Comparative morphometric evaluation of hepatic hemosiderosis in wild Magellanic penguins ( Spheniscus magellanicus ) infected with different Plasmodium spp . subgenera

Avian malaria is one of the most important diseases of captive penguins. We employed morphometric techniques to evaluate hepatic hemosiderosis in rehabilitating wild Magellanic penguins (Spheniscus magellanicus) that were negative (n = 9) or naturally infected by different subgenera of Plasmodium spp. (n = 24), according with: Plasmodium subgenera (Haemamoeba, Huffia, Other lineages, and Unidentified lineages), severity of Plasmodium histopathological lesions, and concurrent diseases, age class (juvenile or adult plumage), sex (male, female or not determined), body score (emaciated, thin, good, excellent, not available), molt, presence or absence of oil contamination upon admission, iron supplementation, and rehabilitation center. The percentage of the area occupied by hemosiderin was called ‘Index of Hepatic Hemosiderosis (IHH)’. Plasmodium-positive females presented significantly higher IHH values (17.53 ± 12.95%) than males (7.20 ± 4.25%; p = 0.041). We observed higher levels of congestion (p = 0.0182) and pneumonia (p = 0.0250) severity between Unidentified lineages vs. Huffia. We believe that the hepatic hemosiderosis observed in this study was multifactorial, the result of pathological processes caused by malaria, molting, hemoglobin and myoglobin catabolism during migration, anemia, concomitant diseases, and iron supplementation, all possibly potentiated by decreased liver mass. Further studies are needed to clarify the mechanisms of these hypotheses.

Iron homeostasis in birds, as in other vertebrates, is regulated mainly through intestinal absorption and excess iron is stored as intracellular aggregations of ferritin or hemosiderin, particularly in hepatic and splenic cells (NORAMBUENA & BOZINOVIC, 2009).Hemosiderosis is the accumulation of iron in hepatic and splenic sinusoidal macrophages and hepatocytes without significant architectural, cellular, or functional alterations (LOWENSTEIN & MUNSON, 1999).Iron absorption is influenced by age, iron and health status, gastrointestinal conditions, the amount and chemical form of the iron ingested, and the amount and proportions of various other organic and inorganic components of the diet (CRISSEY et al., 2000;SHEPPARD & DIERENFELD, 2002;PEREIRA et al., 2010).Although there is no conclusive evidence that hepatic hemosiderosis has any clinical significance in birds, it has been associated with concurrent infections, neoplasia, parasitism, anemia, hepatopathies, intoxications, starvation, oil exposure, trauma, lead poisoning, and the physiological cycle (e.g., egg production and laying, migration, and molting) (CORK et al., 1995;GOTTDENKER et al., 2008;PEREIRA et al., 2010;KLASING et al., 2012).
Hemosiderosis, especially hepatic, has been previously reported in Plasmodium-infected penguins (FIX et al., 1988;GRILO et al., 2016), believed to be caused by the build up of hemosiderin in visceral macrophages following intravascular hemolysis and phagocytosis of parasitized cells (VANSTREELS & PARSONS, 2014).However, these studies did not evaluate if such condition varied among Plasmodium spp.subgenera or considered other possible causes of hemosiderosis aside from Plasmodium spp.infection.Considering the importance of avian malaria to penguin conservation and the limited amount of information regarding iron metabolism and the pathogenesis of avian malaria in birds, our objective was to quantify the hepatic hemosiderosis in Plasmodium spp.-negative and -naturally infected Magellanic penguins that died while under care in rehabilitation centers in Brazil, in regards to the presence of Plasmodium and Plasmodium species/lineages, and within the context of these birds' biology and individual history.
could not be morphologically or molecularly characterized).
The positive birds evaluated in this study either (a) died suddenly, (b) presented antemortem malaria-associated clinical signs, dying after unsuccessful treatment, or (c) were euthanized for reasons unrelated to malaria (e.g., aspergillosis).Nine Plasmodium-negative Magellanic penguins were selected based on individual history and a negative diagnosis for Plasmodium spp. on both nested PCR and blood smears.Selected Plasmodium-negative birds died without presenting any clinical signs of concurrent infections, and remained in captivity for similar periods as the positive birds.
All birds from both groups (positive and negative) were kept under similar infrastructure and husbandry conditions, and diet and iron supplementation varied among rehabilitation facilities (Table 1).All birds were examined postmortem, and liver samples were collected and fixed in 10% formalin for a minimum of 72 hours before processing.Paraffin embedded liver fragments were sectioned at 5 μm and stained with haematoxylin-eosin and Perls stains, as described by Luna (1992) (Figure 1).The Index of Hepatic   Hemosiderosis in penguins with malaria v. 28, n. 1, jan.-mar. 2019 71/79 71 Hemosiderosis (IHH) was evaluated according with the presence of infection, Plasmodium species/lineage, malaria severity (based on the classification of the histopathological findings observed in heart, lung, liver, and spleen samples: mild, mild to moderate, moderate, moderate to severe, and severe -Table 2), concurrent diseases and/or significant lesions, individual history (age class  3).To establish the IHH, slides were evaluated by light microscopy and captured images where analyzed by a computerized image analyses software (Image Pro  Plus, version 5.1.2.59, Media Cybernetics) to determine the percentage of liver tissue occupied by hemosiderin pigment.The intersection of the two longest axis of each bird's hepatic section was determined as the center of the sample, and used as the reference field to capture images at 20X magnification.Another 8 images were captured as TIFF files, at 50 μm of this point, and 45° intervals, under the same lighting conditions and magnification.The malarial pigment, called hemozoin, stains brown with Perls stain, while hemosiderin stains bright turquoise blue, promoting an easy visual distinction between the two pigments.Hemosiderin deposits were semi-automatically delineated on the Perls stained slides, starting at the lowest blue intensity to the most intense, to avoid the inclusion of artifacts and the background.
When overlapping areas of formolic pigment and hemosiderin deposition where present, the areas of formolic pigment deposition where individually marked and deducted from such areas.
In order to standardize the technique, a maximum digital zoom of 50% was stipulated.The average of the 9 microscopic fields was considered the IHH.One-way ANOVA/Kruskal-Wallis test followed by Dunn's multiple comparisons test and Mann-Whitney tests were used to verify differences between the evaluated parameters and groups.The Spearman test was used to evaluate the correlation between quantitative variables.The significance level was defined as 5%.Data were analyzed with GraphPad Prism software (Version 4.02 for Windows, GraphPad Software, Inc.).
All samples used in this study were collected in full compliance with specific federal permits issued by the Brazilian Ministry of Environment and approved by the Biodiversity Information and Authorization System (SISBIO 20825-8).All procedures were performed according to the Ethical Committee in Animal Research of the School of Veterinary Medicine and Animal Sciences, University of São Paulo (Process numbers 1757/2009 and 9411100414).

Results and Discussion
The pulmonary congestion, intense inflammatory response (especially in liver and spleen), and hemosiderosis found in our birds are in agreement with previous microscopic descriptions of Plasmodium lesions in penguins (FIX et al., 1988;VANSTREELS & PARSONS, 2014;GRILO et al., 2016).However, there were no significant differences neither in the presence of moderate to severe hepatic hemosiderosis or in IHH values between Plasmodium-negative vs. Plasmodium-positive individuals or in regards to the Plasmodium species/lineage.The epidemiology and pathogenicity of the Plasmodium lineages previously diagnosed in our birds (VANSTREELS et al., 2015) raised the question on whether the Plasmodium subgenus Haemamoeba was more pathogenic to these penguins than other Plasmodium subgenera.Unfortunately, we were not able to find a clear answer to that question; these lineages are not completely understood and have been recorded only recently.However, we found higher levels of Hemosiderosis in penguins with malaria v. 28, n. 1, jan.-mar. 2019 75/79 75 congestion (p = 0.0182) and pneumonia severity (p = 0.0250) between Unidentified lineages vs. Huffia (Figure 2).
Plasmodium-positive females presented significantly higher IHH values (17.53 ± 12.95%) than males (7.20 ± 4.25%; p = 0.041) (Figure 3), in contrast with findings in toucans and aracaris (Ramphastidae family) (CUBAS, 2008) and in red-spectacled amazons (Amazona pretrei) (Psittacidae family) (PEREIRA et al., 2010), who failed to observe correlation between the rate of hepatic hemosiderosis and sex.Hemosiderosis has been previously associated with egg production (GOTTDENKER et al., 2008), which could not have played a role in our findings based on the age and underdeveloped gonads observed in necropsy.It is not clear why females presented higher IHH values than males; however, one may not eliminate the possibility of sampling influence, either due to the limited number of evaluated birds or higher number of females than males, which reflects previous studies documenting a female-biased sex ratio in Magellanic penguins stranded along the Brazilian coast (VANSTREELS et al., 2013).Comparison between the Pneumonia score (A) and Congestion Score (B) among Plasmodium species/lineages, and IHH in female vs. male Plasmodium-positive Magellanic penguins (Spheniscus magellanicus) (C) are shown in Figure 3.
Concurrent diseases were not statistically relevant.According to the institutions' protocols, many of the birds analyzed in this study were considered anemic, and received iron supplementation, which we believe may have also played a role in the hepatic hemosiderosis observed in this study.The physiological values of circulating iron concentration in Magellanic penguins have been previously evaluated (GHEBREMESKEL et al., 1989); however, to the authors' knowledge, there are no available data regarding iron hepatic levels for this species.Furthermore, the posology of iron supplementation is usually ad-hoc, and might not address the physiological requirements of these birds.In addition, iron represents a critical nutrient for pathogens and tight regulation of iron ("iron-withholding") is a paramount defense strategy of the host's innate immune system (KOSKI & SCOTT, 2003;JOHNSON, 2008;BEERNAERT et al., 2010;CASSAT & SKAAR, 2013).Thus, individuals suffering from iron overload have an enhanced risk of infection (BEERNAERT et al., 2010).Some of our specimens presented gastrointestinal nematodes, air sac parasites, and aspergillosis.It has been suggested that chronic infection or repeated acute infections could lead to greater accumulation of iron in the liver of birds (KLASING et al., 2012).Unfortunately, the low number of concurrent diseases observed in this study prevented statistical evaluation.Aspergillosis and parasitosis could have not only contributed to hemosiderosis, but also thrived upon iron supplementation, further debilitating the affected birds and contributing to their death.Due to all the factors discussed above, we recommend caution in the employment of iron supplementation to Magellanic penguins undergoing rehabilitation.
The IHH values in relation to presence of infection, Plasmodium species/lineage and pathology are summarized in Table 3.
Individual history parameters were not statistically relevant to the IHH values either.Magellanic penguins molt soon after breeding season, immediately prior to their winter migration (SILVA-FILHO & RUOPPOLO, 2014).Anseriformes and Passeriformes of the Sturnidae family showed an association between the hepatic level of hemosiderosis in this same period, with increased amounts of positive iron staining primarily located within lysosomic organelles of parenchymal hepatic cells (WARD et al., 1988;CORK et al., 1995).Such findings are possibly related to an increased hematopoietic activity and consequent erythrocyte production required to enhance oxygen uptake during periods of increased thermogenesis or changes in the levels of thyroid hormone involved in the molting process (WARD et al., 1988;CORK et al., 1995;OSBORN, 1979;CORK, 2000).Although not statistically relevant, some birds (4, 5, 6, 9, 18, 19, and 20) were molting during the study, so it is not possible to completely discard the hypothesis that residual iron from molting might have contributed to the hepatic hemosiderosis observed in our birds.
Fat can be severely depleted during migration, forcing birds to use body protein from the heart, pectoral, and leg muscles and the digestive organs (intestine, gizzard, and liver) as a second fuel option (LINDSTRÖM & PIERSMA, 1993;BATTLEY et al., 2000;BAUCHINGER & BIEBACH, 2001;SCHWILCH et al., 2002;MCWILLIAMS et al., 2004).Protein metabolism stems to a large extent from the skeletal muscles, which in penguins are especially rich in myoglobin, releasing the iron present in heme group of myoglobin, and possibly contributing to the development of hemosiderosis (SCHWARTZ et al., 1973;BAUCHINGER & BIEBACH, 2001;PONGANIS et al., 2010).The majority of migrating Magellanic penguins stranded in Brazil are in poor body condition, with markedly atrophied skeletal muscles and no subcutaneous fat deposits (CARDOSO et al., 2011).We assessed the true nutritional condition of these birds: body mass (ranging from 3.5-5 kg in wild specimens) and body condition score (SILVA-FILHO & RUOPPOLO, 2014).The only 2 emaciated birds (1 and 19), however, presented IHH values within the range found in this study.Nevertheless, increased iron deposition has been reported in malnourished black-necked swans (Cygnus melanocoryphus) (NORAMBUENA & BOZINOVIC, 2009), in the liver of rats subjected to food restriction (ROZENTAL et al., 1967), and in some severe cases of prolonged malnutrition in obese humans subjected to a variety of fasting periods (MITCHELL & JOHNS, 2008).It is important to consider that malnutrition may have a role in hemosiderosis in migratory bird species.
Intoxication (e.g., petroleum products) and hemoparasitosis are known causes of hemolytic anemia (FRY & LOWENSTINE, 1985).Systemic absorption of petroleum and its derivatives following acute or chronic oil exposure causes hemolytic anemia in birds (LEIGHTON, 1986;KHAN & RYAN, 1991;DINHOPL et al., 2011).Oil pollution is an important conservation threat to Magellanic penguins on wintering grounds (GARCÍA-BORBOROGLU et al., 2006).In this study, however, IHH was not significantly different between individuals with and without history of acute oil contamination upon arrival at the rehabilitation centers, suggesting that (a) oil exposure may have been too superficial or rapidly mitigated and therefore systemic absorption was minimal, and/or (b) other factors may have played a greater role in producing/modulating hemosiderosis in a manner that surpassed or contradicted any detectable effects from oiling.Nevertheless, these birds' history during migration is unknown, so we were not able to completely exclude the potential influence of chronic oil exposure in the hepatic hemosiderosis seen in this study.
Although previous studies (CRANFIELD et al., 1994;ONG et al., 2006) suggested that the low parasitaemia levels seen in Plasmodium-infected penguins are not sufficient to cause clinical anemia, Plasmodium-infected erythrocytes may suffer hemolysis or induce erythrocyte sequestration and extravascular hemolysis (VALKIŪNAS, 2005), releasing iron, which is subsequently stored in macrophages, as ferritin and hemosiderin (RATLEDGE, 2007).Eventually, cases 2 and 22 presented spleen hemorrhage and rupture, whereas case 8 presented disseminated intravascular coagulation -acute vascular complications likely associated with avian malaria that if not the primary cause, contributed to the development of secondary hemosiderosis.Another possible cause of anemia was the infestation by gastrointestinal nematodes (cases 4, 10, 15, 18 and 20).
Liver mass changes caused by the spontaneous long-term fasting characteristic of the annual cycle of penguins during reproduction and molt (GROSCOLAS & ROBIN, 2001) and/or due to concurrent diseases could have interfered with the morphometric evaluation of hepatic hemosiderin.It is unclear if changes in digestive organs of birds during migration are related to cell size (hypertrophy/atrophy) and/or cell number (hyperplasia/hypoplasia) (WAGENMAKERS et al., 1984).Further morphometric studies are needed in order to test these possibilities.Both the amount of dietary iron and length of exposure to dietary iron are factors in hepatic iron accumulation in birds (CRISSEY et al., 2000;HELMICK et al., 2011).Unfortunately, data regarding their diet while in the Brazilian continental shelf during the winter migration is limited and there is no such information regarding the species offered to the studied birds while under care.To our knowledge, the only suggested nutritional iron concentration for adult penguin diets in a dry matter basis (60 -80mg/kg) is based on requirements of domestic poultry, cats, and inferences from wild foods composition (WALLACE & WALSH, 2005;AZA Penguin Taxon Advisory Group, 2014).Further studies are needed to determine the nutritional iron requirements of Magellanic penguins.
Finally, our findings show that other factors aside from avian malaria likely played a role in the presence of hepatic hemosiderosis in Plasmodium-infected and Plasmodium-negative Magellanic penguins under rehabilitation in Brazil, and the importance of considering the species life cycle and clinical history when evaluating hepatic hemosiderosis.We propose that the increased hepatic hemosiderosis noted in this study was multifactorial, the result of pathological processes caused by malaria, molting, hemoglobin and myoglobin catabolism promoted by the intense energetic demands of Magellanic penguins' winter migration, anemia, concomitant diseases, and iron supplementation while under care, all possibly potentiated by an altered liver mass.

Figure 3 .
Figure 3.Comparison between the Pneumonia score (A) and Congestion Score (B) among Plasmodium species/lineages, and the index of hepatic hemosiderosis (IHH) in female vs. male Plasmodium-positive Magellanic penguins (Spheniscus magellanicus) (C).

Table 1 .
Plasmodium-positive and control Magellanic penguins (Spheniscus magellanicus) according with rehabilitation center, iron supplementation, administration criteria and diet.

Table 2 .
Plasmodium-positive and control Magellanic penguins according with the Index of Hepatic Hemosiderosis (IHH), study group, Plasmodium lineage, Plasmodium species and malaria severity.

Table 3 .
Plasmodium-positive and control Magellanic penguins according with the Index of Hepatic Hemosiderosis (IHH), rehabilitation center, age class, sex, body mass on admission (g), body score, hematocrit (Ht), presence or absence of oil upon admission (oil), moult, iron supplementation (Fe) and concurrent disease and/or significant lesion.IPRAM = Instituto de Pesquisa e Reabilitação de Animais Marinhos (Espírito Santo state, Brazil); R3 = Associação R3 Animal (Santa Catarina state, Brazil); CRAM = Centro de Recuperação de Animais Marinhos (Rio Grande do Sul state, Brazil); b J = Juvenile; A = adult; c F = female; M = male; ND = not determined; a d N/A = Information not available.