectants 21,23,31 . Maintenance and responsibility of such stations would be essential, especially in the case of disinfectants, as the disinfectant solutions decay over time and become less effective. Furthermore, suitable disposal methods would need to be in place, such as interceptors for treatment water/disinfectant run-off, especially when considering equipment being cleaned prior to entering a site. Although the risk of toxicity to non-target aquatic organisms through disinfectant residues and spills is considered to be low 19 , further examination of low concentration lethality on non-target species is needed. This is of particular importance at biosecurity stations locations where there is a greater opportunity for repeated spills compared to a single-visited area. Biosecurity guidance must highlight the correct disposal of used disinfectant water. Furthermore, the legal issues concerning the use of broad-spectrum disinfectants as biosecurity agents for invasive macroscopic organisms will need to be addressed (e.g. herbicide or insecticide 16,18 ).
The results presented here demonstrate that broad-spectrum disinfectants and direct steam applications could be used as part of effective and efficient biosecurity protocols to prevent the further anthropogenic-mediated spread of D. villosus. Reducing the propagule pressure of this prolific invader is essential. Accordingly, promotion and adoption of these techniques by biosecurity campaigns, stakeholder groups, and practitioners should be encouraged. Furthermore, the requirement to perform and adhere to a biosecurity standard should be incorporated into relevant Codes of Practice, with subsequent enforcement in relation to all water users.

Methods
Specimen collection and maintenance. Dikerogammarus villosus specimens were collected from Grafham Water, Cambridgeshire, UK (52°17′31.2″N, 0°19′23.9″W) and Cardiff Bay (51°27′14.7″N, 3°09′50.4″W). Specimens were transported in source water to the University of Leeds, UK and housed in aerated aquaria filled with dechlorinated tap-water, at a constant temperature of 14 °C under a 12:12 hr light-dark regime. Organisms were acclimated for over one week prior to experimental use. Adult specimens were used as they have been shown to be less susceptible to treatments than juveniles 18 . Specimens from Cardiff Bay were only used for the assessment of Virkon S disinfectant spray treatments, due to a shortage of Grafham Water specimens.
In all cases, groups of ten D. villosus were weighed (mean ± SE individual specimen weight: 115.0 ± 0.9 mg) and briefly maintained (<30 mins) in aerated dechlorinated tap water in circular containers (SA, 548 mm 2 ; volume, 1917 mm 3 ) prior to experimentation. Only active individuals that responded to a stimuli were selected; specimens that displayed visible parasitism or had recently moulted were not used. Using fine-meshed flat-bottomed sieves (SA, 528 mm 2 ; volume, 1848 mm 3 ), treatment groups were immersed in disinfectant solutions for the allotted exposure time. Control groups were likewise immersed in dechlorinated tap water (i.e. 0% solution) for the same exposure times. Following experimental exposure, the fine-mesh sieves containing the ten D. villosus were removed from the experimental solution and re-immersed in dechlorinated tap water for a two-minute period to remove excess disinfectant; this was repeated twice (see Cuthbert et al. 17 ). Following this washing process, www.nature.com/scientificreports www.nature.com/scientificreports/ specimen groups were returned to 200 ml of aerated dechlorinated tap water in their original containers for a 24 hr recovery period (14 °C; 12:12 hr light-dark), after which mortality was assessed. Specimens were considered dead if they did not respond to stimuli and did not hold their pereopoda under their body 13 . Disinfectant spray. Mist-spray applications for all three disinfectants were examined using 1%, 2% or 4% solutions, and a 0% control. Groups of five D. villosus were weighed (107.4 ± 1.1 mg) and briefly maintained in fine-meshed flat-bottomed sieves (SA, 528 mm 2 ; volume, 1848 mm 3 ) within a circular container (SA, 548 mm 2 ; volume, 1917 mm 3 ) in dechlorinated tap water (<30 mins). The sieve was removed from the water and, using a hand-held spray bottle, 2, 5 or 10 spray applications of a disinfectant solution were delivered (n = 3 per experimental group). This were directly applied to treatment groups held within sieves, at a distance of 6-8 cm from the exit-point of the spray bottle. Application of one spray equated to 0.75 ml of solution per 528 mm 2 . The sieves containing the experimental specimens were then left air-exposed for a five minutes period (~20 °C), before being re-immersed in dechlorinated tap water for a period of two minute to removed excess disinfectant. This washing process was repeated twice. Following this, specimen groups were returned to 200 ml of aerated dechlorinated tap water and mortality was assessed following a 24 hr recovery period (as above).
Steam spray. Specimens were directly exposed to a continuous jet of steam for 5 secs, 10 secs, 30 secs, 60 secs, and 120 secs (≥100 °C; Karcher SC3) (n = 3 per experimental group). Groups of ten D. villosus were weighed (111.5 ± 2.4 mg) and briefly maintained in fine-meshed flat-bottomed sieves (SA, 528 mm 2 ; volume, 1848 mm 3 ) in aerated dechlorinated tap water within a larger container (SA, 548 mm 2 ; volume, 1917 mm 3 ) prior to experimentation. Steam was directly applied to groups held within sieves at a distance of 6-8 cm from the exit-point of the lance, ensuring equal application over the surface area of the sieve. Following exposure, groups were air-exposed for a 10 minute period (~20 °C) to allow gradual cooling before being re-immersed in dechlorinated tap water. This was to avoid a second thermal shock occurring if specimens were immediately returned to water after exposure to high temperatures. Control groups were air-exposed for twelve minutes, i.e. the duration of the longest steam exposure and cooling period combined. Mortality was assessed following a 24 hr recovery period (as above).

Data availability
The datasets generated during and analysed during this study are available from the corresponding author on reasonable request.