- Email: [email protected]
Characterization of langerhans cells in epidermal sheets along the body of Armadillo (Dasypus novemcinctus)
Available online at www.sciencedirect.com
Veterinary Immunology and Immunopathology 124 (2008) 220–229 www.elsevier.com/locate/vetimm
Characterization of langerhans cells in epidermal sheets along the body of Armadillo (Dasypus novemcinctus) Fausto Quesada-Pascual b, Rafael Jimenez-Flores a,d, Adriana Flores-Langarica b,1, Aaron Silva-Sanchez b, Juana Calderon-Amador d, Rene Mendez-Cruz a,d, Alberto Y. Limon-Flores d, Sergio Estrada-Parra b, Leopoldo Santos-Argumedo c, Iris Estrada-Garcia b, Leopoldo Flores-Romo d,* b
a Faculty of High Studies Iztacala, The National University of Mexico, UNAM, Mexico City, Mexico Department of Immunology, National School of Biological Sciences ENCB-IPN, Mexico City, Mexico c Department of Molecular Biomedicine, CINVESTAV-IPN, Mexico City, Mexico d Department of Cell Biology, CINVESTAV-IPN, Mexico City, Mexico
Received 26 April 2007; received in revised form 13 March 2008; accepted 19 March 2008
Abstract Armadillos are apparently important reservoirs of Mycobacterium leprae and an animal model for human leprosy, whose immune system has been poorly studied. We aimed at characterizing the armadillo’s langerhans cells (LC) using epidermal sheets instead of tissue sections, since the latter restrict analysis only to cut-traversed cells. Epidermal sheets by providing an en face view, are particularly convenient to evaluate dendritic morphology (cells are complete), spatial distribution (regular vs. clustered), and frequency (cell number/tissue area). Lack of anti-armadillo antibodies was overcome using LC-restricted ATPase staining, allowing assessment of cell frequency, cell size, and dendrites extension. Average LC frequency in four animals was 528 LC/mm2, showing a rather uniform non-clustered distribution, which increased towards the animal’s head, while cell size increased towards the tail; without overt differences between sexes. The screening of antibodies to human DC (MHC-II, CD1a, langerin, CD86) in armadillo epidermal sheets, revealed positive cells with prominent dendritic morphology only with MHC-II and CD86. This allowed us to test DC mobilization from epidermis into dermis under topical oxazolone stimulation, a finding that was corroborated using whole skin conventional sections. We hope that the characterization of armadillo’s LC will incite studies of leprosy and immunity in this animal model. # 2008 Elsevier B.V. All rights reserved. Keywords: Langerhans cells; Armadillo; In vivo; Leprosy; Mycobacterium leprae reservoir; Epithelial sheets
1. Introduction Abbreviations: LC, langerhans cells; DC, dendritic cells; APC, antigen presenting cells. * Corresponding author at: Department of Cell Biology, CINVESTAV-IPN, Av. IPN 2508, Col. San Pedro Zacatenco, 07360 Mexico City, Mexico. Tel.: +52 55 57 47 39 84; fax: +52 55 50 61 33 93. E-mail address: [email protected] (L. Flores-Romo). 1 Present address: Sir William Dunn School of Pathology, South Parks Road, Oxford OX1 3RE, United Kingdom. 0165-2427/$ – see front matter # 2008 Elsevier B.V. All rights reserved. doi:10.1016/j.vetimm.2008.03.011
Since the seminal work of Storrs in 1971 (Colston and Levy, 1987), the armadillo (Dasypus novemcinctus) has become the most appropriate animal model to assess various aspects of leprosy infection in vivo. It seems that the armadillo is even more susceptible to (Lepromatous) leprosy than humans, developing more severe complications (Kirchheimer et al., 1972). Although the precise
F. Quesada-Pascual et al. / Veterinary Immunology and Immunopathology 124 (2008) 220–229
reasons for this are not well known, factors such as a decreased cell-mediated immunity have been advanced, but without much experimental support. Also, the low body temperature (32–35 8C) of armadillos has been put forward, since this feature apparently coincides with the decreased temperature observed in equivalent areas where bacilli can be found in leprosy patients (Kirchheimer and Storrs, 1971). On the other hand, an histological examination of lymphoid organs in armadillo, including thymus, spleen, lymph nodes, tonsils, and bone marrow, neither reveal any ostensible alterations nor absence of any of these organs (Purtilo et al., 1975). Whether leprosy infection occurs naturally in armadillos, or was introduced upon contacts with humans, is still controversial. However, there seem to be agreement and experimental evidences strongly suggest that in certain areas of the southern USA, leprosy might be a zoonotic disease in armadillos, with up to 16–19% of wild animals positive (Lane et al., 2006; Truman et al., 1991). Interestingly, naturally acquired leprosy has also been reported in non-human primates in other regions of the world (Walsh et al., 1988). Further, in low endemic areas, human cases have been newly diagnosed where the subjects had no known human contacts for leprosy, while all had been exposed to armadillos one way or another (Bruce et al., 2000; Lane et al., 2006; Lumpkin et al., 1983; West et al., 1988). The langerhans cells constitute the epidermal chapter of the DC system. This functions to initiate immune responses and is integrated by sentinel cells strategically located at the boundaries with the external world such as the skin (Flores-Romo, 2001), the gastrointestinal (Flores-Langarica et al., 2005; Kiyono and Fukuyama, 2004), the respiratory (Garcia-Romo et al., 2004), and the genital mucosa (Jimenez-Flores et al., 2006; FloresRomo, 2001). It is at these locations that most daily encounters with environmental antigens occur, many of them are microbes and some can be infectious and pathogenic (Garcia-Romo et al., 2004; Jimenez-Flores et al., 2006; Limon-Flores et al., 2005). DC are deemed the most potent antigen presenting cell (APCs) of all, whose functions include Ag capture at peripheral locations, Ag degradation and processing, Ag transportation into the regional lymphoid tissues, and finally, induction of primary immune responses by an efficient Ag presentation (Flores-Romo, 2001). Since the discovery of Mycobacterium leprae, the skin has been considered from time to time as a potential port of entry for leprosy bacilli (Huang, 1980; Pallen and McDermott, 1986). However, perhaps due to the lack of an adequate animal model and also to the technical difficulties
221
(including absence of antibodies and other reagents) to study LC in animal species such as the armadillo, the specific role of these cells has not been assessed in vivo during leprosy infection in animal models. The ectoenzyme ATPase has been quite useful to detect LC in the skin of a wide variety of animals, especially those species for which there are no antibodies currently available. Besides humans, ATPase+ LC have been identified in rat, mice, chicken and turtles (Baker and Habowsky, 1983; Carrillo-Farga et al., 1991; PerezTorres et al., 1995; Perez and Millan Aldaco, 1994). Since this enzymatic marker seems both quite well preserved along the phylogeny as well as quite restricted to LC in the epidermis of various animals, we reasoned that if LC were present in armadillo skin they were likely to express this enzymatic activity. Although there is a seminal histology work performed on armadillo skin by Convit’s group in 1975, they used twelve different histochemical approaches including eight enzymatic stainings (except ATPase), aiming at ‘‘eluicidating macrophage–lymphocyte interactions in the immunity phenomena in leprosy’’(Campo-Aasen and Convit, 1975). However, no ATPase labelling was done, sections were used instead of epidermal sheets, and very few cells were hesitantly described as dendritic (only) when using alpha naphthyl acetate esterase. As a first step, we have undertaken the identification and quantification of epidermal LC along the body of this animal. We hope that these findings will provide a baseline for prospective studies on LC and the immune responses in the armadillo. 2. Material and methods 2.1. Animals These animals are from an armadillo colony of the Department of Immunology in The National School of Biological Sciences (ENCB-IPN), which was initiated in 1980 as part of a program to facilitate leprosy research (Quesada-Pascual et al., 1987). Since armadillos are not easy to keep in captivity (on average, 40% of the captured animals die within the first 3 months of life in captivity), we decided to use only four healthy adult (3 years old) animals, two males and two females. These four animals were already well adapted since they have lived in these facilities for at least 6 months. Upon arrival, animals are carefully examined for the presence of acid-fast bacilli (AFB) and for any other obvious infections and injuries, these exams include blood tests and Zhiel–Nielssen staining in ear imprintings and nasal exudates.
222
F. Quesada-Pascual et al. / Veterinary Immunology and Immunopathology 124 (2008) 220–229
2.2. Skin biopsies Animals were previously anaesthetized with ketamine (10 mg/kg) and the area to be biopsied was carefully cleaned first with soap and water and then with iodine solution. Areas to be sampled in each animal included each side of the nape, the neck, the armpit, the abdomen and the thighs. A new and sterile scalpel was used each time to carefully excise a skin portion of approximately 1 cm2 in each of the four animals, the wounds were closed with stitches or suture, the animals were then allowed to recover from anaesthesia and were examined daily and the suture/stitches were removed usually at day 8 post-surgery. 2.3. Preparation of epidermal sheets for ATPase staining of langerhans cells The ATPase technique routinely used for human and mice skin was adapted to the armadillo’s skin and is detailed elsewhere (Shreedhar et al., 1999). Briefly, freshly taken skin biopsies of armadillos were placed on Petri dishes containing sterile, pyrogen-free saline solution (0.85%) and underlying fat was carefully removed. Skin portions were then stretched to place them on a previously warmed (37 8C) 0.5 M EDTA solution. After incubation, epidermis was separated by firm traction, each portion was washed and then placed in a fixative solution of paraformaldehyde (1%). After washing, epidermal sheets were then incubated with ATP-lead and positive cells were revealed with ammonium sulphide, which reaction provides a dark brown staining. 2.4. Langerhans cells: frequency, body size and dendrites extension We assessed the frequencies of ATPase+ LC per square millimetre of epidermis using a calibrated grid in the microscope, screening at least 10 fields per skin portion, using at least three different portions of the same biopsy (Baker and Habowsky, 1983). Results thus represent the mean of all counts performed and are expressed as LC/mm2 for each sample. The average body size of LC as well as the average extension of dendrites was measured with the program Qwin500 from Leica, using 500 LC, assessing at least 10 different microscopic fields for every epidermal sheet. Results thus express the mean plus standard deviation of all measurements performed for each condition. Statistical analyses were performed with Sigma plot statistics, using one-way ANOVA and the differences between the groups by using Tukey’s tests. ***P < 0.001.
2.5. Crossreactions of antibodies to human dendritic cells in armadillo skin To ascertain markers used for human cutaneous DC regarding their potential crossreactivity in the skin of armadillo, we assessed four monoclonal antibodies to human DC as primary reagents which were incubated overnight at 4 8C: anti-MHC-II (DR) (Dako, dilution 1:50), CD1a (Dako, dilution 1:50), CD86 (BD-Pharmingen, dilution 1:50) and langerin (Laboratory for Immunological Research Dardilly, dilution 1:50). Mouse monoclonal antibodies were revealed using peroxidasetagged horse anti-mouse antibody (Vector, dilution 1:800) incubated for 1 h at room temperature. After extensive washings with PBS-(0.2%) bovine serum albumin (BSA, Sigma St. Louis, MO), positivity was revealed with a substrate yielding a dark blue staining. Isotype-matched control antibodies were included in each set of labellings. Epidermal sheets were prepared as described and conventional skin sections (5 mm thickness) were prepared from frozen skin biopsies using a Leica 1900CM cryostat (Nussloch Germany). Slides were fixed in cold acetone, dried at room temperature for 45 min before rehydration to start the immunolabelling procedures with the indicated antibodies. 2.6. Topical in vivo treatment with oxazolone In order to evaluate the potential response of epidermal DC from armadillo skin to a typical cutaneous irritant, Oxazolone (5%) was applied in olive oil: acetone to an abdominal skin area of approximately 1.0 cm2. Control skin from the contralateral abdominal flank received the application of vehicle only (olive oil: acetone). After 18 h, skin biopsies were taken from both sides previous anaesthesia, as indicated. Skin samples were used to prepare epidermal sheets as described earlier, and to do conventional vertical tissue sections of 5 mm thickness. Primary and secondary antibody reagents were used as indicated for both epidermal sheets and tissue sections. 3. Results 3.1. Epidermal langerhans cells of armadillo express ATPase ATPase staining in intact whole epidermis of armadillo revealed positive cells which displayed the typical dendritic appearance found in other species, it was remarkable that even the delicate cytoplasmic extensions could be individually followed in each single
F. Quesada-Pascual et al. / Veterinary Immunology and Immunopathology 124 (2008) 220–229
223
LC. This is another important advantage of epidermal sheets over conventional tissue sections, since the former provide the whole tissue microenvironment as supporting frame, allowing us to follow and measure these delicate cytoplasmic projections. The relative abundance as well as the relative size of LC in each region can also be noticed. For instance LC frequency appears evidently higher in the nape (Fig. 1E) that in the thighs (Fig. 1A). 3.2. The frequency of epidermal ATPase+ LC increases towards the armadillo’s head When skin LC were quantified along the five different anatomical regions examined, it was evident that their density increased towards the head of the animals. Indeed, the nape showed nearly as many as twofold the average LC counted in the thighs (Fig. 2A). In contrast, the size of LC body was found smaller in the back of the neck than in the thighs (Figs. 1A–E and 3B),
Fig. 1. ATPase+ langerhans cells are present in armadillo’s epidermis. ATPase+ langerhans cells were readily identified in epidermal sheets of the five different anatomical regions examined: (A) thigh, (B) abdomen, (C) armpit, (D) neck and (E) nape. Cells with typical dendritic morphology and a rather regular distribution were observed in the normal epidermis of all anatomical areas sampled. Pictures shown for each anatomical region are representatives of the four armadillos tested.
Fig. 2. Langerhans cells (LC) frequency increases towards the cephalic region of armadillos. LC frequency per area of epidermis was ascertained in each anatomical region described (A) and also between animals of different sexes (B). The frequency of epidermal LC per tissue area increased towards the head of the animals (A). The total average of LC frequencies from the two female vs. the two male armadillos (B), reveals no major differences between sexes (mean). F = female, M = male. Bars represent the mean with standard deviation of the LC frequencies for all animals tested.
224
F. Quesada-Pascual et al. / Veterinary Immunology and Immunopathology 124 (2008) 220–229
the two female versus the two male animals, there were not significant differences found between sexes (Fig. 2B). Statistical analyses were performed as previously indicated. ***P < 0.001. 3.3. Crossreaction of antibodies to human MHC-II (DR) and CD86 in armadillo skin and oxazoloneinduced DC mobilization in vivo
Fig. 3. Body size and dendrites extension of LC in armadillo’s epidermis vary among the anatomical regions examined. (A) The complete LC size (indicated in mm) includes both the body and dendrites, (B) only the body size, and (C) the extension of dendrites only. Results are expressed as the mean of all measurements plus standard deviation performed for each condition. ***P < 0.001.
in an apparently inverse relationship to cell density per tissue area. The average extension of LC dendrites was also scored (Fig. 3A and C), the largest dendrites being the ones in the thighs and neck (Figs. 1 and 3). In contrast, when comparing the average LC density from
Since we lack antibodies to putative DCs from armadillo, we tested in the skin of this animal four monoclonal antibodies which can identify DC in humans: anti MHC-II (DR), CD1a, CD86 and langerin. Of these, a clear positivity in armadillo skin samples was revealed only when using anti MHC-II (DR) and CD86 (Fig. 4). It is worth mentioning that the positive cells identified in armadillo epidermal sheets displayed a prominent dendritic appearance, both for the MHC-II and the CD86 labelling (Fig. 4). Once we had selected at least two antibodies identifying epidermal DC in armadillo, we wanted to use these two reagents to explore whether these putative LC will respond in vivo to a classical cutaneous irritant such as oxazolone, by mobilizing themselves from epidermis into dermis. In epidermal sheets it was readily seen that compared to vehicle treatment (Fig. 5, top panel), topical oxazolone clearly decreased the density per area of both DR+ and especially for CD86+ cells with dendritic morphology (Fig. 5, lower panel). However, since analysis of epidermal sheets showed lower frequency of LC from epidermis but the technique precludes assessment of the dermis; we additionally prepared conventional tissue section of whole skin (which includes both the epidermis and the dermis) to perform immunohistochemistry with these two antibodies. An obvious advantage of whole skin sections over epidermal sheets is that only the former provides a complete overview of both epidermis and dermis at once. This feature facilitates the in vivo study of a basic functional property of cutaneous DC: whether under an appropriate stimulus, DC from the upper skin layers are somehow moving down into the lower, dermal layers. MHC-II labelling of whole skin sections showed that under topical vehicle treatment, DR+ cells were mainly located within the area of basal epidermis (Fig. 6, top panel). In contrast, under oxazolone treatment the main bulk of DR+ cells now appeared displaced into the (upper) dermis (Fig. 6, lower panel). Essentially the same, although perhaps more evident, was found regarding CD86+ cells strongly suggesting that topical oxazolone treatment in vivo indeed seemed to trigger DC mobilization from epidermis into dermis.
F. Quesada-Pascual et al. / Veterinary Immunology and Immunopathology 124 (2008) 220–229
225
Fig. 4. Antibodies to human DR and human CD86 crossreact with epidermal langerhans cells of armadillo skin. Searching for reagents which could help to identify putative LC in armadillo skin, we tested a panel of four monoclonal antibodies used to evaluate DC in humans: anti-MHC-II (DR), CD1a, CD86 and Langerin. Of these, only with antibodies to human MHC-II (DR) (top panel A–D) and to human CD86 (middle and lower panel F– J) we found clearly positive cells (dark blue labelling) in the epidermal sheets of armadillo. Importantly, the positive cells revealed in armadillo epidermis using antibodies to human markers, exhibited a prominent dendritic morphology. Labelling in E was performed with a murine isotypematched irrelevant antibody as a control. Magnifications in A, E, F and H = 20, in B, C, G and I = 40, in D and J = 100.
4. Discussion Though armadillos have been relatively well studied regarding the basic biology such as behaviour, feeding habits, and reproductive cycle (Anderson and Benirschke, 1966; Purtilo et al., 1975; Quesada-Pascual et al., 1987), it is perhaps in immunology where they have a more prominent and well-earned place. This is likely due to its intriguingly unique susceptibility to the experimental infection with M. leprae, the causative agent of leprosy in humans. In the last decade, there have been intriguing reports pointing to a more complex, perhaps unforeseen relation of this animal with M. leprae: (a) using serologic and molecular biology methods, an important proportion (4–19%) of armadillos in the wild have been reported as naturally infected with leprosy in the southern USA, some authors already describe this as a
zoonotic disease (Lane et al., 2006; Truman et al., 1991; Walsh et al., 1988); and (b) the diagnose of leprosy in humans without any known human leprosy contact or exposure, but with well-documented exposure to armadillos (Lane et al., 2006; Walsh et al., 1988). Though mycobacteria have indeed been reported in armadillos in the wild (Rojas-Espinosa and Lovik, 2001), at present it would seem debatable whether those bacilli found represent first, truly M. leprae, and secondly, whether they were indeed acquired as a natural disease. In any case, or precisely because of these controversial topics about leprosy and the armadillo still remain, this animal model definitely merits more attention in the context of both leprosy and basic immunology research to clarify these matters. On the other hand, despite the extensive immunological research on dendritic cells (DCs) in recent years, major gaps still await to be fulfilled. First, most of the
226
F. Quesada-Pascual et al. / Veterinary Immunology and Immunopathology 124 (2008) 220–229
Fig. 5. Topical skin application of oxazolone in vivo decreases dendritic cells in epidermal sheets. To ascertain in vivo whether epidermal DC of armadillo respond to a cutaneous challenge, the topical irritant oxazolone was applied onto armadillo skin (D–F) whereas control (contralateral) skin received topical application of vehicle only (A–C). Skin biopsies were taken 18 h later and processed to prepare epidermal sheets (A–F). In the skin receiving only vehicle (A-C) there were DR+ (A) and CD86+ (B–C) cells with clear dendritic appearance, whose frequency markedly decreased when oxazolone was applied, both for DR (D) and especially for CD86 (E–F). Magnifications in B, D and E = 20, in A, C and F = 40.
Fig. 6. Topical oxazolone in vivo decreases dendritic cells in epidermis and increases their frequency within dermis Conventional sections of whole skin treated with vehicle (A–D) demonstrated positive cells (dark blue labelling) for human DR (A and B) and for human CD86 (C and D) mostly around epidermis (indicated with red arrows). In contrast, whole skin sections of oxazolone treated skin (E–G) clearly revealed an increase of cells positive for both DR (E) and CD86 (G) but now within dermis (indicated with red arrows). An isotype-matched control antibody was used in (F). Magnifications in A, C and F = 10, in B, D, E and G = 20.
F. Quesada-Pascual et al. / Veterinary Immunology and Immunopathology 124 (2008) 220–229
DC work has been performed in vitro and is confined almost exclusively to two animal species: human and mice, although there is some good research in rats and sheep (Gemmell, 1973). An obvious factor strongly limiting the study of DCs in any other species is the lack of suitable reagents like for instance antibodies to identify molecules in rather uncommon laboratory animals. Convit’s group in 1975 performed extensive histochemical stainings in armadillo skin (CampoAasen and Convit, 1975). However, among eight different enzymatic stainings, they did not use ATPase, and sections were used instead of epidermal sheets. Since this pioneer work was not aimed at assessing DC but rather the basic histology of armadillo skin, very few cells were hesitantly described as dendritic, and only when using alpha naphthyl acetate esterase. Luckily, besides humans, in several animal species such as rat, mice, chicken and turtles the ectoenzyme ATPase has proven quite useful to identify putative langerhans cells in skin, especially when epidermal sheets are feasible to obtain. This has made possible to study these cells even when no antibodies to DC are available in some species. Indeed, ATPase+ cells were found in the epidermis of the five different anatomical regions examined along the body of the four armadillos tested, regardless of the sexes. In this regard, the possibility to obtain epidermal sheets from armadillo skin, as the ones obtained from mice or humans, would be a great advantage over classical transversal skin sections, which only allows a very partial view of exclusively those cells sectioned by the cut. The most important limitations of traditional tissue sections are two: cell labelling would not allow to distinguish the complete morphology of a single (cut) LC; and second, it would not permit the quantification of LC per tissue area. Both these limitations are clearly overcome once epidermal sheets are obtained and the ATPase staining proves feasible, since both the morphology of each single LC as well as the overall LC frequency per area are revealed given that cells are viewed en face, from above. Furthermore, since the epidermis is obtained as a complete, intact tissue, this, combined with the ATPase staining, provides a good natural support to preserve such delicate cytoplasmic extensions in their native microenvironment, so as to assess even each single dendrite branching off any given LC. This would be rather difficult to achieve using conventional (vertical) skin sectioning. Nevertheless, sections of whole skin have indeed other, different advantages like for instance providing at once an overview of both the epidermis and the dermis.
227
Initially, we tried different variations of enzymatic treatments to the skin to separate the epidermal sheets, without success, especially with regards to the final quality of the tissue and the ATPase staining. Only a chelant such as EDTA allowed us to obtain the fragile epidermal layers without disrupting their integrity, thus permitting LC staining and quantification along the different anatomical regions of the armadillo’s body. Quantification of ATPase+ LC per area and anatomical regions, revealed that LC frequency increases from the caudal towards the cephalic region, reaching even double the number of LC in the nape with respect to the thighs, which showed the lower LC density. Examination of the epithelial sheets also revealed some apparent morphological differences among LC from different areas of armadillo’s body (Fig. 1A–E). We do not have an explanation for this, but we would like to speculate that this might be related to the differential environmental contacts and perhaps differential antigenic exposure of these anatomical areas. We also measured the body size of LC in each anatomical region examined and found that the average size of LC body appears bigger towards the caudal region, the average largest body size found in the thighs. Apparently, the cell frequency will concur with what is reported in humans (Chen et al., 1985). Also, alike humans (and mice), LC in armadillo’s epidermis exhibited a rather regular and non-clustered tissue distribution. Furthermore, epidermal LC frequency in armadillos would appear not to differ importantly between females and males, though the number of animals examined is clearly small for definitive conclusions. We consider that an important finding regarding DCs is that from a panel of four different antibodies to study these cells in humans; two yielded consistently and clear positive results in armadillo’s samples: MHC-II and CD86. We believe this to be relevant for the potential biological and evolutionary implications of these molecules. Both are mostly involved in crucial cell–cell interactions, one (MHC-II) mainly in Ag presentation, and the other – CD86 – mainly as a costimulatory and activation molecule during adaptive immune responses, at least from what is known in humans and mice. Since ATPase, an evolutionary conspicuous epidermal DC marker, is lost once those DC located at peripheral sites get activated and triggered to move, we could not use this molecule to assess one of the most basic properties of epidermal DC in vivo. But having identified at least two molecules (DR and CD86) positive in DC from intact armadillo epidermis, permitted us to evaluate the potential effects
228
F. Quesada-Pascual et al. / Veterinary Immunology and Immunopathology 124 (2008) 220–229
of the classical cutaneous irritant oxazolone. Indeed, already in epidermal sheets it was clear that compared to topical vehicle treatment, oxazolone induced a reduction of DR+ cells, but more dramatically of the CD86+ cells. However, since by using epidermal sheets we cannot therefore screen the dermis; we turned to whole skin sections after applying either vehicle or oxazolone. Compared to epidermal sheets an advantage of whole skin sections is that the latter gives a complete picture of both epidermis and dermis at once. This facilitates to evaluate whether DC are being mobilized from epidermis into the dermis. Indeed sections of whole armadillo skin confirmed first the presence of DR+ and CD86+ cells confined to the basal epidermis area 18 h after applying topical vehicle. By contrast, under oxazolone treatment most of the DR+/CD86+ cells were identified within dermis, strongly suggesting that these cells seem to be mobilized. This is a crucial in vivo observation, since to accomplish their functions as peripheral sentinels, epidermal DC must first be able to move at least from epidermis into dermis. In summary, we have found that integral epidermal sheets can be successfully separated from armadillo skin using the calcium chelant EDTA instead of enzymes. We believe this is an important advantage over both the enzymatic treatments and the traditional skin sections. EDTA-obtained epidermal sheets are an excellent tool to reveal ATPase positive LC, which can be viewed en face and therefore examined both at the individual level and also to quantify their frequency and spatial distribution per skin area. We also screened in armadillo skin, four different antibodies used to assess human DC (anti-MHC-II [DR], CD1a, CD86 and langerin). Only MHC-II and CD86 revealed positive cells with prominent dendritic morphology, further allowing us to examine in vivo a basic function of cutaneous DC in armadillo. Indeed it was evident first in epidermal sheets, and then verified in whole skin sections (visualizing both epidermis and dermis at once) that compared to vehicle only, topical oxazolone triggered mobilization of DR+/CD86+ (putative) langerhans cells from the epidermis into the dermis. We hope that these baseline findings would incite further studies on LC and the immune responses in armadillos, for instance during naturally acquired or experimental leprosy. Acknowledgements Authors wish to thank Dr. Luz Diaz and the help provided by the lab teams in CINVESTAV, ENCB-IPN,
and IZTACALA. A. Flores-Langarica, A. SilvaSanchez, and A.Y. Limon-Flores were fellow-holders from the National Council for Science and Technology, Conacyt. I. Estrada-Garcia, L. Santos-Argumedo, S. Estrada-Parra and L. Flores-Romo have been partially supported by the National Council for Science and Technology (Conacyt) and are members of the National System of Researchers SNI. References Anderson, J.M., Benirschke, K., 1966. The armadillo, Dasypus novemcinctus, in experimental biology. Lab. Anim. Care 16, 202–216. Baker, K.W., Habowsky, J.E., 1983. EDTA separation and ATPase langerhans cell staining in the mouse epidermis. J. Invest. Dermatol. 80, 104–107. Bruce, S., Schroeder, T.L., Ellner, K., Rubin, H., Williams, T., Wolf Jr., J.E., 2000. Armadillo exposure and Hansen’s disease: an epidemiologic survey in southern Texas. J. Am. Acad. Dermatol. 43, 223–228. Campo-Aasen, I., Convit, J., 1975. The histochemistry of armadillo skin. J. Anat. 120, 337–348. Carrillo-Farga, J., Perez, T.A., Castell, R.A., Antuna, B.S., 1991. Adenosine triphosphatase-positive langerhans-like cells in the epidermis of the chicken (Gallus gallus). J. Anat. 176, 1–8. Chen, H., Yuan, J., Wang, Y., Silvers, W.K., 1985. Distribution of ATPase-positive langerhans cells in normal adult human skin. Br. J. Dermatol. 113, 707–711. Colston, M.J., Levy, L., 1987. Infection of other experimental animals with Mycobacterium leprae. Int. J. Lepr. Other Mycobact. Dis. 55, 896–898. Flores-Langarica, A., Meza-Perez, S., Calderon-Amador, J., EstradaGarcia, T., Macpherson, G., Lebecque, S., Saeland, S., Steinman, R.M., Flores-Romo, L., 2005. Network of dendritic cells within the muscular layer of the mouse intestine. Proc. Natl. Acad. Sci. U.S.A. 102, 19039–19044. Flores-Romo, L., 2001. In vivo maturation and migration of dendritic cells. Immunology 102, 255–262. Garcia-Romo, G.S., Pedroza-Gonzalez, A., Aguilar-Leon, D., Orozco-Estevez, H., Lambrecht, B.N., Estrada-Garcia, I., Flores-Romo, L., Hernandez-Pando, R., 2004. Airways infection with virulent Mycobacterium tuberculosis delays the influx of dendritic cells and the expression of costimulatory molecules in mediastinal lymph nodes. Immunology 112, 661–668. Gemmell, R.T., 1973. Langerhans cells in the ruminal epithelium of the sheep. J. Ultrastruct. Res. 43, 256–259. Huang, C.L., 1980. The transmission of leprosy in man. Int. J. Lepr. Other Mycobact. Dis. 48, 309–318. Jimenez-Flores, R., Mendez-Cruz, R., Ojeda-Ortiz, J., Munoz-Molina, R., Balderas-Carrillo, O., de la Luz Diaz-Soberanes, Lebecque, S., Saeland, S., Daneri-Navarro, A., Garcia-Carranca, A., Ullrich, S.E., Flores-Romo, L., 2006. High-risk human papilloma virus infection decreases the frequency of dendritic langerhans’ cells in the human female genital tract. Immunology 117, 220–228. Kirchheimer, W.F., Storrs, E.E., 1971. Attempts to establish the armadillo (Dasypus novemcinctus Linn) as a model for the study of leprosy. I. Report of lepromatoid leprosy in an experimentally infected armadillo. Int. J. Lepr. Other Mycobact. Dis. 39, 693–702.
F. Quesada-Pascual et al. / Veterinary Immunology and Immunopathology 124 (2008) 220–229 Kirchheimer, W.F., Storrs, E.E., Binford, C.H., 1972. Attempts to establish the Armadillo (Dasypus novemcinctus Linn) as a model for the study of leprosy. II. Histopathologic and bacteriologic postmortem findings in lepromatoid leprosy in the Armadillo. Int. J. Lepr. Other Mycobact. Dis. 40, 229–242. Kiyono, H., Fukuyama, S., 2004. NALT—versus Peyer’s-patchmediated mucosal immunity. Nat. Rev. Immunol. 4, 699–710. Lane, J.E., Walsh, D.S., Meyers, W.M., Klassen-Fischer, M.K., Kent, D.E., Cohen, D.J., 2006. Borderline tuberculoid leprosy in a woman from the state of Georgia with armadillo exposure. J. Am. Acad. Dermatol. 55, 714–716. Limon-Flores, A.Y., Perez-Tapia, M., Estrada-Garcia, I., Vaughan, G., Escobar-Gutierrez, A., Calderon-Amador, J., Herrera-Rodriguez, S.E., Brizuela-Garcia, A., Heras-Chavarria, M., Flores-Langarica, A., Cedillo-Barron, L., Flores-Romo, L., 2005. Dengue virus inoculation to human skin explants: an effective approach to assess in situ the early infection and the effects on cutaneous dendritic cells. Int. J. Exp. Pathol. 86, 323–334. Lumpkin III, L.R., Cox, G.F., Wolf Jr., J.E., 1983. Leprosy in five armadillo handlers. J. Am. Acad. Dermatol. 9, 899–903. Pallen, M.J., McDermott, R.D., 1986. How might Mycobacterium leprae enter the body? Lepr. Rev. 57, 289–297. Perez, T.A., Millan Aldaco, D.A., 1994. Ia antigens are expressed on ATPase-positive dendritic cells in chicken epidermis. J. Anat. 184 (Pt 3), 591–596.
229
Perez-Torres, A., Millan-Aldaco, D.A., Rondan-Zarate, A., 1995. Epidermal langerhans cells in the terrestrial turtle Kinosternum integrum. Dev. Comp. Immunol. 19, 225–236. Purtilo, D.T., Walsh, G.P., Storrs, E.E., Gannon, C., 1975. The immune system of the nine-banded armadillo (Dasypus novemcinctus Linn). Anat. Rec. 181, 725–733. Quesada-Pascual, F., Rojas-Espinosa, O., Santos, A.L., Estrada-Parra, S., 1987. A Mexican armadillo (Dasypus novemcinctus) colony for leprosy research. Int. J. Lepr. Other Mycobact. Dis. 55, 716–718. Rojas-Espinosa, O., Lovik, M., 2001. Mycobacterium leprae and Mycobacterium lepraemurium infections in domestic and wild animals. Rev. Sci. Technol. 20, 219–251. Shreedhar, V., Moodycliffe, A.M., Ullrich, S.E., Bucana, C., Kripke, M.L., Flores-Romo, L., 1999. Dendritic cells require T cells for functional maturation in vivo. Immunity 11, 625–636. Truman, R.W., Kumaresan, J.A., McDonough, C.M., Job, C.K., Hastings, R.C., 1991. Seasonal and spatial trends in the detectability of leprosy in wild armadillos. Epidemiol. Infect. 106, 549–560. Walsh, G.P., Meyers, W.M., Binford, C.H., Gormus, B.J., Baskin, G.B., Wolf, R.H., Gerone, P.J., 1988. Leprosy as a zoonosis: an update. Acta Leprol. 6, 51–60. West, B.C., Todd, J.R., Lary, C.H., Blake, L.A., Fowler, M.E., King, J.W., 1988. Leprosy in six isolated residents of northern Louisiana time-clustered cases in an essentially nonendemic area. Arch. Intern. Med. 148, 1987–1992.
Veterinary Immunology and Immunopathology 124 (2008) 220–229 www.elsevier.com/locate/vetimm
Characterization of langerhans cells in epidermal sheets along the body of Armadillo (Dasypus novemcinctus) Fausto Quesada-Pascual b, Rafael Jimenez-Flores a,d, Adriana Flores-Langarica b,1, Aaron Silva-Sanchez b, Juana Calderon-Amador d, Rene Mendez-Cruz a,d, Alberto Y. Limon-Flores d, Sergio Estrada-Parra b, Leopoldo Santos-Argumedo c, Iris Estrada-Garcia b, Leopoldo Flores-Romo d,* b
a Faculty of High Studies Iztacala, The National University of Mexico, UNAM, Mexico City, Mexico Department of Immunology, National School of Biological Sciences ENCB-IPN, Mexico City, Mexico c Department of Molecular Biomedicine, CINVESTAV-IPN, Mexico City, Mexico d Department of Cell Biology, CINVESTAV-IPN, Mexico City, Mexico
Received 26 April 2007; received in revised form 13 March 2008; accepted 19 March 2008
Abstract Armadillos are apparently important reservoirs of Mycobacterium leprae and an animal model for human leprosy, whose immune system has been poorly studied. We aimed at characterizing the armadillo’s langerhans cells (LC) using epidermal sheets instead of tissue sections, since the latter restrict analysis only to cut-traversed cells. Epidermal sheets by providing an en face view, are particularly convenient to evaluate dendritic morphology (cells are complete), spatial distribution (regular vs. clustered), and frequency (cell number/tissue area). Lack of anti-armadillo antibodies was overcome using LC-restricted ATPase staining, allowing assessment of cell frequency, cell size, and dendrites extension. Average LC frequency in four animals was 528 LC/mm2, showing a rather uniform non-clustered distribution, which increased towards the animal’s head, while cell size increased towards the tail; without overt differences between sexes. The screening of antibodies to human DC (MHC-II, CD1a, langerin, CD86) in armadillo epidermal sheets, revealed positive cells with prominent dendritic morphology only with MHC-II and CD86. This allowed us to test DC mobilization from epidermis into dermis under topical oxazolone stimulation, a finding that was corroborated using whole skin conventional sections. We hope that the characterization of armadillo’s LC will incite studies of leprosy and immunity in this animal model. # 2008 Elsevier B.V. All rights reserved. Keywords: Langerhans cells; Armadillo; In vivo; Leprosy; Mycobacterium leprae reservoir; Epithelial sheets
1. Introduction Abbreviations: LC, langerhans cells; DC, dendritic cells; APC, antigen presenting cells. * Corresponding author at: Department of Cell Biology, CINVESTAV-IPN, Av. IPN 2508, Col. San Pedro Zacatenco, 07360 Mexico City, Mexico. Tel.: +52 55 57 47 39 84; fax: +52 55 50 61 33 93. E-mail address: [email protected] (L. Flores-Romo). 1 Present address: Sir William Dunn School of Pathology, South Parks Road, Oxford OX1 3RE, United Kingdom. 0165-2427/$ – see front matter # 2008 Elsevier B.V. All rights reserved. doi:10.1016/j.vetimm.2008.03.011
Since the seminal work of Storrs in 1971 (Colston and Levy, 1987), the armadillo (Dasypus novemcinctus) has become the most appropriate animal model to assess various aspects of leprosy infection in vivo. It seems that the armadillo is even more susceptible to (Lepromatous) leprosy than humans, developing more severe complications (Kirchheimer et al., 1972). Although the precise
F. Quesada-Pascual et al. / Veterinary Immunology and Immunopathology 124 (2008) 220–229
reasons for this are not well known, factors such as a decreased cell-mediated immunity have been advanced, but without much experimental support. Also, the low body temperature (32–35 8C) of armadillos has been put forward, since this feature apparently coincides with the decreased temperature observed in equivalent areas where bacilli can be found in leprosy patients (Kirchheimer and Storrs, 1971). On the other hand, an histological examination of lymphoid organs in armadillo, including thymus, spleen, lymph nodes, tonsils, and bone marrow, neither reveal any ostensible alterations nor absence of any of these organs (Purtilo et al., 1975). Whether leprosy infection occurs naturally in armadillos, or was introduced upon contacts with humans, is still controversial. However, there seem to be agreement and experimental evidences strongly suggest that in certain areas of the southern USA, leprosy might be a zoonotic disease in armadillos, with up to 16–19% of wild animals positive (Lane et al., 2006; Truman et al., 1991). Interestingly, naturally acquired leprosy has also been reported in non-human primates in other regions of the world (Walsh et al., 1988). Further, in low endemic areas, human cases have been newly diagnosed where the subjects had no known human contacts for leprosy, while all had been exposed to armadillos one way or another (Bruce et al., 2000; Lane et al., 2006; Lumpkin et al., 1983; West et al., 1988). The langerhans cells constitute the epidermal chapter of the DC system. This functions to initiate immune responses and is integrated by sentinel cells strategically located at the boundaries with the external world such as the skin (Flores-Romo, 2001), the gastrointestinal (Flores-Langarica et al., 2005; Kiyono and Fukuyama, 2004), the respiratory (Garcia-Romo et al., 2004), and the genital mucosa (Jimenez-Flores et al., 2006; FloresRomo, 2001). It is at these locations that most daily encounters with environmental antigens occur, many of them are microbes and some can be infectious and pathogenic (Garcia-Romo et al., 2004; Jimenez-Flores et al., 2006; Limon-Flores et al., 2005). DC are deemed the most potent antigen presenting cell (APCs) of all, whose functions include Ag capture at peripheral locations, Ag degradation and processing, Ag transportation into the regional lymphoid tissues, and finally, induction of primary immune responses by an efficient Ag presentation (Flores-Romo, 2001). Since the discovery of Mycobacterium leprae, the skin has been considered from time to time as a potential port of entry for leprosy bacilli (Huang, 1980; Pallen and McDermott, 1986). However, perhaps due to the lack of an adequate animal model and also to the technical difficulties
221
(including absence of antibodies and other reagents) to study LC in animal species such as the armadillo, the specific role of these cells has not been assessed in vivo during leprosy infection in animal models. The ectoenzyme ATPase has been quite useful to detect LC in the skin of a wide variety of animals, especially those species for which there are no antibodies currently available. Besides humans, ATPase+ LC have been identified in rat, mice, chicken and turtles (Baker and Habowsky, 1983; Carrillo-Farga et al., 1991; PerezTorres et al., 1995; Perez and Millan Aldaco, 1994). Since this enzymatic marker seems both quite well preserved along the phylogeny as well as quite restricted to LC in the epidermis of various animals, we reasoned that if LC were present in armadillo skin they were likely to express this enzymatic activity. Although there is a seminal histology work performed on armadillo skin by Convit’s group in 1975, they used twelve different histochemical approaches including eight enzymatic stainings (except ATPase), aiming at ‘‘eluicidating macrophage–lymphocyte interactions in the immunity phenomena in leprosy’’(Campo-Aasen and Convit, 1975). However, no ATPase labelling was done, sections were used instead of epidermal sheets, and very few cells were hesitantly described as dendritic (only) when using alpha naphthyl acetate esterase. As a first step, we have undertaken the identification and quantification of epidermal LC along the body of this animal. We hope that these findings will provide a baseline for prospective studies on LC and the immune responses in the armadillo. 2. Material and methods 2.1. Animals These animals are from an armadillo colony of the Department of Immunology in The National School of Biological Sciences (ENCB-IPN), which was initiated in 1980 as part of a program to facilitate leprosy research (Quesada-Pascual et al., 1987). Since armadillos are not easy to keep in captivity (on average, 40% of the captured animals die within the first 3 months of life in captivity), we decided to use only four healthy adult (3 years old) animals, two males and two females. These four animals were already well adapted since they have lived in these facilities for at least 6 months. Upon arrival, animals are carefully examined for the presence of acid-fast bacilli (AFB) and for any other obvious infections and injuries, these exams include blood tests and Zhiel–Nielssen staining in ear imprintings and nasal exudates.
222
F. Quesada-Pascual et al. / Veterinary Immunology and Immunopathology 124 (2008) 220–229
2.2. Skin biopsies Animals were previously anaesthetized with ketamine (10 mg/kg) and the area to be biopsied was carefully cleaned first with soap and water and then with iodine solution. Areas to be sampled in each animal included each side of the nape, the neck, the armpit, the abdomen and the thighs. A new and sterile scalpel was used each time to carefully excise a skin portion of approximately 1 cm2 in each of the four animals, the wounds were closed with stitches or suture, the animals were then allowed to recover from anaesthesia and were examined daily and the suture/stitches were removed usually at day 8 post-surgery. 2.3. Preparation of epidermal sheets for ATPase staining of langerhans cells The ATPase technique routinely used for human and mice skin was adapted to the armadillo’s skin and is detailed elsewhere (Shreedhar et al., 1999). Briefly, freshly taken skin biopsies of armadillos were placed on Petri dishes containing sterile, pyrogen-free saline solution (0.85%) and underlying fat was carefully removed. Skin portions were then stretched to place them on a previously warmed (37 8C) 0.5 M EDTA solution. After incubation, epidermis was separated by firm traction, each portion was washed and then placed in a fixative solution of paraformaldehyde (1%). After washing, epidermal sheets were then incubated with ATP-lead and positive cells were revealed with ammonium sulphide, which reaction provides a dark brown staining. 2.4. Langerhans cells: frequency, body size and dendrites extension We assessed the frequencies of ATPase+ LC per square millimetre of epidermis using a calibrated grid in the microscope, screening at least 10 fields per skin portion, using at least three different portions of the same biopsy (Baker and Habowsky, 1983). Results thus represent the mean of all counts performed and are expressed as LC/mm2 for each sample. The average body size of LC as well as the average extension of dendrites was measured with the program Qwin500 from Leica, using 500 LC, assessing at least 10 different microscopic fields for every epidermal sheet. Results thus express the mean plus standard deviation of all measurements performed for each condition. Statistical analyses were performed with Sigma plot statistics, using one-way ANOVA and the differences between the groups by using Tukey’s tests. ***P < 0.001.
2.5. Crossreactions of antibodies to human dendritic cells in armadillo skin To ascertain markers used for human cutaneous DC regarding their potential crossreactivity in the skin of armadillo, we assessed four monoclonal antibodies to human DC as primary reagents which were incubated overnight at 4 8C: anti-MHC-II (DR) (Dako, dilution 1:50), CD1a (Dako, dilution 1:50), CD86 (BD-Pharmingen, dilution 1:50) and langerin (Laboratory for Immunological Research Dardilly, dilution 1:50). Mouse monoclonal antibodies were revealed using peroxidasetagged horse anti-mouse antibody (Vector, dilution 1:800) incubated for 1 h at room temperature. After extensive washings with PBS-(0.2%) bovine serum albumin (BSA, Sigma St. Louis, MO), positivity was revealed with a substrate yielding a dark blue staining. Isotype-matched control antibodies were included in each set of labellings. Epidermal sheets were prepared as described and conventional skin sections (5 mm thickness) were prepared from frozen skin biopsies using a Leica 1900CM cryostat (Nussloch Germany). Slides were fixed in cold acetone, dried at room temperature for 45 min before rehydration to start the immunolabelling procedures with the indicated antibodies. 2.6. Topical in vivo treatment with oxazolone In order to evaluate the potential response of epidermal DC from armadillo skin to a typical cutaneous irritant, Oxazolone (5%) was applied in olive oil: acetone to an abdominal skin area of approximately 1.0 cm2. Control skin from the contralateral abdominal flank received the application of vehicle only (olive oil: acetone). After 18 h, skin biopsies were taken from both sides previous anaesthesia, as indicated. Skin samples were used to prepare epidermal sheets as described earlier, and to do conventional vertical tissue sections of 5 mm thickness. Primary and secondary antibody reagents were used as indicated for both epidermal sheets and tissue sections. 3. Results 3.1. Epidermal langerhans cells of armadillo express ATPase ATPase staining in intact whole epidermis of armadillo revealed positive cells which displayed the typical dendritic appearance found in other species, it was remarkable that even the delicate cytoplasmic extensions could be individually followed in each single
F. Quesada-Pascual et al. / Veterinary Immunology and Immunopathology 124 (2008) 220–229
223
LC. This is another important advantage of epidermal sheets over conventional tissue sections, since the former provide the whole tissue microenvironment as supporting frame, allowing us to follow and measure these delicate cytoplasmic projections. The relative abundance as well as the relative size of LC in each region can also be noticed. For instance LC frequency appears evidently higher in the nape (Fig. 1E) that in the thighs (Fig. 1A). 3.2. The frequency of epidermal ATPase+ LC increases towards the armadillo’s head When skin LC were quantified along the five different anatomical regions examined, it was evident that their density increased towards the head of the animals. Indeed, the nape showed nearly as many as twofold the average LC counted in the thighs (Fig. 2A). In contrast, the size of LC body was found smaller in the back of the neck than in the thighs (Figs. 1A–E and 3B),
Fig. 1. ATPase+ langerhans cells are present in armadillo’s epidermis. ATPase+ langerhans cells were readily identified in epidermal sheets of the five different anatomical regions examined: (A) thigh, (B) abdomen, (C) armpit, (D) neck and (E) nape. Cells with typical dendritic morphology and a rather regular distribution were observed in the normal epidermis of all anatomical areas sampled. Pictures shown for each anatomical region are representatives of the four armadillos tested.
Fig. 2. Langerhans cells (LC) frequency increases towards the cephalic region of armadillos. LC frequency per area of epidermis was ascertained in each anatomical region described (A) and also between animals of different sexes (B). The frequency of epidermal LC per tissue area increased towards the head of the animals (A). The total average of LC frequencies from the two female vs. the two male armadillos (B), reveals no major differences between sexes (mean). F = female, M = male. Bars represent the mean with standard deviation of the LC frequencies for all animals tested.
224
F. Quesada-Pascual et al. / Veterinary Immunology and Immunopathology 124 (2008) 220–229
the two female versus the two male animals, there were not significant differences found between sexes (Fig. 2B). Statistical analyses were performed as previously indicated. ***P < 0.001. 3.3. Crossreaction of antibodies to human MHC-II (DR) and CD86 in armadillo skin and oxazoloneinduced DC mobilization in vivo
Fig. 3. Body size and dendrites extension of LC in armadillo’s epidermis vary among the anatomical regions examined. (A) The complete LC size (indicated in mm) includes both the body and dendrites, (B) only the body size, and (C) the extension of dendrites only. Results are expressed as the mean of all measurements plus standard deviation performed for each condition. ***P < 0.001.
in an apparently inverse relationship to cell density per tissue area. The average extension of LC dendrites was also scored (Fig. 3A and C), the largest dendrites being the ones in the thighs and neck (Figs. 1 and 3). In contrast, when comparing the average LC density from
Since we lack antibodies to putative DCs from armadillo, we tested in the skin of this animal four monoclonal antibodies which can identify DC in humans: anti MHC-II (DR), CD1a, CD86 and langerin. Of these, a clear positivity in armadillo skin samples was revealed only when using anti MHC-II (DR) and CD86 (Fig. 4). It is worth mentioning that the positive cells identified in armadillo epidermal sheets displayed a prominent dendritic appearance, both for the MHC-II and the CD86 labelling (Fig. 4). Once we had selected at least two antibodies identifying epidermal DC in armadillo, we wanted to use these two reagents to explore whether these putative LC will respond in vivo to a classical cutaneous irritant such as oxazolone, by mobilizing themselves from epidermis into dermis. In epidermal sheets it was readily seen that compared to vehicle treatment (Fig. 5, top panel), topical oxazolone clearly decreased the density per area of both DR+ and especially for CD86+ cells with dendritic morphology (Fig. 5, lower panel). However, since analysis of epidermal sheets showed lower frequency of LC from epidermis but the technique precludes assessment of the dermis; we additionally prepared conventional tissue section of whole skin (which includes both the epidermis and the dermis) to perform immunohistochemistry with these two antibodies. An obvious advantage of whole skin sections over epidermal sheets is that only the former provides a complete overview of both epidermis and dermis at once. This feature facilitates the in vivo study of a basic functional property of cutaneous DC: whether under an appropriate stimulus, DC from the upper skin layers are somehow moving down into the lower, dermal layers. MHC-II labelling of whole skin sections showed that under topical vehicle treatment, DR+ cells were mainly located within the area of basal epidermis (Fig. 6, top panel). In contrast, under oxazolone treatment the main bulk of DR+ cells now appeared displaced into the (upper) dermis (Fig. 6, lower panel). Essentially the same, although perhaps more evident, was found regarding CD86+ cells strongly suggesting that topical oxazolone treatment in vivo indeed seemed to trigger DC mobilization from epidermis into dermis.
F. Quesada-Pascual et al. / Veterinary Immunology and Immunopathology 124 (2008) 220–229
225
Fig. 4. Antibodies to human DR and human CD86 crossreact with epidermal langerhans cells of armadillo skin. Searching for reagents which could help to identify putative LC in armadillo skin, we tested a panel of four monoclonal antibodies used to evaluate DC in humans: anti-MHC-II (DR), CD1a, CD86 and Langerin. Of these, only with antibodies to human MHC-II (DR) (top panel A–D) and to human CD86 (middle and lower panel F– J) we found clearly positive cells (dark blue labelling) in the epidermal sheets of armadillo. Importantly, the positive cells revealed in armadillo epidermis using antibodies to human markers, exhibited a prominent dendritic morphology. Labelling in E was performed with a murine isotypematched irrelevant antibody as a control. Magnifications in A, E, F and H = 20, in B, C, G and I = 40, in D and J = 100.
4. Discussion Though armadillos have been relatively well studied regarding the basic biology such as behaviour, feeding habits, and reproductive cycle (Anderson and Benirschke, 1966; Purtilo et al., 1975; Quesada-Pascual et al., 1987), it is perhaps in immunology where they have a more prominent and well-earned place. This is likely due to its intriguingly unique susceptibility to the experimental infection with M. leprae, the causative agent of leprosy in humans. In the last decade, there have been intriguing reports pointing to a more complex, perhaps unforeseen relation of this animal with M. leprae: (a) using serologic and molecular biology methods, an important proportion (4–19%) of armadillos in the wild have been reported as naturally infected with leprosy in the southern USA, some authors already describe this as a
zoonotic disease (Lane et al., 2006; Truman et al., 1991; Walsh et al., 1988); and (b) the diagnose of leprosy in humans without any known human leprosy contact or exposure, but with well-documented exposure to armadillos (Lane et al., 2006; Walsh et al., 1988). Though mycobacteria have indeed been reported in armadillos in the wild (Rojas-Espinosa and Lovik, 2001), at present it would seem debatable whether those bacilli found represent first, truly M. leprae, and secondly, whether they were indeed acquired as a natural disease. In any case, or precisely because of these controversial topics about leprosy and the armadillo still remain, this animal model definitely merits more attention in the context of both leprosy and basic immunology research to clarify these matters. On the other hand, despite the extensive immunological research on dendritic cells (DCs) in recent years, major gaps still await to be fulfilled. First, most of the
226
F. Quesada-Pascual et al. / Veterinary Immunology and Immunopathology 124 (2008) 220–229
Fig. 5. Topical skin application of oxazolone in vivo decreases dendritic cells in epidermal sheets. To ascertain in vivo whether epidermal DC of armadillo respond to a cutaneous challenge, the topical irritant oxazolone was applied onto armadillo skin (D–F) whereas control (contralateral) skin received topical application of vehicle only (A–C). Skin biopsies were taken 18 h later and processed to prepare epidermal sheets (A–F). In the skin receiving only vehicle (A-C) there were DR+ (A) and CD86+ (B–C) cells with clear dendritic appearance, whose frequency markedly decreased when oxazolone was applied, both for DR (D) and especially for CD86 (E–F). Magnifications in B, D and E = 20, in A, C and F = 40.
Fig. 6. Topical oxazolone in vivo decreases dendritic cells in epidermis and increases their frequency within dermis Conventional sections of whole skin treated with vehicle (A–D) demonstrated positive cells (dark blue labelling) for human DR (A and B) and for human CD86 (C and D) mostly around epidermis (indicated with red arrows). In contrast, whole skin sections of oxazolone treated skin (E–G) clearly revealed an increase of cells positive for both DR (E) and CD86 (G) but now within dermis (indicated with red arrows). An isotype-matched control antibody was used in (F). Magnifications in A, C and F = 10, in B, D, E and G = 20.
F. Quesada-Pascual et al. / Veterinary Immunology and Immunopathology 124 (2008) 220–229
DC work has been performed in vitro and is confined almost exclusively to two animal species: human and mice, although there is some good research in rats and sheep (Gemmell, 1973). An obvious factor strongly limiting the study of DCs in any other species is the lack of suitable reagents like for instance antibodies to identify molecules in rather uncommon laboratory animals. Convit’s group in 1975 performed extensive histochemical stainings in armadillo skin (CampoAasen and Convit, 1975). However, among eight different enzymatic stainings, they did not use ATPase, and sections were used instead of epidermal sheets. Since this pioneer work was not aimed at assessing DC but rather the basic histology of armadillo skin, very few cells were hesitantly described as dendritic, and only when using alpha naphthyl acetate esterase. Luckily, besides humans, in several animal species such as rat, mice, chicken and turtles the ectoenzyme ATPase has proven quite useful to identify putative langerhans cells in skin, especially when epidermal sheets are feasible to obtain. This has made possible to study these cells even when no antibodies to DC are available in some species. Indeed, ATPase+ cells were found in the epidermis of the five different anatomical regions examined along the body of the four armadillos tested, regardless of the sexes. In this regard, the possibility to obtain epidermal sheets from armadillo skin, as the ones obtained from mice or humans, would be a great advantage over classical transversal skin sections, which only allows a very partial view of exclusively those cells sectioned by the cut. The most important limitations of traditional tissue sections are two: cell labelling would not allow to distinguish the complete morphology of a single (cut) LC; and second, it would not permit the quantification of LC per tissue area. Both these limitations are clearly overcome once epidermal sheets are obtained and the ATPase staining proves feasible, since both the morphology of each single LC as well as the overall LC frequency per area are revealed given that cells are viewed en face, from above. Furthermore, since the epidermis is obtained as a complete, intact tissue, this, combined with the ATPase staining, provides a good natural support to preserve such delicate cytoplasmic extensions in their native microenvironment, so as to assess even each single dendrite branching off any given LC. This would be rather difficult to achieve using conventional (vertical) skin sectioning. Nevertheless, sections of whole skin have indeed other, different advantages like for instance providing at once an overview of both the epidermis and the dermis.
227
Initially, we tried different variations of enzymatic treatments to the skin to separate the epidermal sheets, without success, especially with regards to the final quality of the tissue and the ATPase staining. Only a chelant such as EDTA allowed us to obtain the fragile epidermal layers without disrupting their integrity, thus permitting LC staining and quantification along the different anatomical regions of the armadillo’s body. Quantification of ATPase+ LC per area and anatomical regions, revealed that LC frequency increases from the caudal towards the cephalic region, reaching even double the number of LC in the nape with respect to the thighs, which showed the lower LC density. Examination of the epithelial sheets also revealed some apparent morphological differences among LC from different areas of armadillo’s body (Fig. 1A–E). We do not have an explanation for this, but we would like to speculate that this might be related to the differential environmental contacts and perhaps differential antigenic exposure of these anatomical areas. We also measured the body size of LC in each anatomical region examined and found that the average size of LC body appears bigger towards the caudal region, the average largest body size found in the thighs. Apparently, the cell frequency will concur with what is reported in humans (Chen et al., 1985). Also, alike humans (and mice), LC in armadillo’s epidermis exhibited a rather regular and non-clustered tissue distribution. Furthermore, epidermal LC frequency in armadillos would appear not to differ importantly between females and males, though the number of animals examined is clearly small for definitive conclusions. We consider that an important finding regarding DCs is that from a panel of four different antibodies to study these cells in humans; two yielded consistently and clear positive results in armadillo’s samples: MHC-II and CD86. We believe this to be relevant for the potential biological and evolutionary implications of these molecules. Both are mostly involved in crucial cell–cell interactions, one (MHC-II) mainly in Ag presentation, and the other – CD86 – mainly as a costimulatory and activation molecule during adaptive immune responses, at least from what is known in humans and mice. Since ATPase, an evolutionary conspicuous epidermal DC marker, is lost once those DC located at peripheral sites get activated and triggered to move, we could not use this molecule to assess one of the most basic properties of epidermal DC in vivo. But having identified at least two molecules (DR and CD86) positive in DC from intact armadillo epidermis, permitted us to evaluate the potential effects
228
F. Quesada-Pascual et al. / Veterinary Immunology and Immunopathology 124 (2008) 220–229
of the classical cutaneous irritant oxazolone. Indeed, already in epidermal sheets it was clear that compared to topical vehicle treatment, oxazolone induced a reduction of DR+ cells, but more dramatically of the CD86+ cells. However, since by using epidermal sheets we cannot therefore screen the dermis; we turned to whole skin sections after applying either vehicle or oxazolone. Compared to epidermal sheets an advantage of whole skin sections is that the latter gives a complete picture of both epidermis and dermis at once. This facilitates to evaluate whether DC are being mobilized from epidermis into the dermis. Indeed sections of whole armadillo skin confirmed first the presence of DR+ and CD86+ cells confined to the basal epidermis area 18 h after applying topical vehicle. By contrast, under oxazolone treatment most of the DR+/CD86+ cells were identified within dermis, strongly suggesting that these cells seem to be mobilized. This is a crucial in vivo observation, since to accomplish their functions as peripheral sentinels, epidermal DC must first be able to move at least from epidermis into dermis. In summary, we have found that integral epidermal sheets can be successfully separated from armadillo skin using the calcium chelant EDTA instead of enzymes. We believe this is an important advantage over both the enzymatic treatments and the traditional skin sections. EDTA-obtained epidermal sheets are an excellent tool to reveal ATPase positive LC, which can be viewed en face and therefore examined both at the individual level and also to quantify their frequency and spatial distribution per skin area. We also screened in armadillo skin, four different antibodies used to assess human DC (anti-MHC-II [DR], CD1a, CD86 and langerin). Only MHC-II and CD86 revealed positive cells with prominent dendritic morphology, further allowing us to examine in vivo a basic function of cutaneous DC in armadillo. Indeed it was evident first in epidermal sheets, and then verified in whole skin sections (visualizing both epidermis and dermis at once) that compared to vehicle only, topical oxazolone triggered mobilization of DR+/CD86+ (putative) langerhans cells from the epidermis into the dermis. We hope that these baseline findings would incite further studies on LC and the immune responses in armadillos, for instance during naturally acquired or experimental leprosy. Acknowledgements Authors wish to thank Dr. Luz Diaz and the help provided by the lab teams in CINVESTAV, ENCB-IPN,
and IZTACALA. A. Flores-Langarica, A. SilvaSanchez, and A.Y. Limon-Flores were fellow-holders from the National Council for Science and Technology, Conacyt. I. Estrada-Garcia, L. Santos-Argumedo, S. Estrada-Parra and L. Flores-Romo have been partially supported by the National Council for Science and Technology (Conacyt) and are members of the National System of Researchers SNI. References Anderson, J.M., Benirschke, K., 1966. The armadillo, Dasypus novemcinctus, in experimental biology. Lab. Anim. Care 16, 202–216. Baker, K.W., Habowsky, J.E., 1983. EDTA separation and ATPase langerhans cell staining in the mouse epidermis. J. Invest. Dermatol. 80, 104–107. Bruce, S., Schroeder, T.L., Ellner, K., Rubin, H., Williams, T., Wolf Jr., J.E., 2000. Armadillo exposure and Hansen’s disease: an epidemiologic survey in southern Texas. J. Am. Acad. Dermatol. 43, 223–228. Campo-Aasen, I., Convit, J., 1975. The histochemistry of armadillo skin. J. Anat. 120, 337–348. Carrillo-Farga, J., Perez, T.A., Castell, R.A., Antuna, B.S., 1991. Adenosine triphosphatase-positive langerhans-like cells in the epidermis of the chicken (Gallus gallus). J. Anat. 176, 1–8. Chen, H., Yuan, J., Wang, Y., Silvers, W.K., 1985. Distribution of ATPase-positive langerhans cells in normal adult human skin. Br. J. Dermatol. 113, 707–711. Colston, M.J., Levy, L., 1987. Infection of other experimental animals with Mycobacterium leprae. Int. J. Lepr. Other Mycobact. Dis. 55, 896–898. Flores-Langarica, A., Meza-Perez, S., Calderon-Amador, J., EstradaGarcia, T., Macpherson, G., Lebecque, S., Saeland, S., Steinman, R.M., Flores-Romo, L., 2005. Network of dendritic cells within the muscular layer of the mouse intestine. Proc. Natl. Acad. Sci. U.S.A. 102, 19039–19044. Flores-Romo, L., 2001. In vivo maturation and migration of dendritic cells. Immunology 102, 255–262. Garcia-Romo, G.S., Pedroza-Gonzalez, A., Aguilar-Leon, D., Orozco-Estevez, H., Lambrecht, B.N., Estrada-Garcia, I., Flores-Romo, L., Hernandez-Pando, R., 2004. Airways infection with virulent Mycobacterium tuberculosis delays the influx of dendritic cells and the expression of costimulatory molecules in mediastinal lymph nodes. Immunology 112, 661–668. Gemmell, R.T., 1973. Langerhans cells in the ruminal epithelium of the sheep. J. Ultrastruct. Res. 43, 256–259. Huang, C.L., 1980. The transmission of leprosy in man. Int. J. Lepr. Other Mycobact. Dis. 48, 309–318. Jimenez-Flores, R., Mendez-Cruz, R., Ojeda-Ortiz, J., Munoz-Molina, R., Balderas-Carrillo, O., de la Luz Diaz-Soberanes, Lebecque, S., Saeland, S., Daneri-Navarro, A., Garcia-Carranca, A., Ullrich, S.E., Flores-Romo, L., 2006. High-risk human papilloma virus infection decreases the frequency of dendritic langerhans’ cells in the human female genital tract. Immunology 117, 220–228. Kirchheimer, W.F., Storrs, E.E., 1971. Attempts to establish the armadillo (Dasypus novemcinctus Linn) as a model for the study of leprosy. I. Report of lepromatoid leprosy in an experimentally infected armadillo. Int. J. Lepr. Other Mycobact. Dis. 39, 693–702.
F. Quesada-Pascual et al. / Veterinary Immunology and Immunopathology 124 (2008) 220–229 Kirchheimer, W.F., Storrs, E.E., Binford, C.H., 1972. Attempts to establish the Armadillo (Dasypus novemcinctus Linn) as a model for the study of leprosy. II. Histopathologic and bacteriologic postmortem findings in lepromatoid leprosy in the Armadillo. Int. J. Lepr. Other Mycobact. Dis. 40, 229–242. Kiyono, H., Fukuyama, S., 2004. NALT—versus Peyer’s-patchmediated mucosal immunity. Nat. Rev. Immunol. 4, 699–710. Lane, J.E., Walsh, D.S., Meyers, W.M., Klassen-Fischer, M.K., Kent, D.E., Cohen, D.J., 2006. Borderline tuberculoid leprosy in a woman from the state of Georgia with armadillo exposure. J. Am. Acad. Dermatol. 55, 714–716. Limon-Flores, A.Y., Perez-Tapia, M., Estrada-Garcia, I., Vaughan, G., Escobar-Gutierrez, A., Calderon-Amador, J., Herrera-Rodriguez, S.E., Brizuela-Garcia, A., Heras-Chavarria, M., Flores-Langarica, A., Cedillo-Barron, L., Flores-Romo, L., 2005. Dengue virus inoculation to human skin explants: an effective approach to assess in situ the early infection and the effects on cutaneous dendritic cells. Int. J. Exp. Pathol. 86, 323–334. Lumpkin III, L.R., Cox, G.F., Wolf Jr., J.E., 1983. Leprosy in five armadillo handlers. J. Am. Acad. Dermatol. 9, 899–903. Pallen, M.J., McDermott, R.D., 1986. How might Mycobacterium leprae enter the body? Lepr. Rev. 57, 289–297. Perez, T.A., Millan Aldaco, D.A., 1994. Ia antigens are expressed on ATPase-positive dendritic cells in chicken epidermis. J. Anat. 184 (Pt 3), 591–596.
229
Perez-Torres, A., Millan-Aldaco, D.A., Rondan-Zarate, A., 1995. Epidermal langerhans cells in the terrestrial turtle Kinosternum integrum. Dev. Comp. Immunol. 19, 225–236. Purtilo, D.T., Walsh, G.P., Storrs, E.E., Gannon, C., 1975. The immune system of the nine-banded armadillo (Dasypus novemcinctus Linn). Anat. Rec. 181, 725–733. Quesada-Pascual, F., Rojas-Espinosa, O., Santos, A.L., Estrada-Parra, S., 1987. A Mexican armadillo (Dasypus novemcinctus) colony for leprosy research. Int. J. Lepr. Other Mycobact. Dis. 55, 716–718. Rojas-Espinosa, O., Lovik, M., 2001. Mycobacterium leprae and Mycobacterium lepraemurium infections in domestic and wild animals. Rev. Sci. Technol. 20, 219–251. Shreedhar, V., Moodycliffe, A.M., Ullrich, S.E., Bucana, C., Kripke, M.L., Flores-Romo, L., 1999. Dendritic cells require T cells for functional maturation in vivo. Immunity 11, 625–636. Truman, R.W., Kumaresan, J.A., McDonough, C.M., Job, C.K., Hastings, R.C., 1991. Seasonal and spatial trends in the detectability of leprosy in wild armadillos. Epidemiol. Infect. 106, 549–560. Walsh, G.P., Meyers, W.M., Binford, C.H., Gormus, B.J., Baskin, G.B., Wolf, R.H., Gerone, P.J., 1988. Leprosy as a zoonosis: an update. Acta Leprol. 6, 51–60. West, B.C., Todd, J.R., Lary, C.H., Blake, L.A., Fowler, M.E., King, J.W., 1988. Leprosy in six isolated residents of northern Louisiana time-clustered cases in an essentially nonendemic area. Arch. Intern. Med. 148, 1987–1992.