- Email: [email protected]
Occurrence of adrenergic nerve fibers in human thymus during immune response
Neurochemistry International 40 (2002) 211– 221 www.elsevier.com/locate/neuint
Occurrence of adrenergic nerve fibers in human thymus during immune response Daniela Cavallotti a, Marco Artico b, Giandomenico Iannetti a,c, Carlo Cavallotti c,* a
Department of Neurosciences, Uni6ersity of Rome ‘La Sapienza’, Rome, Italy Chair of Human Anatomy, Faculty of Pharmacy, Uni6ersity of Rome ‘La Sapienza’, Rome, Italy c Department of Cardio6ascular and Respiratory Sciences (Section of Anatomy), Uni6ersity of Rome ‘La Sapienza’, Via A. Borelli, 50 00161 Rome, Italy b
Received 8 June 2001; accepted 16 June 2001
Abstract The adrenergic nerve fibers (ANF), the neuropeptide Y-like immunoreactive nerve fibers (NPY-NF) and the noradrenaline (NA) amount were studied in the human thymus in subjects previously treated or not treated with interferon therapy with the aim to identify the changes due to the interferon therapy. This therapy has been used in patients affected by multiple sclerosis (MS). Biochemical and morphological methods were used associated with quantitative analysis of images. The whole thymuses were removed during autopsies in young and adult patients not treated with interferon. Moreover, samples of thymus were removed from patients, either young or adult who had previously been treated with interferon therapy, and subjected, for diagnostic reasons, to thymic biopsy. All samples of thymus were weighed, measured and dissected. Thymic slices were stained with Eosin-orange for detection of the microanatomical details, or with Bodian’s reaction for recognition of nervous structures. Histofluorescence microscopy was used for detection of ANF, and immunofluorescence microscopy for recognition of NPY-like immunoreactive structures. All morphological results were subjected to quantitative analysis of images. Noradrenaline contained in thymic structures was measured by biochemical methods. Our results only concerned the effects of the therapy and suggested that treatment with interferon therapy induces many changes in the thymic structures: (1) The protein content of thymus is significantly increased; (2) the NA content in the thymus is also significantly increased; (3) NPY-like immunoreactive structures in the thymus are significantly increased; (4) occurrence of NPY-like immunoreactivity is particularly and significantly increased both in thymic microenvironment and in structures resembling nerve fibers; (5) ANF are significantly increased in the same thymic structures in which NPY-like immunoreactivity is also increased (i.e. thymic microenvironment and structures resembling nerve fibers). The morphological and biochemical changes observed can also explain the immunological changes induced in the thymus after immunostimulating therapy. © 2002 Elsevier Science Ltd. All rights reserved. Keywords: Adrenergic nerve fibers; Thymus; Immune response
1. Introduction Primary and secondary lymphoid organs receive extensive sympathetic/noradrenergic innervation (Elenkov et al., 2000). Under stimulation, noradrenaline (NA) is released from the sympathetic nerve terminals in these organs and the target immune cells express adrenoceptors (Elenkov et al., 2000). The noradrenergic innervation of the human thymus was studied with the
* Corresponding author. Tel.: + 39-06-4958291; fax: +39-064957669. E-mail address: [email protected] (C. Cavallotti).
immunohistochemical staining of tyrosine-hydroxylase and of dopamine-B-hydroxylase, considered as specific marker of noradrenergic innervation (Vizi et al., 1995). NA released from neuronal varicosities exerts its effects on thymocytes. Moreover, most studies on the sympathetic innervation of the thymus have been carried out using other techniques on thymus glands from a variety of mammalian species (Bulloch and Moore, 1981, Singh, 1984, Felten et al., 1985, Magni et al., 1987, Kendall and al-Shawaf, 1991, Madden et al., 1997). Only a few investigations have been reported on the autonomic innervation of human thymus (Haar, 1974; Ghali et al., 1980; de Leeuw et al., 1992). In addition to
0197-0186/02/$ - see front matter © 2002 Elsevier Science Ltd. All rights reserved. PII: S 0 1 9 7 - 0 1 8 6 ( 0 1 ) 0 0 0 7 4 - 2
212
D. Ca6allotti et al. / Neurochemistry International 40 (2002) 211–221
the description of adrenergic nerves in rat and human thymus, Kranz and coworkers (Kranz et al., 1997) report the presence of NA in epithelial cells of the thymic medulla. In the present paper we studied the effects of therapy with interferons on adrenergic nerve fibers (ANF), on neuropeptide Y (NPY)-like immunoreactivity and on NA content in the human thymus.
ical involvement during the last two years’’ can be enrolled. For this reason we are unable to furnish further clinical and laboratory data. In order to study the possible involvement of the thymus gland in MS or in other neurological diseases, a great number of patients were submitted to a surgical biopsy of the thymus. For these reasons small fragments of these thymic biopsies drawn from immunostimulated young and adult patients affected by MS were available for our experiments.
2. Experimental procedures
2.2. Treatment of patients The following experimental procedures were performed: (1) Selection of immunostimulating therapy; (2) treatment of patients; (3) thymic biopsies; (4) cutting of thymus; (5) staining of thymus; (6) staining of nerve fibers; (7) histofluorescence microscopy for the staining of adrenergic nerve fibers (ANF); (8) staining of NPYlike immunoreactivity; (9) protein measurement; (10) NA measurement; (11) Quantitative analysis of images (QAI); (12) Statistical analysis of data. Each procedure is briefly explained.
2.1. Selection of immunostimulating therapy b-1a interferon (Avonex, Solvay-Duphar, The Netherlands) belongs to a groups of drugs that modulate the regulation of the human immune system. b-1b interferon (Betaferon, Schering AG, Germany) also modifies the response of the human immune system. Recently, human recombinant b-interferon (Serobif, Serono SPA, Italy) has been introduced into clinical practice clinic as an immunomodulating systemic agent. All these substances belong to the family of cytokines, natural proteins also acting as neurotransmitters/ neuromodulators and exerting an immunostimulating or immunomodulating effect. Cytokines comprehend a number of biologically active substances, including, for instance, the prostaglandins. All the above-mentioned interferons can be alternatively employed in the treatment of patients affected by recurrent– remittent multiple sclerosis (MS), whose clinical history is marked by at least two episodes of neurological involvement during the last 2 years. In our city, many neurological hospitals are authorized as national data points for the study of MS; in these centers, patients affected by MS are enrolled and treated by interferons. As above mentioned the patients were selected and enrolled in our medical protocols according to recommendations of the Health National Service. Such recommendations do not specify the clinical status and the laboratory data of the selected groups, but point out that only those patients affected by MS ‘‘whose clinical history is marked by at least two episodes of neurolog-
Our studies were approved by the Ethical Committees of the hospitals involved and patients gave their personal, written informed consent. In our national clinical centers the following therapeutic procedures were employed for administration of interferon drugs: 36 patients (age range 14–66 years) are randomly distributed in three groups and are treated by three different protocols in order to obtain an immunomodulation.
2.2.1. First group Patients enrolled: 14, treated for at least 6 months by one intramuscular injection (1,600,000 International Units, IUs) of Avonex (b-1a interferon) weekly. 2.2.2. Second group Patients enrolled: 12, treated for at least 6 months by three intradermal injections of Betaferon (b-1b interferon), corresponding to 8,000,000 IUs weekly. 2.2.3. Third group Patients enrolled: 10, treated for at least 3 months by three intradermal or intramuscular injections of Serobif (human recombinant b-interferon; the drug is still completing the official registration course to the Italian Health Authorities), corresponding to 6,000,000 IUs weekly. The efficiency of immunostimulating therapy was tested on each patient both by means of laboratory tests (especially antibody titers) and specific dosage of protein content in thymic samples, as reported in Section 1.9. 2.3. Thymic biopsies from treated and untreated patients Three patients of the first group, two patients of the second group and two patients of the third group also underwent surgical thymic biopsies; these patients were young (n=3) and adult (n= 4). In the course of six autopsies on six subjects who died for non-mediastinal and non-neurological diseases and were not treated by chemo, radio, corticosteroid or immunostimulating therapy, the whole thymuses were
D. Ca6allotti et al. / Neurochemistry International 40 (2002) 211–221
drawn and immediately prefixed by immersion in Bouin’s fluid. Thymus, or residual islands of thymic tissue, after involution (in adult or in aged subjects), were found in the retrosternal adipose tissue after removal of the sternocostal plate. Thymuses were quickly carried under dry ice to our laboratories for all experimental procedures. The samples served as controls and the relative values are described as coming from ‘not treated patients’ (=not treated with interferon therapy) in our results. In this case too, when possible, informed written consent for thymus removal was obtained from the close relatives of deceased patients.
2.4. Sections of thymus Serial sections (10 mm thick) were cut on a cryostat at −20 °C. Each section was mounted on a preweighed slide. The mounted slide was postweighed to determine the weight of each section. Ten consecutive thymic sections were mounted on ten numbered slides. The first was treated with Eosin to provide histological orientation, to identify the microanatomical details and to define the thymic compartments. The second one was treated with Bodian’s method (Bodian, 1936) to recognize the nerve fibers. Another three sections (from third to fifth) were treated for histofluorescence microscopy (blank without glyoxylic acid, blank with denatured section, whole reaction). The last five sections (from sixth to tenth) were used for immunostaining of NPY (four blank plus one slide for whole reaction). After cutting the sections, a small slice (200 mm thick) was cut, weighed and used for the biochemical assay. The same procedure (cutting in serial sections) was used for the whole length of the thymic biopsy.
2.5. Staining of thymus The microanatomical details of thymic tissue were detected with Eosin-orange. After fixation, sections were treated with a working solution of Eosin-orange (Eosin g water soluble 10 g dissolved in 1000 ml of distilled water plus 2 ml of glacial acetic acid; this is the stock solution. Twenty-five milliliters of stock solution + 75 ml of H2O + 0.5 ml of glacial acetic acid is the working solution (Townsend, 1960).
2.6. Staining of ner6e fibers Nervous structures were coloured by Bodian’s method. This method can be used to verify that a stained structure is nervous in nature. In fact, it stains nerve fibers and neurofibrils. After fixation, the sections were treated with: 1) 1% Protargol solution (colloidal silver), 2) reducing solution (hydroquinone+ sodium sulphite), 3) 1% gold chloride solution and 4) 2% oxalic
213
acid solution, and counterstained with 0.03% aniline blue. The nerve fibers and neurofibrils were stained in black. Further details on this staining are reported in Bodian’s cited work (Bodian, 1936).
2.7. Histofluorescence microscopy For the staining of adrenergic nerve fibers a glyoxylic acid-induced fluorescence technique was used, as described by Qayyum and Fatani (1985). Briefly, immediately before use, the staining solution was prepared by adding in a solution of 0.236 M potassium phosphate monobasic (pH 7.4) 0.2 M sucrose and 1% glyoxylic acid. This staining solution is named sucrose, phosphate, glyoxylic (SPG). The slides with thymic samples were immediately dipped in this solution for 5 min. To assure a comparable fluorescence it is important to standardize times and temperatures without intervals. After staining, the sections must be drained, covered with non-autofluorescent immersion oil, heated at 95 °C for 5 min and coverslipped. Then the sections must be immediately observed, analyzed and photographed to prevent the diffusion and the photodecomposition of the fluorescence. The sections were examined and photographed under a Zeiss photomicroscope equipped with exciter and barrier filters and with a mercury lamp for observation of fluorescence.
2.8. Staining of NPY-like immunoreacti6ity The immunohistochemical method used for the detection of the NPY-positive nerve fibers was proposed by Uddman et al. (1985). Owing to the thickness of the sections (10 mm), the samples were incubated for a long time (18–24 h) at room temperature, so that the antibodies completely penetrated the sections, with the rabbit anti-NPY (Cambridge RB-CRB-UK) diluted 1:600 in phosphate-buffered saline (PBS). Five slides (each containing one slice of sample) were used for immunostaining experiments. In the first blank, primary or secondary antiserum was omitted or denatured or previously absorbed with an excess of corresponding peptide; in the second blank, primary or secondary antiserum was replaced by a non-immune serum; in the third blank, the sample was previously fixed by immersion in a 4% solution of formaldehyde in PBS: this treatment does not preserve the immunoreactive sites. In the fourth blank, the sample was denatured with formaldehyde before or after treatment with primary antiserum or before treatment with secondary antiserum. All these procedures resulted in the absence of any immunoreaction. Positive immunostaining was observed only in the fifth slide that contained the normally treated sample. After treatment with the specific antibodies (rabbit anti-NPY) the samples were washed in PBS and incubated with fluorescein isothiocyanate-
214
D. Ca6allotti et al. / Neurochemistry International 40 (2002) 211–221
conjugated antiserum (goat anti-rabbit IgG-Nordic Immunological Reagent: NIR, The Netherlands) diluted 1:100 in PBS for 18– 24 h at room temperature, allowing the complete penetration of the fluorescent IgG into the thick sections (10 mm). The samples were washed in PBS and observed using a Zeiss III photomicroscope equipped with epi-illumination and Neofluar objectives. Once the nerve fibers have been stained with NPY, it is always possible to identify under light microscopy the total fluorescent area of nerve fibers stained by this neurotransmitter. Morphometrical quantification of the density of nerve fibers was performed using a Quantimet Leica® 500 image analyzer (Manual of Methods of Quantimet Leica®, 1997). More sophisticated and precise techniques are now available for the identification and quantitative evaluation of ANF and NPY-like immunoreactivity (immunohistochemistry, immuno-electron microscopy, PCR techniques, immunoblotting, enzyme-linked immunosorbent assay (ELISA), high-performance liquid chromatography (HPLC)). Nevertheless, many of these techniques can only be applied on fresh homogenates and not on small fragments coming from biopsies or from autopsies at least 1 day after death. Moreover, many tissues (including the thymus) require a pre-fixation in Bouin’s fluid that makes it impossible to apply these techniques. For the above reasons we adopted two traditional but less recent techniques for the identification of ANF and of NPY-like immunoreactivity. Nevertheless, as a new finding, we performed quantitative analysis of images after the histofluorescent or immune stainings. The identification of NPY in peripheral tissues such as the thymus requires numerous precautionary controls. Immunohistochemical techniques are able to show the localization in situ of numerous neuropeptides, including NPY. Immunohistochemistry uses fluorescent antibodies to stain many neuropeptides, including NPY, in specific structures or tissue sections. Fluorescein isothiocyanate is a common fluorescent marker used to visualize the immunohistochemical reaction. However, other visualization techniques are also suitable, as markers in immunohistochemistry. Because of the potential for antibody cross-reactivity to chemically-related specific antigens, together with other nonspecific antigens, immunohistochemical staining is never unequivocal and absolute. The specific staining of a neuropeptide, including NPY, requires numerous controls (Coons et al., 1955). After employment of all these controls, as described above (see methods), the absolute identification of a specific neuropeptide still requires a biochemical analysis. For these reasons, in all immunohistochemical results the descriptive suffix ‘NPY-like immunoreactivity’ can be used.
2.9. Measurement of proteins In all these experiments, samples of thymus from biopsies or from autopsies were weighed and placed on dry ice (specimens for histochemical staining) or into an ice-cold homogenisation buffer (samples for estimating the protein content and biochemical activity). Tissue protein concentrations were determined using the method described by Lowry et al. using bovine serum albumin (BSA) as standard and Folin phenol as reagent (Lowry et al., 1951).
2.10. Determination of noradrenaline content The NA content of the tissue was determined by HPLC with chromatography at high resolution as reported by Keller et al. (1976). Briefly, the tissue samples (with previous determined weight and protein content) were homogenized in a solution 1:10 of perchloric acid (0.1 mol/l) with sodium metabisulphite (0.5 mol/l) to prevent NA oxidation. The homogenate was centrifuged at 3000 rpm for 20 min. The supernatant was injected into the chromatographic system in aliquots of 10, 20 and 30 ml. The electrochemical detection was performed using a glassy carbon electrode versus Ag/AgCl reference electrode at 0.75 V. The mobile phase was formed by sodium phosphate (50 mmol/l), citric acid (25 mmol/l, pH= 3.6), EDTA (0.25 mmol/l), octane sulphonic acid (sodium salt 0.75 mmol/l) and 3% acetonitrile. The results are expressed as ng/mg protein9S.D.
2.11. Quantitati6e analysis of images (QAI) In order to evaluate the amount of staining, a quantitative analysis of the intensity of the histofluorescent staining was performed on photographs (to avoid the photo decomposition) by means of a Quantimet analyzer (Leica®). The values of control photographs coming from samples incubated without glyoxylic acid were considered as a ‘zero’. Each photograph was examined separately, evaluating the standard error of mean (S.E.M.). QAI may provide incorrect results. In fact, the main choices (i.e. the instructions for software) are ordered by each research-worker, according to personal preferences. For these reasons the resulting data are partial rather than impartial and it is necessary to follow very careful rules. The counts must be repeated at least three times using the double masked technique. All the counts should be performed by different research-workers, on different analysers, and with samples identified only by a number or by a letter. Final results must be obtained by another research-worker, who examines experimental protocols to identify each sample and attribute specific values. Final values must be submitted
D. Ca6allotti et al. / Neurochemistry International 40 (2002) 211–221
215
to the statistical analysis of data. The values reported in our results represent the intensity of staining for each sample and are expressed in conventional units (CUs) 9S.E.M. Further details on QAI are reported in the manual of methods of Quantimet Leica® 500 image analyzer (Manual of Methods of Quantimet Leica®, 1997).
2.12. Statistical analysis of data The statistical methods used throughout this study must be interpreted as an accurate description of the data rather than a statistical inference of the data itself. The preliminary studies of each value were performed with the use of basic sample statistics. Mean values, maximum and minimum limits, variations, standard deviation (S.D.), standard error of the mean (S.E.M.) and correlation coefficients were performed according to Serio (1986). The relationship between each pair of variables was studied using the respective correlation coefficients grouped in a correlation matrix, thus enabling us to study the existence of a linear (values next + 1 or − 1) or non-linear (values next +0) dependency. Finally, a correlative analysis of the morphological and biochemi-
Fig. 2. Human thymus NPY-positive nerve fibers after treatment with interferon in a young patient (24 years old). The treatment induces an increase of fluorescence in nerve fibers that show thick varicosities and crossings. Magnification 100 ×.
cal data was performed by comparing the significant differences for each group with the corresponding values of the other homogeneous groups. Correlation coefficients denote a significant level less than 0.001 (PB0.001), while the correlation coefficient is not significant when P\ 0.05 (n.s.). This correlation coefficient was calculated according to Castino and Roletto (1992).
3. Results
Fig. 1. Human thymus NPY-positive nerve fibers in normal conditions in a young subject (22 years old). We can observe many nerve fibers in the corticomedullary boundary. These fibers show an irregular course with varicosities, swellings and crossings. Magnification 100× .
The results of our experiments are reported in Figs. 1–6 and summarized in Tables 1–8. As can be seen in Fig. 1, NPY-like immunoreactivity in the thymus of a young man not treated by immunostimulating therapy, shows that the fluorescence is bound to many nerve fibers with varicosities, swellings and crossings on their course. Fig. 2 shows the same corresponding picture in another young man previously treated with interferon therapy. Immunostimulation has induced an increase of fluorescence in nerve fibers that show thick varicosities and crossings. Fig. 3 shows NPY-like immunoreactivity localized in two thin parallel nerve fibers that run in the corticomedullary boundary of the thymus of an adult patient (56 years old) not treated with interferon. These
216
D. Ca6allotti et al. / Neurochemistry International 40 (2002) 211–221
nerve fibers, branching on their course, show varicosities and beads. In Fig. 4 we can observe the same district as Fig. 3 coming from the thymus of another adult male (50 years old), previously treated with interferon therapy. Immunostimulation has induced an increase of the fluorescence in nerve fibers. In fact the nerve fibers appear thicker and more fluorescent than those in Fig. 3. Fig. 5 shows the ANF in the subcapsular zone of the human thymus removed from an adult patient (56 years old) not treated with interferon therapy. A nerve bundle perforates the thymic capsula and penetrates into the thymic parenchyma. Numerous fluorescent nerve fibers form a rich subcapsular plexus organized as a close mesh network: the nerve fibers show beads, swellings and crossings on their course. In the thymus of an adult patient (54 years old), after treatment with interferon, we can observe an increase of fluorescence and/or of ANF (Fig. 6). Numerous fluorescent nerve fibers present beads, swellings and crossings on their course. From a comparative analysis of these figures it is evident that treatment with interferon drugs induces a strong increase both of the NPY-like immunoreactivity and of ANF in the thymic tissue.
Fig. 3. Human thymus NPY-positive nerve fibers in normal conditions in an adult subject (56 years old). Two parallel thin nerve fibers run in the corticomedullary boundary, branching on their course. These nerve fibers show thin beads and numerous varicosities. Magnification 100 × .
Fig. 4. Human thymus NPY-positive nerve fibers after treatment with interferon in an adult patient (50 years old). The treatment induces an increase of fluorescence in nerve fibers that show thick beads and varicosities. These nerve fibers show branches on their course. Magnification 100 ×.
Table 1 contains the values of proteins in thymic tissue homogenates of six young (n= 3 not treated and n=3 treated with interferon) or seven adult (n=3 not treated and n= 4 treated with interferon) subjects. As can be seen, the protein content of thymic tissue decreases with age but increases with treatment by immunostimulating interferons, with a high coefficient of significance (treated vs. not treated, PB 0.001). Table 2 shows the values of NA in thymic tissue supernatant of the same subjects as in Table 1. The amount of NA decreases with age but increases after treatment with interferon drugs. In these experiments the coefficient of significance treated vs. not treated presents p B0.001. In the same subjects of Tables 1 and 2, staining for NPY-like immunoreactivity was also performed, and the occurrence of this reactivity has been quantified with QAI (see Sections 1.8 and 1.11). Table 3 shows that NPY-like immunoreactivity bound to the thymic structures decreases with age, while it increases after treatment with interferon drugs in both classes of age (young and adult). P shows a high significance, and is B0.001. Comparing the statistically significant differences of NPY-like immunoreactivity in the human thymic tissues of young (Table 4) and adult (Table 5) subjects
D. Ca6allotti et al. / Neurochemistry International 40 (2002) 211–221
217
either treated or not treated with immunostimulating therapy, we can seen that in young, treated patients, the NPY-like immunoreactivity increases in the functionally active thymic structures (whole thymus, thymic microenvironment, nerve fiber-like structures, reticular cells and lymphocytes), while it remains unvaried in other thymic structures (Hassall’s corpuscles, arterioles, venules, lymphatic vessels, septa and capsula). In fact, for the first group of thymic structures P was B0.001 treated vs. not treated, while for the second group of thymic structures P was not significant (n.s.). On the contrary, in the adult subjects (Table 5) there was a lower increase in functional active structures of the thymus in comparison to that observed in young subjects. Moreover, the other structures with a lower functional importance present a smaller increase of NPY-like immunoreactivity after immunostimulating therapy. Table 6 deals with quantitative results of QAI on occurrence of ANF in the human thymus in the same subjects (as Tables 1– 3) with or without interferon treatment. In young subjects the values of ANF expressed in CUs9S.E.M. are 47.99 5.3. These values increase after interferon treatment to 52.89 5.6. In Fig. 6. After treatment with interferon the human thymus shows an increase of fluorescence and/or of adrenergic nerve fibers in subcapsular area in an adult patient (54 years old). Numerous fluorescent nerve fibers can be observed. These fibers present beads, swellings, and crossings on their course. Magnification 100 × . Table 1 Protein content in thymic tissue homogenates of subjects of various ages treated or not treated with immunostimulating therapy Age (years)
Not treated (protein mg/g of fresh weight 9 S.D.)
Treated (protein mg/g of fresh weight 9 S.D.)
22–24 50–62
18.2 91.9 (n = 3) 10.8 9 0.8 (n =3)
19.6 90.22 (n = 3)* 12.8 9 0.18 (n = 4)*
Results are expressed as mg/g fresh weight tissue. Each value is the mean value 9S.D. of independent determinations from n patients, carried out in triplicate. Treatment of patients with interferon drugs has been alternatively performed by one of the three procedures reported in methods. P was calculated by comparing the significant differences of treated versus not treated subjects. n = number of subjects. * PB0.001.
Fig. 5. Adrenergic nerve fibers in the subcapsular region of human thymus in an adult subject (56 years old). Nerve bundles (constituted by many parallel nerve fibers) pierce the capsula and penetrate into the thymic tissue. Numerous fluorescent nerve fibers form a rich subcapsular plexus organized as a close mesh network. These nerve fibers present beads, swellings and crossing on their course. Magnification 100 × .
adult subjects not treated with immunostimulating therapy the values of ANF are 29.19 3.1. After treatment with interferon these values rise to 32.59 2.8. In both age-groups (young and adult) the coefficients of significance from statistical analysis of the data considering treated versus not treated are highly positive with PB 0.001. Comparing the statistically significant differences of ANF in the human thymic tissues between young and
D. Ca6allotti et al. / Neurochemistry International 40 (2002) 211–221
218
adult subjects treated and not treated with immunostimulating therapy, we can observe that in young subjects (Table 7) interferon administration is followed by an increase of ANF in the functionally active structures of the thymus (whole thymus, thymic microenvironment, nerve fiber-like structures, reticular cells and lymphocytes), while it remains unvaried in other thymic structures (Hassall’s corpuscles, arterioles, venules, lymphatic vessels, septa and capsula). In fact, for the first group of thymic structures P was B 0.001, considering the values of treated vs. not treated subjects, while for the second group of thymic structures P was not significant (n.s.). On the contrary, in adult subjects (Table 8), we can observe a lower increase in functionally active structures of the thymus in comparison to young subjects. Moreover, the other thymic structures with a lower functional importance show a lower increase of NPY-like immunoreactivity after immunostimulating therapy. All the results obtained by QAI are in accordance with our other morphological and biochemical results and confirm that immunostimulating therapy with interferon induces a substantial increase in the action of the sympathetic nervous system on thymic structures. Table 2 Biochemical values of NA in the human thymus supernatant of subjects of various ages treated or not treated with immunostimulating therapy Age (years)
Not treated (ng/mg protein9S.D.)
Treated (ng/mg protein9 S.D.)
22–24 50–62
184.4 911.5 (n= 3) 153.7 910.2 (n= 3)
224.3 9 12.2 (n= 3)* 167.4 9 10.8 (n= 4)*
Results are expressed as ng/mg protein. Each value is the mean value9 S.D. of independent determinations from n patients, carried out in triplicate. Treatment of patients with interferon drugs has been alternatively performed by one of the three procedures reported in methods. P was calculated by comparing the significant differences of treated versus not treated subjects. n= number of subjects. * PB0.001. Table 3 Occurrence of NPY in the thymus of subjects of various ages treated or not treated with immunostimulating therapy Age (years)
Not treated (CUs9S.E.M.)
Treated (CUs9 S.E.M.)
22–24 50–62
28.2 92.2 (n=3) 20.1 91.7 (n=3)
32.7 9 2.4 (n = 3)* 21.4 9 2.3 (n= 4)*
The values are expressed in CUs 9 S.E.M. The analyzer was calibrated considering ‘zero’ the values of control sections incubated without antibodies. For other precautions used in QAI, see Section 1.11. Each value represents the mean of numerous determinations (at least 50) carried out with double masked technique, performed in triplicate9S.E.M. P was calculated by comparing the significant differences of treated versus not treated subjects. n= number of subjects. * PB0.001.
Table 4 Occurrence of NPY in different structures of the thymus of young subjects treated and not treated with immunostimulating therapy Structures of thymus
Not treated (CUs9S.E.M.)
Treated (CUs9 S.E.M.)
Whole thymus Thymic microenvironment Structures resembling nerve-fibers Reticular cells Lymphocytes Hassall’s corpuscles Arterioles Venules Lymphatic vessels Septa Capsula
28.5 93.1 46.2 93.9
37.4 9 3.1* 56.3 9 4.9*
58.3 94.5
66.2 9 5.3*
23.9 9 2.4 18.6 92.1 42.1 93.1 39.3 9 2.4 26.8 91.9 11.1 9 1.1 9.2 90.9 8.1 90.8
33.1 9 3.2* 28.3 92.5* 41.7 9 3.9 (n.s.) 41.3 93.6 (n.s.) 28.9 92.5 (n.s.) 12.4 9 1.1 (n.s.) 9.5 90.4 (n.s.) 8.7 90.5 (n.s.)
The intensity of staining for NPY was measured in thymus of young subjects (n =6), treated (n =3) and not treated (n = 3) with b-interferon. On the basis of microanatomical details revealed by staining with Eosin-orange we can distinguish all the structures presented in the table. The thymic parenchyma is also called thymic microenvironment, and it includes especially the following structures: nerve fiberlike structures, reticular cells, lymphocytes and Hassall’s corpuscles. All the values arises from Quantimet (Leica®) and are expressed in CUs9S.E.M. P was calculated by comparing the significant differences of treated versus not treated subjects. n =number of subjects; n.s.= not significant. * PB0.001. Table 5 Occurrence of NPY in different structures of the thymus of adult subjects treated and not treated with immunostimulating therapy Structures of thymus
Not treated (CUs9S.E.M.)
Treated (CUs9 S.E.M.)
Whole thymus Thymic microenvironment Structures resembling nerve-fibers Reticular cells Lymphocytes Hassall’s corpuscles Arterioles Venules Lymphatic vessels Septa Capsula
19.2 91.3 33.3 92.4
24.3 9 1.8* 37.9 9 2.3*
53.5 94.1
58.7 9 4.1*
17.7 9 1.1 10.9 9 0.9 41.4 93.8 35.3 9 3.1 22.3 9 2.3 11.1 9 1.2 9.0 90.8 8.7 9 0.7
22.6 91.9* 14.8 9 1.1* 45.2 93.6* 37.1 93.1 (n.s.) 25.3 92.2 (n.s.) 13.6 9 1.1 (n.s.) 9.7 9 0.5 (n.s.) 8.9 90.6 (n.s.)
The intensity of staining for NPY was measured in thymus of adult subjects (n =7), treated (n =3) and not treated (n = 4) with b-interferon. On the basis of microanatomical details revealed by staining with Eosin-orange, we can distinguish all the structures presented in the table. The thymic parenchyma is also called thymic microenvironment, and it includes especially the following structures: nerve fiberslike structures, reticular cells, lymphocytes and Hassall’s corpuscles. All the values arises from Quantimet (Leica®) and are expressed in CUs9S.E.M. P was calculated by comparing the significant differences of treated versus not treated subjects. n =number of subjects; n.s.= not significant. * PB0.001.
D. Ca6allotti et al. / Neurochemistry International 40 (2002) 211–221
4. Discussion The present results provide direct evidence that the immune response increases the level of NPY-like immunoreactivity, of ANF and of NA contained in human thymic structures. Treatment with interferon is also capable of increasTable 6 Occurrence of ANF in the thymus of subjects of various ages treated or not treated with immunostimulating therapy AGE (years)
Not treated (CUs9 S.E.M.)
TREATED (CUs 9 S.E.M.)
22–24 50–62
47.9 9 5.3 (n= 3) 29.1 9 3.1 (n= 3)
52.8 9 5.6 (n = 3)* 32.5 9 2.8 (n = 4)*
The values are expressed in CUs 9S.E.M. The analyzer was calibrated considering ‘zero’ the values of control sections incubated without glyoxylic acid. For other precautions used in QAI, see Section 1.11. Each value represent the mean of numerous determinations (at least 50) carried out with double masked technique by different research-workers, performed in triplicate 9 S.E.M. P was calculated by comparing the significant differences of treated versus not treated subjects. P was calculated by comparing the significant differences of treated versus not treated subjects. n = number of subjects. * PB0.001. Table 7 Occurrence of ANF in different structures of the thymus of young subjects treated and not treated with immunostimulating therapy Structures of thymus
Not treated (CUs9S.E.M.)
Treated (CUs9 S.E.M.)
Whole thymus Thymic microenvironment Structures resembling nerve-fibers Reticular cells Lymphocytes Hassall’s corpuscles Arterioles Venules Lymphatic vessels Septa Capsula
47.39 4.4 61.295.8
59.5 9 3.9* 71.4 9 5.1*
78.6 96.2
78.1 9 5.6 (n.s.)
35.3 93.1 26.4 9 2.1 51.694.3 40.5 93.5 31.1 92.6 21.49 1.9 9.390.8 8.190.7
43.8 9 3.6* 38.6 9 2.9* 55.4 9 4.1* 42.29 3.3 (n.s.) 33.7 9 2.8 (n.s.) 24.5 9 2.1 (n.s.) 10.39 0.9 (n.s.) 8.1 9 0.6 (n.s.)
The intensity of staining for ANF was measured in thymus of young subjects (n =6), treated (n=3) and not treated (n = 3) with b-interferon. On the basis of microanatomical details revealed by staining with Eosin-orange we can distinguish all the structures presented in the table. The thymic parenchyma is also called thymic microenvironment, and it includes especially the following structures: nerve fiberslike structures, reticular cells, lymphocytes and Hassall’s corpuscles. All the values arises from Quantimet (Leica®) and are expressed in CUs9 S.E.M. The analyzer was calibrated considering ‘zero’ the values of control sections incubated without glyoxylic acid. For other precautions used in QAI, see Section 1.11. P was calculated by comparing the significant differences of treated versus not treated subjects. n =number of subjects; n.s. = not significant. * PB0.001.
219
Table 8 Occurrence of NPY in different structures of the thymus of adult subjects treated and not treated with immunostimulating therapy Structures of thymus
Not treated (CUs9S.E.M.)
Treated (CUs9 S.E.M.)
Whole thymus Thymic microenvironment Structures resembling nerve-fibers Reticular cells Lymphocytes Hassall’s corpuscles Arterioles Venules Lymphatic vessels Septa Capsula
24.1 91.9 33.6 93.1
29.4 9 2.1* 38.3 9 3.4*
73.4 96.4
77.8 9 5.9 (n.s.)
18.2 91.7 15.7 9 1.6 61.5 95.3 28.3 9 2.6 23.1 92.1 14.2 91.2 9.4 90.9 8.6 90.5
22.7 92.0* 19.4 9 1.6* 68.3 95.6* 31.1 9 2.1 (n.s.) 25.2 9 1.9 (n.s.) 16.5 9 1.5 (n.s.) 10.4 9 0.8 (n.s.) 9.3 9 0.6 (n.s.)
The intensity of staining for ANF was measured in thymus of adult subjects (n =7), treated (n =3) and not treated (n = 4) with b-interferon. On the basis of microanatomical details revealed by staining with Eosin-orange we can distinguish all the structures presented in table. The thymic parenchyma is also called thymic microenvironment, and it includes especially the following structures: nerve fiberslike structures, reticular cells, lymphocytes and Hassall’s corpuscles. All the values arises from Quantimet (Leica®) and are expressed in CUs9S.E.M. The analyzer was calibrated considering ‘zero’ the values of control sections incubated without glyoxylic acid. For other precautions used in QAI, see Section 1.11. P was calculated by comparing the significant differences of treated versus not treated subjects. n =number of subjects; n.s. =not significant. * PB0.001.
ing the protein content of the human thymus. The biochemical data, together with the histochemical results, provide direct evidence for a specific localization of a sympathetic innervation in the thymic gland. The ANF of the thymus, especially after treatment with interferon, are localized particularly in the thymic microenvironment. Owing to this strong localization of the sympathetic nerve fibers in the microenvironment of the human thymuses treated or not treated with interferon drugs, it is possible to study the relationships between nerve fibers and thymic cells by examining our experimental results obtained by serial sections and by QAI. In conclusion, treatment with interferon drugs induces an immune response. This response can be verified with dosages of elevated antibody titers, as we have performed in each patient (results not reported). In our patients the efficiency of immunostimulating therapy was verified by means of common laboratory tests performed during hospitalization. A significant elevation in thymic levels of proteins was found after treatment with interferon drugs. The possible biological significance of this elevation can be considered. The thymic cells that synthesize thymosin also possess
220
D. Ca6allotti et al. / Neurochemistry International 40 (2002) 211–221
adrenergic receptors. The thymic ANF can act on these receptors. In comparison with other immunostimulating drugs, useful in MS and in other autoimmune neurological diseases, such as thymopoietin and thymosin (whose effects produce a more general stimulation of cellular thymic response), interferon therapy performed in this study induces different modifications on morphological and functional parameters. Our results may lead us to affirm that, while thymosin influences the development of T-cells and thymopoietin induces T-cell maturation from prothymocytes, inhibiting in the meantime B-cell differentiation, interferons act mainly on quantitative increase of morphological and functional thymic parameters (protein content, NPY-like immunoreactivity, ANF, NA content) whose modifications underline the effects of immunostimulating therapy directly on thymic tissues and via the action of the sympathetic nervous system on thymic structures. Our results demonstrate that patients treated with interferon, a potent promoter of the immune response, show, in comparison to untreated patients, high levels of many adrenergic parameters in thymic tissues (i.e. ANF, NA, NPY-like immunoreactivity). Interferon treatment is also capable of increasing the protein content of the thymus. Because many samples (untreated controls) were taken at autopsy, any different expression of NA, NPY and ANF may be due to post-mortem degradation. Nevertheless, the delay following death was minimal, as demonstrated by More and Fatty (1958) for cholinesterase activity. Moreover, in order to prevent oxidation of NA during processing of the tissue, samples of tissues were immersed in a solution 1 mmol of 3-mercapto-propionic acid. Treatment with this drug has been found to be suitable for prevention of post-mortem increase of endogenous levels of neurotransmitters in the brain (van der Heyden and Korf, 1978): it is also able to prevent the oxidation of neurotransmitters in the ovaries and in the uterine tube and does not interfere with NA determination (Erdo et al., 1989). Therefore the samples coming from autopsies can be compared with those are coming from biopsies. In conclusion, treatment with interferon therapy is able to induce many changes in the thymic structures. Our biochemical data together with the histochemical qualitative and quantitative results provide direct evidence for a specific localization of ANF in the thymic gland.
Acknowledgements The authors are greatly indebted to Drs B. Nagar and J. Feher for the kind help furnished in the critical revision of the manuscript. The technical assistance of D. Caporuscio, the photographic service of G.
Leoncini, the excellent secretarial work of S. Casamento and the kind help of S. Hobby in the revision of the English language are also gratefully acknowledged. The present study was supported by grants from MURST and University of Rome ‘La Sapienza’ Ministerio Universita e Ricerca Scientifica e Tecnologica.
References Bodian, D., 1936. A new method for staining nerve fibers and nerve ending in mounted paraffin sections. Anat. Rec. 65, 89 –97. Bulloch, K., Moore, R.Y., 1981. Innervation of the thymus gland by brain stem and spinal cord in mouse and rat. Am. J. Anat. 162 (2), 157 – 166. Castino, M., Roletto, E., 1992. Statistica Applicata. Piccin, Padova. Coons, A.H., Leduc, E.H., Connolly, J.M., 1955. Studies on antibody production. I. A method for the histochemical demonstration of specific antibody and its application to a study of the hyperimmune rabbit. J. Exp. Med. 102, 49 – 60. de Leeuw, F.E., Jansen, G.H., Batanero, E., van Wichen, D.F., Huber, J., Schuurman, H.J., 1992. The neural and neuro-endocrine component of the human thymus. I. Nerve-like structures. Brain Behav. Immun. 6, 234 – 248. Elenkov, I.J., Wilder, R.L., Chrousos, G.P., Vizi, E.S., 2000. The sympathetic nerve —an integrative interface between two supersystems: the brain and the immune system. Pharmacol. Rev. 52 (4), 595 – 638. Erdo, S.L., Kiss, B., Szporny, L., 1989. Age-dependent changes of Gaba and Noradrenaline levels in rat ovary and uterine tubes. In: Cerebrovascolar Neuroeffector Mechanisms in Health and Disease. EUR, Roma, pp. 191 – 202. Felten, D.L., Felten, S.Y., Carlson, S.L., Olschowska, J.A., Livnat, S., 1985. Noradrenergic and peptidergic innervation of lymphoid tissue. J. Immunol. 135 (Suppl. 2), 755s – 765s. Ghali, W.M., Abdel-Rahman, S., Nagib, M., Mahran, Z.Y., 1980. Intrinsic innervation and vasculature of pre- and post-natal human thymus. Acta Anatomica (Basel) 108 (1), 115 – 123. Haar, J.L., 1974. Light and electron microscopy of the human fetal thymus. Anat. Rec. 179, 463 – 476. Keller, R., Oke, A., Mefford, I., Adams, R.N., 1976. Liquid chromatographic analysis of catecholamines. Life Sci. 19, 995 –1004. Kendall, M.D., al-Shawaf, A.A., 1991. Innervation of the rat thymus gland. Brain Behav. Immun. 5, 19 – 28. Kranz, A., Kendall, M.D., von Gaudecker, B., 1997. Studies on rat and human thymus to demonstrate immunoreactivity of calcitonin-gene-related-peptide, tyrosine hydroxylase and neuropeptide Y. J. Anat. 191, 441 – 450. Lowry, O.H., Rosebrough, N.J., Farr, A.L., 1951. Protein measurement with the Folin phenol reagent. J. Biol. Chem. 193, 265 –275. Madden, K.S., Bellinger, D.L., Felten, S.Y., Snyder, E., Maida, M.E., Felten, D.L., 1997. Alterations in sympathetic innervation of thymus and spleen in aged mice. Mech. Ageing Dev. 94, 165 – 175. Magni, F., Bruschi, F., Kasti, M., 1987. The afferent innervation of the thymus gland in the rat. Brain Res. 424, 379 – 385. Manual of Methods of Quantimet Leica® 500 Microsystems Imaging Solutions (1997). Clifton Road, Cambridge. More, E.J., Fatty, C.S., 1958. A note on cholinesterase activity in post-mortem tissues. J. Histochem. Cytochem. 6, 377 – 379. Qayyum, M.A., Fatani, J.A., 1985. Use of glyoxylic acid in the demonstration of autonomic nerve profiles. Experientia 41, 1389 – 1390. Serio, A., 1986. Appunti Dalle Lezioni di Statistica Sanitaria. Kappa, Roma.
D. Ca6allotti et al. / Neurochemistry International 40 (2002) 211–221 Singh, U., 1984. Sympathetic innervation of fetal mouse thymus. Eur. J. Immunol. 14, 757 –759. Townsend, F.M., 1960. Manual of histologic and special staining techniques. McGraw-Hill, New York. Uddman, R., Ekblad, E., Edvinsson, L., Ha˚ kanson, R., Sundler, F., 1985. Neuropeptide Y like immunoreactivity in perivascular nerve fibers of the guinea pig. Regul. Pept. 10, 243 –256.
.
221
van der Heyden, J.A., Korf, J., 1978. Regional levels of Gaba in the brain: rapid semiautomatic assay and prevention of postmortem increase by 3-mercapto-propionic acid. J. Neurochem. 31, 197 –203. Vizi, E.S., Orso´ , E., Osipenko, O.N., Hasko´ , G., Elenkov, I.J., 1995. Neurochemical, electrophysiological and immunocytochemical evidence for a noradrenergic link between the sympathetic nervous system and thymocytes. Neuroscience 68, 1263 – 1269.
Occurrence of adrenergic nerve fibers in human thymus during immune response Daniela Cavallotti a, Marco Artico b, Giandomenico Iannetti a,c, Carlo Cavallotti c,* a
Department of Neurosciences, Uni6ersity of Rome ‘La Sapienza’, Rome, Italy Chair of Human Anatomy, Faculty of Pharmacy, Uni6ersity of Rome ‘La Sapienza’, Rome, Italy c Department of Cardio6ascular and Respiratory Sciences (Section of Anatomy), Uni6ersity of Rome ‘La Sapienza’, Via A. Borelli, 50 00161 Rome, Italy b
Received 8 June 2001; accepted 16 June 2001
Abstract The adrenergic nerve fibers (ANF), the neuropeptide Y-like immunoreactive nerve fibers (NPY-NF) and the noradrenaline (NA) amount were studied in the human thymus in subjects previously treated or not treated with interferon therapy with the aim to identify the changes due to the interferon therapy. This therapy has been used in patients affected by multiple sclerosis (MS). Biochemical and morphological methods were used associated with quantitative analysis of images. The whole thymuses were removed during autopsies in young and adult patients not treated with interferon. Moreover, samples of thymus were removed from patients, either young or adult who had previously been treated with interferon therapy, and subjected, for diagnostic reasons, to thymic biopsy. All samples of thymus were weighed, measured and dissected. Thymic slices were stained with Eosin-orange for detection of the microanatomical details, or with Bodian’s reaction for recognition of nervous structures. Histofluorescence microscopy was used for detection of ANF, and immunofluorescence microscopy for recognition of NPY-like immunoreactive structures. All morphological results were subjected to quantitative analysis of images. Noradrenaline contained in thymic structures was measured by biochemical methods. Our results only concerned the effects of the therapy and suggested that treatment with interferon therapy induces many changes in the thymic structures: (1) The protein content of thymus is significantly increased; (2) the NA content in the thymus is also significantly increased; (3) NPY-like immunoreactive structures in the thymus are significantly increased; (4) occurrence of NPY-like immunoreactivity is particularly and significantly increased both in thymic microenvironment and in structures resembling nerve fibers; (5) ANF are significantly increased in the same thymic structures in which NPY-like immunoreactivity is also increased (i.e. thymic microenvironment and structures resembling nerve fibers). The morphological and biochemical changes observed can also explain the immunological changes induced in the thymus after immunostimulating therapy. © 2002 Elsevier Science Ltd. All rights reserved. Keywords: Adrenergic nerve fibers; Thymus; Immune response
1. Introduction Primary and secondary lymphoid organs receive extensive sympathetic/noradrenergic innervation (Elenkov et al., 2000). Under stimulation, noradrenaline (NA) is released from the sympathetic nerve terminals in these organs and the target immune cells express adrenoceptors (Elenkov et al., 2000). The noradrenergic innervation of the human thymus was studied with the
* Corresponding author. Tel.: + 39-06-4958291; fax: +39-064957669. E-mail address: [email protected] (C. Cavallotti).
immunohistochemical staining of tyrosine-hydroxylase and of dopamine-B-hydroxylase, considered as specific marker of noradrenergic innervation (Vizi et al., 1995). NA released from neuronal varicosities exerts its effects on thymocytes. Moreover, most studies on the sympathetic innervation of the thymus have been carried out using other techniques on thymus glands from a variety of mammalian species (Bulloch and Moore, 1981, Singh, 1984, Felten et al., 1985, Magni et al., 1987, Kendall and al-Shawaf, 1991, Madden et al., 1997). Only a few investigations have been reported on the autonomic innervation of human thymus (Haar, 1974; Ghali et al., 1980; de Leeuw et al., 1992). In addition to
0197-0186/02/$ - see front matter © 2002 Elsevier Science Ltd. All rights reserved. PII: S 0 1 9 7 - 0 1 8 6 ( 0 1 ) 0 0 0 7 4 - 2
212
D. Ca6allotti et al. / Neurochemistry International 40 (2002) 211–221
the description of adrenergic nerves in rat and human thymus, Kranz and coworkers (Kranz et al., 1997) report the presence of NA in epithelial cells of the thymic medulla. In the present paper we studied the effects of therapy with interferons on adrenergic nerve fibers (ANF), on neuropeptide Y (NPY)-like immunoreactivity and on NA content in the human thymus.
ical involvement during the last two years’’ can be enrolled. For this reason we are unable to furnish further clinical and laboratory data. In order to study the possible involvement of the thymus gland in MS or in other neurological diseases, a great number of patients were submitted to a surgical biopsy of the thymus. For these reasons small fragments of these thymic biopsies drawn from immunostimulated young and adult patients affected by MS were available for our experiments.
2. Experimental procedures
2.2. Treatment of patients The following experimental procedures were performed: (1) Selection of immunostimulating therapy; (2) treatment of patients; (3) thymic biopsies; (4) cutting of thymus; (5) staining of thymus; (6) staining of nerve fibers; (7) histofluorescence microscopy for the staining of adrenergic nerve fibers (ANF); (8) staining of NPYlike immunoreactivity; (9) protein measurement; (10) NA measurement; (11) Quantitative analysis of images (QAI); (12) Statistical analysis of data. Each procedure is briefly explained.
2.1. Selection of immunostimulating therapy b-1a interferon (Avonex, Solvay-Duphar, The Netherlands) belongs to a groups of drugs that modulate the regulation of the human immune system. b-1b interferon (Betaferon, Schering AG, Germany) also modifies the response of the human immune system. Recently, human recombinant b-interferon (Serobif, Serono SPA, Italy) has been introduced into clinical practice clinic as an immunomodulating systemic agent. All these substances belong to the family of cytokines, natural proteins also acting as neurotransmitters/ neuromodulators and exerting an immunostimulating or immunomodulating effect. Cytokines comprehend a number of biologically active substances, including, for instance, the prostaglandins. All the above-mentioned interferons can be alternatively employed in the treatment of patients affected by recurrent– remittent multiple sclerosis (MS), whose clinical history is marked by at least two episodes of neurological involvement during the last 2 years. In our city, many neurological hospitals are authorized as national data points for the study of MS; in these centers, patients affected by MS are enrolled and treated by interferons. As above mentioned the patients were selected and enrolled in our medical protocols according to recommendations of the Health National Service. Such recommendations do not specify the clinical status and the laboratory data of the selected groups, but point out that only those patients affected by MS ‘‘whose clinical history is marked by at least two episodes of neurolog-
Our studies were approved by the Ethical Committees of the hospitals involved and patients gave their personal, written informed consent. In our national clinical centers the following therapeutic procedures were employed for administration of interferon drugs: 36 patients (age range 14–66 years) are randomly distributed in three groups and are treated by three different protocols in order to obtain an immunomodulation.
2.2.1. First group Patients enrolled: 14, treated for at least 6 months by one intramuscular injection (1,600,000 International Units, IUs) of Avonex (b-1a interferon) weekly. 2.2.2. Second group Patients enrolled: 12, treated for at least 6 months by three intradermal injections of Betaferon (b-1b interferon), corresponding to 8,000,000 IUs weekly. 2.2.3. Third group Patients enrolled: 10, treated for at least 3 months by three intradermal or intramuscular injections of Serobif (human recombinant b-interferon; the drug is still completing the official registration course to the Italian Health Authorities), corresponding to 6,000,000 IUs weekly. The efficiency of immunostimulating therapy was tested on each patient both by means of laboratory tests (especially antibody titers) and specific dosage of protein content in thymic samples, as reported in Section 1.9. 2.3. Thymic biopsies from treated and untreated patients Three patients of the first group, two patients of the second group and two patients of the third group also underwent surgical thymic biopsies; these patients were young (n=3) and adult (n= 4). In the course of six autopsies on six subjects who died for non-mediastinal and non-neurological diseases and were not treated by chemo, radio, corticosteroid or immunostimulating therapy, the whole thymuses were
D. Ca6allotti et al. / Neurochemistry International 40 (2002) 211–221
drawn and immediately prefixed by immersion in Bouin’s fluid. Thymus, or residual islands of thymic tissue, after involution (in adult or in aged subjects), were found in the retrosternal adipose tissue after removal of the sternocostal plate. Thymuses were quickly carried under dry ice to our laboratories for all experimental procedures. The samples served as controls and the relative values are described as coming from ‘not treated patients’ (=not treated with interferon therapy) in our results. In this case too, when possible, informed written consent for thymus removal was obtained from the close relatives of deceased patients.
2.4. Sections of thymus Serial sections (10 mm thick) were cut on a cryostat at −20 °C. Each section was mounted on a preweighed slide. The mounted slide was postweighed to determine the weight of each section. Ten consecutive thymic sections were mounted on ten numbered slides. The first was treated with Eosin to provide histological orientation, to identify the microanatomical details and to define the thymic compartments. The second one was treated with Bodian’s method (Bodian, 1936) to recognize the nerve fibers. Another three sections (from third to fifth) were treated for histofluorescence microscopy (blank without glyoxylic acid, blank with denatured section, whole reaction). The last five sections (from sixth to tenth) were used for immunostaining of NPY (four blank plus one slide for whole reaction). After cutting the sections, a small slice (200 mm thick) was cut, weighed and used for the biochemical assay. The same procedure (cutting in serial sections) was used for the whole length of the thymic biopsy.
2.5. Staining of thymus The microanatomical details of thymic tissue were detected with Eosin-orange. After fixation, sections were treated with a working solution of Eosin-orange (Eosin g water soluble 10 g dissolved in 1000 ml of distilled water plus 2 ml of glacial acetic acid; this is the stock solution. Twenty-five milliliters of stock solution + 75 ml of H2O + 0.5 ml of glacial acetic acid is the working solution (Townsend, 1960).
2.6. Staining of ner6e fibers Nervous structures were coloured by Bodian’s method. This method can be used to verify that a stained structure is nervous in nature. In fact, it stains nerve fibers and neurofibrils. After fixation, the sections were treated with: 1) 1% Protargol solution (colloidal silver), 2) reducing solution (hydroquinone+ sodium sulphite), 3) 1% gold chloride solution and 4) 2% oxalic
213
acid solution, and counterstained with 0.03% aniline blue. The nerve fibers and neurofibrils were stained in black. Further details on this staining are reported in Bodian’s cited work (Bodian, 1936).
2.7. Histofluorescence microscopy For the staining of adrenergic nerve fibers a glyoxylic acid-induced fluorescence technique was used, as described by Qayyum and Fatani (1985). Briefly, immediately before use, the staining solution was prepared by adding in a solution of 0.236 M potassium phosphate monobasic (pH 7.4) 0.2 M sucrose and 1% glyoxylic acid. This staining solution is named sucrose, phosphate, glyoxylic (SPG). The slides with thymic samples were immediately dipped in this solution for 5 min. To assure a comparable fluorescence it is important to standardize times and temperatures without intervals. After staining, the sections must be drained, covered with non-autofluorescent immersion oil, heated at 95 °C for 5 min and coverslipped. Then the sections must be immediately observed, analyzed and photographed to prevent the diffusion and the photodecomposition of the fluorescence. The sections were examined and photographed under a Zeiss photomicroscope equipped with exciter and barrier filters and with a mercury lamp for observation of fluorescence.
2.8. Staining of NPY-like immunoreacti6ity The immunohistochemical method used for the detection of the NPY-positive nerve fibers was proposed by Uddman et al. (1985). Owing to the thickness of the sections (10 mm), the samples were incubated for a long time (18–24 h) at room temperature, so that the antibodies completely penetrated the sections, with the rabbit anti-NPY (Cambridge RB-CRB-UK) diluted 1:600 in phosphate-buffered saline (PBS). Five slides (each containing one slice of sample) were used for immunostaining experiments. In the first blank, primary or secondary antiserum was omitted or denatured or previously absorbed with an excess of corresponding peptide; in the second blank, primary or secondary antiserum was replaced by a non-immune serum; in the third blank, the sample was previously fixed by immersion in a 4% solution of formaldehyde in PBS: this treatment does not preserve the immunoreactive sites. In the fourth blank, the sample was denatured with formaldehyde before or after treatment with primary antiserum or before treatment with secondary antiserum. All these procedures resulted in the absence of any immunoreaction. Positive immunostaining was observed only in the fifth slide that contained the normally treated sample. After treatment with the specific antibodies (rabbit anti-NPY) the samples were washed in PBS and incubated with fluorescein isothiocyanate-
214
D. Ca6allotti et al. / Neurochemistry International 40 (2002) 211–221
conjugated antiserum (goat anti-rabbit IgG-Nordic Immunological Reagent: NIR, The Netherlands) diluted 1:100 in PBS for 18– 24 h at room temperature, allowing the complete penetration of the fluorescent IgG into the thick sections (10 mm). The samples were washed in PBS and observed using a Zeiss III photomicroscope equipped with epi-illumination and Neofluar objectives. Once the nerve fibers have been stained with NPY, it is always possible to identify under light microscopy the total fluorescent area of nerve fibers stained by this neurotransmitter. Morphometrical quantification of the density of nerve fibers was performed using a Quantimet Leica® 500 image analyzer (Manual of Methods of Quantimet Leica®, 1997). More sophisticated and precise techniques are now available for the identification and quantitative evaluation of ANF and NPY-like immunoreactivity (immunohistochemistry, immuno-electron microscopy, PCR techniques, immunoblotting, enzyme-linked immunosorbent assay (ELISA), high-performance liquid chromatography (HPLC)). Nevertheless, many of these techniques can only be applied on fresh homogenates and not on small fragments coming from biopsies or from autopsies at least 1 day after death. Moreover, many tissues (including the thymus) require a pre-fixation in Bouin’s fluid that makes it impossible to apply these techniques. For the above reasons we adopted two traditional but less recent techniques for the identification of ANF and of NPY-like immunoreactivity. Nevertheless, as a new finding, we performed quantitative analysis of images after the histofluorescent or immune stainings. The identification of NPY in peripheral tissues such as the thymus requires numerous precautionary controls. Immunohistochemical techniques are able to show the localization in situ of numerous neuropeptides, including NPY. Immunohistochemistry uses fluorescent antibodies to stain many neuropeptides, including NPY, in specific structures or tissue sections. Fluorescein isothiocyanate is a common fluorescent marker used to visualize the immunohistochemical reaction. However, other visualization techniques are also suitable, as markers in immunohistochemistry. Because of the potential for antibody cross-reactivity to chemically-related specific antigens, together with other nonspecific antigens, immunohistochemical staining is never unequivocal and absolute. The specific staining of a neuropeptide, including NPY, requires numerous controls (Coons et al., 1955). After employment of all these controls, as described above (see methods), the absolute identification of a specific neuropeptide still requires a biochemical analysis. For these reasons, in all immunohistochemical results the descriptive suffix ‘NPY-like immunoreactivity’ can be used.
2.9. Measurement of proteins In all these experiments, samples of thymus from biopsies or from autopsies were weighed and placed on dry ice (specimens for histochemical staining) or into an ice-cold homogenisation buffer (samples for estimating the protein content and biochemical activity). Tissue protein concentrations were determined using the method described by Lowry et al. using bovine serum albumin (BSA) as standard and Folin phenol as reagent (Lowry et al., 1951).
2.10. Determination of noradrenaline content The NA content of the tissue was determined by HPLC with chromatography at high resolution as reported by Keller et al. (1976). Briefly, the tissue samples (with previous determined weight and protein content) were homogenized in a solution 1:10 of perchloric acid (0.1 mol/l) with sodium metabisulphite (0.5 mol/l) to prevent NA oxidation. The homogenate was centrifuged at 3000 rpm for 20 min. The supernatant was injected into the chromatographic system in aliquots of 10, 20 and 30 ml. The electrochemical detection was performed using a glassy carbon electrode versus Ag/AgCl reference electrode at 0.75 V. The mobile phase was formed by sodium phosphate (50 mmol/l), citric acid (25 mmol/l, pH= 3.6), EDTA (0.25 mmol/l), octane sulphonic acid (sodium salt 0.75 mmol/l) and 3% acetonitrile. The results are expressed as ng/mg protein9S.D.
2.11. Quantitati6e analysis of images (QAI) In order to evaluate the amount of staining, a quantitative analysis of the intensity of the histofluorescent staining was performed on photographs (to avoid the photo decomposition) by means of a Quantimet analyzer (Leica®). The values of control photographs coming from samples incubated without glyoxylic acid were considered as a ‘zero’. Each photograph was examined separately, evaluating the standard error of mean (S.E.M.). QAI may provide incorrect results. In fact, the main choices (i.e. the instructions for software) are ordered by each research-worker, according to personal preferences. For these reasons the resulting data are partial rather than impartial and it is necessary to follow very careful rules. The counts must be repeated at least three times using the double masked technique. All the counts should be performed by different research-workers, on different analysers, and with samples identified only by a number or by a letter. Final results must be obtained by another research-worker, who examines experimental protocols to identify each sample and attribute specific values. Final values must be submitted
D. Ca6allotti et al. / Neurochemistry International 40 (2002) 211–221
215
to the statistical analysis of data. The values reported in our results represent the intensity of staining for each sample and are expressed in conventional units (CUs) 9S.E.M. Further details on QAI are reported in the manual of methods of Quantimet Leica® 500 image analyzer (Manual of Methods of Quantimet Leica®, 1997).
2.12. Statistical analysis of data The statistical methods used throughout this study must be interpreted as an accurate description of the data rather than a statistical inference of the data itself. The preliminary studies of each value were performed with the use of basic sample statistics. Mean values, maximum and minimum limits, variations, standard deviation (S.D.), standard error of the mean (S.E.M.) and correlation coefficients were performed according to Serio (1986). The relationship between each pair of variables was studied using the respective correlation coefficients grouped in a correlation matrix, thus enabling us to study the existence of a linear (values next + 1 or − 1) or non-linear (values next +0) dependency. Finally, a correlative analysis of the morphological and biochemi-
Fig. 2. Human thymus NPY-positive nerve fibers after treatment with interferon in a young patient (24 years old). The treatment induces an increase of fluorescence in nerve fibers that show thick varicosities and crossings. Magnification 100 ×.
cal data was performed by comparing the significant differences for each group with the corresponding values of the other homogeneous groups. Correlation coefficients denote a significant level less than 0.001 (PB0.001), while the correlation coefficient is not significant when P\ 0.05 (n.s.). This correlation coefficient was calculated according to Castino and Roletto (1992).
3. Results
Fig. 1. Human thymus NPY-positive nerve fibers in normal conditions in a young subject (22 years old). We can observe many nerve fibers in the corticomedullary boundary. These fibers show an irregular course with varicosities, swellings and crossings. Magnification 100× .
The results of our experiments are reported in Figs. 1–6 and summarized in Tables 1–8. As can be seen in Fig. 1, NPY-like immunoreactivity in the thymus of a young man not treated by immunostimulating therapy, shows that the fluorescence is bound to many nerve fibers with varicosities, swellings and crossings on their course. Fig. 2 shows the same corresponding picture in another young man previously treated with interferon therapy. Immunostimulation has induced an increase of fluorescence in nerve fibers that show thick varicosities and crossings. Fig. 3 shows NPY-like immunoreactivity localized in two thin parallel nerve fibers that run in the corticomedullary boundary of the thymus of an adult patient (56 years old) not treated with interferon. These
216
D. Ca6allotti et al. / Neurochemistry International 40 (2002) 211–221
nerve fibers, branching on their course, show varicosities and beads. In Fig. 4 we can observe the same district as Fig. 3 coming from the thymus of another adult male (50 years old), previously treated with interferon therapy. Immunostimulation has induced an increase of the fluorescence in nerve fibers. In fact the nerve fibers appear thicker and more fluorescent than those in Fig. 3. Fig. 5 shows the ANF in the subcapsular zone of the human thymus removed from an adult patient (56 years old) not treated with interferon therapy. A nerve bundle perforates the thymic capsula and penetrates into the thymic parenchyma. Numerous fluorescent nerve fibers form a rich subcapsular plexus organized as a close mesh network: the nerve fibers show beads, swellings and crossings on their course. In the thymus of an adult patient (54 years old), after treatment with interferon, we can observe an increase of fluorescence and/or of ANF (Fig. 6). Numerous fluorescent nerve fibers present beads, swellings and crossings on their course. From a comparative analysis of these figures it is evident that treatment with interferon drugs induces a strong increase both of the NPY-like immunoreactivity and of ANF in the thymic tissue.
Fig. 3. Human thymus NPY-positive nerve fibers in normal conditions in an adult subject (56 years old). Two parallel thin nerve fibers run in the corticomedullary boundary, branching on their course. These nerve fibers show thin beads and numerous varicosities. Magnification 100 × .
Fig. 4. Human thymus NPY-positive nerve fibers after treatment with interferon in an adult patient (50 years old). The treatment induces an increase of fluorescence in nerve fibers that show thick beads and varicosities. These nerve fibers show branches on their course. Magnification 100 ×.
Table 1 contains the values of proteins in thymic tissue homogenates of six young (n= 3 not treated and n=3 treated with interferon) or seven adult (n=3 not treated and n= 4 treated with interferon) subjects. As can be seen, the protein content of thymic tissue decreases with age but increases with treatment by immunostimulating interferons, with a high coefficient of significance (treated vs. not treated, PB 0.001). Table 2 shows the values of NA in thymic tissue supernatant of the same subjects as in Table 1. The amount of NA decreases with age but increases after treatment with interferon drugs. In these experiments the coefficient of significance treated vs. not treated presents p B0.001. In the same subjects of Tables 1 and 2, staining for NPY-like immunoreactivity was also performed, and the occurrence of this reactivity has been quantified with QAI (see Sections 1.8 and 1.11). Table 3 shows that NPY-like immunoreactivity bound to the thymic structures decreases with age, while it increases after treatment with interferon drugs in both classes of age (young and adult). P shows a high significance, and is B0.001. Comparing the statistically significant differences of NPY-like immunoreactivity in the human thymic tissues of young (Table 4) and adult (Table 5) subjects
D. Ca6allotti et al. / Neurochemistry International 40 (2002) 211–221
217
either treated or not treated with immunostimulating therapy, we can seen that in young, treated patients, the NPY-like immunoreactivity increases in the functionally active thymic structures (whole thymus, thymic microenvironment, nerve fiber-like structures, reticular cells and lymphocytes), while it remains unvaried in other thymic structures (Hassall’s corpuscles, arterioles, venules, lymphatic vessels, septa and capsula). In fact, for the first group of thymic structures P was B0.001 treated vs. not treated, while for the second group of thymic structures P was not significant (n.s.). On the contrary, in the adult subjects (Table 5) there was a lower increase in functional active structures of the thymus in comparison to that observed in young subjects. Moreover, the other structures with a lower functional importance present a smaller increase of NPY-like immunoreactivity after immunostimulating therapy. Table 6 deals with quantitative results of QAI on occurrence of ANF in the human thymus in the same subjects (as Tables 1– 3) with or without interferon treatment. In young subjects the values of ANF expressed in CUs9S.E.M. are 47.99 5.3. These values increase after interferon treatment to 52.89 5.6. In Fig. 6. After treatment with interferon the human thymus shows an increase of fluorescence and/or of adrenergic nerve fibers in subcapsular area in an adult patient (54 years old). Numerous fluorescent nerve fibers can be observed. These fibers present beads, swellings, and crossings on their course. Magnification 100 × . Table 1 Protein content in thymic tissue homogenates of subjects of various ages treated or not treated with immunostimulating therapy Age (years)
Not treated (protein mg/g of fresh weight 9 S.D.)
Treated (protein mg/g of fresh weight 9 S.D.)
22–24 50–62
18.2 91.9 (n = 3) 10.8 9 0.8 (n =3)
19.6 90.22 (n = 3)* 12.8 9 0.18 (n = 4)*
Results are expressed as mg/g fresh weight tissue. Each value is the mean value 9S.D. of independent determinations from n patients, carried out in triplicate. Treatment of patients with interferon drugs has been alternatively performed by one of the three procedures reported in methods. P was calculated by comparing the significant differences of treated versus not treated subjects. n = number of subjects. * PB0.001.
Fig. 5. Adrenergic nerve fibers in the subcapsular region of human thymus in an adult subject (56 years old). Nerve bundles (constituted by many parallel nerve fibers) pierce the capsula and penetrate into the thymic tissue. Numerous fluorescent nerve fibers form a rich subcapsular plexus organized as a close mesh network. These nerve fibers present beads, swellings and crossing on their course. Magnification 100 × .
adult subjects not treated with immunostimulating therapy the values of ANF are 29.19 3.1. After treatment with interferon these values rise to 32.59 2.8. In both age-groups (young and adult) the coefficients of significance from statistical analysis of the data considering treated versus not treated are highly positive with PB 0.001. Comparing the statistically significant differences of ANF in the human thymic tissues between young and
D. Ca6allotti et al. / Neurochemistry International 40 (2002) 211–221
218
adult subjects treated and not treated with immunostimulating therapy, we can observe that in young subjects (Table 7) interferon administration is followed by an increase of ANF in the functionally active structures of the thymus (whole thymus, thymic microenvironment, nerve fiber-like structures, reticular cells and lymphocytes), while it remains unvaried in other thymic structures (Hassall’s corpuscles, arterioles, venules, lymphatic vessels, septa and capsula). In fact, for the first group of thymic structures P was B 0.001, considering the values of treated vs. not treated subjects, while for the second group of thymic structures P was not significant (n.s.). On the contrary, in adult subjects (Table 8), we can observe a lower increase in functionally active structures of the thymus in comparison to young subjects. Moreover, the other thymic structures with a lower functional importance show a lower increase of NPY-like immunoreactivity after immunostimulating therapy. All the results obtained by QAI are in accordance with our other morphological and biochemical results and confirm that immunostimulating therapy with interferon induces a substantial increase in the action of the sympathetic nervous system on thymic structures. Table 2 Biochemical values of NA in the human thymus supernatant of subjects of various ages treated or not treated with immunostimulating therapy Age (years)
Not treated (ng/mg protein9S.D.)
Treated (ng/mg protein9 S.D.)
22–24 50–62
184.4 911.5 (n= 3) 153.7 910.2 (n= 3)
224.3 9 12.2 (n= 3)* 167.4 9 10.8 (n= 4)*
Results are expressed as ng/mg protein. Each value is the mean value9 S.D. of independent determinations from n patients, carried out in triplicate. Treatment of patients with interferon drugs has been alternatively performed by one of the three procedures reported in methods. P was calculated by comparing the significant differences of treated versus not treated subjects. n= number of subjects. * PB0.001. Table 3 Occurrence of NPY in the thymus of subjects of various ages treated or not treated with immunostimulating therapy Age (years)
Not treated (CUs9S.E.M.)
Treated (CUs9 S.E.M.)
22–24 50–62
28.2 92.2 (n=3) 20.1 91.7 (n=3)
32.7 9 2.4 (n = 3)* 21.4 9 2.3 (n= 4)*
The values are expressed in CUs 9 S.E.M. The analyzer was calibrated considering ‘zero’ the values of control sections incubated without antibodies. For other precautions used in QAI, see Section 1.11. Each value represents the mean of numerous determinations (at least 50) carried out with double masked technique, performed in triplicate9S.E.M. P was calculated by comparing the significant differences of treated versus not treated subjects. n= number of subjects. * PB0.001.
Table 4 Occurrence of NPY in different structures of the thymus of young subjects treated and not treated with immunostimulating therapy Structures of thymus
Not treated (CUs9S.E.M.)
Treated (CUs9 S.E.M.)
Whole thymus Thymic microenvironment Structures resembling nerve-fibers Reticular cells Lymphocytes Hassall’s corpuscles Arterioles Venules Lymphatic vessels Septa Capsula
28.5 93.1 46.2 93.9
37.4 9 3.1* 56.3 9 4.9*
58.3 94.5
66.2 9 5.3*
23.9 9 2.4 18.6 92.1 42.1 93.1 39.3 9 2.4 26.8 91.9 11.1 9 1.1 9.2 90.9 8.1 90.8
33.1 9 3.2* 28.3 92.5* 41.7 9 3.9 (n.s.) 41.3 93.6 (n.s.) 28.9 92.5 (n.s.) 12.4 9 1.1 (n.s.) 9.5 90.4 (n.s.) 8.7 90.5 (n.s.)
The intensity of staining for NPY was measured in thymus of young subjects (n =6), treated (n =3) and not treated (n = 3) with b-interferon. On the basis of microanatomical details revealed by staining with Eosin-orange we can distinguish all the structures presented in the table. The thymic parenchyma is also called thymic microenvironment, and it includes especially the following structures: nerve fiberlike structures, reticular cells, lymphocytes and Hassall’s corpuscles. All the values arises from Quantimet (Leica®) and are expressed in CUs9S.E.M. P was calculated by comparing the significant differences of treated versus not treated subjects. n =number of subjects; n.s.= not significant. * PB0.001. Table 5 Occurrence of NPY in different structures of the thymus of adult subjects treated and not treated with immunostimulating therapy Structures of thymus
Not treated (CUs9S.E.M.)
Treated (CUs9 S.E.M.)
Whole thymus Thymic microenvironment Structures resembling nerve-fibers Reticular cells Lymphocytes Hassall’s corpuscles Arterioles Venules Lymphatic vessels Septa Capsula
19.2 91.3 33.3 92.4
24.3 9 1.8* 37.9 9 2.3*
53.5 94.1
58.7 9 4.1*
17.7 9 1.1 10.9 9 0.9 41.4 93.8 35.3 9 3.1 22.3 9 2.3 11.1 9 1.2 9.0 90.8 8.7 9 0.7
22.6 91.9* 14.8 9 1.1* 45.2 93.6* 37.1 93.1 (n.s.) 25.3 92.2 (n.s.) 13.6 9 1.1 (n.s.) 9.7 9 0.5 (n.s.) 8.9 90.6 (n.s.)
The intensity of staining for NPY was measured in thymus of adult subjects (n =7), treated (n =3) and not treated (n = 4) with b-interferon. On the basis of microanatomical details revealed by staining with Eosin-orange, we can distinguish all the structures presented in the table. The thymic parenchyma is also called thymic microenvironment, and it includes especially the following structures: nerve fiberslike structures, reticular cells, lymphocytes and Hassall’s corpuscles. All the values arises from Quantimet (Leica®) and are expressed in CUs9S.E.M. P was calculated by comparing the significant differences of treated versus not treated subjects. n =number of subjects; n.s.= not significant. * PB0.001.
D. Ca6allotti et al. / Neurochemistry International 40 (2002) 211–221
4. Discussion The present results provide direct evidence that the immune response increases the level of NPY-like immunoreactivity, of ANF and of NA contained in human thymic structures. Treatment with interferon is also capable of increasTable 6 Occurrence of ANF in the thymus of subjects of various ages treated or not treated with immunostimulating therapy AGE (years)
Not treated (CUs9 S.E.M.)
TREATED (CUs 9 S.E.M.)
22–24 50–62
47.9 9 5.3 (n= 3) 29.1 9 3.1 (n= 3)
52.8 9 5.6 (n = 3)* 32.5 9 2.8 (n = 4)*
The values are expressed in CUs 9S.E.M. The analyzer was calibrated considering ‘zero’ the values of control sections incubated without glyoxylic acid. For other precautions used in QAI, see Section 1.11. Each value represent the mean of numerous determinations (at least 50) carried out with double masked technique by different research-workers, performed in triplicate 9 S.E.M. P was calculated by comparing the significant differences of treated versus not treated subjects. P was calculated by comparing the significant differences of treated versus not treated subjects. n = number of subjects. * PB0.001. Table 7 Occurrence of ANF in different structures of the thymus of young subjects treated and not treated with immunostimulating therapy Structures of thymus
Not treated (CUs9S.E.M.)
Treated (CUs9 S.E.M.)
Whole thymus Thymic microenvironment Structures resembling nerve-fibers Reticular cells Lymphocytes Hassall’s corpuscles Arterioles Venules Lymphatic vessels Septa Capsula
47.39 4.4 61.295.8
59.5 9 3.9* 71.4 9 5.1*
78.6 96.2
78.1 9 5.6 (n.s.)
35.3 93.1 26.4 9 2.1 51.694.3 40.5 93.5 31.1 92.6 21.49 1.9 9.390.8 8.190.7
43.8 9 3.6* 38.6 9 2.9* 55.4 9 4.1* 42.29 3.3 (n.s.) 33.7 9 2.8 (n.s.) 24.5 9 2.1 (n.s.) 10.39 0.9 (n.s.) 8.1 9 0.6 (n.s.)
The intensity of staining for ANF was measured in thymus of young subjects (n =6), treated (n=3) and not treated (n = 3) with b-interferon. On the basis of microanatomical details revealed by staining with Eosin-orange we can distinguish all the structures presented in the table. The thymic parenchyma is also called thymic microenvironment, and it includes especially the following structures: nerve fiberslike structures, reticular cells, lymphocytes and Hassall’s corpuscles. All the values arises from Quantimet (Leica®) and are expressed in CUs9 S.E.M. The analyzer was calibrated considering ‘zero’ the values of control sections incubated without glyoxylic acid. For other precautions used in QAI, see Section 1.11. P was calculated by comparing the significant differences of treated versus not treated subjects. n =number of subjects; n.s. = not significant. * PB0.001.
219
Table 8 Occurrence of NPY in different structures of the thymus of adult subjects treated and not treated with immunostimulating therapy Structures of thymus
Not treated (CUs9S.E.M.)
Treated (CUs9 S.E.M.)
Whole thymus Thymic microenvironment Structures resembling nerve-fibers Reticular cells Lymphocytes Hassall’s corpuscles Arterioles Venules Lymphatic vessels Septa Capsula
24.1 91.9 33.6 93.1
29.4 9 2.1* 38.3 9 3.4*
73.4 96.4
77.8 9 5.9 (n.s.)
18.2 91.7 15.7 9 1.6 61.5 95.3 28.3 9 2.6 23.1 92.1 14.2 91.2 9.4 90.9 8.6 90.5
22.7 92.0* 19.4 9 1.6* 68.3 95.6* 31.1 9 2.1 (n.s.) 25.2 9 1.9 (n.s.) 16.5 9 1.5 (n.s.) 10.4 9 0.8 (n.s.) 9.3 9 0.6 (n.s.)
The intensity of staining for ANF was measured in thymus of adult subjects (n =7), treated (n =3) and not treated (n = 4) with b-interferon. On the basis of microanatomical details revealed by staining with Eosin-orange we can distinguish all the structures presented in table. The thymic parenchyma is also called thymic microenvironment, and it includes especially the following structures: nerve fiberslike structures, reticular cells, lymphocytes and Hassall’s corpuscles. All the values arises from Quantimet (Leica®) and are expressed in CUs9S.E.M. The analyzer was calibrated considering ‘zero’ the values of control sections incubated without glyoxylic acid. For other precautions used in QAI, see Section 1.11. P was calculated by comparing the significant differences of treated versus not treated subjects. n =number of subjects; n.s. =not significant. * PB0.001.
ing the protein content of the human thymus. The biochemical data, together with the histochemical results, provide direct evidence for a specific localization of a sympathetic innervation in the thymic gland. The ANF of the thymus, especially after treatment with interferon, are localized particularly in the thymic microenvironment. Owing to this strong localization of the sympathetic nerve fibers in the microenvironment of the human thymuses treated or not treated with interferon drugs, it is possible to study the relationships between nerve fibers and thymic cells by examining our experimental results obtained by serial sections and by QAI. In conclusion, treatment with interferon drugs induces an immune response. This response can be verified with dosages of elevated antibody titers, as we have performed in each patient (results not reported). In our patients the efficiency of immunostimulating therapy was verified by means of common laboratory tests performed during hospitalization. A significant elevation in thymic levels of proteins was found after treatment with interferon drugs. The possible biological significance of this elevation can be considered. The thymic cells that synthesize thymosin also possess
220
D. Ca6allotti et al. / Neurochemistry International 40 (2002) 211–221
adrenergic receptors. The thymic ANF can act on these receptors. In comparison with other immunostimulating drugs, useful in MS and in other autoimmune neurological diseases, such as thymopoietin and thymosin (whose effects produce a more general stimulation of cellular thymic response), interferon therapy performed in this study induces different modifications on morphological and functional parameters. Our results may lead us to affirm that, while thymosin influences the development of T-cells and thymopoietin induces T-cell maturation from prothymocytes, inhibiting in the meantime B-cell differentiation, interferons act mainly on quantitative increase of morphological and functional thymic parameters (protein content, NPY-like immunoreactivity, ANF, NA content) whose modifications underline the effects of immunostimulating therapy directly on thymic tissues and via the action of the sympathetic nervous system on thymic structures. Our results demonstrate that patients treated with interferon, a potent promoter of the immune response, show, in comparison to untreated patients, high levels of many adrenergic parameters in thymic tissues (i.e. ANF, NA, NPY-like immunoreactivity). Interferon treatment is also capable of increasing the protein content of the thymus. Because many samples (untreated controls) were taken at autopsy, any different expression of NA, NPY and ANF may be due to post-mortem degradation. Nevertheless, the delay following death was minimal, as demonstrated by More and Fatty (1958) for cholinesterase activity. Moreover, in order to prevent oxidation of NA during processing of the tissue, samples of tissues were immersed in a solution 1 mmol of 3-mercapto-propionic acid. Treatment with this drug has been found to be suitable for prevention of post-mortem increase of endogenous levels of neurotransmitters in the brain (van der Heyden and Korf, 1978): it is also able to prevent the oxidation of neurotransmitters in the ovaries and in the uterine tube and does not interfere with NA determination (Erdo et al., 1989). Therefore the samples coming from autopsies can be compared with those are coming from biopsies. In conclusion, treatment with interferon therapy is able to induce many changes in the thymic structures. Our biochemical data together with the histochemical qualitative and quantitative results provide direct evidence for a specific localization of ANF in the thymic gland.
Acknowledgements The authors are greatly indebted to Drs B. Nagar and J. Feher for the kind help furnished in the critical revision of the manuscript. The technical assistance of D. Caporuscio, the photographic service of G.
Leoncini, the excellent secretarial work of S. Casamento and the kind help of S. Hobby in the revision of the English language are also gratefully acknowledged. The present study was supported by grants from MURST and University of Rome ‘La Sapienza’ Ministerio Universita e Ricerca Scientifica e Tecnologica.
References Bodian, D., 1936. A new method for staining nerve fibers and nerve ending in mounted paraffin sections. Anat. Rec. 65, 89 –97. Bulloch, K., Moore, R.Y., 1981. Innervation of the thymus gland by brain stem and spinal cord in mouse and rat. Am. J. Anat. 162 (2), 157 – 166. Castino, M., Roletto, E., 1992. Statistica Applicata. Piccin, Padova. Coons, A.H., Leduc, E.H., Connolly, J.M., 1955. Studies on antibody production. I. A method for the histochemical demonstration of specific antibody and its application to a study of the hyperimmune rabbit. J. Exp. Med. 102, 49 – 60. de Leeuw, F.E., Jansen, G.H., Batanero, E., van Wichen, D.F., Huber, J., Schuurman, H.J., 1992. The neural and neuro-endocrine component of the human thymus. I. Nerve-like structures. Brain Behav. Immun. 6, 234 – 248. Elenkov, I.J., Wilder, R.L., Chrousos, G.P., Vizi, E.S., 2000. The sympathetic nerve —an integrative interface between two supersystems: the brain and the immune system. Pharmacol. Rev. 52 (4), 595 – 638. Erdo, S.L., Kiss, B., Szporny, L., 1989. Age-dependent changes of Gaba and Noradrenaline levels in rat ovary and uterine tubes. In: Cerebrovascolar Neuroeffector Mechanisms in Health and Disease. EUR, Roma, pp. 191 – 202. Felten, D.L., Felten, S.Y., Carlson, S.L., Olschowska, J.A., Livnat, S., 1985. Noradrenergic and peptidergic innervation of lymphoid tissue. J. Immunol. 135 (Suppl. 2), 755s – 765s. Ghali, W.M., Abdel-Rahman, S., Nagib, M., Mahran, Z.Y., 1980. Intrinsic innervation and vasculature of pre- and post-natal human thymus. Acta Anatomica (Basel) 108 (1), 115 – 123. Haar, J.L., 1974. Light and electron microscopy of the human fetal thymus. Anat. Rec. 179, 463 – 476. Keller, R., Oke, A., Mefford, I., Adams, R.N., 1976. Liquid chromatographic analysis of catecholamines. Life Sci. 19, 995 –1004. Kendall, M.D., al-Shawaf, A.A., 1991. Innervation of the rat thymus gland. Brain Behav. Immun. 5, 19 – 28. Kranz, A., Kendall, M.D., von Gaudecker, B., 1997. Studies on rat and human thymus to demonstrate immunoreactivity of calcitonin-gene-related-peptide, tyrosine hydroxylase and neuropeptide Y. J. Anat. 191, 441 – 450. Lowry, O.H., Rosebrough, N.J., Farr, A.L., 1951. Protein measurement with the Folin phenol reagent. J. Biol. Chem. 193, 265 –275. Madden, K.S., Bellinger, D.L., Felten, S.Y., Snyder, E., Maida, M.E., Felten, D.L., 1997. Alterations in sympathetic innervation of thymus and spleen in aged mice. Mech. Ageing Dev. 94, 165 – 175. Magni, F., Bruschi, F., Kasti, M., 1987. The afferent innervation of the thymus gland in the rat. Brain Res. 424, 379 – 385. Manual of Methods of Quantimet Leica® 500 Microsystems Imaging Solutions (1997). Clifton Road, Cambridge. More, E.J., Fatty, C.S., 1958. A note on cholinesterase activity in post-mortem tissues. J. Histochem. Cytochem. 6, 377 – 379. Qayyum, M.A., Fatani, J.A., 1985. Use of glyoxylic acid in the demonstration of autonomic nerve profiles. Experientia 41, 1389 – 1390. Serio, A., 1986. Appunti Dalle Lezioni di Statistica Sanitaria. Kappa, Roma.
D. Ca6allotti et al. / Neurochemistry International 40 (2002) 211–221 Singh, U., 1984. Sympathetic innervation of fetal mouse thymus. Eur. J. Immunol. 14, 757 –759. Townsend, F.M., 1960. Manual of histologic and special staining techniques. McGraw-Hill, New York. Uddman, R., Ekblad, E., Edvinsson, L., Ha˚ kanson, R., Sundler, F., 1985. Neuropeptide Y like immunoreactivity in perivascular nerve fibers of the guinea pig. Regul. Pept. 10, 243 –256.
.
221
van der Heyden, J.A., Korf, J., 1978. Regional levels of Gaba in the brain: rapid semiautomatic assay and prevention of postmortem increase by 3-mercapto-propionic acid. J. Neurochem. 31, 197 –203. Vizi, E.S., Orso´ , E., Osipenko, O.N., Hasko´ , G., Elenkov, I.J., 1995. Neurochemical, electrophysiological and immunocytochemical evidence for a noradrenergic link between the sympathetic nervous system and thymocytes. Neuroscience 68, 1263 – 1269.