Critical chronological review of Jorge Lôbo’s disease etiology

Jorge Lôbo, in 1930 [27], reported a new human “blastomycosis,” restricted to subcutaneous tissue and different from the one described 20 years earlier by Adolfo Lutz [1], causing systemic infections in South American patients, which later became known as Paracoccidioides brasiliensis [17,56]. Later, Jorge Lôbo [28] published a detailed description of his finding but did not name the etiologic agent. The patient was a 48-year-old Brazilian man that first noted a small nodule on the lumbosacral anatomical region after a trip to the Amazonas, Brazil. The patient referred that the lesions appeared after he was bitten by a snake. Several months later, a small skin lesion appeared at the bite location. The small lesion was highly pruritic and slowly increased in size; thus, it was surgically removed. However, few months later, the lesion reappeared. When the patient came to Lôbo’s attention in Recife, Brazil, he had carried the infection for the last 19 years. The patient was described as a businessperson, but details on other occupational habits were not identified. Histopathological preparation of the infected tissues showed the presence of numerous uniformly in size yeast-like cells in chains, connected by small bridges, different from the yeast cells developed by P. brasiliensis in infected tissues. Based on this finding, he concluded that it was a new clinical entity, a notion reviewed later in his thesis to become Professor at the Department of Dermatology, Faculty of Medicine Recife, Brazil and in subsequent reviews on the subject [29,30]. Jorge Lôbo inoculated several animals with the yeast-like cells extracted from one of the nodules, without success. He also used P. brasiliensis antigens in a skin test and in complement fixation assay with negative results. These results confirmed that he was facing a new morbidity [29]. Although Jorge Lôbo inoculated culture media with the yeast-like cells extracted from the lesion in the original patient, he was not able to recover the pathogen in culture [28,29]. Therefore, he remitted his patient to the Institute Oswaldo Cruz, Rio de Janeiro, Brazil.

At Institute Oswaldo Cruz, Rio de Janeiro, Brazil, Olympio da Fonseca junior and A.E. de Arêa Leão [32] collected clinical specimens from the patient remitted by Jorge Lôbo, and a filamentous hyaline fungus was isolated from the biopsied tissues. At room temperature, the fungus grew very slow on Sabouraud dextrose agar, forming white colonies containing septate hyaline hyphae with conidia. The authors considered the isolated fungus as the etiologic agent of the disease. They argued, “The new forms of reproduction observed in the cultures of this mushroom demonstrate its great affinities with the producing agents of Gilchrist’s disease or North American blastomycosis and South American blastomycosis.” Based on Nannizzi’s proposal of the new genus, Glenosporella, they classified the fungus recovered from Jorge Lôbo’s patient as Glenosporella loboi and then deposited a sample at the Oswaldo Cruz Institute’s Fungal Collection, as isolate No 1488. In addition, a subculture of this isolate was also submitted to the Tropical Medicine Institute, Faculty of Medicine, São Paulo University, as isolate No 525. On page 12 of their proposal, they provided details of the genus and the new species fulfilling nomenclatural rules in place at that time [32].

As previously mentioned, G. loboi isolate No 1488 was identified later as a P. brasiliensis isolate; therefore, a theory was proposed to explain the meaning of tube No 1488 containing a typical P. brasiliensis isolate [31]. Because Jorge Lôbo’s disease agent cannot be cultured, thus the original isolate recovered by da Fonseca and Arêa Leão [32] probably was an environmental fungal contaminant from the patient skin, as was the norm in subsequent cases [31]. Most likely, the original environmental fungus was mistakenly labelled with a previous P. brasiliensis isolate in the Oswaldo Cruz Institute’s collection, an error common in large culture collections. The binominal G. loboi proposed by da Fonseca and Arêa Leão [32] (MB#286651) was validly published based on the isolate No 1488 = 525 recovered from the original human case diagnosed by Jorge Lôbo. The authors provided a Latin description and deposited a “type” isolate at the Institute Oswaldo Cruz, Rio de Janeiro, Brazil. Since G. loboi was later identified as a typical P. brasiliensis isolate [31] (Table 3), it became conspecific with P. brasiliensis and thus, a heterotypic synonym of the latter species (Fig 3).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 3. Phenotypic and molecular events in the taxonomy of Paracoccidioides species.

Phenotypic and molecular features used to justify the different binomials given to the etiologic agents of Lutz’s and Jorge Lôbo’s diseases involving cultivable and uncultivable Paracoccidioides species. The left chart depicts the synonyms utilized across the years for both diseases, giving the names and type culture used to justify their names (for P. lobogeorgii usually contaminant fungi), as well as the years when the binomials were first introduced. The current accepted taxon for the etiologic agents of both diseases is shown within green rectangles, whereas invalid names appeared within red rectangles as per Vilela and colleagues [26].

https://doi.org/10.1371/journal.pntd.0011220.g003

In 1943, Olympio da Fonseca [33] published the clinical features of a similar case to that of Jorge Lôbo. According to him, “a new type of granulomatose blastomicoide,” clinically different to that reported by Jorge Lôbo, was found, and, therefore, he argued that its etiology should be different. The patient was previously diagnosed by several dermatologists as a “Lôbo-type of blastomycosis, the latter term having been designated the disease described by Jorge Lôbo and whose parasite Arêa Leão and I described under the name of Glenosporella loboi. The clinical and parasitological study of the case in question, led us, however, to different conclusions.” Da Fonseca [33] described the patient as a 46-year-old man from the state of Alagoas, Brazil, whose injuries started 20 years earlier. The patient did not remember any special event but suggested mosquito or hematophagous arthropods bites as the initial event. After collecting specimens from the new case, da Fonseca [33] isolated a fungus developing white/grey colonies with aerial branching septate hyaline hyphae with spores. The isolate was deposited at the Tropical Medicine Institute, Faculty of Medicine, São Paulo University, as isolates No 481 and No 979. In addition, isolate No 481 was also deposited at the Institute of Tropical Medicine, Faculty of Medicine, University of São Paulo, Brazil. He published the new species, and on page 711, he gave details of the proposed genus and species Glenosporopsis amazonica (MB#286652). The proposal, however, lacked a Latin description, which was a requirement of the Code at that time. In the last part of the proposal, he also combined several other names in the new genus Glenosporopsis [33], most of which were later judged to concern unrelated fungi [34,57]. Because of the missing Latin description, all names introduced using da Fonseca [33] proposal are considered invalid. Moreover, the isolate was later found to be A. penicillioides; therefore, da Fonseca’s [33] binomial became conspecific with A. penicillioides and, thus, a heterotypic synonym of the latter species (Fig 3, Table 3).

Regarding the placement of the genus Paracoccidioides as synonym of Blastomyces, this proposal was contested by various authors [31] who argued that “Conant & Howell [24] in 1942, placed the genus Paracoccidioides in the synonymy of Blastomyces, a fact that was rightly contested by others. Under such conditions, we believe that the genus Paracoccidioides described by Almeida and Splendore is valid and does not violate any rules of botanical nomenclature.” Moreover, a similar proposal by Langeron and Vanbreuseghem [35] in 1958 reintroduced again the binomial Blastomyces loboi (MB#492449), but the proposal is questioned, since the authors were not fully convinced of the name change, “il semble qu’on doive la nommer Blastomyces loboi” (it seems we should name it Blastomyces loboi).

In 1948, Floriano de Almeida and Carlos da Silva Lacaz [34] studied histological preparations of several cases of Jorge Lôbo’s diseases, including the 2 isolates recovered earlier by da Fonseca and Arêa Leão as isolate No 525 of Glenosporella loboi [32] and da Fonseca isolate No 979 of Glenosporopsis amazonica [33]. They concluded that da Fonseca and Arêa Leão’s [32] G. loboi in several culture media behaved as P. brasiliensis, especially at 37°C, and that Glenosp. amazonica isolated by da Fonseca [33] does not develop at 37°C. Using the data collected from this study, along with their critical evaluation of several histological preparations from previous cases of Jorge Lôbo’s disease, they concluded, “Mycologically speaking, the fungus that causes Jorge Lobo’s blastomycosis is very similar (in tissue sections) to Paracoccidioides brasiliensis, and should, in our opinion, be included within the same genus—Paracoccidioides, with the species loboi. In our view, there are not enough reasons to place the agent of keloid blastomycosis in a new genus” (Fig 3). Floriano de Almeida and Carlos da Silva Lacaz [34] proposal was considered valid since a Latin diagnosis is not necessary when publishing a new combination. The basionym name was validly published as indicated in the paper. Interestingly, on page 16, the binominal is not given as Paracoccidioides loboi but as “Paracoccidioides, spécie loboi,” a statement that complicate the legitimacy of the proposal. However, de Almeida and Lacaz [34] proposal of P. loboi binomial, followed nomenclatural regulations, and, thus, P. loboi is currently recognized as validly published name (MB#302227).

In 1956, Raffaele Ciferri [36] from the Center for Human and Comparative Medicine, University of Pavia, Italy, along with Paulo Cordeiro de Acevedo, Sylvio Campos, and Luiz Siqueira Carneiro from Pernambuco, Brazil (MB#299852), published a review on the taxonomy of Jorge Lôbo’s etiology using isolate No 525 provided by Dr. Carlos da Silva Lacaz. The isolate was the same identified earlier by da Fonseca and Arêa Leão [32] as G. loboi. Apparently, the study was based on the concept of “cryptosporulation” introduced earlier by Ciferri [58]. They indicated that the phenotypic features of Glenosp. amazonica recovered by da Fonseca [33] should be placed in the genus Aleurisma as defined by Vuillemin [8]. No further comments were presented of the isolate recovered by the latter author. Ciferri and colleagues [36] proposed the binomial Loboa loboi using isolate No 525 (Fig 3). Since Loboa loboi was introduced as a combination based on G. loboi, the name became a heterotypic synonym of P. brasiliensis as isolate No 525 = 1488 (G. loboi) [32], proved to be identical with the latter species.

Two years later, in a one-page note, Dante Borelli [37] reviewed the findings of Siqueira Carneiro [59] regarding isolate No 1488 (G. loboi) recovered by da Fonseca and Arêa Leão [32]. The author stated that the epithet Glenosp. amazonica proposed by da Fonseca [33] was built on an Aspergillus penicillioides isolate and that G. loboi as the etiologic agent of Jorge Lôbo’s disease described by da Fonseca and Arêa Leão [32] (isolate No 1488) was based on a typical isolate of P. brasiliensis. The author’s conclusion was that “…..based in our current knowledge of this disease, the author proposes that it [Jorge Lobo’s disease] should be provisionally and concisely called lobomycosis, and its agent Lobomyces” (Fig 3). The genus Lobomyces, however, was proposed without Latin description and thus invalid. Although, 10 years later, Borelli [60] published a comprehensive review on the synonyms used for the etiologic agent of Jorge Lôbo’s disease, the binomial Lobomyces loboi appeared throughout the paper, but without Latin description. From this perspective, it is an invalid binomial. Oddly, in MycoBank (MB#493012), it appeared as validly published name, but we could not trace the citation to validate the reference in their web site.

Ozorio José de Meneses Fonseca and Carlos da Silva Lacaz in 1971 [40] published an extensive review on the etiology of Jorge Lôbo’s disease, giving details on the main clinical and histopathological features of the isolates so far studied from Brazilian cases. The study included isolates No 1488 = 525 (da Fonseca and Arêa Leão) [32] and 481 (da Fonseca) [33], biochemical tests, experimental infection in animals, and immunological tests such as immunodiffusion, precipitation test, and complement fixation. As a critique to Ciferri and colleagues [36], Fonseca and Lacaz [50] argued that they never saw “cryptosporulation” in the genus Paracoccidioides. Also, these investigators mentioned that in early studies (de Almeida and Lacaz 1948) [34], one of the authors (CSL) stated that the original “….princeps isolate was probably exchanged at the Instituto Oswaldo Cruz, as there is no doubt that the 1488 isolate at this institution fungal collection represents a typical strain of P. brasiliensis, a fungus that has never been isolated from keloid blastomycosis lesions.” To investigate this possibility, the authors studied putative isolates (likely contaminants) recovered from cases of Jorge Lôbo’s disease, kept at the Fungal Culture Collection of the Tropical Medicine Institute, São Paulo, Brazil.

The study included isolates recovered from Jorge Lôbo’s disease cases, as follows (1) isolate No 294 recovered by Silvio Campos in Pernambuco, Recife, Brazil, (2) isolate No 481 by Olympio da Fonseca [33] as Glenosp. amazonica, (3) isolate No 525 (a subculture of isolate No 1488 deposited at the Institute Oswaldo Cruz as G. loboi) isolated by O. da Fonseca and Arêa Leão [32] from the original case published by Jorge Lôbo, (4) isolate No 979 recovered by Siqueira Carneiro [59] in 1952 from the patient A.A.B. in Pernambuco Recife, Brazil, and (5) isolate No 987 = 481 = 979 recovered by Siqueira Carneiro [59] in 1952 from the patient F.V.M. in Pernambuco Recife, Brazil. The experiment results showed that isolates No 294, 481, 979, and 987 did not develop at 37°C and were negative in experimentally infected mice, whereas isolate No 525 grew at 25°C as hyaline septate hyphae with conidia, and at 37°C as multibudding yeast cells. The authors concluded that the isolates that do not developed at 37°C are unlikely the etiologic agent of Jorge Lôbo’s disease. Based on histological examination of several skin specimens from cases of Jorge Lôbo’s disease, they determined that “Clinically, histopathologically, and mycologically, the fungus that causes Jorge Lôbo’s blastomycosis is very similar to P. brasiliensis, and should, as we see it, be included within the same genus—Paracoccidioides, with the species loboi.” Interestingly, these authors make the same mistake as Floriano de Almeida and Carlos da Silva Lacaz [34] by failing to write the full binomial as Paracoccidioides loboi compromising the legitimacy of their proposed name.

To validate O.J. de M. Fonseca and C. da S. Lacaz [60] P. loboi epithet, in 1996, Lacaz [31] published the Latin description of the suggested binomial. He argued that the isolates recovered from the original Jorge Lôbo’s case were all later identified as isolates of P. brasiliensis or A. penicillioides; therefore, P. loboi has priority (Fig 3). In addition, he argued that several lines of evidence, including (1) electron microscopic studies, (2) the cell wall polysaccharides and oligosaccharides, (3) antigenic serology, and (4) the fact that P. loboi resists culturing, suggested P. brasiliensis and P. loboi, are 2 different species. He also mentioned that de Almeida and Lacaz [34] in 1948 had submitted several “neotype” tissue sections from a patient with Jorge Lôbo’s disease (J.P.S.) to 9 different herbaria around the globe to support the species “loboi.” Lacaz [31] Latin description for the binomial P. loboi was not necessary as it was introduced as a new combination. Moreover, the basionym name was not directly indicated in the paper, only indirectly by referring to the original authors of G. loboi [32]. The designated neotype was not Code compliant as a holotype was indicated in the original publication of G. loboi [32].

Three years later, Taborda and colleagues [38], using several histological preparations from cases of Jorge Lôbo’s disease, stated that, “Based upon our study of the lectotype (BPI 792295, U.S. National Fungus Collections) and tissue sections from 35 patients living in the Amazon region of Brazil, we concluded that no existing genus can accommodate this taxon. We propose a new genus and binomial for the obligate pathogen that causes lobomycosis.” Therefore, they introduced the binomial Lacazia loboi, dedicating the genus names to Dr. Carlos da Silva Lacaz for his many contributions on the etiology of Jorge Lôbo’s disease (Fig 3). Although there are several nomenclatural problems with the proposal, the name was validly published. Both an English and Latin description were provided for the new genus Lacazia, but Taborda and colleagues [38] published the type species Lacazia loboi as a new combination citing the invalid name Paracoccidioides loboi (O.J.M. Fonseca et Lacaz) [40] as basionym, but without giving a full and direct reference to the original author and place of publication as required by the Code [16]. Nonetheless, Lacazia loboi was validly published under Art. 38.1 of the Code (https://www.iapt-taxon.org/nomen/pages/main/art_38.html). The use of the term “lectotype” to denote what is, in fact, a holotype was corrected under Art. 9.10 of the Shenzhen Code (https://www.iapt-taxon.org/nomen/pages/main/art_9.html).

Almost concomitantly, Herr and colleagues [46], using molecular methodologies, sequenced several protein coding and noncoding genes using the genomic DNA recovered from yeast-like cells of a Brazilian patient with the disease. Phylogenetic analysis of the investigated genes revealed that L. loboi was a taxon independent from P. brasiliensis, supporting Taborda and colleagues’ [38] proposal. Based on Herr and colleagues’ [46] phylogenetic results, Vilela and colleagues [42] explored the possibility of isolate No 525, originally identified as G. loboi by da Fonseca and Arêa Leão [32], and labelled as P. brasiliensis by traditional methods, being the etiologic agent of Jorge Lôbo’s disease (Table 3). Using molecular tools, 9 isolates previously identified by traditional methods, including No 294 = Sterigmatomyces halophilus; No 481 = Glenosp. amazonica; No 525, and No 1488 G. loboi [Loboa loboi]; 979 and 852 = Paracoccidioides loboi, and isolates No 987 = 755 = 767 = P. loboi were investigated (Table 3). The isolates were recovered from cases of Jorge Lôbo’s disease and kept at the Institute of Tropical Medicine, Faculty of Medicine, University of São Paulo, Brazil. Phylogenetic analysis found that isolate No 294 was indeed the saprophytic yeast S. halophilus; isolates No 525 and No 1488 were typical P. brasiliensis isolates; isolates No 481, and 979 = 852 = 987 = 755 = 767 were all identified as A. penicillioides environmental contaminants (Table 3). This study further validated the genus Lacazia proposed by Taborda and colleagues [38]. Immunological and phylogenetic studies, using the Gp43 antigen and well-known conserved DNA coding regions of P. brasiliensis, further confirmed that the etiologic agent of Jorge Lôbo’s disease was an independent species closely related to cultivable Paracoccidioides species (Fig 1) [42,61,62].

Rotstein and colleagues [49] sequenced the 28S LSrDNA of a dolphin with “lobomycosis” captured in the USA coast and found that it possesses 97% identity with the DNA sequences of P. brasiliensis. This report was followed by other studies linking the dolphin pathogen to P. brasiliensis rather than to L. loboi DNA sequences [50–54]. To investigate this inconsistency, Vilela and colleagues [52] sequenced several coding DNA sequences, including the DNA sequences encoding Gp43, Kex, and chitin synthase 4, from the yeast-like cells recovered from 4 infected dolphins. Based upon the analyses, they proposed that the etiology on the skin granulomas in dolphin was caused by a variety of P. brasiliensis [55]. To validate their studies, they conducted new phylogenetic and population genetic analyses including new DNA sequences from several infected dolphins [26]. Population genetic analysis using STRUCTURE software placed the etiologic agent of dolphin skin granulomas in a monophyletic cluster sister to P. americana and L. loboi from humans formed another monophyletic cluster sister to P. lutzii, and both clusters grouped sister to P. ceti, P. americana, and P. brasiliensis (Fig 3) [26]. They argued that previous phylogenetic analysis, without the inclusion of DNA sequences from the dolphin pathogen, incorrectly placed P. loboi away from Paracoccidioides species. Furthermore, the authors argued that “The placement of P. ceti sister to P. americana DNA sequences in this study, indicates the use of phenotypic or phylogenetic characteristics without the inclusion of anomalous species, can lead to inaccuracies in the taxonomic and phylogenetic classification of these type of microbes.”

Taxonomy of Jorge Lôbo’s disease etiology

According to the findings in this review [26,31,34,38,42,54,60], a replacement name for the etiologic agent of Jorge Lôbo’s disease is proposed as follows:

Paracoccidioides lobogeorgii Vilela, de Hoog, Bagagli, and Mendoza, nom. nov.

MycoBank MB 844093.

Replaced synonym: Lacazia loboi Taborda, V.A. Taborda and McGinnis, J. Clin. Microbiol. 37: 2031 (1999) [38], non Paracoccidioides loboi (da Fonseca and Leão) F.P. Almeida and C da S Lacaz (1949) [MycoBank MB 302227].

Holotype: Brazil, tissue section from a Brazilian man with Jorge Lôbo’s disease, BPI 792295, slide preserved at BPI [38].

Etymology: Dedicated to Dr. Jorge Lôbo, who first described the disease in 1931 [28,29].

Description: As per Taborda et al. (1999) [38].

Disease nomenclature: The disease names retained are Jorge Lôbo’s disease and paracoccidioidomycosis loboi [55]. Names such as lobomycosis, lacaziosis, Lôbo’s disease, and many others based on obsolete epithets are no longer supported [55].

Regarding the current nomenclatural status of cultivable Paracoccidioides species, it was found that the proposed species were invalidly published [10,15], as their holotypes were not correctly indicated, as required by Arts 40.7 and 40.8 of the Code [16]. Therefore, the species P. americana and P. lutzii are validated here in. For P. brasiliensis, a neotype is designated. The type material preserved in a metabolically inactive stage has been deposited at the Instituto Adolfo Lutz (IAL), São Paulo, Brazil (see below).

Paracoccidioides americana Vilela, de Hoog, Bagagli and L. Mend., sp. nov. MycoBank MB 846034.

For a detailed description, see Turissini et al., Fungal Genetics and Biology 106: 22 (2017) [15].

Holotype: Brazil, Botucatu, São Paulo State, isolated from a chronic PCM patient, Pb03 preserved in a metabolically inactive state at the Núcleo de Coleção de Micro-organismos do Instituto Adolfo Lutz (IAL), São Paulo, Brazil, ID number = IAL 9802. Ext-type: Corporación para Investigaciones Biologicas, CIB_Pb03 = B26.

Paracoccidioides brasiliensis (Splend.) F.P. Almeida, Anais Fac. Med. Univ. S. Paulo 5: 134 (1930) MycoBank MB 258811.

For a detailed description, see Turissini et al., Fungal Genetics and Biology 106: 22 (2017) [15].

Holotype: not preserved.

Neotype designated here (MBT 10010030): Brazil, Botucatu, São Paulo State, isolated from a chronic PCM patient, Pb18 preserved in a metabolically inactive state at the Núcleo de Coleção de Micro-organismos do Instituto Adolfo Lutz (IAL), São Paulo, Brazil, ID number = IAL 9803. Ex-type: Corporación para Investigaciones Biologicas, CIB_Pb18 = B17.

Paracoccidioides lutzii Vilela, de Hoog, Bagagli and L. Mend., sp nov. MycoBank MB 846033.

For a detailed description, see Teixeira et al., Med Mycol 52: 26 (2014) [10].

Holotype: Brazil, Goiás state, Goiânia, IPTSP, Universidade Federal de Goiás, clinical isolate, 1992, M.R.R. Silva Pb01, preserved in a metabolically inactive state at the Núcleo de Coleção de Micro-organismos do Instituto Adolfo Lutz (IAL), São Paulo, Brazil, ID number = IAL 9804. Ex-type: American Type Culture Collection, ATCC MYA-826.