1. Filariae systematics

A large number of species from around the world were described or redescribed by Odile Bain. She published a dichotomous key of the Filarioidea, together with hypotheses on the evolution of the superfamily, including the position of the human filariae, using, initially, traditional morphological methods and, later, supported by the first molecular phylogeny of these parasites. She described new filarial species, and she largely studied genera and species of the Onchocercidae such as:

• The genus Onchocerca
• The Dipetalonema lineage
• The genus Dipetalonema sensu stricto parasite of South American monkeys
• The genus Mansonella and its sub-genera
• The genus Molinema from caviomorph rodents
• The genus Cercopithifiaria from mammals worldwide
• The genus Litomosoides from neotropical rodents, marsupials, and bats
• The genera Litomosa and Josefilaria from old-world bats
• The filariae Edesonfilaria from Dermaptera, Chabfilaria from Xenarthra, and Cherylia from South American marsupials
• Filariae of birds: Dessetfilaria, Andersonfilaria, and Versternema.
• Filariae of reptiles
• Filariae of amphibians
• The Stephanofilaria, skin parasites of ungulates
• Brugia malayi: geographical strains identified by discriminative characteristics based on cuticular ornamentation of the male posterior end

She also largely contributed to the study of filarial zoonoses, such as the ones due to Meningonema perruzzi, a parasite of cercopithecid monkey; Mansonella rodhaini, a parasite of anthropoid apes; Onchocerca dewittei japonica, a parasite of wild boars in Japan; and more recently, Onchocerca lupi, a parasite of dogs.

In the last decade, she was studying the relationship between Wolbachia and filariae and their coevolution.

She also worked on other orders of nematodes, including:

• Muspiceoidea, with the reinterpretation of the anatomy of Riouxgolvania and Dioctowittus; new descriptions of the genera Lukonema, Pennisia, and Lappnema; and muspiceoid infective larva in Simulium; a dichotomous key has also been published.
• Trichinelloidea (Capillaria sensu lato, Trichomosoides nasalis)
• Rhabdiasidae and other parasites of terrestrial cold-blooded vertebrates (Chabirenia, Rhabdias)
• Oxyuroidea and Oxyuridae, with a focus on traumatic insemination, the organ of De Mann, and Gyrinicola from tadpoles

2. Vectors and transmission of filariae

One of her earliest interests was filarial morphogenesis in the vector. She clarified the fate of the microfilarial cells called R1, R2, R3, and R4 and discovered that R2, R3, and R4 cells form the rectal glands and do not divide. The R1 is a primary l mesenchymal cell that divides and produces muscle cells. The morphology and development of microfilariae and their organs were determined through comparative studies of filariae from amphibians, reptiles, birds, and mammals.

She also focused on identification of infective larvae from vectors, a difficult task, but one that was essential for estimating transmission intensity in endemic areas. At the request of WHO, she published an atlas of infective stages of filariae, using simple characteristics as a mean of identification that can be used in the field.

Her third interest was the regulation of infection in the vector, after ingestion of microfilariae or during development.

Odile established that in many filaria–vector relationships, the number of infective larvae produced is not proportional to the number of microfilariae ingested. She described two opposing types of regulation processes: limitation versus facilitation. Limitation means that the proportion of microfilariae arriving in the haemocoel decreases when the number of microfilariae ingested increases. Such examples are numerous and include Onchocerca volvulus in Simulium sirbanum and S. damnosum in the African savannah; filariae of lizards, marsupials, and South American rodents in various strains of Aedes aegypti; and Wuchereria bancrofti in A. polynesiensis. Facilitation indicates that the proportion of microfilariae arriving in the hemocoel increases with the number of microfilariae ingested, as for W. bancrofti in Anopheles gambiae A in West Africa. Both types of regulation are due to changes in the vector's stomach wall in response to stimulation by the microfilariae, causing an increased secretion of the peritrophic membrane (S. damnosum, S. sirbanum), and either death (A. aegypti) or hypertrophy of the digestive cells affected by microfilariae. These reactions suggest a comprehensive immune response in the stomach wall. Ivermectin administered to the host modifies these vector–filariae relationships. Other patterns of regulations were observed during the ingestion of microfilariae, particularly following ivermectin treatment for onchocerciasis.

Odile also demonstrated different interactions within the adipose tissue where the parasites develop in mosquitoes (Molinema dessetae and Breinlia booliati). For the first species, blood meals activate both the infected syncytium and healthy adipocytes (vitellogenesis cycle is unchanged). In B. booliati, the infected syncytium is unresponsive to the blood meal stimulus. The yield of M. dessetae infective larvae is better when mosquitoes continue to take blood meals after ingestion of microfilariae.

3. Models of filariasis

Filariae infect all vertebrate phyla except fishes, and many hundreds of species from 94 currently known onchocercid genera have been described. Most species exhibit a marked host specificity, and this includes the important medical filariae such as Onchocerca volvulus, the causative agent of river blindness, and Wuchereria bancrofti, one of the filariae responsible for lymphatic filariasis or elephantiasis. This situation prompted the search for animal models that mimic the various biological and developmental characteristics of human parasites and could provide a screening system for drugs and, eventually, experimental vaccines.

Odile's research led to the discovery of:

• Diurnal periodicity of microfilariae of Molinema dessetae in the blood of Proechimys oris, the para spiny rat from Brazil.
• Description of dermal lesions associated with skin-dwelling microfilariae of Monanema martini in Lemniscomys striatus and Arvicanthis niloticus, Cercopithifilaria roussilhoni from Atherurus africanus (the African bush-tailed porcupine), and C. johnstoni from Australian rodents and marsupials. These models have facilitated the study of filarial pathogenesis and complemented work with bovine Onchocerca species. Both rodent and bovid modeles have been used in drug trials.
• The observation that the L3 to L4 moult of Onchocerca and Dirofilaria filariae occurs just two days after infection of the vertebrate host, contrasting with the later moulting seen at seven days post infection in most other filariae, including Brugia, Wuchereria, Litomosoides, Acanthocheilonema, Monanema, and Loa. These biological events, as well as genome sequence data and the typology of endosymbiont Wolbachia, now question the validity of subfamily status of Onchocercinae and Dirofilariinae.
• Detailed description of biological features to analyse filarial infection, such as kinetics of the recovery of filariae in relation to the number of infective larvae inoculated in rodents, the relationship of filariae with the lymphatic system and cardiopulmonary circulation, the blood-feeding behaviour by filariae in the host, etc.

4. The Litomosoides sigmodontis murine model

Odile's greatest contribution to the development of filarial models was the demonstration that Litomosoides sigmodontis, a natural parasite of South American cretids, can produce patent infections in laboratory BALB/c mice.

L. sigmodontis started its “scientific career” in 1945 as L. carinii. Earlier, a filarial parasite was discovered and described by Chandler in 1931 in a Muridae Sigmodontinae, the cotton rat, Sigmodon hispidus, captured in Houston, Texas. Chandler used this material to describe the type species of a new genus, Litomosoides, in which was noted a uniquely long buccal capsule. This author transferred to this new genus a filaria described by Travassos (1919), Filaria carinii, that had been recovered from an undetermined Sciuridae, Sciurus sp. (Phyrhonotus?), São Paulo, Brazil. Shortly after, Vaz synonymized all the species of Litomosoides, including L. carinii and L. sigmodontis, into a single species referred to as L. carinii.

However, in the late 80's, Odile studied anew the nomenclatural type specimens of L. carinii from squirrels and L. sigmodontis from cotton rats and confirmed their distinctive morphological characters. As a consequence, the filaria now being maintained in laboratory mice has reverted to its original name of L. sigmodontis Chandler, 1931.

The genus Litomosoides is very diverse, with 33 species distributed in the New World. A closely related genus, Litomosa York and Maplestone, 1926, is largely restricted to the Old World. Both genera are parasitic in bats, but Litomosoides' host range also includes rodents and some marsupials. Litomosoides species from rodents and marsupials are morphologically very similar, indicating a recent diversification of Litomosoides in these zoologically distant hosts. Odile suggested that Litomosoides, like Litomosa, are primarily parasites of bats. Chiropterans are ancient (fossils were found as early as the Eocene), and they are present on the South American continent. This continent was isolated during much of the Tertiary, but during the Pleistocene era, the junction of the two Americas was achieved. Holarctic fauna, e.g. rodents, migrated to the south, where they diversified. This event promoted the capture by these newcomers of paleo-endemic Litomosoides from bats. Parasitism of marsupials by Litomosoides seems contemporary, and not prior to this second wave of diversification, because only the smaller species are parasitized.

L. sigmodontis is now established as a model of filarial infections in mice, where it presents with a spectrum of parasitological and immunological presentations that mimic those seen in human infections. BALB/c mice are susceptible to infection and can produce patent infections, although some BALB/c can be amicrofilaraemic. Other mouse strains, such as C57BL/6, are resistant to infection, and thus when Odile and her colleagues published these observations in 1992, the research community was finally provided with a tractable system for investigation of the immunology of filarial infections.

The following 20 years saw publications of detailed descriptions of the development, maturation, and behaviour of L. sigmodontis in mice, while immunological studies have defined immune responses evoked by filarial infections and the regulatory processes that lead to expression of protective immunity. Proof-of-principle of immunisation against filarial parasites has been obtained following vaccination with irradiated attenuated L3 larvae, recombinant protein, and recombinant DNA.

These achievements are all owed to Odile's ability to integrate classical zoology with high technology. Today, her generosity and enthusiasm continue to drive her colleagues along the road to elimination of filarial infections as public health problems, while maintaining a fascination and curiosity for a unique group of animals.