- Email: [email protected]
Lethal and sub-lethal yessotoxin dose-induced morpho-functional alterations in intraperitoneal injected Swiss CD1 mice
Toxicon 44 (2004) 83–90 www.elsevier.com/locate/toxicon
Lethal and sub-lethal yessotoxin dose-induced morpho-functional alterations in intraperitoneal injected Swiss CD1 mice A. Franchinia, E. Marchesinia, R. Polettib, E. Ottaviania,* a
Department of Animal Biology, University of Modena and Reggio Emilia, Via Campi 213/D, 41100 Modena, Italy b Centro Ricerche Marine, Cesenatico, Italy
Abstract Histological and immunocytochemical investigations were performed on different organs (brain, duodenum and thymus) of mice following lethal (420 mg/kg) or sublethal (10 mg/kg) intraperitoneal injection of yessotoxin (YTX). No morpho-functional modifications were observed in large neurons of the cerebral and cerebellar cortex with the sub-lethal dose, nor in the cerebral cortex with the lethal dose. The duodenum also did not show significant alterations. However, there was an inflammation response to the toxin, in which blood cells and cytokines were involved. This was more evident with the lethal YTX dose. The thymus and, in general, the immune system are the main targets of YTX at both the concentrations used. Furthermore, the alterations present in the thymus may support tumorigenic implications. q 2004 Elsevier Ltd. All rights reserved. Keywords: Yessotoxin; Swiss CD1 mice; Brain; Duodenum; Thymus
1. Introduction
2. Materials and methods
Yessotoxin (YTX) has been reported to exert serious cytotoxic effects (Malaguti et al., 2002; Franchini et al., 2004). We have found that a dose of 420 mg/kg injected intraperitoneally in Swiss CD1 mice induces alterations in the Purkinje cells of the cerebellum. Modifications are seen in the Ca2þ-binding proteins, b-tubulin and neurofilaments, suggesting that the toxin may be involved in neurological disorders (Franchini et al., 2004). In vitro studies have shown that YTX induces an alteration in the E-cadherin system in MCF-7 cells (Pierotti et al., 2003), supporting a correlation between the loss of E-cadherin function and tumour induction (Christofori and Semb, 1999). Against this background, the present investigation extends the study of the biological effects of the intraperitoneal injection of YTX to other areas of the brain, duodenum and thymus, as main component of the immune system. Two different YTX concentrations are used: 420 and 10 mg/kg.
2.1. Animals and YTX treatment
* Corresponding author. Tel.: þ39-59-205-5536; fax: þ 39-59205-5548. E-mail address: [email protected] (E. Ottaviani). 0041-0101/$ - see front matter q 2004 Elsevier Ltd. All rights reserved. doi:10.1016/j.toxicon.2004.04.012
Male, Swiss CD1 mice weighing 17 – 19 g obtained from Morini (San Polo d’ Enza, RE, Italy) were used. Two groups of five animals received intraperitoneal injection with 420 or 10 mg/kg YTX, respectively. Immediately following death (after 2 h), the animals from the first group were dissected, and different organs, i.e. brain, thymus and duodenum were fixed in 10% neutral formol and Bouin’s mixture for the histological, histochemical and immunocytochemical procedures. The animals from the second group were sacrificed after 24 h and the same protocol as for the first group was followed. A further five animals were used as controls. The choice of two YTX concentrations, 420 and 10 mg/kg, was based on the data present in literature. The higher dose is about half that (750 mg/kg), which, although causing animal death, does not provoke any significant necroscopic changes in the injected mice (Tubaro et al., 2003). The lower dose is about ten-fold less than 1 mouse unit, i.e. the amount of toxin that kills a mouse of 20 g body weight in 24 h (Ciminiello et al., 1997), corresponding to the lethal YTX dose of 100 mg/kg found by Murata et al. (1987) and Ogino et al. (1997).
84
A. Franchini et al. / Toxicon 44 (2004) 83–90
2.2. Histological, histochemical and immunocytochemical procedures The specimens were embedded in paraffin wax and cut into 7 mm sections. The following stains were performed (Bancroft and Stevens, 1996): hematoxylin/eosin and trichrome stainings for general morphology; toluidine blue reaction for Nissl bodies; Feulgen method for DNA, PAS/hematoxylin and Alcian blue pH 2.5 for carbohydrates. For the immunocytochemical procedure, the sections were treated overnight at 4 8C with the following primary antibodies: anti-calbindin D-28K polyclonal antibody (pAb) (1:1000) (Chemicon, Temecula, CA, USA), anti-S-100 protein monoclonal antibody (mAb) (1:250) (Chemicon), anti-b-tubulin mAb (1:500) (Chemicon), anti-neurofilament 160 mAb (1:250) (Sigma Chemical Comp., St Louis, MO, USA), anti-cytokeratin 8 mAb (5 mg/ml) (Roche Molecular Bichemicals, Germany), anti-cytokeratin 18 mAb (5 mg/ml) (Roche Molecular Bichemicals), anti-cytokeratin 1 mAb (1:100) (Novocastra Laboratories, Newcastle, UK), anti-cytokeratin 1/5/10/14 mAb (1:250) (Novocastra), anti-vimentin mAb (1:500) (Novocastra), anti-IL-1a pAb (1:1000) (Santa Cruz Biotechnology, Santa Cruz, CA, USA), anti-IL-1a mAb (1:1000) (Santa Cruz), anti-IL-6 pAb (1:1000) (Santa Cruz), anti-IL-8 mAb (1:1000) (Santa Cruz) and anti-TNF-a pAb (1:2000) (Santa Cruz). The labelling sites were revealed using avidin – biotin– peroxidase complex (ABC). Nuclei were counterstained with hematoxylin. Negative controls were performed by substituting the primary antibodies with non-immune sera. A detailed immunocytochemical protocol is described elsewhere (Ottaviani et al., 1995). 2.3. Quantification of apoptotic and mitotic cells in the thymus The number of apoptotic and mitotic cells present in thymic sections was recorded in 100 cortical and medullary microscopic fields. The differences between control and treated samples were analysed statistically by ANOVA.
3. Results The morphological and immunocytochemical investigations performed on the large neurons of the cerebral and cerebellar cortex of mice injected with the sub-lethal (10 mg/kg) YTX dose and on the cerebral cortex of mice injected with the lethal (420 mg/kg) YTX dose did not reveal any significant morpho-functional modifications. Indeed, the morphological stainings showed no signs of sufferance at either cytoplasmic or nuclear level. The immunocytochemical reactions for the S-100, the calbindin D-28K, the neurotubule and neurofilament proteins showed the same patterns in neurons from controls
and from YTX injected animals. Here, it should be remembered that in a previous study on the cerebellar cortex of mice receiving a lethal dose of YTX by intraperitoneal injection, severe damage to the Purkinje cells was observed (Franchini et al., 2004). With regard to the duodenum and thymus, both organs were seen to be affected by the lethal and the sub-lethal YTX doses. Specifically, the duodenum was more sensitive to the higher dose, while the thymus reacted to both doses, but particularly to the sub-lethal concentration. The general structure of the duodenum of mice receiving a lethal dose seems preserved, although a higher number of infiltrated blood cells between the epithelial cells of villi was observed (Fig. 1a and b). Moreover, some blood cells, mainly lymphocytes, located in the connective tissue and epithelial layer of the mucosa and in lymphoid Payer’s patches showed apoptotic phenotypes (Fig. 1c and d). The tested cytokines presented different behaviour. In control samples, cells immunoreactive to IL-6 were distributed in the loose connective tissue of duodenum mucosa and were mainly granulocytes. After YTX treatment, a larger number of strongly stained immunoreactive cells was observed, and these were granulocytes and macrophages. TNF-a immunoreactive cells were also seen in the connective tissue of controls, while in treated samples, the number of immunoreactive cells increased, and intraepithelial cells were also positive (Fig. 1e and f). The number of cells immunoreactive to anti-IL-8 antibody decreased after YTX treatment (Fig. 1g and h). No evident histological modifications were observed with the sub-lethal YTX dose, while the cytokine responses were still detected with the same trend as for the lethal dose. Histological studies of the thymus revealed morphological modifications with both YTX doses. Lethal treatment provoked structural changes in the cortex region that appeared less compact with light areas containing a reduced number of thymocytes and large, pale epithelial cells (Fig. 2a – c). The number of apoptotic phenotypes significantly increased and was higher in the thymic medullary area than the cortex (Figs. 2d and 3). The cell types undergoing cell death were mostly thymocytes. In contrast, mitosis was more stimulated in the cortex than in the medulla, and the proliferation rate was about five times greater than controls (Fig. 4). More severe structural damage was observed with the sub-lethal dose. In particular, in the cortico-medullary junction and in the medulla, there was a significantly higher number of apoptotic cells, and this response was more conspicuous than with the lethal treatment (Figs. 2e and 3). With regard to mitosis, a significantly larger number of mitotic cells was found, showing the same trend observed with higher dose (Fig. 4). Groups of hypertrophic and flattened medullary epithelial cells formed single or clustered round structures that resembled Hassall’s corpuscles and contained Alcian blue-PAS positive heterogeneous secretory material, cell debris and necrotic nuclei
A. Franchini et al. / Toxicon 44 (2004) 83–90
85
Fig. 1. Cross sections of duodenum from control (a, e, g), 420 mg/kg (b, c, d, h) and 10 mg/kg (f) YTX injected mice. Hematoxylin/eosin (a, b, c) and PAS/hematoxylin (d) stains: intraepithelial blood cells (arrows), apoptotic cells (arrowheads). Immunocytochemical reactions with anti-TNF-a pAb (e, f) and with anti-IL-8 mAb (g, h). Bar ¼ 10 mm.
(Fig. 2f and g). The dendritic cells showed cytological features of active phagocytic cells (Fig. 2h). The medullary epithelial cells were the most affected cell population. These cells were arranged in a regular reticulum of stellate cells immunoreactive to high MW cytokeratins (Fig. 5a – c). After YTX treatment, some cells decreased their
immunoreactivity and others withdrew cytoplasmic projections becoming rounded and strongly immunopositive (Fig. 5d and e). The outer layer or the core of the newly formed medullary structures was also immunoreactive to higher MW cytokeratins (Fig. 5f). Apart from subcapsular epithelial cells positive to high MW cytokeratins (Fig. 5g),
86
A. Franchini et al. / Toxicon 44 (2004) 83–90
Fig. 2. Thymic sections from control (a), 420 mg/kg (b,c,d) and 10 mg/kg (e,f,g,h) YTX injected mice. Trichrome (a, c), hematoxylin/eosin (b, d, f, g) stains and PAS/hematoxylin reaction (e). Note isolated (arrows) (d), and grouped (arrowheads) (e) apoptotic cells. The newly formed Hassall’s corpuscle-like bodies are shown (f, g). Active phagocytic dendritic cells are immunostained with anti-vimentin mAb (arrows) (h). Cortex (C); medulla (M); cortico-medullary junction (CMJ). Bar ¼ 10 and 50 mm (b).
cortical epithelial cells showed small amounts of low MW cytokeratins in controls (Fig. 6a). After YTX treatment, a lower expression of low MW cytokeratins was observed (Fig. 6b) and some single cortical cells were positive to high MW cytokeratins (Fig. 5h). With regard to cytokine response, changes in comparison to controls were observed with both YTX concentrations. There were more cells immunoreactive to IL-6 located at the cortico-medullary junction and the medulla, and these were mostly dendritic cells (Fig. 6c– f). The number of IL-1a
and IL-8 immunoreactive cells observed in the cortex decreased (Fig. 6g and h).
4. Discussion Our findings show that YTX mainly affects the immune system of mice, with alterations in both the cellular and humoral component, as well as in the lymphoid organ
A. Franchini et al. / Toxicon 44 (2004) 83–90
87
Fig. 3. Number of apoptotic cells recorded in thymic sections from control, 420 and 10 mg/kg YTX injected mice. Statistical significance is determined by ANOVA (*, ** p , 0:05 vs control; ** p , 0:05 vs *).
thymus. With regard to the cellular component, blood cells show apoptotic phenotypes in the duodenum and thymus of treated animals. The different cytokines tested, mediators of the humoral component, reacted to the marine toxin. IL-6, and TNF-a immunoreactive cells increased, while IL-1a and IL-8 decreased in an organ-independent manner. However, the most severe morpho-functional damages are observed in the thymus, and the medulla is the most affected thymic area with evident alterations. The population of the medullary epithelial cells changes in morphology, modifies
the expression of cytokeratins and contributes to the formation of many Hassall’s corpuscle-like bodies that show heterogeneous structures probably due to different maturational stages. This last observation suggests a faster epithelial differentiation into keratinized epithelial cells and this is also indicated by the medullary phenotype, with high MW cytokeratins, of some cortical epithelial cells. On the whole, these changes in the thymic microenvironment organization may alter the interactions between epithelial cells and thymocytes through surface molecules
Fig. 4. Number of mitotic cells recorded in thymic sections from control, 420 and 10 mg/kg YTX injected mice. Statistical significance is determined by ANOVA (*, ** p , 0:05 vs control; ** p , 0:05 vs *).
88
A. Franchini et al. / Toxicon 44 (2004) 83–90
Fig. 5. Sections of thymus from control (a,b,c,g) and 10 mg/kg (d,e,f,h) YTX injected mice immunostained with anti-cytokeratin 1/5/10/14 mAb. Note the changes in organization of the medullary epithelial compartment with modifications in cell morphology and cytokeratin immunoreactivity. Subcapsular epithelial cells (arrowheads); cortex (C); medulla (M); cortico-medullary junction (CMJ). Bar ¼ 10 mm.
and soluble products that are critical for organ physiology (Savino and Dardenne, 2000). In this respect, we observed modifications in cytokine responses. Moreover, thymic epithelial cell organization and thymocyte development involve E-cadherin interactions (Muller et al., 1997), and in vitro studies have demonstrated that YTX induces selective disruption of the E-cadherin-catenin system (Ronzitti et al., 2004). The altered thymic function is also indicated by the larger number of apoptotic cells found in the cortico-medullary
junction and in the medulla, resulting in a reduced mature thymocyte population leaving the thymus and moving to peripheral lymphoid organs. In vitro studies have characterized YTX as a potent cytotoxic compound, and sub-nanomolar concentrations are able to induce cell death by involving specific caspase isoforms (Malaguti et al., 2002). The critical role played by active cell death (apoptosis), is now accepted not only in physiological functions of the immune system, but also in toxicities and diseases (Corcoran et al., 1994; Islam et al., 1998).
A. Franchini et al. / Toxicon 44 (2004) 83–90
89
Fig. 6. Sections of thymus from control (a, c, g), 420 mg/kg (d, e) and 10 mg/kg (b, f, h) YTX injected mice immunostained with anti-cytokeratin 18 mAb (a, b), anti-IL-6 pAb (c –f) and with anti-IL-8 mAb (g, h). Dendritic cells (arrows); epithelial cells (arrowheads); granulocyte ( p ). Bar ¼ 10 and 50 mm (c, d).
The morpho-functional alterations found in the thymus of treated mice are compatible with those reported in thymic tumours (Ring and Addis, 1986; Fukai et al., 1993; Palestro et al., 1998; Kuo, 2000) suggesting a possible tumorigenic property of YTX. These findings represent an important signal of risk for the consumers of shellfish, and in this context, we have planned experiments with particular attention to the thymus from mice fed with YTX contaminated mussels. Recently, it has been reported that the oral administration of a sub-lethal dose of azaspiracid in
mice strongly indicates the tumorigenic property of this marine toxin (Ito et al., 2002). The study of the cortex of the cerebrum and the cerebellum from YTX injected mice does not show alterations, indicating a time- and dose-dependent effect. Indeed, only at the lethal dose did we see sufferance in the large neurons of the cerebellum, i.e. Purkinje cells (Franchini et al., 2004). The duodenum also shows no significant alterations. The only phenomenon observed is the inflammation response to the toxin. This is more evident
90
A. Franchini et al. / Toxicon 44 (2004) 83–90
at the lethal YTX dose, in which the main actors of this process are involved, i.e. blood cells and cytokines. The present results and others previously published (Franchini et al., 2004) suggest that the main target of YTX is the thymus. This organ is sensitive even to the sub-lethal YTX concentration, while alterations to the cerebellum are detected only at the lethal dose. Furthermore, the alterations present in the thymus of animals receiving intraperitoneal YTX injection may support tumorigenic implications. Acknowledgements This work was supported by a MIUR (Italy) grant to E.O. References Bancroft, J.D., Stevens, A., 1996. Theory and Practice of Histological Techniques, Churchill Livingstone, Edinburgh. Christofori, G., Semb, H., 1999. The role of the cell-adhesion molecule E-cadherin as a tumour-suppressor gene. Trends Biochem. Sci. 24, 73 –76. Ciminiello, P., Fattorusso, E., Forino, M., Magno, S., Poletti, R., Satake, M., Viviani, R., Yasumoto, T., 1997. Yessotoxin in mussels of the northern adriatic sea. Toxicon 35, 177 –183. Corcoran, G.B., Fix, L., Jones, D.P., Moslen, M.T., Nicotera, P., Oberhammer, F.A., Buttyan, R., 1994. Apoptosis: molecular control point in toxicity. Toxicol. Appl. Pharmacol. 128, 169–181. Franchini, A., Marchesini, E., Poletti, R., Ottaviani, E., 2004. Acute toxic effect of the algal yessotoxin on Purkinje cells from the cerebellum of Swiss CD1 mice. Toxicon 43, 347–352. Fukai, I., Masaoka, A., Hashimoto, T., Yamakawa, Y., Mizuno, T., Tanamura, O., 1993. Cytokeratins in normal thymus and thymic epithelial tumors. Cancer 71, 99–105. Islam, Z., Nagase, M., Yoshizawa, T., Yamauchi, K., Sakato, N., 1998. T-2 toxin induces thymic apoptosis in vivo in mice. Toxicol. Appl. Pharmacol. 148, 205– 214. Ito, E., Satake, M., Ofuji, K., Higashi, M., Harigaya, K., McMahon, T., Yasumoto, T., 2002. Chronic effects in mice caused by oral
administration of sublethal doses of azaspiracid, a new marine toxin isolated from mussels. Toxicon 40, 193–203. Kuo, T., 2000. Cytokeratin profiles of the thymus and thymomas: histogenetic correlations and proposal for a histological classification of thymomas. Histopathology 36, 403 –414. Malaguti, C., Ciminiello, P., Fattorusso, E., Rossini, G.P., 2002. Caspase activation and death induced by yessotoxin in HeLa cells. Toxicol. In vitro 16, 357–363. Muller, K.M., Luedecker, C.J., Udey, M.C., Farr, A.G., 1997. Involvement of E-cadherin in thymus organogenesis and thymocyte maturation. Immunity 6, 257–264. Murata, M., Kumagai, M., Lee, J.S., Yasumoto, T., 1987. Isolation and structure of yessotoxin, a novel polyether compound implicated in diarrhetic shellfish poisoning. Tetrahedron Lett. 28, 5869–5872. Ogino, H., Kumagai, M., Yasumoto, T., 1997. Toxicological evaluation of yessotoxin. Nat. Toxins 5, 255 –259. Ottaviani, E., Franchini, A., Franceschi, C., 1995. Evidence for the presence of immunoreactive pro-opiomelanocortin-derived peptides and cytokines in the thymus of the goldfish (Carassius c. auratus). Histochem. J. 27, 597–601. Palestro, G., Novero, D., Geuna, M., Chiarle, R., Chiusa, L., Pagano, M., Pich, A., 1998. Role of the perivascular epithelium in the histogenesis of Hassall’s corpuscles. A morphologic and immunohistological study. Int. J. Surg. Pathol. 6, 213–222. Pierotti, S., Malaguti, C., Milandri, A., Poletti, R., Rossini, G.P., 2003. Functional assay to measure yessotoxins in contaminated mussel samples. Anal. Biochem. 312, 208 –216. Ring, N.P., Addis, B.J., 1986. Thymoma: an integrated clinicopathological and immunohistochemical study. J. Pathol. 149, 327 –337. Ronzitti, G., Callegari, F., Malaguti, C., Rossini, G.P., 2004. Selective disruption of the E-cadherin-catenin system by algal toxin. Br. J. Cancer 90, 1100–1107. Savino, W., Dardenne, M., 2000. Neuroendocrine control of thymus physiology. Endocr. Rev. 21, 412 –443. Tubaro, A., Sosa, S., Carbonatto, M., Altinier, G., Vita, F., Melato, M., Satake, M., Yasumoto, T., 2003. Oral and intraperitoneal acute toxicity studies of yessotoxin and homoyessotoxins in mice. Toxicon 41, 783 –792.
Lethal and sub-lethal yessotoxin dose-induced morpho-functional alterations in intraperitoneal injected Swiss CD1 mice A. Franchinia, E. Marchesinia, R. Polettib, E. Ottaviania,* a
Department of Animal Biology, University of Modena and Reggio Emilia, Via Campi 213/D, 41100 Modena, Italy b Centro Ricerche Marine, Cesenatico, Italy
Abstract Histological and immunocytochemical investigations were performed on different organs (brain, duodenum and thymus) of mice following lethal (420 mg/kg) or sublethal (10 mg/kg) intraperitoneal injection of yessotoxin (YTX). No morpho-functional modifications were observed in large neurons of the cerebral and cerebellar cortex with the sub-lethal dose, nor in the cerebral cortex with the lethal dose. The duodenum also did not show significant alterations. However, there was an inflammation response to the toxin, in which blood cells and cytokines were involved. This was more evident with the lethal YTX dose. The thymus and, in general, the immune system are the main targets of YTX at both the concentrations used. Furthermore, the alterations present in the thymus may support tumorigenic implications. q 2004 Elsevier Ltd. All rights reserved. Keywords: Yessotoxin; Swiss CD1 mice; Brain; Duodenum; Thymus
1. Introduction
2. Materials and methods
Yessotoxin (YTX) has been reported to exert serious cytotoxic effects (Malaguti et al., 2002; Franchini et al., 2004). We have found that a dose of 420 mg/kg injected intraperitoneally in Swiss CD1 mice induces alterations in the Purkinje cells of the cerebellum. Modifications are seen in the Ca2þ-binding proteins, b-tubulin and neurofilaments, suggesting that the toxin may be involved in neurological disorders (Franchini et al., 2004). In vitro studies have shown that YTX induces an alteration in the E-cadherin system in MCF-7 cells (Pierotti et al., 2003), supporting a correlation between the loss of E-cadherin function and tumour induction (Christofori and Semb, 1999). Against this background, the present investigation extends the study of the biological effects of the intraperitoneal injection of YTX to other areas of the brain, duodenum and thymus, as main component of the immune system. Two different YTX concentrations are used: 420 and 10 mg/kg.
2.1. Animals and YTX treatment
* Corresponding author. Tel.: þ39-59-205-5536; fax: þ 39-59205-5548. E-mail address: [email protected] (E. Ottaviani). 0041-0101/$ - see front matter q 2004 Elsevier Ltd. All rights reserved. doi:10.1016/j.toxicon.2004.04.012
Male, Swiss CD1 mice weighing 17 – 19 g obtained from Morini (San Polo d’ Enza, RE, Italy) were used. Two groups of five animals received intraperitoneal injection with 420 or 10 mg/kg YTX, respectively. Immediately following death (after 2 h), the animals from the first group were dissected, and different organs, i.e. brain, thymus and duodenum were fixed in 10% neutral formol and Bouin’s mixture for the histological, histochemical and immunocytochemical procedures. The animals from the second group were sacrificed after 24 h and the same protocol as for the first group was followed. A further five animals were used as controls. The choice of two YTX concentrations, 420 and 10 mg/kg, was based on the data present in literature. The higher dose is about half that (750 mg/kg), which, although causing animal death, does not provoke any significant necroscopic changes in the injected mice (Tubaro et al., 2003). The lower dose is about ten-fold less than 1 mouse unit, i.e. the amount of toxin that kills a mouse of 20 g body weight in 24 h (Ciminiello et al., 1997), corresponding to the lethal YTX dose of 100 mg/kg found by Murata et al. (1987) and Ogino et al. (1997).
84
A. Franchini et al. / Toxicon 44 (2004) 83–90
2.2. Histological, histochemical and immunocytochemical procedures The specimens were embedded in paraffin wax and cut into 7 mm sections. The following stains were performed (Bancroft and Stevens, 1996): hematoxylin/eosin and trichrome stainings for general morphology; toluidine blue reaction for Nissl bodies; Feulgen method for DNA, PAS/hematoxylin and Alcian blue pH 2.5 for carbohydrates. For the immunocytochemical procedure, the sections were treated overnight at 4 8C with the following primary antibodies: anti-calbindin D-28K polyclonal antibody (pAb) (1:1000) (Chemicon, Temecula, CA, USA), anti-S-100 protein monoclonal antibody (mAb) (1:250) (Chemicon), anti-b-tubulin mAb (1:500) (Chemicon), anti-neurofilament 160 mAb (1:250) (Sigma Chemical Comp., St Louis, MO, USA), anti-cytokeratin 8 mAb (5 mg/ml) (Roche Molecular Bichemicals, Germany), anti-cytokeratin 18 mAb (5 mg/ml) (Roche Molecular Bichemicals), anti-cytokeratin 1 mAb (1:100) (Novocastra Laboratories, Newcastle, UK), anti-cytokeratin 1/5/10/14 mAb (1:250) (Novocastra), anti-vimentin mAb (1:500) (Novocastra), anti-IL-1a pAb (1:1000) (Santa Cruz Biotechnology, Santa Cruz, CA, USA), anti-IL-1a mAb (1:1000) (Santa Cruz), anti-IL-6 pAb (1:1000) (Santa Cruz), anti-IL-8 mAb (1:1000) (Santa Cruz) and anti-TNF-a pAb (1:2000) (Santa Cruz). The labelling sites were revealed using avidin – biotin– peroxidase complex (ABC). Nuclei were counterstained with hematoxylin. Negative controls were performed by substituting the primary antibodies with non-immune sera. A detailed immunocytochemical protocol is described elsewhere (Ottaviani et al., 1995). 2.3. Quantification of apoptotic and mitotic cells in the thymus The number of apoptotic and mitotic cells present in thymic sections was recorded in 100 cortical and medullary microscopic fields. The differences between control and treated samples were analysed statistically by ANOVA.
3. Results The morphological and immunocytochemical investigations performed on the large neurons of the cerebral and cerebellar cortex of mice injected with the sub-lethal (10 mg/kg) YTX dose and on the cerebral cortex of mice injected with the lethal (420 mg/kg) YTX dose did not reveal any significant morpho-functional modifications. Indeed, the morphological stainings showed no signs of sufferance at either cytoplasmic or nuclear level. The immunocytochemical reactions for the S-100, the calbindin D-28K, the neurotubule and neurofilament proteins showed the same patterns in neurons from controls
and from YTX injected animals. Here, it should be remembered that in a previous study on the cerebellar cortex of mice receiving a lethal dose of YTX by intraperitoneal injection, severe damage to the Purkinje cells was observed (Franchini et al., 2004). With regard to the duodenum and thymus, both organs were seen to be affected by the lethal and the sub-lethal YTX doses. Specifically, the duodenum was more sensitive to the higher dose, while the thymus reacted to both doses, but particularly to the sub-lethal concentration. The general structure of the duodenum of mice receiving a lethal dose seems preserved, although a higher number of infiltrated blood cells between the epithelial cells of villi was observed (Fig. 1a and b). Moreover, some blood cells, mainly lymphocytes, located in the connective tissue and epithelial layer of the mucosa and in lymphoid Payer’s patches showed apoptotic phenotypes (Fig. 1c and d). The tested cytokines presented different behaviour. In control samples, cells immunoreactive to IL-6 were distributed in the loose connective tissue of duodenum mucosa and were mainly granulocytes. After YTX treatment, a larger number of strongly stained immunoreactive cells was observed, and these were granulocytes and macrophages. TNF-a immunoreactive cells were also seen in the connective tissue of controls, while in treated samples, the number of immunoreactive cells increased, and intraepithelial cells were also positive (Fig. 1e and f). The number of cells immunoreactive to anti-IL-8 antibody decreased after YTX treatment (Fig. 1g and h). No evident histological modifications were observed with the sub-lethal YTX dose, while the cytokine responses were still detected with the same trend as for the lethal dose. Histological studies of the thymus revealed morphological modifications with both YTX doses. Lethal treatment provoked structural changes in the cortex region that appeared less compact with light areas containing a reduced number of thymocytes and large, pale epithelial cells (Fig. 2a – c). The number of apoptotic phenotypes significantly increased and was higher in the thymic medullary area than the cortex (Figs. 2d and 3). The cell types undergoing cell death were mostly thymocytes. In contrast, mitosis was more stimulated in the cortex than in the medulla, and the proliferation rate was about five times greater than controls (Fig. 4). More severe structural damage was observed with the sub-lethal dose. In particular, in the cortico-medullary junction and in the medulla, there was a significantly higher number of apoptotic cells, and this response was more conspicuous than with the lethal treatment (Figs. 2e and 3). With regard to mitosis, a significantly larger number of mitotic cells was found, showing the same trend observed with higher dose (Fig. 4). Groups of hypertrophic and flattened medullary epithelial cells formed single or clustered round structures that resembled Hassall’s corpuscles and contained Alcian blue-PAS positive heterogeneous secretory material, cell debris and necrotic nuclei
A. Franchini et al. / Toxicon 44 (2004) 83–90
85
Fig. 1. Cross sections of duodenum from control (a, e, g), 420 mg/kg (b, c, d, h) and 10 mg/kg (f) YTX injected mice. Hematoxylin/eosin (a, b, c) and PAS/hematoxylin (d) stains: intraepithelial blood cells (arrows), apoptotic cells (arrowheads). Immunocytochemical reactions with anti-TNF-a pAb (e, f) and with anti-IL-8 mAb (g, h). Bar ¼ 10 mm.
(Fig. 2f and g). The dendritic cells showed cytological features of active phagocytic cells (Fig. 2h). The medullary epithelial cells were the most affected cell population. These cells were arranged in a regular reticulum of stellate cells immunoreactive to high MW cytokeratins (Fig. 5a – c). After YTX treatment, some cells decreased their
immunoreactivity and others withdrew cytoplasmic projections becoming rounded and strongly immunopositive (Fig. 5d and e). The outer layer or the core of the newly formed medullary structures was also immunoreactive to higher MW cytokeratins (Fig. 5f). Apart from subcapsular epithelial cells positive to high MW cytokeratins (Fig. 5g),
86
A. Franchini et al. / Toxicon 44 (2004) 83–90
Fig. 2. Thymic sections from control (a), 420 mg/kg (b,c,d) and 10 mg/kg (e,f,g,h) YTX injected mice. Trichrome (a, c), hematoxylin/eosin (b, d, f, g) stains and PAS/hematoxylin reaction (e). Note isolated (arrows) (d), and grouped (arrowheads) (e) apoptotic cells. The newly formed Hassall’s corpuscle-like bodies are shown (f, g). Active phagocytic dendritic cells are immunostained with anti-vimentin mAb (arrows) (h). Cortex (C); medulla (M); cortico-medullary junction (CMJ). Bar ¼ 10 and 50 mm (b).
cortical epithelial cells showed small amounts of low MW cytokeratins in controls (Fig. 6a). After YTX treatment, a lower expression of low MW cytokeratins was observed (Fig. 6b) and some single cortical cells were positive to high MW cytokeratins (Fig. 5h). With regard to cytokine response, changes in comparison to controls were observed with both YTX concentrations. There were more cells immunoreactive to IL-6 located at the cortico-medullary junction and the medulla, and these were mostly dendritic cells (Fig. 6c– f). The number of IL-1a
and IL-8 immunoreactive cells observed in the cortex decreased (Fig. 6g and h).
4. Discussion Our findings show that YTX mainly affects the immune system of mice, with alterations in both the cellular and humoral component, as well as in the lymphoid organ
A. Franchini et al. / Toxicon 44 (2004) 83–90
87
Fig. 3. Number of apoptotic cells recorded in thymic sections from control, 420 and 10 mg/kg YTX injected mice. Statistical significance is determined by ANOVA (*, ** p , 0:05 vs control; ** p , 0:05 vs *).
thymus. With regard to the cellular component, blood cells show apoptotic phenotypes in the duodenum and thymus of treated animals. The different cytokines tested, mediators of the humoral component, reacted to the marine toxin. IL-6, and TNF-a immunoreactive cells increased, while IL-1a and IL-8 decreased in an organ-independent manner. However, the most severe morpho-functional damages are observed in the thymus, and the medulla is the most affected thymic area with evident alterations. The population of the medullary epithelial cells changes in morphology, modifies
the expression of cytokeratins and contributes to the formation of many Hassall’s corpuscle-like bodies that show heterogeneous structures probably due to different maturational stages. This last observation suggests a faster epithelial differentiation into keratinized epithelial cells and this is also indicated by the medullary phenotype, with high MW cytokeratins, of some cortical epithelial cells. On the whole, these changes in the thymic microenvironment organization may alter the interactions between epithelial cells and thymocytes through surface molecules
Fig. 4. Number of mitotic cells recorded in thymic sections from control, 420 and 10 mg/kg YTX injected mice. Statistical significance is determined by ANOVA (*, ** p , 0:05 vs control; ** p , 0:05 vs *).
88
A. Franchini et al. / Toxicon 44 (2004) 83–90
Fig. 5. Sections of thymus from control (a,b,c,g) and 10 mg/kg (d,e,f,h) YTX injected mice immunostained with anti-cytokeratin 1/5/10/14 mAb. Note the changes in organization of the medullary epithelial compartment with modifications in cell morphology and cytokeratin immunoreactivity. Subcapsular epithelial cells (arrowheads); cortex (C); medulla (M); cortico-medullary junction (CMJ). Bar ¼ 10 mm.
and soluble products that are critical for organ physiology (Savino and Dardenne, 2000). In this respect, we observed modifications in cytokine responses. Moreover, thymic epithelial cell organization and thymocyte development involve E-cadherin interactions (Muller et al., 1997), and in vitro studies have demonstrated that YTX induces selective disruption of the E-cadherin-catenin system (Ronzitti et al., 2004). The altered thymic function is also indicated by the larger number of apoptotic cells found in the cortico-medullary
junction and in the medulla, resulting in a reduced mature thymocyte population leaving the thymus and moving to peripheral lymphoid organs. In vitro studies have characterized YTX as a potent cytotoxic compound, and sub-nanomolar concentrations are able to induce cell death by involving specific caspase isoforms (Malaguti et al., 2002). The critical role played by active cell death (apoptosis), is now accepted not only in physiological functions of the immune system, but also in toxicities and diseases (Corcoran et al., 1994; Islam et al., 1998).
A. Franchini et al. / Toxicon 44 (2004) 83–90
89
Fig. 6. Sections of thymus from control (a, c, g), 420 mg/kg (d, e) and 10 mg/kg (b, f, h) YTX injected mice immunostained with anti-cytokeratin 18 mAb (a, b), anti-IL-6 pAb (c –f) and with anti-IL-8 mAb (g, h). Dendritic cells (arrows); epithelial cells (arrowheads); granulocyte ( p ). Bar ¼ 10 and 50 mm (c, d).
The morpho-functional alterations found in the thymus of treated mice are compatible with those reported in thymic tumours (Ring and Addis, 1986; Fukai et al., 1993; Palestro et al., 1998; Kuo, 2000) suggesting a possible tumorigenic property of YTX. These findings represent an important signal of risk for the consumers of shellfish, and in this context, we have planned experiments with particular attention to the thymus from mice fed with YTX contaminated mussels. Recently, it has been reported that the oral administration of a sub-lethal dose of azaspiracid in
mice strongly indicates the tumorigenic property of this marine toxin (Ito et al., 2002). The study of the cortex of the cerebrum and the cerebellum from YTX injected mice does not show alterations, indicating a time- and dose-dependent effect. Indeed, only at the lethal dose did we see sufferance in the large neurons of the cerebellum, i.e. Purkinje cells (Franchini et al., 2004). The duodenum also shows no significant alterations. The only phenomenon observed is the inflammation response to the toxin. This is more evident
90
A. Franchini et al. / Toxicon 44 (2004) 83–90
at the lethal YTX dose, in which the main actors of this process are involved, i.e. blood cells and cytokines. The present results and others previously published (Franchini et al., 2004) suggest that the main target of YTX is the thymus. This organ is sensitive even to the sub-lethal YTX concentration, while alterations to the cerebellum are detected only at the lethal dose. Furthermore, the alterations present in the thymus of animals receiving intraperitoneal YTX injection may support tumorigenic implications. Acknowledgements This work was supported by a MIUR (Italy) grant to E.O. References Bancroft, J.D., Stevens, A., 1996. Theory and Practice of Histological Techniques, Churchill Livingstone, Edinburgh. Christofori, G., Semb, H., 1999. The role of the cell-adhesion molecule E-cadherin as a tumour-suppressor gene. Trends Biochem. Sci. 24, 73 –76. Ciminiello, P., Fattorusso, E., Forino, M., Magno, S., Poletti, R., Satake, M., Viviani, R., Yasumoto, T., 1997. Yessotoxin in mussels of the northern adriatic sea. Toxicon 35, 177 –183. Corcoran, G.B., Fix, L., Jones, D.P., Moslen, M.T., Nicotera, P., Oberhammer, F.A., Buttyan, R., 1994. Apoptosis: molecular control point in toxicity. Toxicol. Appl. Pharmacol. 128, 169–181. Franchini, A., Marchesini, E., Poletti, R., Ottaviani, E., 2004. Acute toxic effect of the algal yessotoxin on Purkinje cells from the cerebellum of Swiss CD1 mice. Toxicon 43, 347–352. Fukai, I., Masaoka, A., Hashimoto, T., Yamakawa, Y., Mizuno, T., Tanamura, O., 1993. Cytokeratins in normal thymus and thymic epithelial tumors. Cancer 71, 99–105. Islam, Z., Nagase, M., Yoshizawa, T., Yamauchi, K., Sakato, N., 1998. T-2 toxin induces thymic apoptosis in vivo in mice. Toxicol. Appl. Pharmacol. 148, 205– 214. Ito, E., Satake, M., Ofuji, K., Higashi, M., Harigaya, K., McMahon, T., Yasumoto, T., 2002. Chronic effects in mice caused by oral
administration of sublethal doses of azaspiracid, a new marine toxin isolated from mussels. Toxicon 40, 193–203. Kuo, T., 2000. Cytokeratin profiles of the thymus and thymomas: histogenetic correlations and proposal for a histological classification of thymomas. Histopathology 36, 403 –414. Malaguti, C., Ciminiello, P., Fattorusso, E., Rossini, G.P., 2002. Caspase activation and death induced by yessotoxin in HeLa cells. Toxicol. In vitro 16, 357–363. Muller, K.M., Luedecker, C.J., Udey, M.C., Farr, A.G., 1997. Involvement of E-cadherin in thymus organogenesis and thymocyte maturation. Immunity 6, 257–264. Murata, M., Kumagai, M., Lee, J.S., Yasumoto, T., 1987. Isolation and structure of yessotoxin, a novel polyether compound implicated in diarrhetic shellfish poisoning. Tetrahedron Lett. 28, 5869–5872. Ogino, H., Kumagai, M., Yasumoto, T., 1997. Toxicological evaluation of yessotoxin. Nat. Toxins 5, 255 –259. Ottaviani, E., Franchini, A., Franceschi, C., 1995. Evidence for the presence of immunoreactive pro-opiomelanocortin-derived peptides and cytokines in the thymus of the goldfish (Carassius c. auratus). Histochem. J. 27, 597–601. Palestro, G., Novero, D., Geuna, M., Chiarle, R., Chiusa, L., Pagano, M., Pich, A., 1998. Role of the perivascular epithelium in the histogenesis of Hassall’s corpuscles. A morphologic and immunohistological study. Int. J. Surg. Pathol. 6, 213–222. Pierotti, S., Malaguti, C., Milandri, A., Poletti, R., Rossini, G.P., 2003. Functional assay to measure yessotoxins in contaminated mussel samples. Anal. Biochem. 312, 208 –216. Ring, N.P., Addis, B.J., 1986. Thymoma: an integrated clinicopathological and immunohistochemical study. J. Pathol. 149, 327 –337. Ronzitti, G., Callegari, F., Malaguti, C., Rossini, G.P., 2004. Selective disruption of the E-cadherin-catenin system by algal toxin. Br. J. Cancer 90, 1100–1107. Savino, W., Dardenne, M., 2000. Neuroendocrine control of thymus physiology. Endocr. Rev. 21, 412 –443. Tubaro, A., Sosa, S., Carbonatto, M., Altinier, G., Vita, F., Melato, M., Satake, M., Yasumoto, T., 2003. Oral and intraperitoneal acute toxicity studies of yessotoxin and homoyessotoxins in mice. Toxicon 41, 783 –792.