Trypanosoma cruzi in marsupial didelphids ( Philander frenata and Didelhis marsupia lis ) : differences in the humoral immune response in natural and experimental infections

Philander frenata and Didelphis marsupialis harbor parasitism by Trypanosoma cruzi without developing any apparent disease and on the contrary to D. marsupialis, P. frenata maintains parasitism by T. cruzi II subpopulations. Here we compared the humoral immune response of the two didelphids naturally and experimentally infected with T. cruzi II group, employing SDS-PAGE/Western blot techniques and by an Indirect immunofluorescence assay.We also studied the histopathological pattern of naturally and experimentally infected P. frenata with T. cruzi. P. frenata sera recognized more antigens than D. marsupialis, and the recognition pattern did not show any change over the course of the follow up of both didelphid species. Polypeptides of 66 and 90kDa were the most prominent antigens recognized by both species in the soluble and enriched membrane fractions. P. frenata recognized intensely also a 45kDa antigen. Our findings indicate that: 1) there were no quantitative or qualitative differences in the patent or subpatent phases in the recognition pattern of P. frenata; 2) the significant differences in the recognition pattern of parasitic antigens by P. frenata and D. marsupialis sera suggest that they probably “learned” to live in harmony with T. cruzi by different strategies; 3) although P. frenata do not display apparent disease, tissular lesions tended to be more severe than has been described in D. marsupialis; and 4) Both didelphids probably acquired infection by T. cruzi after their evolutionary divergence. Key-words: Trypanosoma cruzi. Humoral response. Marsupial didelphid. Resumo Philander frenata e Didelphis marsupialis mantém Trypanosoma cruzi sem desenvolver aparentemente nenhuma doença. Ao contrário de Didelphis marsupialis, Philander frenata mantém subpopulações do grupo T. cruzi II. Aqui, nós comparamos a resposta imune humoral de dois didelfídeos natural e experimentalmente infectados com o grupo T. cruzi II, empregando as técnicas de SDS-PAGE/Western blot e por um método de imunofluorescência indireta. Também estudamos os padrões histopatológicos de P. frenata infectado natural e experimentalmente por T. cruzi. O soro de P. frenata reconheceu mais antígenos que o de D. marsupialis. Polipeptídeos de 66 e 90kDa foram os antígenos mais reconhecidos por ambas espécies nas frações enriquecidas de membranas solúveis. P. frenata reconheceu intensamente o antígeno de 45kDa. Nossos resultados indicam que: não existem diferenças quantitativas e qualitativas nas fases subpatente e patente no padrão de reconhecimento de P. frenata; as diferenças no padrão de reconhecimento de antígenos de parasitas pelos soros de P. frenata e D. marsupialis sugerem que provavelmente essas espécies “aprenderam” a viver em harmonia com T. cruzi por estratégias distintas; embora P. frenata não apresente doença aparente, lesões tissulares tendem a ser mais severas do que as descritas para D.marsupialis; ambos didelfídeos provavelmente adquiriram a infecção por T. cruzi depois de sua divergência evolucionária. Palavras-chaves: Trypanosoma cruzi. Resposta humoral. Marsupial didelfídeo. 1. Laboratório de Biologia de Tripanosomatídeos. 2. Laboratório de Neurovirulência, Biomanguinhos. 3. Laboratório de Bioquímica de Tripanosomatídeos do Instituto Oswaldo Cruz da Fundação Oswaldo Cruz, Rio de Janeiro, Brazil. Address to: Dr Ana Maria Jansen. Lab. de Biologia de Tripanosomatídeos/Dept de Protozoologia/IOC/ FIOCRUZ. Av. Brasil 4365, 21045-900 Rio de Janeiro, RJ, Brasil Fax: 55-21-2560-6572 e-mail: jansen@gene.dbbm.fiocruz.br Recebido para publicação em 11/3/2002.

Trypanosoma cruzi circulates in sylvatic environments among dozens of species of mammalian reservoirs in complex transmission cycles 9 15 .The significant heterogeneity of the species observed based on molecular biological and biochemical markers recognized in the subpopulations of T. cruzi two main groups: T. cruzi I (associated mainly to the sylvatic transmission cycle) and T. cruzi II (associated mainly to the domestic transmission cycle) 2 .
The marsupials Philander frenata and Didelphis marsupialis (Marsupialia, Didelphidae), are probably the most ancient sylvatic reservoirs of Trypanosoma cruzi 23 .In nature these two didelphid marsupials share the same ecotope and display similar alimentary habits.P. frenata and D. marsupialis are both endemic species from the Atlantic Coastal Rain Forest.Both species maintain the parasitism by T. cruzi from an early age onward 7 14 , without clinical signs of disease.Moreover, D. marsupialis support without tissular lesions T. cruzi II infection and only scarce monocytic infiltrates were described in T. cruzi I experimentally infected D. marsupialis 5 .Nevertheless significant differences have been described in the course of the natural and experimental infections by T. cruzi of D. marsupialis and P. frenata 19 .D. marsupialis controls experimental infections with Y strain (T.cruzi II) from an early age while still marsupium dependent.Indeed, the T. cruzi II strains experimentally infected animals display a patent parasitemia only detectable during a short period through rare positive fresh blood smears examinations and hemocultures become rapidly and consistently negative a few weeks after inoculation.No parasitism in the scent glands is observed and the serological titers evaluated by Indirect Fluorescent Antibody Test (IFAT) are low 12 13 .Moreover some animals are able to eliminate the infection.Another picture is observed in D. marsupialis experimentally infected with F and other T. cruzi I strains.In this case the animals present a stable and long-lasting infection with high parasitemias and percentages of positive hemocultures during follow up.Furthermore, parasitism of the scent glands is detected in high percentages.A correlation between the rise of serological antibodies and the drop of parasitemia (that reaches a peak of 10 6 parasites/ml) was observed and suggested to the authors that the early control of Y strain infection was probably achieved independently from the humoral immune response, but this was important for the control of infections with T. cruzi I strains 13 .The same authors observed also through immunoblottings of sera of experimentally and naturally infected animals that D.marsupialis recognized significantly fewer T. cruzi antigens than conventional laboratory animals and humans.
P. frenata was demonstrated to be able to maintain both T. cruzi I and T. cruzi II.It seems therefore that this species is not so strict a selective filter as D. marsupialis.Experimental infections can result in high parasitemia but parasitism in scent glands had never been recorded 19 .
In order to better understand the different features of the interaction of T. cruzi with P. frenata and D. marsupialis its probable most ancient reservoir, we endeavored to compare the humoral immune response of both species in natural and experimental infections.We also studied the histopathological pattern of naturally and experimentally infected P. frenata.

Animals.
Young recently weaned P. frenata and D. marsupialis 100 to 120-day-old animals reared in our laboratory were used in experimental infections.Naturally infected P. frenata and D. marsupialis were captured in the Caleme area near Teresópolis (Rio de Janeiro, Brazil).
Parasites.The Trypanosoma cruzi strains used in the experiments were as follows: MHOM/BR/53/Y (T.cruzi II), isolated from a chagasic patient 22 , MDID/BR/ 92/C13 (T.cruzi II), isolated from a naturally infected P. frenata, Teresópolis, RJ, MDID/BR/92/G645 (T.cruzi I) isolated from a naturally infected D. marsupialis, M000/ BR/00/F (T.cruzi I).The epimastigote forms were grown in LIT medium (Liver Infusion Tryptose Medium) supplemented with 10% fetal calf serum.For antigen preparation epimastigotes were harvested in the log phase.The metacyclic trypomastigote forms were obtained after spontaneous metacyclogenesis in Agar-LIT, counted in Neubauer chamber and adjusted so as to yield 500 or 1,000 parasites per gram of body weight to inoculate in didelphids.The molecular characterization of the isolates (T.cruzi I and II groups) were done by Dr. Octavio Fernandes, Departamento de Medicina Tropical/IOC/FIOCRUZ. Antigen to IFAT.The antigen, consisting of epimastigote forms of T.cruzi F strain, incubated at 27 o C, were centrifuged (3,000g, 20 minutes), the sediment washed twice in PBS (Phosphate Buffered Saline-0.15M,pH 7.2), re-suspended in about the initial culture volume of one percent formalin in PBS.The stock antigen was kept in refrigerator and adjusted to 40 parasites per microscopic field examination (40x) in final preparations.

Inoculations.
All animals were subcutaneously inoculated in the inner part of the right thigh.One litter (five specimens) of P. frenata were inoculated with C13 (T.cruzi II) metacyclic trypomastigotes (500 parasites per gram of body weight), one litter (five specimens) of P.frenata inoculated with Y (T.cruzi II) metacyclic trypomastigotes and one litter (five specimens) of D. marsupialis were inoculated with C13 strain (T.cruzi II) (1,000 parasites per gram of body weight).(Observation-In another paper, Jansen et al 10 14 studied opossums inoculated with Y strain, so we did not repeat that experiment.In this paper we inoculated another strain (C13) in D. marsupialis belonging to the same group (T.cruzi II).
Parasitological follow-up.Parasitemia was followed up by fresh blood smears examination every other day.Monthly hemocultures were performed in NNN with a LIT overlay.Weekly examination for parasites of the scent glands was performed by gentle manual squeezing of the glands.
Sera.P. frenata and D. marsupialis sera from natural and experimental infections were obtained from blood samples taken from the tail vein.The sera from experimental animals were collected before inocula and at 30-day intervals pos inocula during the follow up.The sera were stored frozen at -20 o C and used in Indirect Fluorescent Antibody Test and Western Blot assays.
Indirect Fluorescent Antibody Test (IFAT).An IFAT was performed according to Camargo et al 4 .Positive c o n t r o l s e r u m w a s o b t a i n e d f r o m P. f r e n a t a/ D. marsupialis immunized with T. cruzi antigens, and negative control serum was obtained from uninfected animals, born and kept in captivity.Diluted (1:10 up to 1:5120) naturally and experimentally infected and control didelphids sera, rabbit antisera (1:40) to P. frenata/ D. marsupialis immunoglobulins (Ig) and a fluorescein conjugated anti-rabbit Ig (Sigma) diluted with Evans blue (1mg%) were used.The sera, antisera and conjugate were incubated for 40 minutes at 37 o C, twice washed in PBS for five minutes and mounted with a coverslip and buffered glycerin (pH 8.0).The fluorescent reactions were read under a 40x objective binocular microscopic Zeiss (HBO50W) and barrier filters.
Histopathology.The naturally and experimentally infected P. frenata by T. cruzi were killed by injection of Ketalar (ketamine cholridrate.The following viscera were collected: heart, stomach, striated muscle, liver.Tissue samples were fixed in phosphate-buffer formaldehyde, processed for paraffin embedding and stained with hematoxylin/eosin (HE).

RESULTS
Parasitological Follow Up.Experimental infections in P. frenata with C13 strain resulted in a long period of patent parasitemia (70 days): the parasitemia peak reached 2x10 4 parasites/ml and 33% of the hemocultures performed during the follow up of the subpatent phase were positive (Table 2).No patent parasitemia was observed in P. frenata inoculated with Y strain but 95% of the hemocultures performed during the follow up were positive (Table 1).Experimental infection in D. marsupialis with C13 resulted in patent parasitemia detectable only through rare positive fresh blood smears examination.Only 16.6% of the hemocultures performed during the follow up were positive (Table 2).
Naturally infected P. frenata and D. marsupialis (Table 2) showed 45% and 16.6% positivity, respectively, in the hemocultures performed during the follow up.No parasites were observed in the lumen of the scent glands.

Reactivity of infected marsupial didelphids sera with T. cruzi antigens.
Although both didelphids species recognized antigens in a range of 29-116kDa of T. cruzi I and T. cruzi II isolates, the pattern of recognition was significantly different.P. frenata recognized more intensively a larger number of T. cruzi antigens than D. marsupialis (Figures 1 and 2).Concerning the number of bands and intensity of recognition, both didelphids marsupial species recognized more antigens in experimental infections (Figure 1) than natural ones (Figure 2).Experimentally and naturally infected P. frenata and D. marsupialis by T. cruzi recognized more antigens from soluble fraction than membrane fraction ones (data not shown).Sera from P. frenata and D. marsupialis recognized more intensively the following antigens: 66 and 90kDa in both soluble and enriched membrane fractions and 45kDa on soluble fraction.
Histopathology.Natural infections: a moderate interstitial myocarditis with involvement of the subendothelial layer was observed in one of the naturally infected P. frenata.The inflammatory reaction predominated in the striated muscular fibers and was focal mononuclear cells without fibrosis and not related to the tissue parasitism.Experimental infections: the experimentally infected P. frenata with Y strain displayed

DISCUSSION
Marsupials are known to be one of the most ancient mammals, since they appeared in the late cretaceous period (80 million years ago) and are probably the most ancient reservoirs of T. cruzi 21 .In the sylvatic environment P. frenata and D. marsupialis share the same habitat but only D. marsupialis frequent human dwellings.Although previous studies described marsupials as dull and primitive mammals with a weak immune response, recent reports together with the present study do not confirm these aspects.Indeed opossums are very adaptable animals since they resist several pathogenic microorganisms, new world snake venoms and take advantage of human colonization.Their immune response is also comparable to that of placental as described in more recent reports 13 18 .
Although the two species present subpatent parasitemia, when inoculated with Y strain the follow up of P. frenata resulted in significantly higher hemoculture rates than was described for D. marsupialis 13 14 .
Although some common features were seen, significant differences in the recognition pattern of T. cruzi antigens between D. marsupialis and P. frenata could also be observed.
A noticeable difference in T. cruzi infection antigen recognition pattern between P. frenata and D. marsupialis was that D.marsupialis species recognized less intensively a fewer number of antigens after a two fold higher inoculum of a T. cruzi II strain.A previous work 13 , demonstrated that D. marsupialis recognizes fewer antigens on Y strain, also a T. cruzi II strain, than on F strain (T.cruzi I).This intense antigen recognition by P. frenata sera could explain the more severe lesions observed in histopathological studies in this species (Figure 3).
The more intense recognition of T. cruzi antigens in the experimentally infected didelphids than in the naturally infected animals is probably due to a larger inoculum in experimental conditions than one would expect to occur in nature.Moreover it should also be considered that the route of infection is a determinant of differences in the recognition pattern -it is worthy of mention that in nature animals are probably infected by predatory habits.
A common feature was the early recognition of the total spectrum of Trypanosoma cruzi antigens although experimentally infected P. frenata recognized twofold m o r e a n t i g e n s t h a n e x p e r i m e n t a l l y i n f e c t e d D. marsupialis.Similar experiments in mice concluded that distinct mouse strains differed significantly concerning the spectrum of recognized antigens during the course of infection.Moreover, the authors correlated these differences with susceptibility or resistance 3 10 .Significant differences in the antigen recognition pattern between early infection stage sera samples and late

Table 1 -
Serological and parasitological data on P. frenata and D. marsupialis subcutaneously inoculated with culture metacycle forms of Trypanosoma cruzi (200 metacyclics per gram of body weigth).

Table 2 -
Serological data and parasitological follow up of naturally infected P. frenata and D. marsupialis by T. cruzi.