Diatom Epibionts on Amphipod Crustaceans: A Possible Vector for Co-introductions?

Abstract

Epibiotic associations can result in co-introductions of non-indigenous species, which may affect ecosystems in several ways. In fouling communities of three estuaries in southern Brazil, a number of amphipods was found to harbour a dense coverage of epibionts. Three different species, the two globally widespread caprellids Caprella equilibra and Paracaprella pusilla, as well as the ischyrocerid Jassa valida, had been colonised by diatoms. Further scanning electron microscope analyses assigned these diatoms to 14 different species that had previously been reported from benthic habitats. This is one of the scarce records of diatoms attached to amphipods. The occurrence of the diatom Amphora helenensis represents the first report for Brazilian waters as well as the second record for the whole SW Atlantic Ocean. As some diatoms were associated with common fouling amphipods, a possible regional spread aided by these crustaceans seems likely. Possible effects of this amphipod-diatom association on the animals and their implications for the underlying ecosystems of this remain to be elucidated.

1. Introduction

Human-based transport mechanisms of non-indigenous species (NIS) are often investigated, but processes of co-introduction through associations like parasitism or epibiosis are still understudied [1,2]. These associations allow for simultaneous multiple introductions that can consequently affect ecosystems on several levels, and may further facilitate secondary invasions (i.e., invasion meltdown hypothesis) [3]. For example, the canopy-forming marine macroalga Sargassum muticum (Yendo) Fensholt, 1955, was probably introduced to North American and European shores with the Pacific oyster Magallana gigas (Thunberg, 1793) (formerly known as Crassostrea gigas), which was intentionally introduced for aquaculture purposes. Moreover, oyster trade is the presumed vector for the spread of different NIS around the world, including different macroalgae, ascidians, and even amphipods [4,5].

Marine ecosystems face an increased risk of the spread of NIS because even geographically distant locations are virtually connected through the oceans with intensive human shipping activities. Shipping facilitates the dispersal of species through hull fouling, ballast water, and other vessel components, e.g., [6,7]. Besides, ports meet most of the requirements for the successful introduction and establishment of NIS. Ports are subjected to continuous pressure of propagules in usually eutrophic environments and are modified by artificial structures where native species can lose their potential competitive advantage [8,9]. Hard substrates in ports are usually characterised by fouling communities consisting of r-strategists, filter feeders, and scavengers [10]. These communities are known to be comparably rich in NIS, which frequently outcompete native species in the colonisation of these substrates [11].

Amphipod crustaceans can dominate marine and estuarine fouling communities in terms of abundance [12,13]. Amphipod fouling assemblages are mainly characterised by either tubiculous or free-living suspension feeders, but different combinations of functional traits are also common including grazers, carnivores, and epibionts associated to other organisms [14]. Common tube constructors such as ischyrocerids (e.g., Jassa, Ericthonius), different corophiids (e.g., Monocorophium, Corophium, Apocorophium), aorids (e.g., Grandidierella, Microdeutopus), and photids (e.g., Photis, Gammaropsis) can dominate these communities along with crawling/clinging taxa such as caprellids (e.g., Caprella, Paracaprella), stenothoids (e.g., Stenothoe), and podocerids (e.g., Podocerus). Moreover, at least 55 species of Amphipoda are recognised as introduced or invasive in at least one marine region around the world, of which more than half are directly associated with biofouling [15]. Many species are recognised as neocosmopolitan (i.e., human-mediated cosmopolitan species; see [16]) as well as widespread in invaded regions [17,18,19,20]. Some amphipods are also known to bear different epibionts that, in some cases, would benefit from their basibiont movement without harming the host [21,22,23,24]. Accordingly, amphipods might be effective vectors for some organisms that are directly associated with them.

In a field experiment designed to study fouling communities related to ports, artificial plates of polyethene and rope (nylon rope wrapped around a squared iron frame) were deployed, at an average depth of 1.5 m, in three estuaries of southern Brazil: the Cananéia-Iguape estuarine complex, the Paranaguá estuarine complex, and the Guaratuba Bay. All investigations were carried out between June 2017 and June 2018. After three months, the plates were collected and separately preserved in 96% ethanol. At the site of the experimental set-up, additional samples of the fouling were scraped from buoys and preserved in ethanol for further comparison. All amphipod individuals were sorted and identified to species level. During this step, some amphipods displayed conspicuous levels of epibiosis and were separated for further examination with a scanning electron microscope (SEM). These specimens were mounted on aluminium stubs with a small piece of adhesive carbon tape, completely air-dried at room temperature, and kept in an air desiccator jar. The stubs were then sputter-coated with gold in a Baltec metallizer and observed in a JEOL JSM 6360-LV electron microscope at an acceleration voltage of 15 Kv and 7 mm working distance (Figure 1A). As no further frustule cleaning with chemicals could be applied, many diatom cells from different samples (i.e., amphipod specimens) were photographed, measured, and preliminary separated into morphotypes to ease identification. The definite identification was then determined by carefully comparing their frustule morphologies and dimensions.

2. Results & Discussion

A total of 14 different diatom species belonging to seven families was identified (

Table 1. Morphometric data of the 14 epibiotic diatom species recorded on the amphipods, sampled in estuaries of South Brazil.

Species	Length/Breath (µm)	Striae Pattern	N° Striae in 10 µm	Other Features
Amphora helenensis Giffen, 1973	22–16/2.5–2.2	radiate	19–22	apices rounded
Amphora ovalis var. pediculus (Kutzing) Van Heurck, 1885	16–6/2.2–2.0	parallel to radiate	24–28	apices rounded
Halamphora coffaeiformis (C. A. Agardh) Levikov, 2009	25–11	radiate	20–22	apices subcapitate
Cocconeis dirupta Gregory, 1857	16–12/9–6	parallel	16	striae with 2 rows of poroids
Cocconeis sp. 1	14–12/5	slightly radiate	28	striae with 1 row of poroids
Licmophora cf. abbreviata C. Agardh, 1831	75–40/18–16	parallel	11–12	12–14 slits/multiscissura
Licmophora ehrenbergii (Kützing) Grunow, 1867	75–40/18–16	parallel	12–14	14–16 slits/multiscissura
Melosira moniliformis (O. Müller) Agardh, 1824	22–20 (diameter)	radial		central umbillicus
Navicula platyventris Meister, 1934 (1935)	14	radiate	7–5	5–7 areolae/striae
Navicula sp. 1	22–20/5–4	parallel	18	8–9 areolae/stria
Navicula sp. 2	12–10/2.5	parallel	16–18	3–4 areolae/stria
Rhoicosphenia marina (Kutzing) M.Schmidt, 1889	20–13/8–6	radiate	18–20	raphe incipient in one valve
Tabularia affinis (Kutzing) Snoeijs, 1992	52–32/4.0–3.5	parallel	22	biseriate striae
Tabularia fasciculata (C. Agardh) D.M.Williams & Round, 1986	80–40/5.5–4.5	parallel	15–18	uniseriate striae

). Most of the taxa recorded fell into frustule size ranges reported in the literature, except for two species, which presented smaller valves than previously reported. Cocconeis dirupta Gregory, 1857 valves were shorter (length 16–12 μm, Table 1) than elsewhere (e.g., length 24–17 μm in [25]). The same deviation was found in Halamphora coffaeiformis (C. Agardh) Mereschkowsky, 1903, in which valves were 25–11 μm long, but usually ranging from 30 μm to 53 μm, even when our material is compared to another variety of this species, A. coffeaeformis var. aponina (Kützing) Archibald & Schoeman, 1984 (length 39–15 μm in [26]). The families Catenulaceae and Naviculaceae were the most species-rich with three species each. The diatoms found are typically benthic species, i.e., attached to substrates employing a variety of mucilage secretions. The most common structures were mucilaginous peduncles in Tabularia spp. and Licmophora spp. (Figure 1I and Figure 2B), along with mucilage pads as in Amphora helenensis M. H. Giffen, 1973, H. coffaeiformis, and Cocconeis spp. (Figure 1E,G and Figure 2I). Other species, such as Navicula spp., seemed to attach to amphipod individuals using mucilaginous secretions of the raphe (Figure 2L,O). Although no quantitative analysis was performed, some diatom taxa were dominant such as Tabularia affinis (Kützing) Snoeijs, 1992, Licmophora spp., H. coffaeiformis, and A. helenensis, comprising about 50% of the total diatom composition in the majority of analysed amphipod individuals. While some diatoms are known to be specialised epizoic on planktonic copepods [27], only a few studies have dealt with the growth of diatoms on benthic invertebrates. Usually diatoms of the genera Navicula, Amphora, and Cocconeis are known to occur on several species of cnidarians [28], sponges [29], and gastropods [30,31]. Despite some species being known to display specificity to some taxa, all the diatoms found here have been reported elsewhere from other substrates such as macroalgae and rocks [32,33], indicating that these species are rather opportunistic to grow on invertebrates. The fouling communities of the artificial plates inhabited by the amphipods were characterised by dense assemblages of hydrozoans (e.g., Clytia spp., Ectopleura crocea (Agassiz, 1862), Obelia spp., and Eudendrium carneum Clarke, 1882) and barnacles (e.g., Megabalanus coccopoma Darwin, 1854, Amphibalanus amphitrite (Darwin, 1854)), which are both recognised basibionts of either diatoms or amphipods, e.g., [34,35,36,37,38], which would make them the most probable vector of diatoms for the amphipods. Concomitantly, species inhabiting soft bottoms have been reported to host relatively few epibionts in comparison to hard-substrate species [39].

Of the 15 analysed amphipod specimens, twelve individuals bore epibiotic diatoms (

Table 2. Overview of amphipod and diatom species recorded for each analysed sample. Substrates: PP—polyethene plate; B—buoy; RP—rope plate.

Estuary	Substrate	Lat S	Long W	Amphipod Species	Sex	Amphora helenensis	Amphora pediculus	Halamphora coffaeiformis	Cocconeis sp.1	Cocconeis dirupta	Licmophora cf. abbreviata	Licmophora ehrenbergii	Melosira moniliformis	Navicula platyventris	Navicula sp.1	Navicula sp.2	Rhoicosphenia marina	Tabularia affinis	Tabularia fasciculata
Cananéia-Iguape estuarine complex	PP	25.021	47.918	Paracaprella pusilla	Male			x	x							x
		25.027	47.914		Juvenile											x
Paranaguá estuarine complex	B	25.467	48.663	Paracaprella pusilla	Male					x	x	x					x
	PP	25.473	48.649	Caprella equilibra	Female					x								x	x
	B	25.497	48.491	Caprella equilibra	Male						x	x		x	x
	RP	25.549	48.389	Jassa valida	Juvenile	x		x										x
					Male			x
				Paracaprella pusilla	Male			x							x			x
	PP	25.549	48.388	Paracaprella pusilla	Female	x	x	x		x
					Male	x	x	x	x	x				x				x
Guaratuba Bay	PP	25.875	48.589	Caprella equilibra	Juvenile	x	x						x
					Juvenile			x

). The amphipod specimens belonged to only three species: the two globally distributed caprellids Caprella equilibra Say, 1818, and Paracaprella pusilla Mayer, 1890, and the ischyrocerid Jassa valida (Dana, 1853). On a single specimen of Ericthonius brasiliensis (Dana 1853), as well as on other specimens of J. valida and P. pusilla, non-identified apostomid ciliates were found, but no diatoms. The highest diversity of epibionts was found on a male specimen of P. pusilla with seven different diatom species on its cuticula (Table 2). Three diatom species were each found only once on caprellids: Melosira moniliformis C. Agardh, 1824, Rhoicosphenia marina (Kützing) M.Schmidt, 1889, and Tabularia fasciculata (C. Agardh) D. M. Williams & Round, 1986. Although males generally seemed to display higher epibiont diversity compared to females, only two female individuals were available for analysis. Possible sex-specific differences in epibiotic colonisation may relate to differences in the moulting cycle. However, due to the low replicate number, any assumptions must be regarded with caution. The diatoms were distributed over the whole body of the animals, but the appendages seemed to exhibit a higher coverage with epibionts than the rest of the body. This appears to be counterintuitive as antennas and pereopods are easily and frequently cleaned by the animals, suggesting that the amphipods might be unaffected by the diatoms as reported for epizoic ciliates on the amphipod Bathyporeia spp. [40]. This is in contrast to other crustaceans where epibionts have been shown to cause stress to the crustacean host individuals [41]. In addition, the possibility that the amphipods may even benefit from epibiotic diatoms as an additional food source cannot be ruled out as they are known to be omnivores and diatoms were frequently found in amphipod digestive tracts [42].

Nevertheless, potential interactions of diatoms with their amphipod hosts (e.g., commensality) are completely unknown. Although it could be possible that high diatom densities on fragile body parts such as the gills or sensory appendages (Figure 1C,H) may lead to handicaps (e.g., interferences in gas exchange, reproduction, or sensorial activity) [43,44], the amphipods and the diatoms seemed to display a basibiont-epibiont association.

The biraphid diatom Amphora helenensis found in this work represents the first on record for Brazilian coasts. This species has been recorded repeatedly around the world outside of its type locality (South Africa). Recent records include the coastal waters of Australia and New Zealand [45], Europe (The Netherlands [46]), China (Yellow Sea [33]), and Mexico (Pacific Ocean [47]). We found A. helenensis in the Guaratuba Bay and in the Paranaguá estuarine complex. The latter site is well-studied and has been subject to extensive taxonomic diatom studies since 1960 (see [48]). This suggests that A. helenensis was introduced recently to South Brazil. However, it cannot be ruled out that this species had been overlooked or misidentified with other closely related species at other locations in the SW Atlantic, where it was recorded only once in Argentina [49]. Therefore, A. helenensis can be classified as cryptogenic.

The presence of diatoms on amphipods might not be a newly recorded association, although it was never further investigated and only briefly mentioned. Previous records on species of the genera Leucothoe and Caprella are documented, either in the descriptions of amphipods or unintentionally photographed (e.g., [50], see Figures 5 and 6 in [51]). As benthic amphipods can be associated with other sessile fauna (like in this case) or macroalgae [52], they are susceptible for colonisation by motile diatom species (e.g., Navicula spp.) or even sedentary species (e.g., Tabularia spp., Licmophora spp.) if these are abundant on the surface of the shared host. If the new basibiont is abundant (especially as part of a fouling community), this may result in a rapid spread of the diatom. In fact, regional boating is known to aid the spread of NIS among marinas and ports by allocating the whole communities on the ship hulls [53,54]. Diatoms may thus not only benefit from direct human-mediated transport, but also from the active movement of basibiont amphipods to new natural or artificial substrates as well as from nutrients provided by the host.

3. Conclusions

Our findings represent the first documented analysis of amphipod-associated epibiotic diatoms. In the current case, common benthic diatom species were found on the cuticle of three widespread amphipods, suggesting a rather opportunistic epibiosis due to the epifouling lifestyle of these crustaceans. Especially in fouling communities, amphipods can be potent vectors for epibiotic organisms such as diatoms and ciliates (e.g., [22]) aiding the spread of newly introduced epibionts to neighbouring habitats.