Species recorded in penguins

Five species of Plasmodium have been demonstrated to infect penguins through both morphological and genetic evidence: Plasmodium (Haemamoeba) relictum (Fantham and Porter, Reference Fantham and Porter1944), Plasmodium (Huffia) elongatum (Huff and Shiroishi, Reference Huff and Shiroishi1962), Plasmodium (Haemamoeba) tejerai (Silveira et al. Reference Silveira, Belo, Lacorte, Kolesnikovas, Vanstreels, Steindel, Catão-Dias, Valkiūnas and Braga2013), Plasmodium (Haemamoeba) cathemerium and Plasmodium (Novyella) nucleophilum (Vanstreels et al. Reference Vanstreels, Silva-Filho, Kolesnikovas, Bhering, Ruoppolo, Epiphanio, Amaku, Ferreira-Junior, Braga and Catão-Dias2015a). Additionally, Plasmodium (Novyella) unalis was identified through genetic evidence (Vanstreels et al. Reference Vanstreels, Silva-Filho, Kolesnikovas, Bhering, Ruoppolo, Epiphanio, Amaku, Ferreira-Junior, Braga and Catão-Dias2015a). There is a record of Plasmodium (Bennettinia) juxtanucleare infecting penguins (Grim et al. Reference Grim, van der Merwe, Sullivan, Parsons, McCutchan and Cranfield2003), however that report is problematic (see Appendix 2).

Because a number of studies have not conclusively identified all Plasmodium lineages that were detected (see Appendix 1), it is probable that many other species of Plasmodium have yet to be recorded in penguins. Additionally, concomitant infection by two or more Plasmodium lineages is not uncommon (Huff and Shiroishi, Reference Huff and Shiroishi1962; Fleischman et al. Reference Fleischman, Squire, Sladen and Moore1968a, Reference Fleischman, Squire, Sladen and Melbyb; Herman et al. Reference Herman, Gray, Knisley and Kocan1974; Sladen et al. Reference Sladen, Gailey-Phipps and Divers1979; Stoskopf and Beier, Reference Stoskopf and Beier1979; Beier and Stoskopf, Reference Beier and Stoskopf1980; Beier and Trpis, Reference Beier and Trpis1981; Vanstreels et al. Reference Vanstreels, Kolesnikovas, Sandri, Silveira, Belo, Ferreira-Junior, Epiphanio, Steindel, Braga and Catão-Dias2014a, Reference Vanstreels, Silva-Filho, Kolesnikovas, Bhering, Ruoppolo, Epiphanio, Amaku, Ferreira-Junior, Braga and Catão-Dias2015a, Reference Vanstreels, Capellino, Silveira, Braga, Rodríguez-Heredia, Loureiro and Catão-Diasin press).

Distribution among penguin hosts

Thirteen species have been shown to be susceptible to Plasmodium in the wild or in captivity: king (Scott, Reference Scott1927), Humboldt (Rodhain, Reference Rodhain1939), African, northern rockhopper, yellow-eyed (Fantham and Porter, Reference Fantham and Porter1944), Snares (Laird, Reference Laird1950) (see Appendix 2), chinstrap (Rodhain and Andrianne, Reference Rodhain and Andrianne1952), little, gentoo (Griner and Sheridan, Reference Griner and Sheridan1967), Macaroni (Herman et al. Reference Herman, Gray, Knisley and Kocan1974), Magellanic (Fix et al. Reference Fix, Waterhouse, Greiner and Stoskopf1988), Galapagos (Levin et al. Reference Levin, Outlaw, Vargas and Parker2009) and Southern rockhopper penguins (Dinhopl et al. Reference Dinhopl, Mostegl, Richter, Nedorost, Maderner, Fragner and Weissenböck2011). There are only five penguin species in which Plasmodium infection was never documented: Adélie, emperor, erect-crested, Fiordland and Royal penguins. These species are inhabitants of remote sub-Antarctic and Antarctic environments, were seldom examined for blood parasites in the wild and were either never maintained in captivity or are generally maintained in vector-free acclimatized enclosures. It is therefore probable that the lack of records in these species is not due to a particular resilience to these parasites, but instead reflects a lack of studies in the wild and/or lack of exposure to environmental conditions that might allow their infection ex situ.

Invertebrate hosts

It is well established that avian plasmodia are transmitted exclusively by mosquitoes (Culicidae), particularly Culex spp., Mansonia spp., Culiseta spp. and Aedeomyia spp. Additionally, Aedes spp., Anopheles spp. and Armigeres spp. can also be competent hosts in laboratory experiments (Valkiūnas, Reference Valkiūnas2005; Atkinson, Reference Atkinson, Atkinson, Thomas and Hunter2008b).

Studies in zoos indicate that Culex spp. play a key role in the transmission of Plasmodium sp. to captive penguins, particularly C. pipiens (Rodhain, Reference Rodhain1939; Raethel, Reference Raethel1960; Grünberg and Kutzer, Reference Grünberg and Kutzer1963; Beier and Trpis, Reference Beier and Trpis1981), Culex quinquefasciatus (=Culex fatigans) (Laird and Van Riper, Reference Laird, Van Riper and Canning1981), Culex tarsalis (Huff and Shiroishi, Reference Huff and Shiroishi1962), Culex restuans (Beier and Trpis, Reference Beier and Trpis1981) and C. (Culex) sp. (Bueno et al. Reference Bueno, Lopez, Menezes, Costa-Nascimento, Lima, Araújo, Guida and Kirchgatter2010). Fantham and Porter (Reference Fantham and Porter1944) found Plasmodium sp. in C. quinquefasciatus collected at Saldanha Bay, South Africa, where a wild penguin was found dead with a P. relictum infection. The invasive C. quinquefasciatus is likely involved in the transmission of Plasmodium in New Zealand and in the Galapagos Archipelago (Tompkins and Gleeson, Reference Tompkins and Gleeson2006; Levin et al. Reference Levin, Outlaw, Vargas and Parker2009; Levin and Parker, Reference Levin, Parker, Miller and Fowler2011). In the Galapagos Archipelago, however, Aedes taeniorhynchus still has to be investigated as a potential host since this species has been recorded in Mexico carrying Plasmodium lineages closely related to those identified in Galapagos penguins (Levin et al. Reference Levin, Zwiers, Deem, Geest, Higashiguchi, Iezhova, Jiménez-Uzcátegui, Kim, Morton, Perlut, Renfrew, Sari, Valkiūnas and Parker2013). In New Zealand, C. quinquefasciatus is restricted to the North Island (White, Reference White1989; Holder, Reference Holder1999) and Culex pervigilans is suspected to be responsible for the transmission of Plasmodium sp. at South Island and other islands (Holder, Reference Holder1999; Sturrock and Tompkins, Reference Sturrock and Tompkins2008). Ochlerotatus australis (=Aedes australis) is an invasive species in New Zealand (Holder, Reference Holder1999; Snell, Reference Snell2005) that could also play a role in the transmission of avian malarial parasites.

Geographic distribution

Plasmodium relictum was documented in wild penguins in South Africa (Saldanha Bay), New Zealand (Campbell Island, Fouveaux Strait, Tiritiri Matangi Island, Snares Island) and Gough Island (Fantham and Porter, Reference Fantham and Porter1944; Laird, Reference Laird1950), and in captive penguins in Europe (Rodhain, Reference Rodhain1939; Fantham and Porter, Reference Fantham and Porter1944), North America (Griner and Sheridan, Reference Griner and Sheridan1967; Stoskopf and Beier, Reference Stoskopf and Beier1979), Hawaii (Laird and Van Riper, Reference Laird, Van Riper and Canning1981), Eastern Asia (Bak et al. Reference Bak, Park and Lim1984), South Africa (Penrith et al. Reference Penrith, Huchzermeyer, Wet and Penrith1994) and at rehabilitation centres in South Africa (Brossy et al. Reference Brossy, Plös, Blackbeard and Kline1999) and Chile (Carvajal and Alvarado, Reference Carvajal and Alvarado2009). Plasmodium elongatum was documented infecting penguins at zoos in North America (Huff and Shiroishi, Reference Huff and Shiroishi1962; Beier and Stoskopf, Reference Beier and Stoskopf1980), Europe (Dinhopl et al. Reference Dinhopl, Mostegl, Richter, Nedorost, Maderner, Fragner and Weissenböck2011) and rehabilitation centres in Brazil (Vanstreels et al. Reference Vanstreels, Kolesnikovas, Sandri, Silveira, Belo, Ferreira-Junior, Epiphanio, Steindel, Braga and Catão-Dias2014a, Reference Vanstreels, Silva-Filho, Kolesnikovas, Bhering, Ruoppolo, Epiphanio, Amaku, Ferreira-Junior, Braga and Catão-Dias2015a). Plasmodium tejerai was identified in penguins undergoing rehabilitation in Brazil (Silveira et al. Reference Silveira, Belo, Lacorte, Kolesnikovas, Vanstreels, Steindel, Catão-Dias, Valkiūnas and Braga2013; Vanstreels et al. Reference Vanstreels, Kolesnikovas, Sandri, Silveira, Belo, Ferreira-Junior, Epiphanio, Steindel, Braga and Catão-Dias2014a, Reference Vanstreels, Silva-Filho, Kolesnikovas, Bhering, Ruoppolo, Epiphanio, Amaku, Ferreira-Junior, Braga and Catão-Dias2015a) and Argentina (Vanstreels et al. Reference Vanstreels, Capellino, Silveira, Braga, Rodríguez-Heredia, Loureiro and Catão-Diasin press). Plasmodium cathemerium, P. nucleophilum and P. unalis were reported only in penguins undergoing rehabilitation in Brazil (Vanstreels et al. Reference Vanstreels, Silva-Filho, Kolesnikovas, Bhering, Ruoppolo, Epiphanio, Amaku, Ferreira-Junior, Braga and Catão-Dias2015a). Additionally, unidentified lineages of Plasmodium have been detected in wild penguins at the Galapagos Archipelago (Isabela, Fernandina, Las Marielas and Bartolomé Islands) (Levin et al. Reference Levin, Outlaw, Vargas and Parker2009, Reference Levin, Zwiers, Deem, Geest, Higashiguchi, Iezhova, Jiménez-Uzcátegui, Kim, Morton, Perlut, Renfrew, Sari, Valkiūnas and Parker2013), as well as at zoos in North and South America, Europe and Asia, and at rehabilitation centres in South Africa, Argentina and Brazil (see Appendix 1).

All ex situ and most in situ records of Plasmodium spp. infecting penguins are within the distribution range of C. pipiens, C. quinquefasciatus and C. pervigilans or, in Southern New Zealand, O. australis (Fig. 2C). The record of P. relictum in a Northern rockhopper penguin at Gough Island (Fantham and Porter, Reference Fantham and Porter1944) is a surprising exception, considering this is an extremely remote island where mosquitoes are absent (Gaston et al. Reference Gaston, Jones, Hänel and Chown2003); the only neighbouring archipelago, Tristan da Cunha, is also mosquito-free (Medlock et al. Reference Medlock, Schaffner and Fontenille2010). A possible explanation is that this penguin was exposed to P. relictum while being vagrant in South Africa (see Rollinson et al. Reference Rollinson, Reynolds and Paijmans2013).

The geographic distribution of Culex mosquitoes overlaps with the breeding habitat of penguins in Peru, Chile, Namibia and Australia (Fig. 2C), and it is therefore plausible that wild penguins in these countries may be infected by Plasmodium sp. Most sub-Antarctic islands probably do not provide environmental conditions compatible with the transmission of Plasmodium spp. Gough, South Georgia, Marion, Macquarie and Tristan da Cunha Islands are reportedly free from mosquitoes (Laird, Reference Laird1952; Hänel et al. Reference Hänel, Chown and Davies1998; Medlock et al. Reference Medlock, Schaffner and Fontenille2010) and the climate of South Georgia, South Sandwich, Bouvet, Amsterdam, Saint Paul, Crozet, Kerguelen and Peter I Islands is likely too adverse for mosquitoes (Medlock et al. Reference Medlock, Schaffner and Fontenille2010). It is therefore reasonable to assume that there is no Plasmodium sp. transmission in those locations, even if there have been little to no studies on blood parasites of penguins. The Falkland Islands are reportedly free from mosquitoes (Medlock et al. Reference Medlock, Schaffner and Fontenille2010), however DNA from P. relictum was detected in the blood of a thin-billed prion (Pachyptila belcheri) at New Island; because this a pelagic seabird that only comes to land in the breeding season, it is reasonable to suspect that infection occurred on the island (Quillfeldt et al. Reference Quillfeldt, Martínez, Hennicke, Ludynia, Gladbach, Masello, Riou and Merino2010). The harsh climate of Antarctica and the South Shetland Islands probably also precludes the occurrence of Plasmodium sp., as corroborated by blood parasite studies in the region (e.g. Laird, Reference Laird1961; Becker and Holloway, Reference Becker and Holloway1968; Jones and Shellam, Reference Jones and Shellam1999a; González-Acuña et al. Reference González-Acuña, Hernández, Moreno, Herrman, Palma, Latorre, Medina-Vogel, Kinsella, Martín, Araya, Torres, Fernandez and Olsén2013; Vanstreels et al. Reference Vanstreels, Miranda, Ruoppolo, Reis, Costa, Pessôa, Torres, Cunha, Piuco, Valiati, González-Acuña, Labruna, Petry, Epiphanio and Catão-Dias2014b).

Epidemiology and pathology

In wild penguins, the prevalence of Plasmodium sp. varies considerably. Plasmodium sp. was detected in the blood smears of 0·7% of African penguins at Dyer Island, 3% at Saldanha Bay, 9% at Stony Point, 11% at Robben Island, and 34% at Dassen Island (Fantham and Porter, Reference Fantham and Porter1944; Brossy, Reference Brossy1992; Thiart, Reference Thiart2005). In New Zealand, blood smears revealed P. relictum infection in 10% of yellow-eyed penguins at Fouveaux Strait, 10·7% of Snares penguins at Snares Island, and one of two yellow-eyed penguins at Campbell Island (Fantham and Porter, Reference Fantham and Porter1944; Laird, Reference Laird1950). Furthermore, one of five Northern rockhopper penguins from Gough Island was blood smear-positive to P. relictum (Laird, Reference Laird1950). Using molecular methods, Levin et al. (Reference Levin, Outlaw, Vargas and Parker2009, Reference Levin, Zwiers, Deem, Geest, Higashiguchi, Iezhova, Jiménez-Uzcátegui, Kim, Morton, Perlut, Renfrew, Sari, Valkiūnas and Parker2013) detected Plasmodium sp. in the blood of 5·4% of Galapagos penguins, with prevalence varying between 2·1 and 42·9% among islands. None of the wild penguins in which Plasmodium sp. was detected had external signs of disease, and parasitaemia was generally low or undetectable in blood smears (Fantham and Porter, Reference Fantham and Porter1944; Laird, Reference Laird1950; Brossy, Reference Brossy1992; Levin et al. Reference Levin, Outlaw, Vargas and Parker2009). Fantham and Porter (Reference Fantham and Porter1944) detected P. relictum in a deceased wild African penguin; however, because the penguin had multiple traumatic lesions it was not clear to what extent avian malaria may have contributed to its death.

It is well established that avian malaria outbreaks in zoos result from local mosquitoes inoculating penguins with Plasmodium sp. acquired from the native birds in the surroundings of the penguin exhibit (Beier and Trpis, Reference Beier and Trpis1981; Ejiri et al. Reference Ejiri, Sato, Sawai, Sasaki, Matsumoto, Ueda, Higa, Tsuda, Omori, Murata and Yukawa2009; Bueno et al. Reference Bueno, Lopez, Menezes, Costa-Nascimento, Lima, Araújo, Guida and Kirchgatter2010; Leclerc et al. Reference Leclerc, Chavatte, Landau, Snounou and Petit2014; Dinhopl et al. Reference Dinhopl, Nedorost, Mostegl, Weissenbacher-Lang and Weissenböck2015). Because mosquito abundance is markedly seasonal, cases of avian malaria in captive penguins tend to concentrate in spring-summer, particularly late summer (Grünberg and Kutzer, Reference Grünberg and Kutzer1963; Griner and Sheridan, Reference Griner and Sheridan1967; Sladen et al. Reference Sladen, Gailey-Phipps and Divers1979; Beier and Stoskopf, Reference Beier and Stoskopf1980; Vanstreels et al. Reference Vanstreels, Silva-Filho, Kolesnikovas, Bhering, Ruoppolo, Epiphanio, Amaku, Ferreira-Junior, Braga and Catão-Dias2015a). Mosquitoes are most active in penguin enclosures at night (Beier and Trpis, Reference Beier and Trpis1981). At zoos that recorded avian malaria outbreaks, the prevalence of Plasmodium sp. in mosquitoes near penguin exhibits is generally low (<5%) (Beier and Trpis, Reference Beier and Trpis1981; Bueno et al. Reference Bueno, Lopez, Menezes, Costa-Nascimento, Lima, Araújo, Guida and Kirchgatter2010; Ejiri et al. Reference Ejiri, Sato, Sawai, Sasaki, Matsumoto, Ueda, Higa, Tsuda, Omori, Murata and Yukawa2009, Reference Ejiri, Sato, Kim, Hara, Tsuda, Imura, Murata and Yukawa2011), and similar results were obtained in studies at locations where Plasmodium sp. was reported in wild penguins (Fantham and Porter, Reference Fantham and Porter1944).

Outbreaks of avian malaria in permanently captive penguins usually occur suddenly and/or in successive waves. Mortality might depend on the Plasmodium species/lineage involved, whether there was prior exposure to Plasmodium sp., and on the administration of drug treatment, with between 10 and 83% penguins dying within a few weeks or months (Fleischman et al. Reference Fleischman, Squire, Sladen and Moore1968a; Stoskopf and Beier, Reference Stoskopf and Beier1979; Fix et al. Reference Fix, Waterhouse, Greiner and Stoskopf1988; Cranfield et al. Reference Cranfield, Graczyk, Beall, Ialeggio, Shaw and Skjoldager1994; Graczyk et al. Reference Graczyk, Cranfield, Mccutchan and Bicknese1994a; Bueno et al. Reference Bueno, Lopez, Menezes, Costa-Nascimento, Lima, Araújo, Guida and Kirchgatter2010). A similar pattern is observed for penguins kept in temporary captivity while receiving rehabilitation care in South America (Carvajal and Alvarado, Reference Carvajal and Alvarado2009; Vanstreels et al. Reference Vanstreels, Kolesnikovas, Sandri, Silveira, Belo, Ferreira-Junior, Epiphanio, Steindel, Braga and Catão-Dias2014a, Reference Vanstreels, Silva-Filho, Kolesnikovas, Bhering, Ruoppolo, Epiphanio, Amaku, Ferreira-Junior, Braga and Catão-Dias2015a). On the other hand, avian malaria is enzootic to African penguins undergoing rehabilitation, with 30–35% of penguins being positive (blood smears) upon admission (Parsons and Underhill, Reference Parsons and Underhill2005).

Most penguins with avian malaria in captivity are in good body condition and do not present clinical signs, dying suddenly. When clinical signs are present, they are not specific and may include: anorexia, depression, lethargy, weakness, regurgitation, green faeces, hyperthermia, pale mucosae, and dyspnoea (Rodhain, Reference Rodhain1939; Griner and Sheridan, Reference Griner and Sheridan1967; Sladen et al. Reference Sladen, Gailey-Phipps and Divers1979; Stoskopf and Beier, Reference Stoskopf and Beier1979; Bak et al. Reference Bak, Park and Lim1984; Fix et al. Reference Fix, Waterhouse, Greiner and Stoskopf1988; Bueno et al. Reference Bueno, Lopez, Menezes, Costa-Nascimento, Lima, Araújo, Guida and Kirchgatter2010; Vanstreels et al. Reference Vanstreels, Kolesnikovas, Sandri, Silveira, Belo, Ferreira-Junior, Epiphanio, Steindel, Braga and Catão-Dias2014a). Haematology may reveal leucocytosis with lymphocytosis and/or monocytosis (Stoskopf and Beier, Reference Stoskopf and Beier1979; Fix et al. Reference Fix, Waterhouse, Greiner and Stoskopf1988; Graczyk et al. Reference Graczyk, Cranfield, Mccutchan and Bicknese1994a). Infected penguins often have low parasitaemia (<2%) (Stoskopf and Beier, Reference Stoskopf and Beier1979; Graczyk et al. Reference Graczyk, Cranfield, Mccutchan and Bicknese1994a); however, occasionally much higher parasitaemia may be observed, with up to 80% of erythrocytes parasitized and multiple parasites per erythrocyte (Fantham and Porter, Reference Fantham and Porter1944; Bueno et al. Reference Bueno, Lopez, Menezes, Costa-Nascimento, Lima, Araújo, Guida and Kirchgatter2010; Vanstreels et al. Reference Vanstreels, Kolesnikovas, Sandri, Silveira, Belo, Ferreira-Junior, Epiphanio, Steindel, Braga and Catão-Dias2014a).

Captive penguins deceased due to avian malaria typically present hepatomegaly, splenomegaly, lung congestion and hydropericardium (Rodhain, Reference Rodhain1939; Bak et al. Reference Bak, Park and Lim1984; Fix et al. Reference Fix, Waterhouse, Greiner and Stoskopf1988; Graczyk et al. Reference Graczyk, Cranfield, Mccutchan and Bicknese1994a; Grim et al. Reference Grim, van der Merwe, Sullivan, Parsons, McCutchan and Cranfield2003; Ko et al. Reference Ko, Kang, Jung, Bae and Kim2008; Carvajal and Alvarado, Reference Carvajal and Alvarado2009). Tissue meronts are present in multiple tissues and concentrate especially in the lungs, kidneys, brain, heart, liver and spleen (Rodhain, Reference Rodhain1939; Fleischman et al. Reference Fleischman, Squire, Sladen and Melby1968b; Bak et al. Reference Bak, Park and Lim1984; Fix et al. Reference Fix, Waterhouse, Greiner and Stoskopf1988; Graczyk et al. Reference Graczyk, Cranfield, Mccutchan and Bicknese1994a; Grim et al. Reference Grim, van der Merwe, Sullivan, Parsons, McCutchan and Cranfield2003; Ko et al. Reference Ko, Kang, Jung, Bae and Kim2008; Silveira et al. Reference Silveira, Belo, Lacorte, Kolesnikovas, Vanstreels, Steindel, Catão-Dias, Valkiūnas and Braga2013; Vanstreels et al. Reference Vanstreels, Kolesnikovas, Sandri, Silveira, Belo, Ferreira-Junior, Epiphanio, Steindel, Braga and Catão-Dias2014a, Reference Vanstreels, Silva-Filho, Kolesnikovas, Bhering, Ruoppolo, Epiphanio, Amaku, Ferreira-Junior, Braga and Catão-Dias2015a). Concurrent diseases are not uncommon, and aspergillosis is frequently reported in captive penguins that died from avian malaria (Scott, Reference Scott1927; Rodhain, Reference Rodhain1939; Rewell, Reference Rewell1948; Grünberg and Kutzer, Reference Grünberg and Kutzer1963; Griner and Sheridan, Reference Griner and Sheridan1967; Sladen et al. Reference Sladen, Gailey-Phipps and Divers1979; Fix et al. Reference Fix, Waterhouse, Greiner and Stoskopf1988; Carvajal and Alvarado, Reference Carvajal and Alvarado2009; Grilo, Reference Grilo2014; Vanstreels et al. Reference Vanstreels, Silva-Filho, Kolesnikovas, Bhering, Ruoppolo, Epiphanio, Amaku, Ferreira-Junior, Braga and Catão-Dias2015a). Septicaemia (Grünberg and Kutzer, Reference Grünberg and Kutzer1963), enteritis/diarrhoea (Scott, Reference Scott1927; Fix et al. Reference Fix, Waterhouse, Greiner and Stoskopf1988), infestation with gastrointestinal helminthes (Rodhain and Andrianne, Reference Rodhain and Andrianne1952; Fix et al. Reference Fix, Waterhouse, Greiner and Stoskopf1988; Vanstreels et al. Reference Vanstreels, Silva-Filho, Kolesnikovas, Bhering, Ruoppolo, Epiphanio, Amaku, Ferreira-Junior, Braga and Catão-Dias2015a), clostridiosis (Penrith et al. Reference Penrith, Huchzermeyer, Wet and Penrith1994), babesiosis (Yabsley et al. Reference Yabsley, Parsons, Horne, Shock and Purdee2012), poxvirosis and infestation with lung or liver helminthes (Vanstreels et al. Reference Vanstreels, Kolesnikovas, Sandri, Silveira, Belo, Ferreira-Junior, Epiphanio, Steindel, Braga and Catão-Dias2014a, Reference Vanstreels, Silva-Filho, Kolesnikovas, Bhering, Ruoppolo, Epiphanio, Amaku, Ferreira-Junior, Braga and Catão-Dias2015a) have also been documented concurrently with avian malaria.

Serological studies

Graczyk et al. (Reference Graczyk, Cranfield and Shift1993, Reference Graczyk, Cranfield, Skjoldager and Shaw1994b) developed an indirect enzyme-linked immunosorbent assay (ELISA) that was extensively used to test penguins for antibodies against Plasmodium spp. (Graczyk et al. Reference Graczyk, Cranfield, Mccutchan and Bicknese1994a, Reference Graczyk, Cranfield, Skjoldager and Shawb, Reference Graczyk, Shaw, Cranfield and Beallc, Reference Graczyk, Cranfield, Shaw and Craigd, Reference Graczyk, Brossy, Plös and Stoskopf1995a, Reference Graczyk, Cockrem, Cranfield, Darby and Mooreb, Reference Graczyk, Cranfield, Brossy, Cockrem, Jouventin and Seddonc; Botes, Reference Botes2004; Thiart, Reference Thiart2005; McDonald, Reference McDonald2012; Palmer et al. Reference Palmer, McCutchan, Vargas, Deem, Cruz, Hartman and Parker2013). Seroprevalence for Plasmodium sp. was 29–52% in wild African penguin in South Africa (Graczyk et al. Reference Graczyk, Brossy, Plös and Stoskopf1995a, Reference Graczyk, Cockrem, Cranfield, Darby and Mooreb), 33% in Gentoo and 58% in king penguins at Kerguelen and Crozet Islands (Graczyk et al. Reference Graczyk, Cranfield, Brossy, Cockrem, Jouventin and Seddon1995c), 23–100% in yellow-eyed penguins in New Zealand (Graczyk et al. Reference Graczyk, Cockrem, Cranfield, Darby and Moore1995b, Reference Graczyk, Cranfield, Brossy, Cockrem, Jouventin and Seddonc; McDonald, Reference McDonald2012) and 91–100% in Galapagos penguins at the Galapagos Archipelago (Palmer et al. Reference Palmer, McCutchan, Vargas, Deem, Cruz, Hartman and Parker2013). No Adélie penguins were seropositive at Ross Island, Antarctica (Graczyk et al. Reference Graczyk, Cranfield, Brossy, Cockrem, Jouventin and Seddon1995c).

Penguins in captivity or undergoing rehabilitation have also been tested for antibodies against Plasmodium sp. using this assay. At a rehabilitation centre in Cape Town, South Africa, oiled African penguins had higher seroprevalence to Plasmodium sp. upon admission (62%) than those that had been in rehabilitation for at least two weeks (38%) or those permanently captive (20%) (Graczyk et al. Reference Graczyk, Brossy, Plös and Stoskopf1995a). Seroprevalence to Plasmodium sp. was 92% in little penguins captive at Napier, New Zealand, and 43% in Magellanic penguins captive at San Diego, USA. Furthermore, a few studies have applied this ELISA to study the epidemiology of avian malaria in African penguins in captivity (Graczyk et al. Reference Graczyk, Cranfield, Mccutchan and Bicknese1994a, Reference Graczyk, Cranfield, Skjoldager and Shawb, Reference Graczyk, Shaw, Cranfield and Beallc) and undergoing rehabilitation (Botes, Reference Botes2004; Thiart, Reference Thiart2005).

However, several authors have noted that there is a considerable discrepancy between the high seroprevalence to Plasmodium sp. detected by this assay and the rarity of individuals with detectable Plasmodium sp. parasitaemia in blood smears and PCR tests in the same populations (Sturrock and Tompkins, Reference Sturrock and Tompkins2007; Hill, Reference Hill2008; McDonald, Reference McDonald2012; Palmer et al. Reference Palmer, McCutchan, Vargas, Deem, Cruz, Hartman and Parker2013). Some have interpreted this discrepancy as an indication of inaccuracy of the serological test (Sturrock and Tompkins, Reference Sturrock and Tompkins2007; McDonald, Reference McDonald2012), whereas others considered it an indication of poor sensitivity of PCR tests (Palmer et al. Reference Palmer, McCutchan, Vargas, Deem, Cruz, Hartman and Parker2013) or to be due to parasite latency in tissues (Hill, Reference Hill2008; Palmer et al. Reference Palmer, McCutchan, Vargas, Deem, Cruz, Hartman and Parker2013). It remains to be tested whether or not this assay cross-reacts with Leucocytozoon spp., which is plausible considering their shared phylogenetic history and genetic similarities (Cosgrove et al. Reference Cosgrove, Day and Sheldon2006; Martinsen et al. Reference Martinsen, Perkins and Schall2008). Cross-reactivity with viruses (Greenberg et al. Reference Greenberg, Schable, Sulzer, Collins and Nguyhen-Dinh1986) and helminths (Naus et al. Reference Naus, Jones, Satti, Joseph, Riley, Kimani, Mwatha, Kariuki, Ouma, Kabatereine, Vennervald and Dunne2003) have also been shown to occur in serological tests targeting Plasmodium sp. in humans. The results of these serological tests should therefore be interpreted cautiously until detailed studies explain the discrepancies in the results of serological and direct diagnosis tests and to determine if cross-reactivity may have influenced the serological results.

Implications for public health and conservation

There is no evidence to indicate that avian-infecting Plasmodium spp. can infect humans (Valkiūnas, Reference Valkiūnas2005). Plasmodium spp. are recognized as conservation-threatening pathogens due to their well-documented impacts to the Hawaiian avifauna (Van Riper III et al. Reference Van Riper, Van Riper, Goff and Laird1986; Atkinson and Lapointe, Reference Atkinson and Lapointe2009). The high susceptibility of Hawaiian native birds and penguins is thought to result from a lack of physiological/immune adaptations to deal with the infection, as they did not co-evolve with these parasites (Valkiūnas, Reference Valkiūnas2005). The high morbidity and mortality observed in penguins when they are exposed to avian plasmodia in captivity has led to concern that the introduction of mosquitoes to penguin breeding habitats where they had historically been absent could ensue in substantial morbidity and mortality (Jones and Shellam, Reference Jones and Shellam1999b; Miller et al. Reference Miller, Hofkin, Snell, Hahn and Miller2001; Meile et al. Reference Meile, Lacy, Vargas and Parker2013). This is acutely concerning as climate change increases the pressure imposed by Plasmodium sp. on birds (Garamszegi, Reference Garamszegi2011).

In particular, Plasmodium sp. may constitute a significant conservation threat to the African, Galapagos and yellow-eyed penguins, three endangered species with relatively narrow geographic distribution (IUCN, 2015) in which infection has already been documented in the wild (Fantham and Porter, Reference Fantham and Porter1944; Levin et al. Reference Levin, Outlaw, Vargas and Parker2009). Fortunately the Plasmodium sp. lineages detected at the Galapagos Archipelago so far have failed to become established and produce significant disease in Galapagos penguins (Levin et al. Reference Levin, Outlaw, Vargas and Parker2009, Reference Levin, Zwiers, Deem, Geest, Higashiguchi, Iezhova, Jiménez-Uzcátegui, Kim, Morton, Perlut, Renfrew, Sari, Valkiūnas and Parker2013); however, this could change if more pathogenic lineages are introduced to the archipelago in the future. Plasmodium sp. appears to be enzootic in African and yellow-eyed penguins; however, these species' populations are already declining due to a variety of environmental impacts and pathogens (Crawford et al. Reference Crawford, Altwegg, Barham, Durant, Dyer, Geldenhuys, Makhado, Pichegru, Ryan, Underhill, Upfold, Visagie, Waller and Whittington2011; King et al. Reference King, Harper, Wright, McInnes, van der Lubbe, Dobbins and Murray2012), and avian malaria could synergize with these existing threats. Furthermore, penguin populations at other areas with relatively warm climate such as Peru, Chile and Tristan da Cunha and Gough Islands could also become at risk if mosquitoes become successfully established near penguin breeding habitat, particularly near freshwater deposits associated with human communities. Other populations of penguins that have relatively narrow geographic distributions, such as Fiordland and Snares penguins could also be at risk, since mosquitoes are already present in their breeding habitat (Fantham and Porter, Reference Fantham and Porter1944; Laird, Reference Laird1950).

Species recorded in penguins

Leucocytozoon (Leucocytozoon) tawaki was described from Fiordland penguins (Fallis et al. Reference Fallis, Bisset and Allison1976; Allison et al. Reference Allison, Desser and Whitten1978). When Earlé et al. (Reference Earlé, Bennett and Brossy1992) and Peirce et al. (Reference Peirce, Greenwood and Stidworthy2005) observed similar parasites in other species of penguins in Europe and South Africa they did not hesitate to attribute these records to L. tawaki. On the other hand, other authors have documented leucocytozooids in yellow-eyed penguins at locations in New Zealand but preferred not to comment on the species involved (Hill, Reference Hill2008; Hill et al. Reference Hill, Howe, Gartrell and Alley2010; Argilla et al. Reference Argilla, Howe, Gartrell and Alley2013). Phylogenetic analyses indicated these lineages from yellow-eyed penguins belonged to the subgenus Leucocytozoon, and that there might be at least two distinct phylogenetic groups: cluster A is limited to Enderby Island and might be more pathogenic than cluster B, which was detected at Enderby, Campbell, South and Stewart Islands (Argilla et al. Reference Argilla, Howe, Gartrell and Alley2013). It is unclear whether these phylogenetic clusters are variations within the same morphospecies or correspond to different species, and their relationship to the L. tawaki originally described in Fiordland penguins is also not clear.

Distribution among penguin hosts

Leucocytozoon spp. have been detected in wild Fiordland (Fallis et al. Reference Fallis, Bisset and Allison1976; Allison et al. Reference Allison, Desser and Whitten1978) and yellow-eyed penguins (Hill et al. Reference Hill, Howe, Gartrell and Alley2010; Argilla et al. Reference Argilla, Howe, Gartrell and Alley2013), as well as in African (Earlé et al. Reference Earlé, Bennett and Brossy1992) and Fiordland penguins undergoing rehabilitation (Hill, Reference Hill2008). Additionally, the infection was documented in Macaroni penguins in captivity (Peirce et al. Reference Peirce, Greenwood and Stidworthy2005). Allison et al. (Reference Allison, Desser and Whitten1978) demonstrated that little penguins can develop the infection when forcibly exposed to black flies near L. tawaki-infected Fiordland penguins. There is also evidence to suggest that Leucocytozoon sp. could infect wild and captive little penguins (see Appendix 2). It is worth noting that Snares, erect-crested and little penguins live in close proximity to populations of Fiordland and yellow-eyed penguins that were found to be infected with Leucocytozoon sp., and it is reasonable to presume they are exposed to these parasites in the wild.

Invertebrate hosts

L. (Leucocytozoon) spp. are transmitted by black flies (Simuliidae), particularly Simulium spp. and Prosimulium spp., but also Cnephia spp. and Austrosimulium spp. (Valkiūnas, Reference Valkiūnas2005; Forrester and Greiner, Reference Forrester, Greiner, Atkinson, Thomas and Hunter2008). Cnephia spp., Simulium spp. and Prosimulium spp. are not present in New Zealand (Dumbleton, Reference Dumbleton1963), whereas Austrosimulium australense, Austrosimulium dumbletoni and Austrosimulium ungulatum are abundant and were shown to be competent in the transmission of L. tawaki at South Island (Fallis et al. Reference Fallis, Bisset and Allison1976; Allison et al. Reference Allison, Desser and Whitten1978; Desser and Allison, Reference Desser and Allison1979). Austrosimulium ungulatum is also very abundant in Stewart and South Islands, New Zealand, and could be involved in the transmission of Leucocytozoon sp. to yellow-eyed penguins (Hill et al. Reference Hill, Howe, Gartrell and Alley2010; Argilla et al. Reference Argilla, Howe, Gartrell and Alley2013). On the other hand, Austrosimulium campbellense and Austrosimulium vexans are thought to be respectively responsible for the transmission at Campbell and Auckland Islands, New Zealand (Argilla et al. Reference Argilla, Howe, Gartrell and Alley2013). Austrosimilium spp. do not occur in South Africa (Dumbleton, Reference Dumbleton1963) and thus other simulid flies must be involved in the transmission of this parasite to African penguins (Earlé et al. Reference Earlé, Bennett and Brossy1992); Cnephia spp. and Simulium spp. are present in the region (Dumbleton, Reference Dumbleton1963).

Geographic distribution

Leucocytozoon tawaki is known from South Island, New Zealand (Kaikoura, Jackson Head) (Fallis et al. Reference Fallis, Bisset and Allison1976; Allison et al. Reference Allison, Desser and Whitten1978). Leucocytozoon sp. has been documented at South Island, New Zealand (Otago Peninsula and Catlins) (Argilla et al. Reference Argilla, Howe, Gartrell and Alley2013) and at Campbell, Enderby and Stewart Islands (Hill et al. Reference Hill, Howe, Gartrell and Alley2010; Argilla et al. Reference Argilla, Howe, Gartrell and Alley2013). Additionally, Leucocytozoon sp. has been documented in penguins undergoing rehabilitation in South Africa (Cape Town) (Earlé et al. Reference Earlé, Bennett and Brossy1992) and North Island, New Zealand (Auckland) (Hill, Reference Hill2008), and captive in England (Peirce et al. Reference Peirce, Greenwood and Stidworthy2005). Because the blood smears examined by Earlé et al. (Reference Earlé, Bennett and Brossy1992) were prepared between 5 and 24 days after admission to the rehabilitation centre in South Africa, it is possible that infection occurred in the wild. The geographic distribution of black flies overlaps with that of penguin breeding colonies in several regions where Leucocytozoon sp. has not yet been reported in penguins, such as Peru, Chile, Argentina, Namibia, Australia and New Zealand (Fig. 2B).

Epidemiology and pathology

L. tawaki prevalence is very high in Fiordland penguins at South Island, New Zealand (Jackson Head) (blood smears: 77–94%) (Fallis et al. Reference Fallis, Bisset and Allison1976; Allison et al. Reference Allison, Desser and Whitten1978). Leucocytozoon sp. prevalence is more variable in yellow-eyed penguins in New Zealand, being lower at South Island (Otago Peninsula and Catlins) (PCR: 11%) and Campbell Island (PCR: 21%), and higher at Enderby (PCR: 66%) and Stewart Islands (PCR: 83%) (Hill et al. Reference Hill, Howe, Gartrell and Alley2010; Argilla et al. Reference Argilla, Howe, Gartrell and Alley2013). Both Fiordland and yellow-eyed penguins are infected only when they are 3-weeks-old or older, with the infection being acute and disseminated in older chicks then progressing to a subclinical chronic infection in adulthood (Allison et al. Reference Allison, Desser and Whitten1978; Hill et al. Reference Hill, Howe, Gartrell and Alley2010; Argilla et al. Reference Argilla, Howe, Gartrell and Alley2013). Although prevalence is similar in older chicks and adults (and possibly highest in moulting adults), parasitaemia tends to be lower in adults (Fallis et al. Reference Fallis, Bisset and Allison1976; Allison et al. Reference Allison, Desser and Whitten1978). In fact, parasitaemia in adults may be so low as to be undiagnosed or substantially underestimated by blood smears in comparison with molecular methods (Hill et al. Reference Hill, Howe, Gartrell and Alley2010; Argilla et al. Reference Argilla, Howe, Gartrell and Alley2013). Leucocytozoon sp. occurs at low prevalence amongst African penguins undergoing rehabilitation (blood smears: 0·75%) (Earlé et al. Reference Earlé, Bennett and Brossy1992). Because it is generally accepted that leucocytozooids are not transmitted among birds of different taxonomic orders (Valkiūnas, Reference Valkiūnas2005), it is unlikely that birds other than penguins can serve as reservoirs of infection.

Leucocytozoon sp. can be occasionally pathogenic for penguin chicks. One yellow-eyed penguin chick found dead at Enderby Island (n = 19) and two at Stewart Island, New Zealand (n = 14), were considered to have died from leucocytozoonosis. Necropsy findings included disseminated petechial and ecchymotic haemorrhage, hepatomegaly, splenomegaly and hydropericardium; megalomeronts were abundant in the liver, spleen, kidneys, lungs and other tissues (Hill et al. Reference Hill, Howe, Gartrell and Alley2010; Argilla et al. Reference Argilla, Howe, Gartrell and Alley2013). The tissues of an additional seven yellow-eyed penguins were PCR-positive for Leucocytozoon sp. at Stewart Island; however, it was not determined whether leucocytozoonosis was the cause of death or not (Hill et al. Reference Hill, Howe, Gartrell and Alley2010). Furthermore, a juvenile Fiordland penguin found at North Island, New Zealand (Muriwai beach), died during rehabilitation after having been positive to Leucocytozoon sp. on blood smears, but it was not possible to determine if leucocytozoonosis was the cause of death (Hill, Reference Hill2008). The health effects of the infection in African and Macaroni penguins are not known (Earlé et al. Reference Earlé, Bennett and Brossy1992; Peirce et al. Reference Peirce, Greenwood and Stidworthy2005).

Implications for public health and conservation

There is no evidence to indicate that Leucocytozoon spp. could infect humans (Valkiūnas, Reference Valkiūnas2005). Although Leucocytozoon sp. appears to have limited impacts to the health of adult penguins, this can be a considerably pathogenic parasite to penguin chicks and juveniles (Fallis et al. Reference Fallis, Bisset and Allison1976; Allison et al. Reference Allison, Desser and Whitten1978; Hill, Reference Hill2008; Hill et al. Reference Hill, Howe, Gartrell and Alley2010; Argilla et al. Reference Argilla, Howe, Gartrell and Alley2013). This is particularly troublesome for yellow-eyed penguins, an endangered species that has faced substantial population decrease in the past decades (IUCN, 2015). Yellow-eyed penguin chicks already face a variety of stressors and diseases (Alley et al. Reference Alley, Morgan, Gill and Hocken2004; Hocken, Reference Hocken2005; Browne et al. Reference Browne, Lalas, Mattern and van Heezik2011; Buckle et al. Reference Buckle, Young and Alley2014), and Leucocytozoon sp. might be an additional factor contributing to decrease the species' chick survival (King et al. Reference King, Harper, Wright, McInnes, van der Lubbe, Dobbins and Murray2012). In the case of African penguins, which are also endangered (IUCN, 2015), additional studies are urgent to bring better understanding on the epidemiology and pathology of this parasite. There is also evidence to suspect that wild little penguin chicks might also die as a result from leucocytozoonosis (see Appendix 2), and therefore an investigation on the occurrence of Leucocytozoon sp. in this species, particularly in Western Australia, would be valuable.

Species recorded in penguins

Haemoproteus sp. detected in penguins have not been morphologically characterized, hence their identity has not been conclusively established. Phylogenetic analyses indicate however that the lineages identified in Galapagos and Humboldt penguins belong to the subgenus Parahaemoproteus and are closely related to lineages found in passerines (Levin et al. Reference Levin, Outlaw, Vargas and Parker2009; Sallaberry-Pincheira et al. Reference Sallaberry-Pincheira, González-Acuña, Herrera-Tello, Dantas, Luna-Jorquera, Frere, Valdés-Velasquez, Simeone and Vianna2015) (see Appendix 2).

Distribution among penguin hosts

DNA from Haemoproteus sp. has been detected in the blood of wild Galapagos (Levin et al. Reference Levin, Outlaw, Vargas and Parker2009) and Humboldt penguins (Sallaberry-Pincheira et al. Reference Sallaberry-Pincheira, González-Acuña, Herrera-Tello, Dantas, Luna-Jorquera, Frere, Valdés-Velasquez, Simeone and Vianna2015). There is a report of Haemoproteus sp. infection in wild little penguins (Cannell et al. Reference Cannell, Krasnec, Campbell, Jones, Miller and Stephens2013), however that record is problematic (see Appendix 2).

Invertebrate hosts

Haemoproteus (Parahaemoproteus) spp. are transmitted by biting midges Culicoides spp. (Ceratopogonidae) (Valkiūnas, Reference Valkiūnas2005; Atkinson, Reference Atkinson, Atkinson, Thomas and Hunter2008a). Eleven species of Culicoides spp. have been associated with the transmission of these parasites (Valkiūnas, Reference Valkiūnas2005).

Culicoides pusillus is the only species of its genus that occurs at the Galapagos Archipelago (Sinclair, Reference Sinclair, Bungartz, Herrera, Jaramillo, Tirado, Jiménez-Uzcátegui, Ruiz, Guézou and Ziemmeck2014), whereas a broad variety of species occurs in Peru (Wirth and Felippe-Bauer, Reference Wirth and Felippe-Bauer1989; Borkent, Reference Borkent2013). It is worth noting that the Haemoproteus-positive penguin identified at the Galapagos Archipelago was sampled at western Isabela Island (I. I. Levin, personal communication), whereas C. pusillus has been recorded only at Santa Cruz Island (Sinclair, Reference Sinclair, Bungartz, Herrera, Jaramillo, Tirado, Jiménez-Uzcátegui, Ruiz, Guézou and Ziemmeck2014). Even though DNA from Haemoproteus sp. was identified in the blood meals of mosquitoes Aedes taeniorhynchus at the Galapagos Archipelago, this was most likely an incidental finding and probably did not correspond to actual infections (Bataille et al. Reference Bataille, Fournié, Cruz, Cedeño, Parker, Cunningham and Goodman2012). It is not clear which species of biting midges were involved in the transmission of Haemoproteus sp. to Humboldt penguins at Punta San Juan (Sallaberry-Pincheira et al. Reference Sallaberry-Pincheira, González-Acuña, Herrera-Tello, Dantas, Luna-Jorquera, Frere, Valdés-Velasquez, Simeone and Vianna2015), however Culicoides spp. are not uncommon in Peru (Tabachnick, Reference Tabachnick2004; Felippe-Bauer et al. Reference Felippe-Bauer, Cáceres, Silva, Valderrama-Bazan, Gonzales-Perez and Costa2008) and the coastal range of Peru provides suitable climatic conditions to these insects (Guichard et al. Reference Guichard, Guis, Tran, Garros, Balenghien and Kriticos2014).

Geographic distribution

Haemoproteus (Parahaemoproteus) spp. were detected in penguins in the Galapagos Archipelago (Isabela Island) (Levin et al. Reference Levin, Outlaw, Vargas and Parker2009) and Peru (Punta San Juan) (Sallaberry-Pincheira et al. Reference Sallaberry-Pincheira, González-Acuña, Herrera-Tello, Dantas, Luna-Jorquera, Frere, Valdés-Velasquez, Simeone and Vianna2015). The distribution of biting midges overlaps that of penguins breeding in Namibia, South Africa and Australia, possibly providing opportunities for H. (Parahaemoproteus) sp. inoculation. Similarly, captive penguins in areas of North and South America, southern Europe, Africa, Asia and Oceania could also be exposed (Fig. 2A).

Epidemiology and pathology

The Haemoproteus-positive penguins studied by Levin et al. (Reference Levin, Outlaw, Vargas and Parker2009) and Sallaberry-Pincheira et al. (Reference Sallaberry-Pincheira, González-Acuña, Herrera-Tello, Dantas, Luna-Jorquera, Frere, Valdés-Velasquez, Simeone and Vianna2015) had no external signs of illness. Considering that no parasites were seen in blood smears and that molecular tests may produce false-positive results if DNA of recently inoculated sporozoites is present in the blood even when infection was not developed (Levin et al. Reference Levin, Zwiers, Deem, Geest, Higashiguchi, Iezhova, Jiménez-Uzcátegui, Kim, Morton, Perlut, Renfrew, Sari, Valkiūnas and Parker2013; Valkiūnas et al. Reference Valkiūnas, Palinauskas, Ilgūnas, Bukauskaitė, Dimitrov, Bernotienė, Zehtindjiev, Ilieva and Iezhova2014), it is possible that these Haemoproteus spp. were not truly capable of infecting penguins and instead represent incidental findings (i.e. abortive infections). The report of lethal haemoproteosis in wild little penguins (Cannell et al. Reference Cannell, Krasnec, Campbell, Jones, Miller and Stephens2013) is problematic, and is addressed in detail in Appendix 2.

Implications for public health and conservation

There is no evidence to indicate that Haemoproteus spp. could infect humans (Valkiūnas, Reference Valkiūnas2005). Haemoproteus spp. are generally considered the least pathogenic of avian haemosporidians; however, there are exceptional cases of lethal haemoproteosis (Atkinson and Van Riper III, Reference Atkinson, Van Riper, Loye and Zuk1991; Donovan et al. Reference Donovan, Schrenzel, Tucker, Pessier and Stalis2008). In the case of penguins, however, it is still unclear whether Haemoproteus sp. sporozoites are able to infect and develop in penguin cells (see Levin et al. Reference Levin, Zwiers, Deem, Geest, Higashiguchi, Iezhova, Jiménez-Uzcátegui, Kim, Morton, Perlut, Renfrew, Sari, Valkiūnas and Parker2013; Valkiūnas et al. Reference Valkiūnas, Palinauskas, Ilgūnas, Bukauskaitė, Dimitrov, Bernotienė, Zehtindjiev, Ilieva and Iezhova2014) and, until this has been conclusively demonstrated, it seems unlikely that these parasites pose a significant threat for their conservation.

Species recorded in penguins

Babesia peircei was described from African penguins (Earlé et al. Reference Earlé, Huchzermeyer, Bennett and Brossy1993). It is not clear whether the Babesia sp. reported in little penguins in Australia corresponds to B. peircei or to a different species (Cunningham et al. Reference Cunningham, Gibbs, Rogers, Spielman and Walraven1993; Vanstreels et al. Reference Vanstreels, Woehler, Ruoppolo, Vertigan, Carlile, Priddel, Finger, Dann, Vinette-Herrin, Thompson, Ferreira-Junior, Braga, Hurtado, Epiphanio and Catão-Dias2015b). The remarkable morphological and/or genetic similarities between B. peircei, Babesia sp. of little penguins, Babesia poelea (parasite of boobies Sula spp.) and Babesia uriae (parasite of common murres Uria aalge) has led to speculation that these taxa could in fact correspond to a single seabird-infecting species (Peirce, Reference Peirce2000; Vanstreels et al. Reference Vanstreels, Woehler, Ruoppolo, Vertigan, Carlile, Priddel, Finger, Dann, Vinette-Herrin, Thompson, Ferreira-Junior, Braga, Hurtado, Epiphanio and Catão-Dias2015b).

Distribution among penguin hosts

Babesia peircei is known from African penguins in the wild and in rehabilitation (Coles, Reference Coles1941; Earlé et al. Reference Earlé, Huchzermeyer, Bennett and Brossy1993; Brossy et al. Reference Brossy, Plös, Blackbeard and Kline1999; Yabsley et al. Reference Yabsley, Parsons, Horne, Shock and Purdee2012) and Babesia sp. was documented in wild little penguins (Cunningham et al. Reference Cunningham, Gibbs, Rogers, Spielman and Walraven1993; Vanstreels et al. Reference Vanstreels, Woehler, Ruoppolo, Vertigan, Carlile, Priddel, Finger, Dann, Vinette-Herrin, Thompson, Ferreira-Junior, Braga, Hurtado, Epiphanio and Catão-Dias2015b).

Invertebrate hosts

It is generally accepted that hard ticks (Ixodidae) are the most relevant invertebrate hosts of avian Babesia spp., but soft ticks (Argasidae) are thought to play a significant role for colonial ground-nesting birds (Peirce, Reference Peirce2000). Hard ticks, particularly Ixodes uriae, are the most probable vectors of B. peircei to African penguins (Earlé et al. Reference Earlé, Huchzermeyer, Bennett and Brossy1993; Peirce, Reference Peirce2000), but the soft tick Carios capensis has also been speculated to play a role in the transmission (Brossy et al. Reference Brossy, Plös, Blackbeard and Kline1999). The hard tick Ixodes kohlsi may play a key role in the transmission to little penguins at New South Wales, Australia (Cunningham et al. Reference Cunningham, Gibbs, Rogers, Spielman and Walraven1993), and both soft and hard ticks were observed on Babesia-infected little penguins in Tasmania, Australia (Vanstreels et al. Reference Vanstreels, Woehler, Ruoppolo, Vertigan, Carlile, Priddel, Finger, Dann, Vinette-Herrin, Thompson, Ferreira-Junior, Braga, Hurtado, Epiphanio and Catão-Dias2015b).

Geographic distribution

Babesia peircei was documented infecting penguins in Namibia (Ichaboe Island), South Africa (Western Cape and Eastern Cape) and Babesia sp. infects penguins in Australia (New South Wales, Victoria and Tasmania) (Cunningham et al. Reference Cunningham, Gibbs, Rogers, Spielman and Walraven1993; Earlé et al. Reference Earlé, Huchzermeyer, Bennett and Brossy1993; Vanstreels et al. Reference Vanstreels, Woehler, Ruoppolo, Vertigan, Carlile, Priddel, Finger, Dann, Vinette-Herrin, Thompson, Ferreira-Junior, Braga, Hurtado, Epiphanio and Catão-Dias2015b; Parsons et al. in preparation). Seabird ticks, both soft and hard, are broadly distributed around the world, overlapping with the distribution of penguins in numerous sub-Antarctic islands, Peru, Chile, Argentina, New Zealand, Antarctic Peninsula and at some locations in the Antarctic mainland (Fig. 3A); the occurrence of Babesia sp. in penguins at these locations is therefore plausible.

Epidemiology and pathology

Babesia sp. infects 1·6 to 4·8% (blood smears; Cunningham et al. Reference Cunningham, Gibbs, Rogers, Spielman and Walraven1993) and 2·7% (blood smears; Vanstreels et al. Reference Vanstreels, Woehler, Ruoppolo, Vertigan, Carlile, Priddel, Finger, Dann, Vinette-Herrin, Thompson, Ferreira-Junior, Braga, Hurtado, Epiphanio and Catão-Dias2015b) of wild little penguins in southeastern Australia. B. peircei is endemic at low prevalence in wild African penguins in Namibia and South Africa (blood smears: 1–4%) (Brossy, Reference Brossy1992; Parsons et al. in preparation), whereas a higher prevalence (blood smears: 11–15%) was observed in African penguins undergoing rehabilitation (Brossy, Reference Brossy1992). Because it is not known whether penguin-infecting Babesia sp. and B. poelea are the same species or not, it is not clear if other seabirds can serve as reservoirs of infections for penguins and vice-versa.

The clinical and pathological effects of Babesia spp. infections are not clear. Infected little penguins can present mild regenerative anemia, but did not show any evident signs of illness (Cunningham et al. Reference Cunningham, Gibbs, Rogers, Spielman and Walraven1993; Sergent et al. Reference Sergent, Rogers and Cunningham2004). Brossy et al. (Reference Brossy, Plös, Blackbeard and Kline1999) considered that B. peircei ‘does not cause overt clinical symptoms except under stress or in association with other debilitating diseases’. On the other hand, Parsons et al. (in preparation) found that B. peircei-infected wild African penguins had signs of regenerative response of the erythrocytic lineage and haematological indications of active inflammatory response and hepatic function impairment.

Furthermore, approximately 50% of Borrelia-infected African penguins undergoing rehabilitation in South Africa are also co-infected with B. peircei (Yabsley et al. Reference Yabsley, Parsons, Horne, Shock and Purdee2012), which could reflect: (a) transmission by a shared invertebrate host; (b) Babesia sp. infections predispose penguins or ticks to Borrelia sp. infections or vice versa; or (c) the poor health and immune status of penguins in rehabilitation predispose them to both of these pathogens.

Serological studies

The indirect ELISA designed for Plasmodium by Graczyk et al. (Reference Graczyk, Cranfield and Shift1993, Reference Graczyk, Cranfield, Skjoldager and Shaw1994b) was adapted to test penguins for antibodies against Babesia sp., and showed that 18–22% of wild African penguins in South Africa were seropositive (47% in oiled birds) (Graczyk et al. Reference Graczyk, Brossy, Sanders, Dubey, Plös and Stoskopf1996). However, the limitations and concerns raised regarding the use of this assay to test for Plasmodium sp. may also apply to its application for Babesia sp.

Implications for public health and conservation

There is no evidence to indicate that avian-infecting Babesia spp. can infect humans (Peirce, Reference Peirce2000). There is evidence that Babesia sp. infections significantly affect the health of penguins (Cunningham et al. Reference Cunningham, Gibbs, Rogers, Spielman and Walraven1993; Parsons et al. in preparation), which is concerning because this pathogen is not uncommon in African penguins, an endangered species whose population has been steadily decreasing (Crawford et al. Reference Crawford, Altwegg, Barham, Durant, Dyer, Geldenhuys, Makhado, Pichegru, Ryan, Underhill, Upfold, Visagie, Waller and Whittington2011; IUCN, 2015). Epidemiological and pathological studies of Babesia sp., particularly in African penguins, will therefore be important to clarify its potential conservation impacts.

Species recorded in penguins

Only one species, Trypanosoma eudyptulae, has been reported in penguins (Jones and Woehler, Reference Jones and Woehler1989). This parasite has not been reported in other avian hosts.

Distribution among penguin hosts

Trypanosoma sp. has only been reported in wild little penguins (Jones and Woehler, Reference Jones and Woehler1989).

Invertebrate hosts

There is no information regarding which invertebrates are involved in the transmission of T. eudyptulae. Black flies (Metacnephia lyra, Simulium spp. and Prosimulium decemarticulatum), mosquitoes (Aedes aegypti), louse flies (Ornithomya avicularia) and mites (Dermanyssus gallinae) have been found to be competent hosts of other avian trypanosomes (Molyneux, Reference Molyneux1977; Reeves et al. Reference Reeves, Adler, Rätti, Malmqvist and Strasevicius2007). Of those, A. aegypti is absent in Tasmania (Kearney et al. Reference Kearney, Porter, Williams, Ritchie and Hoffmann2009), whereas Dermanyssus spp. (including D. gallinae) and Ornithomya spp. (including O. avicularia) are present (Domrow, Reference Domrow1979; ALA, 2014). Furthermore, other species of black flies (Austrosimulium spp. and Cnephia spp.) also occur in Tasmania (Dumbleton, Reference Dumbleton1963) and could be plausible hosts.

Geographic distribution

Trypanosoma eudyptulae was originally described at a little penguin colony on Marion Bay in Tasmania, Australia (Fig. 3C) (Jones and Woehler, Reference Jones and Woehler1989). However, that colony was destroyed during a fire in 1994 and has not been recolonized since (Stevenson and Woehler, Reference Stevenson and Woehler2007; E. J. Woehler, personal communication). Recent efforts to detect this parasite in breeding colonies near Marion Bay have failed (Vanstreels et al. Reference Vanstreels, Woehler, Ruoppolo, Vertigan, Carlile, Priddel, Finger, Dann, Vinette-Herrin, Thompson, Ferreira-Junior, Braga, Hurtado, Epiphanio and Catão-Dias2015b). Because this parasite's invertebrate hosts remain unknown, it is difficult to speculate on its potential distribution.

Epidemiology and pathology

Despite having been observed with a relatively high prevalence (blood smears: 15·8%), T. eudyptulae was present only with low parasitaemia (often only one parasite per blood smear) (Jones and Woehler, Reference Jones and Woehler1989), which suggests chronic infection. Infected penguins presented no external signs of illness.

It is worth noting that Jones and Woehler (Reference Jones and Woehler1989) obtained blood samples by superficially scraping the skin near the brachial vein on the flipper with razorblades then collecting a drop of blood with a capillary tube or glass slide (E. J. Woehler, personal communication). This method would result in the collection of blood from capillary vessels, as opposed to blood from larger vessels as is obtained through venipuncture. This may be relevant because it has been shown that mammal-infecting trypanosomes tend to concentrate in capillaries rather than larger blood vessels (Hornby and Bailey, Reference Hornby and Bailey1931; Banks, Reference Banks1978). It is unclear whether or not avian trypanosomes behave similarly (Holmstad et al. Reference Holmstad, Anwar, Iezhova and Skorping2003), but there is evidence to suggest that these parasites concentrate in the bone marrow of birds rather than in their circulating blood (Diamond and Herman, Reference Diamond and Herman1954). For these reasons, it is possible that studies using blood smears from samples collected by venipuncture may have systematically underestimated the occurrence of trypanosomatids in penguins. Molecular methods could also enhance the detection of these parasites in the future (see Sehgal et al. Reference Sehgal, Jones and Smith2001).

Implications for public health and conservation

There is no evidence to indicate that avian-infecting Trypanosoma spp. can infect humans (Molyneux, Reference Molyneux1977). Avian trypanosome infections are not usually regarded as pathogenic, but in some circumstances these parasites may have mild health impacts (Molyneux et al. Reference Molyneux, Cooper and Smith1983; Merino et al. Reference Merino, Barbosa, Moreno and Potti1996; Sehgal et al. Reference Sehgal, Jones and Smith2001). Because of how little is known about T. eudyptulae, it is difficult to evaluate the impacts it could have on the conservation of little penguins, if any.

Species recorded in penguins

Borrelia sp. strains detected in king penguins had a restriction fragment length polymorphism profile identical to that of B. garinii, and are therefore thought to belong to the LDB group (Schramm et al. Reference Schramm, Gauthier-Clerc, Fournier, McCoy, Barthel, Postic, Handrich, Le Maho and Jaulhac2014). On the other hand, the strains identified in African penguins are phylogenetically most related to B. parkeri and B. turicatae, both of which are classified as RFB (Yabsley et al. Reference Yabsley, Parsons, Horne, Shock and Purdee2012). Coles (Reference Coles1941) observed spirochetes in the blood smear of a wild African penguin chick at Dassen Island, and discarded them from being B. anserina; considering that RFB were later found to infect African penguins at that region (Yabsley et al. Reference Yabsley, Parsons, Horne, Shock and Purdee2012), it is reasonable to assume these corresponded to similar strains.

Distribution among penguin hosts

RFB has been documented in African penguins in the wild (Coles, Reference Coles1941; Parsons et al. in preparation) and undergoing rehabilitation (Yabsley et al. Reference Yabsley, Parsons, Horne, Shock and Purdee2012; Parsons et al. in preparation). LDB was recorded in wild king penguins (Schramm et al. Reference Schramm, Gauthier-Clerc, Fournier, McCoy, Barthel, Postic, Handrich, Le Maho and Jaulhac2014); it is reasonable to presume that Gentoo, Macaroni and Southern rockhopper penguins breeding near king penguins at Crozet Archipelago (IUCN, 2015) are also exposed to LDB.

Vectors

With the exception of B. recurrentis, which is transmitted to humans by lice, all Borrelia spp. are transmitted by ticks. LDB are transmitted by hard ticks Ixodes spp., RFB are transmitted by soft ticks Carios (=Ornithodoros) spp., and B. anserina is transmitted by soft ticks Argas spp. (Barbour and Hayes, Reference Barbour and Hayes1986; Olsén, Reference Olsén, Thomas, Hunter and Atkinson2007; Elbir et al. Reference Elbir, Raoult and Drancourt2013). Additionally, B. anserina can be transmitted through the ingestion or inoculation of faeces, fluids, and tissues (Olsén, Reference Olsén, Thomas, Hunter and Atkinson2007).

Soft ticks, particularly C. capensis, are commonly found on wild African penguins (Clarke and Kerry, Reference Clarke and Kerry1993), and are likely responsible for the transmission of RFB to those birds (Yabsley et al. Reference Yabsley, Parsons, Horne, Shock and Purdee2012). On the other hand, the hard ticks Ixodes kerguelensis and I. uriae are abundant in sub-Antarctic islands and are through to play a key role in the transmission of LDB to king penguins (Olsén et al. Reference Olsén, Duffy, Jaenson, Gylfe, Bonnedahl and Bergström1995b; Gauthier-Clerc et al. Reference Gauthier-Clerc, Jaulhac, Frenot, Bachelard, Monteil, Le Maho and Handrich1999; Schramm et al. Reference Schramm, Gauthier-Clerc, Fournier, McCoy, Barthel, Postic, Handrich, Le Maho and Jaulhac2014).

Geographic distribution

RFB infects penguins in South Africa (Cape Town, Dassen Island) (Coles, Reference Coles1941; Yabsley et al. Reference Yabsley, Parsons, Horne, Shock and Purdee2012; Parsons et al. in preparation), and LDB is present in king penguins at Crozet Archipelago (Possession Island) (Olsén et al. Reference Olsén, Duffy, Jaenson, Gylfe, Bonnedahl and Bergström1995b; Schramm et al. Reference Schramm, Gauthier-Clerc, Fournier, McCoy, Barthel, Postic, Handrich, Le Maho and Jaulhac2014). The distribution of Ixodes spp. overlaps with penguin breeding habitat in Southern South America, Antarctic Peninsula, South Africa, Australia, New Zealand and at a number of sub-Antarctic islands (Fig. 3B), and LDB strains are broadly distributed in seabirds at a number of these locations (Olsén et al. Reference Olsén, Duffy, Jaenson, Gylfe, Bonnedahl and Bergström1995b; Olsén, Reference Olsén, Thomas, Hunter and Atkinson2007). Similarly, the distribution of Carios spp. overlaps with breeding colonies of penguins in the Galapagos Archipelago, Peru, Tristan da Cunha Archipelago, South Africa, Amsterdam and Saint-Paul Islands, Southeastern Australia, New Zealand and Chatham Islands (Fig. 3B).

Epidemiology and pathology

RFB occurs at low prevalence (blood smears: 0·9–1·1%) in African penguins undergoing rehabilitation (Yabsley et al. Reference Yabsley, Parsons, Horne, Shock and Purdee2012; Parsons et al. in preparation); infection is more frequent in chicks (3·6%) than in juveniles (0·83%) and adults (0·14%). As previously discussed, approximately 50% of RFB-infected African penguins undergoing rehabilitation in South Africa are co-infected by B. peircei (Yabsley et al. Reference Yabsley, Parsons, Horne, Shock and Purdee2012).

In only one RFB-infected African penguin studied by Yabsley et al. (Reference Yabsley, Parsons, Horne, Shock and Purdee2012) death was considered to result from Borrelia infection; that penguin presented signs of neurological disease (unsteady gait, circling, torticollis) and died after four days. On post-mortem examination, splenomegaly and hepatomegaly were noted and histological findings were consistent with relapsing fever: splenic reticuloendothelial hyperplasia with haemosiderosis, lung congestion, and lymphocytic meningoencephalitis. On the other hand, LDB are generally considered non-pathogenic to seabirds (Olsén et al. Reference Olsén, Jaenson and Bergström1995a, Reference Olsén, Duffy, Jaenson, Gylfe, Bonnedahl and Bergströmb; Olsén, Reference Olsén, Thomas, Hunter and Atkinson2007), and no external signs of illness were observed in LDB-infected king penguins (Schramm et al. Reference Schramm, Gauthier-Clerc, Fournier, McCoy, Barthel, Postic, Handrich, Le Maho and Jaulhac2014).

Serological studies

Gauthier-Clerc et al. (Reference Gauthier-Clerc, Jaulhac, Frenot, Bachelard, Monteil, Le Maho and Handrich1999) tested king penguins with a passive haemagglutination commercial kit developed to detect human antibodies against LDB. No additional studies have employed this serological assay to test other penguin species, and it is unknown whether the assay cross-reacts with other Borrelia spp. or other spirochetes (see Magnarelli et al. Reference Magnarelli, Anderson and Johnson1987).

Implications for public health and conservation

Relapsing fever is a relevant disease for humans worldwide, including in South Africa, but it is presently unknown whether the RFB strains that infect African penguins could be the same involved in any of the few human cases recorded in South Africa (Yabsley et al. Reference Yabsley, Parsons, Horne, Shock and Purdee2012; Elbir et al. Reference Elbir, Raoult and Drancourt2013). Similarly, Lyme disease is relevant for humans, and seabirds are thought to play a role in the maintenance and transmission of LDB to humans and other mammals, especially at high latitudes (Olsén et al. Reference Olsén, Duffy, Jaenson, Gylfe, Bonnedahl and Bergström1995b). It seems probable that domestic and synanthropic animals play a much more significant role than penguins in transmitting this infection to humans (Gauthier-Clerc et al. Reference Gauthier-Clerc, Jaulhac, Frenot, Bachelard, Monteil, Le Maho and Handrich1999; Yabsley et al. Reference Yabsley, Parsons, Horne, Shock and Purdee2012), however it is plausible that humans entering penguin colonies for research, tourism or guano/egg exploitation, or handling these birds in rehabilitation centers could be at risk of exposure to RFB or LDB.

LDB are generally non-pathogenic to seabirds (Olsén, Reference Olsén, Thomas, Hunter and Atkinson2007) and therefore are unlikely to be a conservation threat to penguins. On the other hand, RFB have been reported to cause mortality of an African penguin (Yabsley et al. Reference Yabsley, Parsons, Horne, Shock and Purdee2012), an endangered species (IUCN, 2015); studies on the epidemiology and pathology of this pathogen in African penguins could therefore help clarifying its conservation significance.

Species recorded in penguins

Phylogenetic analysis of microfilariae from the blood of Galapagos penguins revealed this is the same species as the one present in the blood of flightless cormorants (Phalacrocorax harrisi) at the Galapagos Archipelago. This parasite was closely related to mammalian-infecting Onchocercidae, but could not be conclusively identified (Merkel et al. Reference Merkel, Jones, Whiteman, Gottdenker, Vargas, Travis, Miller and Parker2007).

There are other instances in which Onchocercidae have been reported infecting penguins: adult Paronchocerca straeleni in the heart of a captive Galapagos penguin in the USA (Chabaud and Ball, Reference Chabaud and Ball1964; Bartlett and Anderson, Reference Bartlett and Anderson1986), adult D. immitis in the heart of a captive Humboldt penguin in Japan (Sano et al. Reference Sano, Aoki, Takahashi, Miura, Komatsu, Abe, Kakino and Itagaki2005), and multiple unidentified microfilariae in the eyelid skin of a Magellanic penguin undergoing rehabilitation in Brazil (Vanstreels et al. Reference Vanstreels, Silva-Filho, Kolesnikovas, Bhering, Ruoppolo, Epiphanio, Amaku, Ferreira-Junior, Braga and Catão-Dias2015a). In these cases, even though microfilariae were not observed in blood smears, they could have been present in the blood stream at some stage of the infection.

Distribution among penguin hosts

Microfilariae have only been observed in wild Galapagos penguins (Harmon et al. Reference Harmon, Harbecker and Clark1985; Merkel et al. Reference Merkel, Jones, Whiteman, Gottdenker, Vargas, Travis, Miller and Parker2007). However, other life stages of onchocercid worms have been documented in captive Galapagos and Humboldt penguins (Chabaud and Ball, Reference Chabaud and Ball1964; Sano et al. Reference Sano, Aoki, Takahashi, Miura, Komatsu, Abe, Kakino and Itagaki2005) and in a Magellanic penguin undergoing rehabilitation (Vanstreels et al. Reference Vanstreels, Silva-Filho, Kolesnikovas, Bhering, Ruoppolo, Epiphanio, Amaku, Ferreira-Junior, Braga and Catão-Dias2015a).

Vectors

The following insects have been incriminated in the transmission of avian-infecting onchocercids: biting midges (Culicoides spp.), chewing lice (Austromenopon spp., Pseudomenopon pilosum, Trinoton anserinum), mosquitoes (Aedes taeniorhynchus, Armigeres subalbatus, Culex spp., Mansonia crassipes), and black flies (Simulium spp.) (Anderson, Reference Anderson2000; Bartlett, Reference Bartlett, Atkinson, Thomas and Hunter2008; Manrique-Saide et al. Reference Manrique-Saide, Bolio-González, Sauri-Arceo, Dzib-Florez and Zapata-Peniche2008). Ecological modelling suggests that Aedes taeniorhynchus is the most probable vector of microfilariae to Galapagos penguins (Siers et al. Reference Siers, Merkel, Batailler, Vargas and Parker2010; Bataille et al. Reference Bataille, Fournié, Cruz, Cedeño, Parker, Cunningham and Goodman2012); this is corroborated by the detection of DNA from nematodes in blood meals of Aedes taeniorhynchus (Bataille et al. Reference Bataille, Fournié, Cruz, Cedeño, Parker, Cunningham and Goodman2012).

Geographic distribution

Microfilariae were documented in the blood of penguins at the Galapagos Archipelago (Fernandina and Isabela Islands) (Harmon et al. Reference Harmon, Harbecker and Clark1985; Merkel et al. Reference Merkel, Jones, Whiteman, Gottdenker, Vargas, Travis, Miller and Parker2007). However Aedes taeniorhynchus, its most probable vector, is distributed in salt marshes along the tropical and temperate coast of the Americas, including Peru, and could transmit onchocercid worms to wild and captive penguins in the region (Fig. 3C). Adult Onchocercidae have been reported infecting captive penguins in the USA (Chabaud and Ball, Reference Chabaud and Ball1964) and Japan (Sano et al. Reference Sano, Aoki, Takahashi, Miura, Komatsu, Abe, Kakino and Itagaki2005), and microfilariae were reported in the skin of penguins undergoing rehabilitation in southern Brazil (Vanstreels et al. Reference Vanstreels, Silva-Filho, Kolesnikovas, Bhering, Ruoppolo, Epiphanio, Amaku, Ferreira-Junior, Braga and Catão-Dias2015a).

Epidemiology and pathology

Local prevalence of microfilariae in Galapagos penguins ranges from 5·3 to 50% among locations (blood smears) (Merkel et al. Reference Merkel, Jones, Whiteman, Gottdenker, Vargas, Travis, Miller and Parker2007; Siers et al. Reference Siers, Merkel, Batailler, Vargas and Parker2010). Infection rate was higher in males than in females and was positively correlated to ambient temperature, precipitation and dry-season vegetation, whilst being negatively correlated to elevation and slope (Siers et al. Reference Siers, Merkel, Batailler, Vargas and Parker2010). The parasite often occurred with higher prevalence in sympatric flightless cormorants, suggesting this species might act as a reservoir of infection for penguins (Merkel et al. Reference Merkel, Jones, Whiteman, Gottdenker, Vargas, Travis, Miller and Parker2007). Infection rate of Aedes taeniorhynchus at these sites was relatively low (0·15%) (Manrique-Saide et al. Reference Manrique-Saide, Bolio-González, Sauri-Arceo, Dzib-Florez and Zapata-Peniche2008; Bataille et al. Reference Bataille, Fournié, Cruz, Cedeño, Parker, Cunningham and Goodman2012).

Parasitaemia varied greatly among individuals, ranging from 0·04 to 12 parasites per low magnification microscope field (10× objective lens) (Merkel et al. Reference Merkel, Jones, Whiteman, Gottdenker, Vargas, Travis, Miller and Parker2007). With few exceptions in which they cause vasculitis, microfilariae are seldom pathogenic per se, and the most significant health implications tend to derive from the adult parasites (Bartlett, Reference Bartlett, Atkinson, Thomas and Hunter2008). Because the infection site of the adult onchocercids recorded in Galapagos penguins is unknown, it is not currently possible to evaluate the health implications of these infections.

Implications for public health and conservation

The microfilariae detected in Galapagos penguins remain unidentified, but it seems unlikely that it could infect humans since the only onchocercid worm to infect both birds and humans is D. immitis (Bartlett, Reference Bartlett, Atkinson, Thomas and Hunter2008), which produces pulmonary disease in the latter (Simón et al. Reference Simón, López-Belmonte, Marcos-Atxutegi, Morchón and Martín-Pacho2005). However, it seems unlikely that penguins play a significant role as reservoirs of infection to humans, considering there is only one documented case of this parasite in a penguin, which is also the only known case of D. immitis in a bird (Sano et al. Reference Sano, Aoki, Takahashi, Miura, Komatsu, Abe, Kakino and Itagaki2005). Considering the high prevalence and parasitaemia with which microfilariae were observed in Galapagos penguins, an endangered species (IUCN, 2015), studies to determine the identity, adult infection site and health effects of these worms are urgent to determine their relevance as a conservation threat.