Invasive parasites and global change: Evidence for the recent and rapid spillover of a potential pathogen of tilapias with a complex, three-host life cycle

Abstract

Biological invasions pose a serious threat to local flora and fauna and have negative impacts on ecosystems. Invasive parasites can also cause severe losses in aquaculture. In this article, we provide evidence of the recent spillover of an African parasite with a complex, three-host life cycle that has rapidly and successfully established itself in the Middle East, most likely due to the recent migration of its final hosts (great cormorant) from Africa. This case of parasite introduction into a country with intensive aquaculture is also important from an economic point of view, since large (up to 2 cm long) larvae of this parasite, the cyclophyllidean tapeworm Amirthalingamia macracantha (Cestoda) localised in the liver, can be pathogenic to their fish hosts, including farmed and wild fish, as shown by our histopathological examination of heavily infected fish. Since its first detection in Israel in November 2020, the parasite has spread rapidly and is currently found in both migratory (great cormorant, Phalacrocorax carbo) and non-migratory birds (pygmy cormorant, Microcarbo pygmaeus), as well as in fish intermediate hosts, including farmed tilapia in several farms in Israel and wild cichlids. There are numerous examples of the spillover of introduced parasites, including those that parasitise fish of commercial importance, but have a direct life cycle or use only a single intermediate host. Tilapines are the second most important group of farmed fish in the world after carps and are produced mainly in Southeast Asia, Central and South America. The global spread of great cormorants and the early evidence that pygmy cormorant may also harbour A. macracantha pose the risk of further spread of this invasive parasite to other countries and areas. In addition, global warming and reductions in foraging and resting areas near these waters may allow the parasite to complete its life cycle in new hosts.

1. Introduction

Current global change has multiple implications for the biota, including the functioning of ecosystems and the composition of local flora and fauna. One of the most important aspects of global change and ecosystem degradation is biological invasions, which have been accelerated by human population growth, increasing transport capacity, and economic globalisation [1]. An invasive species is an organism that is not indigenous, or native, to a particular area. Invasive species, including parasites, can cause great economic and environmental harm to the new area [2]. Co-introduced parasites are defined as those which have been transported with an alien host to a new locality, outside of their natural range, and co-invading parasites as those which have been co-introduced and then spread to new, native hosts [1]. Invasive parasites pose a serious problem for conservation biology, but also have practical implications for agriculture and aquaculture. Invasive parasites can pose a serious threat to native plant and animal communities and their health.

When alien hosts introduce new parasites, they can be transmitted to native hosts, leading to the emergence of new diseases in native hosts; this phenomenon is referred to as spillover or pathogen pollution [[3], [4], [5]]. Several examples of successful spillover of metazoan parasites have been documented, but the vast majority of these cases involved monoxenic parasites, i.e., parasites with simple (direct) life cycles that do not involve an intermediate host [6,7]. Most studies on biological invasion have focused on invasions of animals and plants in terrestrial environments, while there are few studies on aquatic environments [8].

In this report, we provide evidence of a recent, very rapid spillover of a potentially pathogenic helminth parasite. In 2020, larvae (metacestodes) of the cyclophyllidean tapeworm Amirthalingamia macracantha (Joyeux et Baer, 1935), a parasite of cormorants (Phalacrocoracidae) and tilapia in Africa [9,10], were found for the first time in tilapia hybrids at three farms in Israel [11]. Very soon after the first detection of this parasite, the first author and his collaborators found its larvae in other fish farms, as well as in wild tilapia in different waters in Israel. In addition, adult A. macracantha were found in migratory and resident cormorants, which serve as new definitive hosts of this invasive parasite. Considering the size of A. macracantha larvae and their occurrence in the liver of cultured tilapias and wild, often endangered, cichlids, the possible pathogenic effect of this parasite on its fish hosts in the newly colonised area was also investigated histopathologically.

2. Materials and methods

2.1. Fish sampling

In 2021, farmed tilapia (Oreochromis spp.) of market-size (200–900 g) from 20 farms in northern Israel (Fig. 1) were anesthetised and euthanised by immersion in ice water slurry under the supervision of an official veterinarian at the five sorting sites. Thirty randomly selected fish from each shipment were subjected to routine visual inspection, including macroscopic examination for the presence of zoonotic parasites in accordance with the procedure for pre-market inspection of locally grown food fish [12]. If no zoonotic parasites were found in the fish of a shipment, the fish were approved for marketing; if one or more parasites were found, an additional 30 randomly selected individuals were tested by public veterinarians. In total 5329 tonnes of market-sized tilapia (9690 fish; average weight 550 g) were tested in 2021. In addition, wild tilapia were collected from five sites in Israel (Fig. 1) and processed fresh under permit 2020/42686 from the Israel Nature and Parks Authority.

Fig. 1.

Northern Israel with sampling localities.

A total of 9963 fish of 7 species (3 species of farmed fish and 4 species of wild fish) were examined for parasites (see Table 1 for number and origin of fish examined). The body cavity of the fish was opened and the parasites encapsulated in the liver were isolated and placed in saline. All parasites found were preserved in 70% ethanol for morphological and molecular analyses [11]. Mean intensity (MI) and mean abundance (MA) were calculated according to Ref. [13].

Table 1.

Occurrence of larvae of Amirthalingamia macracantha (Joyeux et Baer, 1935) in tilapia farms monitored in 2021.

Farm	Tilapia species	Total weight of fish examined and marketed (tons)	No. of larvae collected	Prevalence (%)	Mean intensity	Mean abundance
A	Oreochromis niloticus × O. aureus	139	5	1.7–3.3	1	0.02–0.03
B	O. niloticus × O. aureus	143	3	3.3–6.7	1	0
C	O. niloticus × O. aureus	386	1	6.3	1	0.06
D	O. niloticus	162	111	1.2–83.3	1–8	0.01–0.83
E	O. niloticus	9	0	0	0	0
F	Oreochromis sp. O. niloticus × O. aureus	65	0	0	0	0
		61	0	0	0	0
G	Oreochromis sp.	661	1	1.7	1	0.02
H	Oreochromis sp. O. niloticus × O. aureus	88	0	0	0	0
		207	0	0	0	0
I	O. niloticus × O. aureus	175	0	0	0	0
J	O. niloticus × O. aureus	534	2	1.7–3.3	1	0.02–0.03
K	O. niloticus × O. aureus	360	0	0	0	0
L	O. niloticus × O. aureus	494	2	3.3	1	0.03
M	O. niloticus × O. aureus O. niloticus	448	0	0	0	0
		10	0	0	0	0
N	O. niloticus × O. aureus	663	0	0	0	0
O	O. niloticus × O. aureus	262	0	0	0	0
P	O. niloticus × O. aureus	343	0	0	0	0
Q	O. niloticus × O. aureus	117	0	0	0	0
R	O. niloticus × O. aureus	2	0	0	0	0
S	O. niloticus × O. aureus	2	0	0	0	0

2.2. Sampling of birds

In total 29 adults of A. macracantha were collected from three pygmy cormorants, Microcarbo pygmaeus (Pallas, 1773), from the area of Farms C and D (Fig. 1). An additional nine adults of A. macracantha were collected from 12 great cormorants, Phalacrocorax carbo (Linnaeus, 1758), from the area of Farms I, L, N, and O (Fig. 1). Cormorants were collected under permits 2020/42659 and 2021/42855 issued by the Israel Nature and Parks Authority. Tapeworms were preserved as described above.

2.3. Genotyping

Tissue samples (middle part of the body) from larval and adult cestodes were collected for DNA extraction, which was performed using a commercial kit (DNeasy Blood & Tissue, QIAGEN, Hilden, Germany) according to the manufacturer's instructions. A fragment of approximately 1500 base pairs from the large nuclear subunit of the ribosomal RNA gene (lsrDNA) and the cytochrome c oxidase subunit I (cox1) of approximately 550 base pairs was amplified using primers LSU5 and 1500R [14], and PBI-cox1F_PCR and PBI-cox1R_PCR [15], respectively. The resulting amplicons were resolved on a 1.5% agarose gel and sequenced (Hylabs, Israel) using the two primers mentioned above for the lsrDNA amplicons and the primer PBI-cox1F_seq for the cox1 amplicons. The resulting sequences were manually checked, aligned using MEGA 11 [16], and compared with the entries in GenBank at BLAST (https://blast.ncbi.nlm.nih.gov/Blast.cgi). Phylogenetic analysis of the resulting lsrDNA sequences in relation to selected published sequences of gryporhynchid tapeworms, was performed using the maximum likelihood model implemented in PhyML [17]. To assess confidence in the nodes, 1000 bootstrap replicates were performed.

2.4. Histopathological examination

Macroscopically affected hepatopancreas tissues from farmed fish, Nile tilapia, Oreochromis niloticus (Linnaeus, 1758) and whole specimens of wild fish, redbelly tilapia, Coptodon zillii (Gervais, 1848), were fixed in 10% buffered formalin. Hepatopancreas tissues and longitudinal sections of whole fish were dehydrated through a graded ethanol-xylene series and embedded in Paraplast®. Sections of 3 μm were first deparaffinized, rehydrated, then stained with Mayer's hematoxylin-eosin (H & E) and embedded in Eukitt® medium (Kaltek, Italy). The slides were observed with a Nikon H550L microscope at 40–1000×. Digital images were captured using a Nikon DS-Ri2 integrated camera and NIS Elements 5.30 software (Nikon, Japan).

3. Results

3.1. Occurrence of the parasite in fish intermediate hosts

In total 125 larvae of A. macracantha were found in the following cultured tilapia: 1 Oreochromis niloticus (intensity of infection 111 larvae), 1 Oreochromis sp. (1 larva), and 5 hybrids of Oreochromis niloticus × O. aureus (13 larvae; mean intensity of infection 2.6 larvae/infected fish) (Table 1).

In addition, larvae of Amirthalingamia macracantha were found in 12 wild cichlids of three species: 10 Redbelly tilapia, Coptodon zillii (Gervais, 1848), 1 Blue tilapia, Oreochromis aureus (Steindachner, 1864), and 1 Tvarnum simon, Tristramella simonis (Günther, 1864), with prevalence ranging from 12 to 65% (mean intensity 1.0–4.5 and mean abundance 0.12–1.59) (Table 2).

Table 2.

Wild cichlids sampled in 2021 (positive hosts in bold).

Location	Cichlid species	Sampling date	No. of fish infected per fish examined	No. of larvae collected	Prevalence (%)	Mean intensity	Mean abundance
Hod HaSharon park	Coptodon zillii	August	0/60	0	0	0	0
Sea of Galilee (Lake Kinneret)	C. zillii Sarotherodon galilaeus Tristramella simonis Oreochromis aureus	January–December	0/11	0	0	0	0
			0/92	0	0	0	0
			1/6	1	17	1	0.17
			0/29	0	0	0	0
Ein Afek	C. zillii S. galilaeus	May	1/25	2	4	2	0.08
			0/4	0	0	0	0
Einot Tzukim (Ein Feshkha)	C. zillii O. aureus	October	6/17	27	35	5	1.59
			1/3	1	33	1	0.33
Tel Saharon	C. zillii	July	3/26	6	12	1	0.12

3.2. Occurrence of the parasite in definitive hosts

A total of 40 adult tapeworms were found in two of three pygmy cormorants examined (prevalence 67%; mean intensity of infection 15 specimens/infected host; range 2–29 worms) and in five of 12 great cormorants (prevalence 42%; mean intensity of infection 1.8 specimens/infected host; range 1–4 worms). Detailed data on infection of cormorants at individual sites are presented in Table 3.

Table 3.

Fish-eating birds collected in 2021 (positive hosts in bold).

Location	Bird species	Sampling date	No. of birds infected per birds examined	No. of Amirthalingamia macracantha collected	Prevalence (%)	Mean intensity	Mean abundance
Farm D	Microcarbo pygmaeus	August	2/2	29	100	15	14.5
Farm C	M. pygmaeus	August	0/1	0	0	0	0
Farm O	Phalacrocorax carbo	November	1/3	2	33	2	0.7
Farm L	P. carbo	November–December	1/4	2	25	2	0.5
Farm N	P. carbo	November–December	2/2	4	100	2	2.0
Farm I	P. carbo	December	1/2	1	50	1	0.5
Farm T	P. carbo	December	0/1	0	0	0	0

3.3. Genotyping of larvae and adults

DNA amplification was successful for both targets in all samples tested, yielding good quality amplicons. Larval and adult worm sequences for each of the two targets were all identical, regardless of parasite developmental stage or host identity. The lsrDNA sequences of the new isolates from Israel were 99.9–100% identical to those of A. macracantha from two wild cichlids in Zimbabwe (southern Africa) and to those of the larva of an O. aureus × O. niloticus hybrid in Israel (accession number MW418305). The cox1 sequences were 84.5–87.0% identical to various cestodes, mainly taeniids, including Taenia twitchelli (accession number AB731759.1) and Taenia pisiformis (accession number MW350140.1). Representative lsrDNA and cox1 sequences from a single larva and a single adult worm were deposited in Genbank under accession numbers OP900557.1, OP900967.1 and OQ054802.1, OQ054987.1, respectively. Phylogenetic analysis (Fig. 2) confirmed that all new and previously published lsrDNA sequences of A. macracantha from adults and larvae belong to a particular node that is distinctly different from other gryporhynchids.

Fig. 2.

Rooted maximum-likelihood phylogenetic tree indicating the position of the lsrDNA sequence of larvae of Amirthalingamia macracantha (Joyeux et Baer, 1935) found in this study in relation to other gryporhynchids (Cestoda: Cyclophyllidea), based on a general time reversible and gamma-distributed rate heterogeneity (GTR_G) model of nucleotide substitution. Mesocestoides sp. and Dilepis undula were used as outgroups. Scale bar indicates estimated nucleotide substitutions. Only bootstrap values greater than 70% are shown.

3.4. Histopathological examination

In adult O. niloticus specimens, A. macracantha larvae were found encapsulated at the periphery of the hepatopancreas, resulting in a focal alteration of the normal architecture of the organ (Fig. 3A–D). Extensive fibrosis, oedematous degeneration of the connective tissue (capsule) surrounding the parasite, and mild inflammatory infiltration (mainly macrophages and lymphocytes) were observed (Fig. 3C). Abundant glycogen deposits with lipofuscin accumulation were present in the adjacent hepatocytes, which may be attributed to the high-energy diet during in the outgrowth phase. The final location of the parasite and fibrotic scarring on the surface of the hepatopancreas are consistent with larval migration from the gastrointestinal tract through the coelomic cavity into the liver parenchyma.

Fig. 3.

Histopathological observations of the hepatopancreas of farmed (Oreochromis niloticus from farm D; A–D) and wild tilapia (Coptodon zillii juvenile from Tel Saharon; E, F) in Israel infected with larvae (merocercoids) of the tapeworm Amirthalingamia macracantha (Joyeux et Baer, 1935). A, B, D – Encapsulation of coiled merocercoid localised on the surface of the hepatopancreas, enclosed by a granulomatous response (arrow) and edematous degeneration of hepatocytes surrounding the parasite. Moderate lipofuscin accumulation in the hepatopancreas and abundant glycogen storage. Merocercoid (arrowhead) with irregularly arranged annular folds (H & E). C – Detail of the encapsulated merocercoid; multiple calcareous corpuscles (arrowheads) are scattered within the mesenchyme of the larva, which shows a 20 μm-thick, eosinophilic, smooth surfaced tegument. Mild inflammatory infiltrate, mostly macrophages and lymphocytes and fibroblast proliferation compose the granulomatous host response (capsule, arrow). Hydropic degeneration of adjacent hepatocytes (*); moderate lipofuscin accumulation and abundant glycogen storage of peripheral hepatocytes (H & E). E, F – Alteration of the normal architecture of the hepatopancreas caused by encapsulation of a single merocercoid with extensive oedematous degeneration of connective tissue (capsule) surrounding the parasite (arrows). Mild inflammatory infiltration, mostly macrophages and lymphocytes. Merocercoid (arrowhead) presents no segmentation but evident annular folds (H & E).

In juvenile C. zillii collected in the wild, extensive alteration of the architecture of the hepatopancreas was caused by migration of the parasite and subsequent encapsulation of a single metacestode larva that lacked segmentation but exhibited distinct annular folds (Fig. 3E and F). Fibrotic and oedematous connective tissue (capsule) surrounding the parasite accounts for nearly half of the hepatopancreatic volume (Fig. 3F).

4. Discussion

We live in a changing, globalised world where the negative impacts of human activities, such as the extensive exploitation of non-renewable resources and the drastic decline of biodiversity associated with the destruction of ecosystems, are rapidly increasing. The distribution of species is changing at an unprecedented rate due to human activities, resulting in habitat changes, climate change, biological invasions, and emerging diseases. Invasive parasites may also play an important role in these often irreversible processes, and there is strong evidence of their negative impacts on natural ecosystems and human activities [18].

The success of invasive species, including parasites, depends on a variety of abiotic and biotic factors, particularly their transmission patterns, life cycles, and degree of host specificity [19,20]. Overall, generalists (parasites with a broad host range, i.e., a lower degree of host specificity) have a greater chance of successful introduction than specialists (more specific parasites) because they tend to be more resilient to environmental change and may be better competitors than native parasites [[21], [22], [23]]. The life cycle of the parasite may also influence its establishment, as monoxenic parasites are more adaptable than parasites with complex life cycles [5,19,24].

The present report provides direct evidence of the successful and exceptionally rapid spread of a parasite with a complex three-host life cycle, the tapeworm A. macracantha, originally found only in Africa [9,10,25]. This case of parasite spillover is exceptional because the life cycle of this tapeworm involves two intermediate hosts that live in freshwaters and a final host that eats fish (Fig. 4). The presence of metacestodes in a variety of native fish intermediate hosts and adults in two cormorant species is evidence that this invasive parasite has quickly and successfully established itself in a region far from its original range.

Fig. 4.

Life cycle of Amirthalingamia macracantha (Joyeux et Baer, 1935). Tapeworms sexually mature in cormorants (above) and shed eggs into the water. Planktonic crustaceans (copepods) become infected after eating eggs floating in the water; larvae called procercoids develop in copepod body cavity (bottom right). Second intermediate hosts, tilapias, become infected after eating copepods with procercoids; larvae called merocercoids develop in the hepatopancreas of infected fish. Courtesy of Mona Luo.

Such rapid and successful establishment of a new invasive parasite with an aquatic life cycle is remarkable, especially in Israel, a relatively arid to semi-desert country with a limited number and area of water bodies where aquatic parasites can complete their developmental cycle. In particular, the parasite has adapted very rapidly to a local and non-migratory final host, the pygmy cormorant, whose range does not include Africa (Fig. 5, Fig. 6); this new final host contributes to the dispersal of parasite eggs in the newly colonised region. The increased population density and more intense migration of piscivorous birds may itself pose a serious threat to local fish stocks, including farmed tilapia [26].

Fig. 5.

Distribution of cormorants in the Middle East. Courtesy of eBird data.

Fig. 6.

Global distribution of cormorants. Courtesy of eBird data.

The alien parasite must successfully undergo the processes of introduction, establishment, and dispersal with its original host and then switch to a native host species to become a co-invador, i.e., a parasite that was co-introduced and then switches to native hosts in the new region [1]. However, in the present case, the parasite was most probably introduced by a native, i.e., non-invasive, host, great cormorants. They occurred naturally in the Middle East before the tapeworm A. macracantha was introduced. It seems most plausible that this parasite was previously repeatedly introduced to the Middle East as an adult worm with its final hosts, birds, and that gravid tapeworms released their eggs into local waters. However, the cycle of the alien parasite could not be completed, as indicated by the absence of records of its occurrence in Israel in previous parasitological surveys. Given the size of parasite larvae in fish and the intensive research on parasites of farmed and wild fish in recent years, it is unlikely that the parasite was overlooked in previous parasitological surveys in the Middle East [[27], [28], [29], [30], [31]]. More recently, more favourable conditions, i.e., the availability of susceptible second intermediate hosts (cichlids; Fig. 4), have allowed the successful establishment of the parasite and native fish acquired an alien parasite. Potential first intermediate hosts, planktonic crustaceans (Copepoda) [32,33], are available in water bodies everywhere.

Goedknegt et al. (2016) [5] listed 35 co-introduced marine parasite species reported to spill over to native host species; nearly two-thirds of these were macroparasites, particularly Monogenea and Copepoda. The majority of parasite species (74%) have a direct life cycle, making invasion more likely because these parasites do not rely on the presence of all hosts in their life cycle to become established [34]. Of the nine parasites with indirect life cycles, only one species has a three-host cycle: the diphyllobothriid tapeworm Ligula intestinalis uses copepods and fish as intermediate hosts and piscivorous birds as definitive hosts [5,35].

In the case of A. macracantha, the key factor for the spread of the species seems to have been the availability of cichlids in the new region. Cichlids have been reported as second intermediate hosts of A. macracantha in Africa (Democratic Republic of Congo, Kenya, Zimbabwe), namely C. zillii, Pharyngochromis acuticeps (Steindachner, 1866), Pseudocrenilabrus philander (Weber, 1897), and Tilapia sparrmanii Smith, 1840 [10]. The present data show that the parasite adapted to other cichlids of two genera (Oreochromis and Tristramella) as well as to farmed tilapia within a very short time.

Another favourable circumstance for the successful colonisation of the new region by A. macracantha was the susceptibility of local, non-migratory populations of pygmy cormorant (Fig. 5), as they contribute to the greater dispersal of parasite eggs in the waters of Israel. The ecological and evolutionary significance of an invasive parasite successfully spilling over on a naive native species lies in its high dispersal potential in the introduced range and selection for lower host specificity [5]. The current case of fast spillover of A. macracantha in Israel fully supports this assumption.

Regarding the potential impact of the newly established and spreading parasite on its fish host, it is well known that cestode larvae can be harmful to fish, especially those that migrate through internal organs of their intermediate hosts [[36], [37], [38], [39]]. This deleterious interaction often triggers an inflammatory response associated with cell proliferation leading to fibrosis [37]. The pathogenicity of some tapeworm larvae also depends on host defence mechanisms [38], with fish that elicit a strong cellular encapsulation response having lower mortality, while hosts with weak or no cellular response have more lesions and higher mortality [40,41]. The lack of a strong cellular response in the liver of cichlids infected with larvae of A. macracantha indicates the higher pathogenicity of this parasite.

Control of cestode larvae is more problematic than that of adults. Therefore, prevention is critical to avoid losses, especially when new pathogens are introduced with imported fish [19,42]. Preventive measures should include regular parasitological examinations of farmed and ornamental fish, especially for fish to be used in different regions [43].

Central to managing invasive species is risk assessment to identify high-risk scenarios, target resources accordingly, and apply appropriate measures to mitigate the negative impacts of parasites on their non-native hosts (see Fig. 2 in Ref. [18]). However, these measures are generally ineffective when the invasive parasite escapes into wild populations of new intermediate and definitive hosts, which is also the case for A. macracantha in Israel. It is true that this parasite can be easily detected by non-specialists during routine parasitological examination of fish, but parasite detection can be complicated by their low prevalence (usually <5%). Therefore, protocols must be adapted to ensure the effectiveness of quarantine measures.

5. Limitations of the study

This study has few limitations because it presents empirical data. It is unclear if the invasive parasite studied is unique in its ability to spread so rapidly or if its invasive success is related to current climate change. Another limitation is the absence of data on the specific first intermediate hosts (planktonic crustaceans) of the parasite in Israel.

Funding

This study was supported by the Israeli Veterinary Service and the Institute of Parasitology, BC CAS (RVO: 60077344).

Author contribution statement

Nadav Davidovich: Conceived and designed the experiments; Analyzed and interpreted the data; Contributed reagents, materials, analysis tools or data; Wrote the paper.

Daniel Yasur-Landau; Adi Behar: Performed the experiments; Analyzed and interpreted the data.

Tobia Pretto: Performed the experiments; Contributed reagents, materials, analysis tools or data; Wrote the paper.

Tomas Scholz: Analyzed and interpreted the data; Contributed reagents, materials, analysis tools or data; Wrote the paper.

Data availability statement

Data included in article/supplementary material/referenced in article.

Additional information

No additional information is available for this paper.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.