Unionid mussels are globally threatened by several human disturbances, including the introduction of non-native species. Among these, biofouling zebra and quagga mussels of Ponto-Caspian origin are considered to be especially detrimental to unionid locomotion, filtration and physical condition. The aim of our study was to determine and compare the impact of dreissenid fouling and/or presence on locomotion and burrowing of the native Unio tumidus and invasive Sinanodonta woodiana, a novel invader expanding its range in Europe in recent decades. We tested unionids collected from Lake Balaton (central Europe) that were fouled by dreissenids (zebra and quagga mussels mixed), cleaned of fouling or non-fouled (collected without any signs of dreissenid fouling). Moreover, unionids were tested in the presence or absence of other fouled individuals and dreissenids isolated in mesh bags to determine the influence of direct fouling and presence of dreissenids in the environment on unionid behaviour. Movement initiation time, locomotion distance and burrowing level were retrieved from videos recorded for 24 hours. Direct fouling affected only the behaviour of U. tumidus, limiting their burrowing and delaying movements. After removal of fouling, movement timing returned back to normal, but mussels still burrowed less than the control non-fouled individuals, indicating persisting effects of fouling on physical condition. Moreover, U. tumidus reduced their locomotion in the presence of fouled unionids. Sinanodonta woodiana responded to the presence of dreissenids (especially quagga mussels) with increased burrowing. These different responses of the two unionid species to Dreissena spp. indicate that biofoulers may influence biotic interactions between the unionids by promoting the invasive species (less susceptible to negative effects of fouling). Moreover, S. woodiana may indirectly affect U. tumidus through apparent competition, constituting an environmental reservoir of biofoulers exerting a stronger impact on the native species.

Biofouling, biological invasions, Bivalvia, burrowing, ecosystem engineers, interspecific interactions, locomotion, Sinanodonta woodiana, Unio tumidus, Unionidae

Bivalves are ecosystem engineers altering ecosystem structure and function by increasing water clarity and modifying the bottom quality (Strayer et al. 1999; Karatayev et al. 2002; Sousa et al. 2009). Their reported effects consist in declines in plankton abundance (Kissman et al. 2010; Karatayev et al. 2023), decreases in young fish survival (Irwin et al. 2009), changes in fish feeding patterns (Mayer et al. 2001; Balogh et al. 2022), increases in benthic invertebrate abundance (Novais et al. 2015; Ilarri et al. 2018), and development of submerged aquatic vegetation (Chu et al. 2004; Zhu et al. 2006), resulting in a more benthic-oriented food web structure (Mills et al. 2003; Miehls et al. 2009).

Bivalve species richness and abundance decrease all over the world due to ongoing climate change, increasing water pollution and habitat destruction (Strayer and Malcom 2007). This particularly applies to freshwater mussels of the family Unionidae, with dozens of already extinct species and hundreds more heavily endangered (Aldridge et al. 2023; Sousa et al. 2023). Due to their important environmental functions and ecosystem services they provide, further unionid extinction may lead to irreversible changes and effects on the entire aquatic community (Vaughn 2018; Zieritz et al. 2022).

One of the greatest threats to native unionid mussels is the spread of invasive mussel species (Gutiérrez et al. 2014; Douda et al. 2024), including biofouling dreissenids: the zebra mussel Dreissena polymorpha (Pallas, 1771) and the quagga mussel Dreissena rostriformis bugensis Andrusov, 1897. The dreissenids, originating from the Ponto-Caspian region (catchments of the Black, Azov and Caspian Seas and lowland parts of the inflowing rivers) have successfully invaded freshwater ecosystems in Europe and North America (Matthews et al. 2014; Collas et al. 2018). Dreissena polymorpha has spread in Western Europe since the beginning of the 19^th century (Harzhauser and Mandic 2010), whereas D. r. bugensis started to spread in the 20^th century (Therriault et al. 2005) and its invasion is ongoing on both continents. This has led to the co-occurrence of both species in more and more invaded locations (Zhulidov et al. 2004; Grutters et al. 2012).

Dreissena spp. rapidly develop large populations and settle on unionids in high numbers, especially when other hard substrata are scarce, and the mass of fouling dreissenids can exceed the biomass of their host (Burlakova et al. 2000). This impairs physiological condition of unionids, as Dreissena spp. can block their siphons preventing them from filtering and therefore breathing and feeding (Ricciardi et al. 1996; Sousa et al. 2011; Bódis et al. 2014). Dreissenids can also inhibit unionid locomotion (Van Appledorn and Bach 2007) and/or burrowing, making them vulnerable to unfavourable environmental conditions or predation (Saloom and Duncan 2005). In addition, they may compete for food with the unionids, causing severe declines in phytoplankton (Higgins and Zanden 2010). As a result, the spread of Dreissena species may further endanger or lead to extinction of already threatened unionid species (Ricciardi and Whoriskey 2004).

An exception in the generally endangered Unionidae family is the invasive Chinese pond mussel Sinanodonta woodiana (Lea, 1834), native to Eastern Asia (Lopes-Lima et al. 2020), but in recent decades (since 1959) spreading in Europe (Douda et al. 2024). This is a fast-growing and fast-reproducing (Douda et al. 2021) species of high ecological plasticity, tolerant to wider ranges of environmental conditions in comparison to native European Unionidae (Kraszewski and Zdanowski 2007; Poznańska-Kakareko et al. 2021; Zieritz et al. 2021; Dobler et al. 2022). It can co-occur with native mussels and outcompete them for food, fish hosts for their larvae, and living space (Douda and Čadková 2018).

Both native and invasive Unionidae are commonly fouled by dreissenids (Bódis et al. 2014; Balogh et al. 2024). Dreissenids can interfere with the invasion of S. woodiana, with the final outcome depending on the relative susceptibility of S. woodiana and native unionids to fouling and its effects (Bódis et al. 2014). High vulnerability of the invasive S. woodiana to fouling would negatively affect its establishment in areas occupied by dreissenids, whereas its resistance would facilitate competitive exclusion of other unionid species (Sousa et al. 2011).Therefore, the aim of our study was to determine and compare the effects of fouling and/or presence of the invasive mussels D. r. bugensis and D. polymorpha (responses to own fouling, to the presence of fouled conspecifics and to the presence of dreissenids on other substratum in the environment) on the locomotion and burrowing of two unionid mussels: the native Unio tumidus Philipsson, 1788 and the invasive S. woodiana. These species often co-occur in areas invaded by S. woodiana in Europe (e.g. in Lake Balaton; Benkő-Kiss et al. 2013) and can interact with each other, which can be additionally modulated by biofouling. For our study, the mussels were collected from a single location in Lake Balaton (Hungary) to make sure that potential differences in local living conditions of particular species would not affect the results. We hypothesized that: (1) fouling reduces unionid locomotion and burrowing; (2) this effect can be purely mechanical (only when the fouling is directly present), or (3) can persist after fouling removal (due to worsened physiological condition of the host). Additionally, we hypothesized that (4) unionids can detect and respond by behaviour to chemical cues released by other unionids fouled by dreissenids or directly by Dreissena spp. individuals.

Consistently with our first hypothesis, dreissenid fouling affected the behaviour of unionid mussels. Sinanodonta woodiana responded only with changed burrowing level. In Experiment 1, we observed shallower burrowing of S. woodiana immediately after fouling removal, but, over time, burrowing returned to the level exhibited by the control, non-fouled mussels. On the other hand, in Experiment 2, S. woodiana burrowed deeper in the presence of quagga mussels compared to the control. This might be a defensive response of S. woodiana to the presence of dreissenids, consisting in increased burrowing. It is worth noticing that a similar, though marginally non-significant tendency for increased burrowing was observed in S. woodiana exposed to fouled unionids (variants CwF vs F/C1, Fig. 2, Table 4).

The pattern observed for S. woodiana in Experiment 1 can be explained by contrasting effects of fouling (mechanical obstacle to efficient burrowing) and defensive responses to Dreissena spp. presence (stimulating burrowing). For fouled individuals, burrowing was more difficult, but they kept trying to defend themselves from fouling, resulting in a similar level as that showed by the control mussels. Immediately after the fouling removal, with no dreissenids present in the environment (variant F/C1, no need for anti-fouling defence), they burrowed more shallowly due to recently experienced problems related to fouling (e.g. exhaustion), but, with time (the next experimental round, F/C2), their burrowing returned to the control level. Nevertheless, this effect of fouling persisting after cleaning, though short-timed, suggests that the impact of dreissenids on their hosts was not purely mechanical, but also affected their condition.

Increased burrowing is a natural defence mechanism (Saloom and Duncan 2005) and may indicate an attempt to avoid fouling mussels, which, once attached, could have a more negative impact on their host condition than when acting only indirectly by their presence nearby (Sousa et al. 2011). Mussels are known to sense the presence of predators (Reimer and Harms-Ringdahl 2001; Meira et al. 2024) and parasites (Selbach and Mouritsen 2020) from chemical cues in the environment, as well as detect alarm signals from damaged conspecifics (Leonard et al. 1999), which results in the induction of various defence mechanisms, such as increased burrowing (Griffiths and Richardson 2006). The occurrence of a defence mechanism in S. woodiana may give this species an advantage in unfavourable conditions, such as the presence of biofouling Dreissena spp. The deeper the mussel burrows, the less exposed its shell is to the fouling dreissenids, which directly affects the degree of fouling, as shown by Urbańska et al. (2019). Moreover, existing fouling can also be reduced by the burrowing of the host unionid, as dreissenids are known to detach from unsuitable (e.g. burrowed) substrata and move in search of a better attachment site (Dzierżyńska-Białończyk et al. 2018b; Balogh et al. 2024). Dzierżyńska-Białończyk et al. (2018a) and Balogh et al. (2024) have shown that burrowed S. woodiana (but not the native unionid species, including U. tumidus) were less fouled by dreissenids than individuals exposed on the surface, confirming a defensive potential of this behaviour.

In Experiment 2, S. woodiana responded significantly only to the presence of quagga mussels, which confirmed that unionids can detect and respond to chemical cues of other bivalves in the environment (our fourth hypothesis). However, it should be noted that a similar tendency was also apparent in the presence of zebra mussels, thus the support for the hypothesis that unionid responses to fouling depended on the fouler species remains weak. Anyway, it is possible that S. woodiana is more familiar with quagga mussels, which are more abundant in Lake Balaton, although this ratio may vary in different parts of the lake (Balogh et al. 2008; Balogh et al. 2024). Moreover, quagga mussels are less associated with their substratum (D’Hont et al. 2021) and easier to detach due to their weaker adhesion strength (Balogh et al. 2019), thus burrowing can be a more effective defence mechanism against them compared to more strongly and more permanently attached zebra mussels.

In Experiment 1, fouled U. tumidus showed a delayed movement initiation time and shallower burrowing compared to the control, non-fouled individuals. After the fouling removal, the movement initiation time of U. tumidus (variant F/C1) returned to the control level, but the burrowing remained weakened. Therefore, changes in the behaviour of U. tumidus were driven by the mechanical effect of fouling present on their shells, as well as by the impaired condition of fouled mussels. This supported our third hypothesis, as the fouling effect persisted after the fouling removal.

We did not observe any effects of direct fouling on the locomotion, in contrast to Van Appledorn and Bach (2007) working on North American unionids Ligumia nasuta Say, 1817 and Anodonta grandis Say, 1829. However, in accordance with our fourth hypothesis, U. tumidus responded to the presence of other fouled individuals by stopping any horizontal movements and delayed movement initiation (here: burrowing start, as locomotion did not occur at all) (CwF vs F/C1). Theoretically, this difference might be an artefact resulting from the fact that mussels from F/C1 were tested in the experimental arena for the second time, whereas for those from CwF it was the first experimental round. However, the comparison between animals tested repeatedly in consecutive experimental rounds under the same conditions (F/C1 vs F/C2) showed no significant differences and symptoms of exhaustion or habituation due to repeated testing, thus this explanation seems unlikely. On the other hand, the observed responses of U. tumidus to the presence of fouled individuals may represent a general reaction to stressed (here: fouled) conspecifics secreting some infochemicals indicating stressful environmental conditions. Definitely, U. tumidus did not respond to direct cues from dreissenids, as shown by Experiment 2.

All changes in the behaviour of U. tumidus induced by fouling or nearby presence of fouled individuals seem negative, exposing them to a number of environmental threats, including predation, parasites, dislodgement by water movements, desiccation during droughts, and further fouling by dreissenids (Bowers et al. 2005; Lymbery et al. 2021).

Our study showed that the behaviour of the native U. tumidus in the presence of fouling Dreissena spp. mussels was modified to a higher extent than that of the invasive S. woodiana. Moreover, the responses of S. woodiana appeared to have defensive and preventive effects against dreissenid fouling, whereas the behaviour of U. tumidus seemed to be impaired compared to their normal (control) activity. Thus, despite a similar relatively moderate level of fouling of both species in our study, dreissenids presence had stronger negative effects on U. tumidus. Due to large body size of S. woodiana and its weaker burrowing (Poznańska-Kakareko et al. 2021), more dreissenids, especially zebra mussels, can attach to and persistently stay on their surface than to smaller native mussels. Therefore, S. woodiana can act as a reservoir for dreissenids colonizing the native species, with the native species suffering more severe negative consequences of fouling. This suggests yet another potential effect of invasive dreissenids on native communities: their presence can influence interactions between native and invasive species of Unionidae, promoting the latter. On the other hand, S. woodiana can provide suitable substratum for dreissenid development. Moreover, we have shown that, beside parasites (McElwain 2019) and predators (Meira et al. 2024), dreissenids can constitute another driver of apparent competition between invasive and native species, when the invader poses a threat to the native organism by increasing the abundance and availability of biofoulers. This effect is likely to be most pronounced in locations where shells of unionid mussels constitute the primary source of hard substratum available to dreissenids, such as soft lake bottoms, or dam reservoirs and river pools with limited flows and accumulated sediments, missing rocks and artificial embankments (Dzierżyńska-Białończyk et al., 2018a; Balogh et al. 2024). Nevertheless, such places are common in the environment and are frequently used by unionids. Therefore, dreissenid mussels can extend their distribution ranges as they are able to use their shells as attachment sites.

It should be noted that our research was conducted using mussels collected from a single location in Lake Balaton. This allowed us to eliminate potential confounding effects related to different conditions experienced by experimental animals before their sampling. We checked the situation taking place within the same natural community, living under the same conditions. However, we must admit that conducting the same research using mussels from other regions of the world may result in different results. To better understand the influence of dreissenids on Unionidae, it would be useful to conduct similar comparative studies with mussels from other lakes or rivers.

The field collections and experiments were conducted under permission OKTF-KP/12363/2015 issued by the Hungarian National Inspectorate of Environment and Nature Protection.