- Email: [email protected]
Occurrence of γ-aminobutyric acid-transaminase activity in nerve fibers of human thymus
Occurrence of ␥-Aminobutyric Acid-Transaminase Activity in Nerve Fibers of Human Thymus Daniela Cavallotti, Marco Artico, Stefano De Santis, and Carlo Cavallotti ABSTRACT: The specific localization of ␥-aminobutyric acid-transaminase (GABA-t) in the thymus of young and elderly men was studied. Our results show a specific vascular localization of GABA-t in the human thymus, and deal with the amount and distribution of GABA-t and its changes with age. Samples of human thymus were harvested throughout of 12 autopsies in infants (n ⫽ 3), as well as young (n ⫽ 3), adult (n ⫽ 3) and elderly (n ⫽ 3) men. Histologic staining of the human thymus was performed with eosin-orange, while histologic staining of nerve fibers was performed with the Bodian method. Histochemical and biochemical demonstration of GABA-t, including protein dosage, was performed by the methods of Van Gelder and Jung, respectively. Finally, quantitative analysis of images was performed. Staining with eosin-orange reveals the micro-anatomical details of the thymic micro-environment. The Bodian method shows the nerve fibers and neurofibrils. Histochemical
ABBREVIATIONS GABA-t ␥-aminobutyric acid-transaminase BSA bovine serum albumin NADH nicotinamide adenine dinucleotidedehydrogenase
INTRODUCTION The human thymus is richly innervated. In 1918 Crotti [1] and in 1935 Hammer [2] were the first to study thymic innervation. Other studies substantiated their
From the Neurologic Clinic (D.C.), Department of Cardiovascular and Respiratory Sciences (C.C.), and Chair of Human Anatomy (M.A., S.D.S.), Faculty of Pharmacy, University “La Sapienza,” Rome, Italy. Supported by grants from MURST, CNR, and University of Rome “La Sapienza” (CAP-DIP 98.043.14). Address reprint requests to Prof. Dr. Carlo Cavallotti, Department of Cardiovascular and Respiratory Sciences, University of Roma “La Sapienza,” Via A. Borelli, 50, 00161 Rome, Italy; Tel: 0039/06/4958291; Fax: 0039/06/4957669. Received June 8, 1999; accepted August 11, 1999. Human Immunology 60, 1072–1079 (1999) © American Society for Histocompatibility and Immunogenetics, 1999 Published by Elsevier Science Inc.
staining for GABA-t shows an increase of this enzyme with age and a marked localization in the nerve fibers of the thymus in infant, young, adult, and elderly men, as well as specific vascular localization of this enzyme. These biochemical data are in accordance with the histoenzymatic results and confirm all of our previous observations. Finally, quantitative analysis of images performed on slices let us confirm all the morphological changes induced by age. We can conclude that GABA is an inhibitory neurotransmitter of the human thymus, while GABA-t plays an important role in GABA metabolism. Human Immunology 60, 1072–1079 (1999). © American Society for Histocompatibility and Immunogenetics, 1999. Published by Elsevier Science Inc. KEYWORDS: gabatransaminase; thymus; quantitative image analysis; histoenzymatic staining; biochemical dosage; aging; nerve fibers
NAD QAI CU
nicotinamide adenine dinucleotide quantitative analysis of images conventional units
initial results [3– 6] but did not distinguish between the sympathetic or parasympathetic nature of thymic innervation. Later studies provided direct evidence of the neurological control of the immunologic functions of the thymus [7, 8], and revealed that the thymus is innervated by both sympathetic and parasympathetic nerve fibers [9, 10]. Our recent studies using histoenzymatic microscopy and biochemical methods have revealed the presence of ␥-aminobutyric acid-transaminase (GABA-t) activity in the human thymus that increases during the immune response [11]. We also found GABA-t in the thymus of rats in both normal conditions [12] and after treatment 0198-8859/99/$–see front matter PII S0198-8859(99)00107-X
GABA-t in Human Thymus
with interleukins [13]. To improve knowledge of the presence of GABA-t in the nerve fibers of the human thymus, we analyzed in this study: (1) whether GABA-t is localized in the nerve fibers of the thymus of infant, young, adult, and elderly men; (2) whether GABA-t has a specific vascular localization in the human thymus; and (3) the amount and distribution of GABA-t in the human thymus and how it changes with age. MATERIALS AND METHODS The experimental procedures performed in this study include: (1) harvesting of the thymus at autopsy; (2) staining of the thymus; (3) staining of nerve fibers; (4) histochemical demonstration of GABA-t; (5) estimation of the protein content; (6) biochemical dosage of GABA-t; (7) quantitative analysis of images (QAI); and (8) statistical analysis of data. Each procedure is briefly explained. Harvesting of Thymus Samples in the Course of Autopsy All experiments were subject to the approval of the Ethical Committee at our university. In the course of 12 autopsies of 12 subjects aged between 8 – 82 years who had died from non-mediastinal and non-neurological diseases, small fragments of thymus were obtained. Thymus or residual islands of thymic tissue after involution in adult or in elderly subjects were found in the retrosternal adipose tissue after removing the sternocostal plate. Thymic tissue was recognized by classic histological staining, as reported above. Thymic fragments were rapidly transported to our laboratories for experimental procedures.
1073
solution with counterstaining with 0.03 aniline blue. The nerve fibers and neurofibrils are colored in black. Histochemical Demonstration of GABA-t Activity The Van Gelder method was used for the histochemical detection of GABA-t [16], with the slight modifications described in our previous report to minimize enzyme diffusion [17]. Briefly, each slice of thymus was stretched on pre-weighed slides. Subsequently, the slides were newly weighed to calculate the weight of the sample. Specimens were treated by histoenzymatic staining that involves the reduction of tetrazolium salts. Control sections were incubated without substrates or using 50 M of amino-oxalacetic acid, 50 M of gabaculline, 200 M of acetylenic GABA, or 100 M of ethanolamine sulphate as inhibitors, one at a time or simultaneously. Estimation of Protein Content In all of these experiments, autopsy samples of thymus were weighed and placed on dry ice (specimens for GABA-t histochemical staining) or into an ice-cold homogenization buffer (samples for estimating the protein content and GABA-t activity). Tissue protein concentrations were determined by the method described by Lowry et al. [18] using bovine serum albumin (BSA) as the standard.
Staining of Thymus As aforementioned, thymic tissues were identified by classic staining with eosin-orange. Water soluble eosin ␥ (10 gm) was dissolved in 1000 ml distilled water and 2 ml glacial acetic acid was added. This is the stock 1% aqueous eosin solution. One part of stock solution plus three parts of distilled water form the working eosin solution. If a deeper shade of red is desired (eosin-orange) in staining, 0.5 ml of glacial acetic acid is added to each 100 ml of solution [14].
Biochemical Dosage of GABA-t The biochemical dosage of GABA-t activity (EC 2.6.1.19 4-aminobutyrate 2-oxoglutarate aminotransferase) was performed using the method of Jung et al. [19] employing 6 M GABA plus 5 M 2-oxoglutarate as substrate and determining the formation of NADH starting by NAD. Specific GABA-t activity was calculated by subtracting the blank values (measured in the presence of 50 M of amino-oxalacetic acid as the inhibitor or in presence of other inhibitors or in the absence of substrate) from the total activity. Other GABA-t inhibitors used were 50 M gabaculline, 200 M acetylenic GABA or 100 M ethanolamine-o-sulfate. The biochemical values obtained using all these inhibitors (one at a time or simultaneously) can be considered as blank values and must be subtracted from the values of total activity.
Staining of Nerve Fibers These structures were colored by the method of Bodian [15]. This method can be used to verify that a stained structure is nervous in nature. In fact, it stains myelinic and non-myelinic nerve fibers and neurofibrils. After fixation in Bouin’s fluid, 6 –10 sections were treated with: (1) 1% Protargol solution (colloidal silver), (2) reducing solution (Hydroquinone plus sodium sulphite), (3) 1% Gold chloride solution, and (4) 2% oxalic acid
Quantitative Analysis of Images To evaluate the histoenzymatic levels of GABA-t, a quantitative analysis of the intensity of the histochemical staining was performed on slides by means of a Quantimet Analyzer (Leica威). The values of control sections, incubated without substrates, were considered as zero. Each slice was examined separately while evaluating SEM. QAI may provide incorrect results. In fact, the main
1074
choices (that is, instructions for software) are selected by each research worker, according to personal preference. For these reasons, the resulting data are more partial than impartial and it is necessary to follow rules very carefully. 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 analyzers 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 to the statistical analysis of data. The values reported represent the intensity of staining for each sample and are expressed in conventional units (CU) ⫾ SEM. Statistical Analysis of Data The statistical methods used throughout this study must be interpreted as an accurate description of the data more than a statistical inference of such data. The preliminary studies of each value were performed with the use of basic sample statistics. Mean values, maximum and minimum limits, variations, SD, SEM, and correlation coefficients were determined according to Serio [20]. 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 lineal (values next ⫹1 or ⫺1) or non-lineal (values next ⫹0) dependency. Finally, a correlative analysis of the morphological and biochemical data was performed by comparing the significant differences for each age group with the corresponding values in the other homogeneous groups. Correlation coefficients denote a significant level less than 0.01 (p ⬍ 0.01), and a significant level less than 0.005 (p ⬍ 0.005), while the correlation coefficient is not significant when p ⬎ 0.05 or not significant. This correlation coefficient was calculated according to Castino and Roletto [21]. RESULTS Our results are reported in Figs. 1–5 and are summarized in Tables 1– 4. Figure 1 shows the immunohistochemical staining of GABA-t in the human thymus of a 10-year-old boy. Reticular cells and thymocytes show a positive reaction. The wall of arterioles and of lymphatic vessels show a positive reaction. At this age, the thymus is not involuted. Figure 2 shows the structure of the thymus in a young 23-year-old man stained by the immunohistochemical method for GABA-t. As this age, the thymus shows an involution of structure. In fact, large parts of the parenchyma are replaced by adipose tissue. The thymocytes
D. Cavallotti et al.
FIGURE 1 Immunohistochemical staining of GABA-t activity in the thymus of a 10-year-old boy. The walls of two arterioles (A) and six lymphatic vessels (L) show a positive reaction. Many reticular cells (R) and lymphocytes (thymocytes) (T) also show a strong positive reaction. Magnification ⫻250.
have decreased in number. The reticular cells, the wall of arteriolar and lymphatic vessels, and many structures resembling nerve fibers show a slight positive reaction. Structures resembling nerve fibers are recognized as nerve fibers by specific staining by the method of Bodian [15]. Figure 3 shows the human thymus of a 50-year-old man stained by the immunohistochemical method for GABA-t. The thymus is clearly involuted. Only small islands of recognizable thymic tissue are contained in a body of fibrofatty tissue. Only the walls of arterioles, FIGURE 2 Immunohistochemical staining of GABA-t activity in the human thymus of a young 23-year-old man. The wall of an arteriolar vessel (A) and of a lymphatic vessel (L) show a positive reaction. Many reticular cells (R) also show a positive reaction, as do many structures resembling nerve fibers (arrows). The total number of thymocytes is decreased. Magnification ⫻250.
GABA-t in Human Thymus
1075
FIGURE 3 Immunohistochemical staining of GABA-t activity in involuted thymus of a 50-year-old man. One arteriola (A), two venules (V) and three lymphatic vessels (L) show a positive reaction of their walls. Many reticular cells (R) and many structures resembling nerve fibers (arrows) show a slight positive reaction. Magnification ⫻250.
FIGURE 5 Immunohistochemical staining of GABA-t activity in thymus of an elderly 82-year-old man. The walls of an arteriola (A) a venula (V), and a lymphatic vessel (L) show a positive reaction. A small number of reticular cells (R) show a mild positive reaction and many structures resembling nerve fibers (arrows) show a positive reaction. Magnification ⫻500.
venules, and lymphatic vessels show a positive reaction. A small number of reticular cells and many structures resembling nerve fibers (recognized as aforementioned) show a mild positive reaction. Figure 4 shows a portion of Fig. 3 at high magnification (⫻500). The walls of two arterioles, a venula and a lymphatic vessel show a positive reaction. Only a small number of reticular cells show a poor reaction. Many structures resembling nerve fibers and recognizable by Bodian staining show a mild positive reaction. Due to involution of the thymus, fibrofatty tissue entirely surrounds small islands of recognizable thymic tissue.
Figure 5 shows immunohistochemical staining for GABA-t activity in the thymic tissue residues of an elderly 82-year-old man. At this age, the thymus becomes a fibrous pad containing only adipose cells with a few small residual islands of poorly recognizable thymic tissue. In Fig. 5 we can observe an arteriola, a venula and a lymphatic vessel showing positive staining of the walls. A small number of reticular cells show a mild positive reaction, while many structures resembling nerve fibers and recognized by Bodian staining show a positive reaction. All of our morphological results show that GABA-t activity, an enzyme that metabolizes GABA (an inhibitory neurotransmitter of numerous nervous and nonnervous peripheral structures) is present in the thymus. Table 1 deals with our biochemical findings regarding the amounts of GABA-t activity contained in the different structures of the human thymus in 12 subjects 8 to 82 years old. The highest mean amount of GABA-t
FIGURE 4 Immunohistochemical staining of GABA-t activity in thymus of a 50-year-old man showing a portion of Fig. 3 at higher magnification. Two arterioles (A), a venula (V), and a lymphatic vessel (L) show a positive reaction of their walls. Only a small number of reticular cells (R) show a poor reaction. Many structures resembling nerve fibers (arrows) also show a positive reaction. Magnification ⫻500.
TABLE 1 Biochemical values of GABA-t activity in the human thymus Tissue
GABA-ta (nM/mg protein/h)
Whole thymusb Arteriesc Veinsd
28.3 ⫾ 1.59 63.1 ⫾ 3.16 42.1 ⫾ 1.68
Mean value ⫾ SD of three independent determinations in 12 subjects. Fragment of thymus harvested from autopsies. c Large arterial vessels that richly supply the thymus. d Extraparenchymal venous vessels coming from thymus. GABA-t, ␥-aminobutyric acid-transaminase a
b
1076
D. Cavallotti et al.
TABLE 2 Amount of GABA-t (biochemical dosage) in whole thymus of young and elderly men
GABA-ta (nM/mg protein/h)
Young (n ⫽ 3)
Elderly (n ⫽ 3)
32.4 ⫾ 1.26
47.3 ⫾ 1.18b
Mean value ⫾ SD of three independent determinations in 3 subjects in each age group. b p ⬍ 0.001. GABA-t, ␥-aminobutyric acid-transaminase a
activity (63.1 ⫾ 3.16 nM/mg protein/h) is contained in the wall of the thymic arteries, followed by that in the wall of the thymic veins (42.1 ⫾ 1.68 nM/mg protein/h). Finally, the whole thymus contains only a small amount of GABA-t activity (28.3 ⫾ 1.59 nM/mg protein/h). Table 2 indicates the amounts of GABA-t activity contained in the whole thymus of 3 young and 3 elderly subjects. The young thymus contains 32.4 ⫾ 1.26 nM of GABA-t/mg protein/h, while the old thymus contains a higher amount of GABA-t activity (47.3 ⫾ 1.18). Statistical data analysis gives a high significance with p ⬍ 0.001. QAI after histoenzymatic staining reveals that in structures resembling nerve fibers and recognized after post-coloration with the Bodian method the occurrence of GABA-t is very high. Table 3 demonstrates, in fact, that GABA-t is contained in nerve fibers of the thymus. Its amount is 26.6 ⫾ 1.4 conventional units (CU) (see Methods). It varies in infant, young, adult and elderly subjects from 25.2 ⫾ 1.2 to 27.5 ⫾ 1.4 in young, 31.2 ⫾ 1.6 in adult, and 44.8 ⫾ 1.8 in elderly patients. These results are expressed in CU and all values ⫾ SEM were obtained using a Quantimet image analyzer. In this case, too, statistical analysis of data shows a high index of significance for adult and elderly subjects with p ⬍ TABLE 3 Occurrence of GABA-t in structures resembling nerve fibers (histoenzymatic staining and QAI) Age Group (years)
Number of Subjects
Staining for GABA-ta,b (mean CU ⫾ SEM)
p Value
8–82 Total 8–12 Infant 22–24 Young 50–62 Adult 77–82 Elderly
12 3 3 3 3
26.6 ⫾ 1.4 25.2 ⫾ 1.2 27.5 ⫾ 1.4 31.2 ⫾ 1.6 44.8 ⫾ 1.8
n. s. n. s. ⬍0.001 ⬍0.001
a The intensity of staining for GABA-t was measured in whole fragments of human thymus obtained at autopsy. b All values were obtained using a Quantimet (see Methods). GABA-t, ␥-aminobutyric acid-transaminase; QAI, quantitative analysis of images; n. s., not significant
TABLE 4 Occurrence of GABA-t in different structures of the thymus (histoenzymatic staining and QAI) Structures of Thymus
Staining of GABA-ta,b (mean CU ⫾ SEM)
Arterioles Venules Lymphatic vessels Structures resembling nerve fibers Reticular cells Lymphocytes Corpuscles of Hassal Septa Capsula Total thymic micro-environmentc Whole thymus
61.3 ⫾ 2.5 48.4 ⫾ 1.9 32.1 ⫾ 1.6 26.6 ⫾ 1.4 14.1 ⫾ 1.5 13.2 ⫾ 1.6 9.3 ⫾ 0.8 6.6 ⫾ 0.6 5.4 ⫾ 0.6 13.6 ⫾ 1.8 29.6 ⫾ 2.8
a Intensity of staining for GABA-t was measured in the thymus of 12 subjects 8 – 82 years old. On the basis of micro-anatomical details revealed by staining with eosin-orange we distinguished all indicated structures. b All values were obtained using Quantimet (see Methods). c Parenchyma of the thymus, which includes structures resembling nerve fibers, reticular cells, lymphocytes, and corpuscles of Hassal. GABA-t, ␥-aminobutyric acid-transaminase; QAI, quantitative analysis of images
0.001, while for infant and young subjects the values are not significant. Table 4 indicates all the structures contained in the thymic fragments susceptible to recognition by means of eosin-orange staining. For all of these structures, QAI performed after histoenzymatic staining for GABA-t is able to reveal the quantitative intensity of staining expressed in CU ⫾ SEM (see Methods). From a detailed examination of this table, it becomes apparent that the structures with the highest occurrence of GABA-t activity (61.3 ⫾ 2.5 CU) are the walls of the thymic arterioles, followed by the wall of the venules (48.4 ⫾ 1.9 CU). The lymphatic vessels also show strong levels of GABA-t activity (32.1 ⫾ 1.6 CU). Nerve fibers, recognized by specific Bodian staining, only contain 26.6 ⫾ 1.4 CU of GABA-t activity. The total thymic micro-environment, formed substantially by T lymphocytes or thymocytes, by reticular cells, and by corpuscles of Hassal, contains only 13.6 ⫾ 1.8 CU of GABA-t activity. The total thymic micro-environment forms the parenchyma of the thymus. Septa and capsula of the thymus have a low GABA-t content (only 6.6 – 5.4 ⫾ 0.6 CU). All values indicate the intensity of staining for GABA-t and were measured in thymic fragments of 12 subjects between 8 – 82 years of age. DISCUSSION Concerning the possible biological significance of GABA-t in the nerve fibers of the human thymus, the
GABA-t in Human Thymus
occurrence of ␥-aminobutyric acid (GABA) in the human tissues, including peripheral nerve fibers, was first reported three decades ago by Zachmann et al. [22]. Since then, convincing evidence has accumulated that not only GABA, but also enzymes involved in its biosynthesis and catabolism (that is glutamate decarboxylase and GABAtransaminase, respectively) are present in the peripheral tissues of different mammals [23]. Moreover, a high-affinity GABA uptake system has been identified in the mammalian periphery [24], including the endocrine pancreas [25]. However, despite of the large body of evidence concerning the existence of a GABAergic system in the many peripheral tissues, to our knowledge no information is yet available concerning the significance of GABA-t activity, as demonstrated by pharmacohistochemical methods [26]. In our previous studies it was pointed out that GABA-t activity is present in the cerebellar cortex and is stimulated following  estradiol treatment [17]. Moreover, in convoluted tubules of the cortex of the rat kidney, the presence of H3 muscimol, a specific marker of GABA-A receptors [27] was demonstrated. Thymic Microenvironment and its Innervation Comparing our experimental morphological and biochemical results with those reported by others [28, 29], confirms that the human thymus is a central lymphoid organ which plays a key role in cell-mediated immunity. It weighs about 15–20 gm at birth, increases to approximately 30 gm at puberty, and thereafter progressively atrophies to become a fibrofatty tissue of approximately 10 gm containing only small islands of recognizable thymic tissue in the adult. In elderly subjects the thymus becomes a fibrous pad containing only adipose cells with a few small residual islands of recognizable thymic tissue. Two lobes surrounded by a connectival capsule form the thymus. From this capsule originate the fibrous septa that divide each lobe into lobules. The lobules present a peripheral cortex and a central medulla. The cortex is composed of small closely packed lymphocytes, termed thymocytes, belonging to the T-cell systems. Scattered among these lymphocytes are larger, paler cells capable of active phagocytosis termed reticular cells. Due to the abundant cytoplasm and pale vesicular nuclei, the reticular cells are termed reticulo-epithelial or epithelial cells. On the contrary, the medulla is composed of numerous reticular cells with a few lymphocytes between them. In addition, the medulla contains some layered keratinizing epithelial cells known as corpuscles of Hassal. All of these structures form the thymic microenvironment [30, 31]. In particular, it should be noted that normal human thymus does not contain lymphoid
1077
follicles; their appearance is interpreted as thymic hyperplasia. Innervation of the thymic micro-environment arises from the parasympathetic nuclei localized in brain stem and in the spinal cord through the X cranial nerve [32]. Cholinergic innervation induces an increase of the immunoreactivity of thymic lymphocytes [33]. Cyclic nucleotides are also capable of increasing the activation and function of lymphocytes [34]. The thymus is innervated by AchE-positive fibers of the vagus, the recurrent laryngeal, and phrenic nerves. Catecholaminergic innervation derives from the stellate ganglia and other small ganglia of the thoracic sympathetic nervous system chain. Innervation of the thymic micro-environment also arises from the sympathetic system by means of trunks belonging to the cardiac plexuses surrounding the origin of the main vessels emerging from the heart wall [35]. The phrenic nerve also provides nerve fibers for the thymus. The thymic nerves run parallel to the vessels, forming periadventitial plexi and, with the vasculature, perforate the capsule of the thymus, spreading into the connective tissue and passing between lobules. Few thin nerve fibers penetrate the medulla when they end with varicosities. The thymic cortex does not present nerve fibers. Notwithstanding this, some specific coloration techniques appear to demonstrate that many types of peptidergic nerve fibers end their course in the thymic cortex [36]. Sympathetic innervation of the thymus is inhibitory to the immunoreactivity of T lymphocytes and to the proliferative response of the thymus [37– 40]. GABAergic innervation of the thymus (and in this experiment GABA-t is an enzyme linked to the metabolism of GABA) is also inhibitory to the immune response of the thymus [11], but further experiments are needed to clarify the role and the function of GABA-t activity in the human thymus. ACKNOWLEDGMENTS
The authors are greatly indebted to Drs. B. Nagar and J. Grasso for their suggestions and criticisms. The Medline, Internet and other informatic consulting services of Drs. M. Cameroni and F.M. Tranquilli Leali are gratefully acknowledged. The technical assistance of Mr. Dario Caporuscio, the photographic service of Mr. Giuseppe Leoncini, the excellent secretarial work of Ms. Silvana Casamento, and the kind help of Mrs. Sharon Hobby for revising the English version of the manuscript are also gratefully acknowledged.
REFERENCES 1. Crotti A: Thyroid and Thymus. Philadelphia, Lea & Febiger, 1918. 2. Hammer JA: Von Innervationsverhaltnisse der in Krelor-
1078
3.
4. 5. 6.
7.
8.
9. 10.
11.
12.
13.
14. 15.
16.
17.
18.
19.
gans der Thymus bis den 4 Fetal Monat. Zeitsch Mikrosk Anat Forsch 8:253, 1935. Fujiwara M, Muryobayasaky T, Shenamoto K: Histochemical demonstration of monoamines in the thymus of rat. Jap J Pharm 16:493, 1966. Fillenz M: The innervation of the cat spleen. Proc Roy Soc London 174:459, 1970. Sergeeva VE: Histotopography of catecholamines in the mammalian thymus. Bull Exp Biol Med 77:456, 1974. Reilly FD, McCuskey PA, Meineke HW: Studies of the hemopoietic microenvironment. VIII. Adrenergic and cholinergic innervation of the murine spleen. Anat Rec 185:109, 1976. Besedovsky HO, Ray A, Del Sorkin E, Prada MD, Keller HH: Immunoregulation mediated by the sympathetic nervous system. Cell Immunol 48:346, 1979. Williams JM, Peterson RJ, Shea PA, Schmedtje JF, Bauer DC, Felton DL: Sympathetic innervation of murine thymus and spleen. Evidence for a functional link between the nervous and immune systems. Brain Res Bull 6:83, 1981. Singh U: Sympathetic innervation of fetal mouse thymus. Eur J Immunol 14:757, 1984. Raymond F, Morel E, Bach JF: Evidence for the presence of immunoreactive acetylcholine receptors on human thymus cells. J Neuroimmunol 6:31, 1984. Cavallotti D, Artico M, D’Andrea V, Cavallotti C: Gabatransaminase activity in the human thymus during immunoresponse. Neuropeptides, in press. Cavallotti C, Artico M, De Santis S: Specific vascular localisation of Gabatransaminase in the thymus gland of juvenile and aged rats. Eur J Histochem, in press. Cavallotti D, Artico M, Cavallotti C, De Santis S, Tranquilli Leali FM: Interleukines and Gabatransaminase activity in rat thymus. Int J Immunopharmacol, in press. Townsend FM: Manual of Histologic and Special Staining Techniques. New York, McGraw-Hill, 1960. Bodian D: A new method for staining nerve fibers and nerve ending in mounted paraffin sections. Anat Rec 65:89, 1936. Van Gelder NM: The histochemical demonstration of ␥-aminobutyric acid metabolism by reduction of tetrazolium salts. J Neurochem 12:231, 1965. Cavallotti C, Jacopino L, Amenta F: Stimulatory effect of  estradiol treatment on Gaba degratative enzymes within rat cerebellar cortex. Neurosci Lett 39:205, 1983. Lowry OH, Rosebrough NJ, Farr AL, Randall RJ: Protein measurement with the Folin phenol reagent. J Biol Chem 193:265, 1951. Jung MJ, Lippert B, Metcalf BW, Schechter PJ, Bohlen P, Sjoerdsma A: The effect of 4-amino hex 5-ynoic acid (␥-acetylenic GABA y-ethynyl GABA) a catalytic inhibitor of GABA transaminase, on brain GABA metabolism in vivo. J Neurochem 28:717, 1977.
D. Cavallotti et al.
20. Serio A: Appunti Dalle Lezioni di Statistica Sanitaria. Rome, Kappa, 1986. 21. Castino M, Roletto E: Statistica Applicata. Padua, Piccin, 1992. 22. Zachmann M, Tocci P, Nyhan WL: The occurrence of gamma-aminobutyric acid in human tissues other than in brain. J Biol Chem 241:1355, 1966. 23. Wu JY, Lin H, Lin Y, Xu GW, Lin BH, Whang BH, Wei SP: Immunochemical characterization and visualization of glutamate decarboxylase and GABA-transaminase in peripheral tissues. In Erdo SL, Bowery NG (eds): GABAergic Mechanisms in the Mammalian Periphery. New York, Raven Press, 1986. 24. Amenta F: Autoradiographic localization of GABA receptor sites in peripheral tissues. In Erdo SL, Bowery NG (eds): GABAergic Mechanisms in the Mammalian Periphery. New York, Raven Press, 1986. 25. Gerber JC, Hare TA: Gamma-aminobutyric acid in peripheral tissue, with emphasis on the endocrine pancreas. Diabetes 28:1073, 1979. 26. Vincent SR, Kimura H, McGeer EG: The pharmacohistochemical demonstration of GABA-transaminase. Neurosci Lett 16:345, 1980. 27. Cavallotti C, Amenta F, Jacopino L, Erdo SL: Autoradiographic localization of the GABAA receptor against 3Hmuscimol within rat kidney. Pharmacology 36:390, 1988. 28. Simpson JG, Gray ES, Beck JS: Age involution in the normal human adult thymus. Clin Exp Immunol 19:261, 1975. 29. Singh J, Singh AK: Age related changes in human thymus. Clin Exp Immunol 37:507, 1979. 30. Haynes BF: The human thymic microenvironment. Adv Immunol 36:87, 1984. 31. Kendall MD: Functional anatomy of the thymic microenvironment. J Anat 117:1, 1991. 32. Bulloch K, Moore RY: Innervation of the thymus gland by brain stem and spinal cord in mouse and rat. Am J Anat 162:157, 1981. 33. Strom TB, Stykowski AJ, Carpenter CB, Merril JP: Cholinergic augmentation of lymphocyte mediated cytotoxicity. A study of the cholinergic receptors on cytotoxic lymphocytes. Proc Natl Acad Sci USA 71:1330, 1974. 34. Strom TB, Lundin AP, Carpenter CB: The role of cyclic nucleotides in lymphocyte activation and function. Proc Soc Clin Immunol 3:115, 1977. 35. Bulloch K, Pomerantz W: Autonomic nervous system innervation of thymic-related lymphoid tissue in wildtype and nude mice. J Comp Neurol 228:157, 1984. 36. Felten DL, Felten SY, Carlson SL, Olshowka JA, Livnat S: Noradrenergic and peptidergic innervation of lymphoid tissue. J Immunol 135(2 Suppl):755, 1985. 37. Singh U: Effect of catecholamines on lymphopoiesis in fetal mouse thymic explants. J Anat 129:279, 1979.
GABA-t in Human Thymus
38. Singh U: Effect of sympathectomy on the maturation of fetal thymocytes grown within the anterior eye chambers in mice. Proceedings of the VIII International Congress of Lymphatic tissue Germinal Centres, Cambridge. Adv Exp Med Biol 186:349, 1985.
1079
39. Singh U: Lymphopoiesis in the nude fetal thymus following sympathectomy. J Cell Immunol 93:222, 1958. 40. Singh U, Millson DS, Smith PA, Owen JJT: Identification of adrenoreceptors during thymocyte ontogeny in mice. Eur J Immunol 9:31, 1979.
ABBREVIATIONS GABA-t ␥-aminobutyric acid-transaminase BSA bovine serum albumin NADH nicotinamide adenine dinucleotidedehydrogenase
INTRODUCTION The human thymus is richly innervated. In 1918 Crotti [1] and in 1935 Hammer [2] were the first to study thymic innervation. Other studies substantiated their
From the Neurologic Clinic (D.C.), Department of Cardiovascular and Respiratory Sciences (C.C.), and Chair of Human Anatomy (M.A., S.D.S.), Faculty of Pharmacy, University “La Sapienza,” Rome, Italy. Supported by grants from MURST, CNR, and University of Rome “La Sapienza” (CAP-DIP 98.043.14). Address reprint requests to Prof. Dr. Carlo Cavallotti, Department of Cardiovascular and Respiratory Sciences, University of Roma “La Sapienza,” Via A. Borelli, 50, 00161 Rome, Italy; Tel: 0039/06/4958291; Fax: 0039/06/4957669. Received June 8, 1999; accepted August 11, 1999. Human Immunology 60, 1072–1079 (1999) © American Society for Histocompatibility and Immunogenetics, 1999 Published by Elsevier Science Inc.
staining for GABA-t shows an increase of this enzyme with age and a marked localization in the nerve fibers of the thymus in infant, young, adult, and elderly men, as well as specific vascular localization of this enzyme. These biochemical data are in accordance with the histoenzymatic results and confirm all of our previous observations. Finally, quantitative analysis of images performed on slices let us confirm all the morphological changes induced by age. We can conclude that GABA is an inhibitory neurotransmitter of the human thymus, while GABA-t plays an important role in GABA metabolism. Human Immunology 60, 1072–1079 (1999). © American Society for Histocompatibility and Immunogenetics, 1999. Published by Elsevier Science Inc. KEYWORDS: gabatransaminase; thymus; quantitative image analysis; histoenzymatic staining; biochemical dosage; aging; nerve fibers
NAD QAI CU
nicotinamide adenine dinucleotide quantitative analysis of images conventional units
initial results [3– 6] but did not distinguish between the sympathetic or parasympathetic nature of thymic innervation. Later studies provided direct evidence of the neurological control of the immunologic functions of the thymus [7, 8], and revealed that the thymus is innervated by both sympathetic and parasympathetic nerve fibers [9, 10]. Our recent studies using histoenzymatic microscopy and biochemical methods have revealed the presence of ␥-aminobutyric acid-transaminase (GABA-t) activity in the human thymus that increases during the immune response [11]. We also found GABA-t in the thymus of rats in both normal conditions [12] and after treatment 0198-8859/99/$–see front matter PII S0198-8859(99)00107-X
GABA-t in Human Thymus
with interleukins [13]. To improve knowledge of the presence of GABA-t in the nerve fibers of the human thymus, we analyzed in this study: (1) whether GABA-t is localized in the nerve fibers of the thymus of infant, young, adult, and elderly men; (2) whether GABA-t has a specific vascular localization in the human thymus; and (3) the amount and distribution of GABA-t in the human thymus and how it changes with age. MATERIALS AND METHODS The experimental procedures performed in this study include: (1) harvesting of the thymus at autopsy; (2) staining of the thymus; (3) staining of nerve fibers; (4) histochemical demonstration of GABA-t; (5) estimation of the protein content; (6) biochemical dosage of GABA-t; (7) quantitative analysis of images (QAI); and (8) statistical analysis of data. Each procedure is briefly explained. Harvesting of Thymus Samples in the Course of Autopsy All experiments were subject to the approval of the Ethical Committee at our university. In the course of 12 autopsies of 12 subjects aged between 8 – 82 years who had died from non-mediastinal and non-neurological diseases, small fragments of thymus were obtained. Thymus or residual islands of thymic tissue after involution in adult or in elderly subjects were found in the retrosternal adipose tissue after removing the sternocostal plate. Thymic tissue was recognized by classic histological staining, as reported above. Thymic fragments were rapidly transported to our laboratories for experimental procedures.
1073
solution with counterstaining with 0.03 aniline blue. The nerve fibers and neurofibrils are colored in black. Histochemical Demonstration of GABA-t Activity The Van Gelder method was used for the histochemical detection of GABA-t [16], with the slight modifications described in our previous report to minimize enzyme diffusion [17]. Briefly, each slice of thymus was stretched on pre-weighed slides. Subsequently, the slides were newly weighed to calculate the weight of the sample. Specimens were treated by histoenzymatic staining that involves the reduction of tetrazolium salts. Control sections were incubated without substrates or using 50 M of amino-oxalacetic acid, 50 M of gabaculline, 200 M of acetylenic GABA, or 100 M of ethanolamine sulphate as inhibitors, one at a time or simultaneously. Estimation of Protein Content In all of these experiments, autopsy samples of thymus were weighed and placed on dry ice (specimens for GABA-t histochemical staining) or into an ice-cold homogenization buffer (samples for estimating the protein content and GABA-t activity). Tissue protein concentrations were determined by the method described by Lowry et al. [18] using bovine serum albumin (BSA) as the standard.
Staining of Thymus As aforementioned, thymic tissues were identified by classic staining with eosin-orange. Water soluble eosin ␥ (10 gm) was dissolved in 1000 ml distilled water and 2 ml glacial acetic acid was added. This is the stock 1% aqueous eosin solution. One part of stock solution plus three parts of distilled water form the working eosin solution. If a deeper shade of red is desired (eosin-orange) in staining, 0.5 ml of glacial acetic acid is added to each 100 ml of solution [14].
Biochemical Dosage of GABA-t The biochemical dosage of GABA-t activity (EC 2.6.1.19 4-aminobutyrate 2-oxoglutarate aminotransferase) was performed using the method of Jung et al. [19] employing 6 M GABA plus 5 M 2-oxoglutarate as substrate and determining the formation of NADH starting by NAD. Specific GABA-t activity was calculated by subtracting the blank values (measured in the presence of 50 M of amino-oxalacetic acid as the inhibitor or in presence of other inhibitors or in the absence of substrate) from the total activity. Other GABA-t inhibitors used were 50 M gabaculline, 200 M acetylenic GABA or 100 M ethanolamine-o-sulfate. The biochemical values obtained using all these inhibitors (one at a time or simultaneously) can be considered as blank values and must be subtracted from the values of total activity.
Staining of Nerve Fibers These structures were colored by the method of Bodian [15]. This method can be used to verify that a stained structure is nervous in nature. In fact, it stains myelinic and non-myelinic nerve fibers and neurofibrils. After fixation in Bouin’s fluid, 6 –10 sections were treated with: (1) 1% Protargol solution (colloidal silver), (2) reducing solution (Hydroquinone plus sodium sulphite), (3) 1% Gold chloride solution, and (4) 2% oxalic acid
Quantitative Analysis of Images To evaluate the histoenzymatic levels of GABA-t, a quantitative analysis of the intensity of the histochemical staining was performed on slides by means of a Quantimet Analyzer (Leica威). The values of control sections, incubated without substrates, were considered as zero. Each slice was examined separately while evaluating SEM. QAI may provide incorrect results. In fact, the main
1074
choices (that is, instructions for software) are selected by each research worker, according to personal preference. For these reasons, the resulting data are more partial than impartial and it is necessary to follow rules very carefully. 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 analyzers 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 to the statistical analysis of data. The values reported represent the intensity of staining for each sample and are expressed in conventional units (CU) ⫾ SEM. Statistical Analysis of Data The statistical methods used throughout this study must be interpreted as an accurate description of the data more than a statistical inference of such data. The preliminary studies of each value were performed with the use of basic sample statistics. Mean values, maximum and minimum limits, variations, SD, SEM, and correlation coefficients were determined according to Serio [20]. 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 lineal (values next ⫹1 or ⫺1) or non-lineal (values next ⫹0) dependency. Finally, a correlative analysis of the morphological and biochemical data was performed by comparing the significant differences for each age group with the corresponding values in the other homogeneous groups. Correlation coefficients denote a significant level less than 0.01 (p ⬍ 0.01), and a significant level less than 0.005 (p ⬍ 0.005), while the correlation coefficient is not significant when p ⬎ 0.05 or not significant. This correlation coefficient was calculated according to Castino and Roletto [21]. RESULTS Our results are reported in Figs. 1–5 and are summarized in Tables 1– 4. Figure 1 shows the immunohistochemical staining of GABA-t in the human thymus of a 10-year-old boy. Reticular cells and thymocytes show a positive reaction. The wall of arterioles and of lymphatic vessels show a positive reaction. At this age, the thymus is not involuted. Figure 2 shows the structure of the thymus in a young 23-year-old man stained by the immunohistochemical method for GABA-t. As this age, the thymus shows an involution of structure. In fact, large parts of the parenchyma are replaced by adipose tissue. The thymocytes
D. Cavallotti et al.
FIGURE 1 Immunohistochemical staining of GABA-t activity in the thymus of a 10-year-old boy. The walls of two arterioles (A) and six lymphatic vessels (L) show a positive reaction. Many reticular cells (R) and lymphocytes (thymocytes) (T) also show a strong positive reaction. Magnification ⫻250.
have decreased in number. The reticular cells, the wall of arteriolar and lymphatic vessels, and many structures resembling nerve fibers show a slight positive reaction. Structures resembling nerve fibers are recognized as nerve fibers by specific staining by the method of Bodian [15]. Figure 3 shows the human thymus of a 50-year-old man stained by the immunohistochemical method for GABA-t. The thymus is clearly involuted. Only small islands of recognizable thymic tissue are contained in a body of fibrofatty tissue. Only the walls of arterioles, FIGURE 2 Immunohistochemical staining of GABA-t activity in the human thymus of a young 23-year-old man. The wall of an arteriolar vessel (A) and of a lymphatic vessel (L) show a positive reaction. Many reticular cells (R) also show a positive reaction, as do many structures resembling nerve fibers (arrows). The total number of thymocytes is decreased. Magnification ⫻250.
GABA-t in Human Thymus
1075
FIGURE 3 Immunohistochemical staining of GABA-t activity in involuted thymus of a 50-year-old man. One arteriola (A), two venules (V) and three lymphatic vessels (L) show a positive reaction of their walls. Many reticular cells (R) and many structures resembling nerve fibers (arrows) show a slight positive reaction. Magnification ⫻250.
FIGURE 5 Immunohistochemical staining of GABA-t activity in thymus of an elderly 82-year-old man. The walls of an arteriola (A) a venula (V), and a lymphatic vessel (L) show a positive reaction. A small number of reticular cells (R) show a mild positive reaction and many structures resembling nerve fibers (arrows) show a positive reaction. Magnification ⫻500.
venules, and lymphatic vessels show a positive reaction. A small number of reticular cells and many structures resembling nerve fibers (recognized as aforementioned) show a mild positive reaction. Figure 4 shows a portion of Fig. 3 at high magnification (⫻500). The walls of two arterioles, a venula and a lymphatic vessel show a positive reaction. Only a small number of reticular cells show a poor reaction. Many structures resembling nerve fibers and recognizable by Bodian staining show a mild positive reaction. Due to involution of the thymus, fibrofatty tissue entirely surrounds small islands of recognizable thymic tissue.
Figure 5 shows immunohistochemical staining for GABA-t activity in the thymic tissue residues of an elderly 82-year-old man. At this age, the thymus becomes a fibrous pad containing only adipose cells with a few small residual islands of poorly recognizable thymic tissue. In Fig. 5 we can observe an arteriola, a venula and a lymphatic vessel showing positive staining of the walls. A small number of reticular cells show a mild positive reaction, while many structures resembling nerve fibers and recognized by Bodian staining show a positive reaction. All of our morphological results show that GABA-t activity, an enzyme that metabolizes GABA (an inhibitory neurotransmitter of numerous nervous and nonnervous peripheral structures) is present in the thymus. Table 1 deals with our biochemical findings regarding the amounts of GABA-t activity contained in the different structures of the human thymus in 12 subjects 8 to 82 years old. The highest mean amount of GABA-t
FIGURE 4 Immunohistochemical staining of GABA-t activity in thymus of a 50-year-old man showing a portion of Fig. 3 at higher magnification. Two arterioles (A), a venula (V), and a lymphatic vessel (L) show a positive reaction of their walls. Only a small number of reticular cells (R) show a poor reaction. Many structures resembling nerve fibers (arrows) also show a positive reaction. Magnification ⫻500.
TABLE 1 Biochemical values of GABA-t activity in the human thymus Tissue
GABA-ta (nM/mg protein/h)
Whole thymusb Arteriesc Veinsd
28.3 ⫾ 1.59 63.1 ⫾ 3.16 42.1 ⫾ 1.68
Mean value ⫾ SD of three independent determinations in 12 subjects. Fragment of thymus harvested from autopsies. c Large arterial vessels that richly supply the thymus. d Extraparenchymal venous vessels coming from thymus. GABA-t, ␥-aminobutyric acid-transaminase a
b
1076
D. Cavallotti et al.
TABLE 2 Amount of GABA-t (biochemical dosage) in whole thymus of young and elderly men
GABA-ta (nM/mg protein/h)
Young (n ⫽ 3)
Elderly (n ⫽ 3)
32.4 ⫾ 1.26
47.3 ⫾ 1.18b
Mean value ⫾ SD of three independent determinations in 3 subjects in each age group. b p ⬍ 0.001. GABA-t, ␥-aminobutyric acid-transaminase a
activity (63.1 ⫾ 3.16 nM/mg protein/h) is contained in the wall of the thymic arteries, followed by that in the wall of the thymic veins (42.1 ⫾ 1.68 nM/mg protein/h). Finally, the whole thymus contains only a small amount of GABA-t activity (28.3 ⫾ 1.59 nM/mg protein/h). Table 2 indicates the amounts of GABA-t activity contained in the whole thymus of 3 young and 3 elderly subjects. The young thymus contains 32.4 ⫾ 1.26 nM of GABA-t/mg protein/h, while the old thymus contains a higher amount of GABA-t activity (47.3 ⫾ 1.18). Statistical data analysis gives a high significance with p ⬍ 0.001. QAI after histoenzymatic staining reveals that in structures resembling nerve fibers and recognized after post-coloration with the Bodian method the occurrence of GABA-t is very high. Table 3 demonstrates, in fact, that GABA-t is contained in nerve fibers of the thymus. Its amount is 26.6 ⫾ 1.4 conventional units (CU) (see Methods). It varies in infant, young, adult and elderly subjects from 25.2 ⫾ 1.2 to 27.5 ⫾ 1.4 in young, 31.2 ⫾ 1.6 in adult, and 44.8 ⫾ 1.8 in elderly patients. These results are expressed in CU and all values ⫾ SEM were obtained using a Quantimet image analyzer. In this case, too, statistical analysis of data shows a high index of significance for adult and elderly subjects with p ⬍ TABLE 3 Occurrence of GABA-t in structures resembling nerve fibers (histoenzymatic staining and QAI) Age Group (years)
Number of Subjects
Staining for GABA-ta,b (mean CU ⫾ SEM)
p Value
8–82 Total 8–12 Infant 22–24 Young 50–62 Adult 77–82 Elderly
12 3 3 3 3
26.6 ⫾ 1.4 25.2 ⫾ 1.2 27.5 ⫾ 1.4 31.2 ⫾ 1.6 44.8 ⫾ 1.8
n. s. n. s. ⬍0.001 ⬍0.001
a The intensity of staining for GABA-t was measured in whole fragments of human thymus obtained at autopsy. b All values were obtained using a Quantimet (see Methods). GABA-t, ␥-aminobutyric acid-transaminase; QAI, quantitative analysis of images; n. s., not significant
TABLE 4 Occurrence of GABA-t in different structures of the thymus (histoenzymatic staining and QAI) Structures of Thymus
Staining of GABA-ta,b (mean CU ⫾ SEM)
Arterioles Venules Lymphatic vessels Structures resembling nerve fibers Reticular cells Lymphocytes Corpuscles of Hassal Septa Capsula Total thymic micro-environmentc Whole thymus
61.3 ⫾ 2.5 48.4 ⫾ 1.9 32.1 ⫾ 1.6 26.6 ⫾ 1.4 14.1 ⫾ 1.5 13.2 ⫾ 1.6 9.3 ⫾ 0.8 6.6 ⫾ 0.6 5.4 ⫾ 0.6 13.6 ⫾ 1.8 29.6 ⫾ 2.8
a Intensity of staining for GABA-t was measured in the thymus of 12 subjects 8 – 82 years old. On the basis of micro-anatomical details revealed by staining with eosin-orange we distinguished all indicated structures. b All values were obtained using Quantimet (see Methods). c Parenchyma of the thymus, which includes structures resembling nerve fibers, reticular cells, lymphocytes, and corpuscles of Hassal. GABA-t, ␥-aminobutyric acid-transaminase; QAI, quantitative analysis of images
0.001, while for infant and young subjects the values are not significant. Table 4 indicates all the structures contained in the thymic fragments susceptible to recognition by means of eosin-orange staining. For all of these structures, QAI performed after histoenzymatic staining for GABA-t is able to reveal the quantitative intensity of staining expressed in CU ⫾ SEM (see Methods). From a detailed examination of this table, it becomes apparent that the structures with the highest occurrence of GABA-t activity (61.3 ⫾ 2.5 CU) are the walls of the thymic arterioles, followed by the wall of the venules (48.4 ⫾ 1.9 CU). The lymphatic vessels also show strong levels of GABA-t activity (32.1 ⫾ 1.6 CU). Nerve fibers, recognized by specific Bodian staining, only contain 26.6 ⫾ 1.4 CU of GABA-t activity. The total thymic micro-environment, formed substantially by T lymphocytes or thymocytes, by reticular cells, and by corpuscles of Hassal, contains only 13.6 ⫾ 1.8 CU of GABA-t activity. The total thymic micro-environment forms the parenchyma of the thymus. Septa and capsula of the thymus have a low GABA-t content (only 6.6 – 5.4 ⫾ 0.6 CU). All values indicate the intensity of staining for GABA-t and were measured in thymic fragments of 12 subjects between 8 – 82 years of age. DISCUSSION Concerning the possible biological significance of GABA-t in the nerve fibers of the human thymus, the
GABA-t in Human Thymus
occurrence of ␥-aminobutyric acid (GABA) in the human tissues, including peripheral nerve fibers, was first reported three decades ago by Zachmann et al. [22]. Since then, convincing evidence has accumulated that not only GABA, but also enzymes involved in its biosynthesis and catabolism (that is glutamate decarboxylase and GABAtransaminase, respectively) are present in the peripheral tissues of different mammals [23]. Moreover, a high-affinity GABA uptake system has been identified in the mammalian periphery [24], including the endocrine pancreas [25]. However, despite of the large body of evidence concerning the existence of a GABAergic system in the many peripheral tissues, to our knowledge no information is yet available concerning the significance of GABA-t activity, as demonstrated by pharmacohistochemical methods [26]. In our previous studies it was pointed out that GABA-t activity is present in the cerebellar cortex and is stimulated following  estradiol treatment [17]. Moreover, in convoluted tubules of the cortex of the rat kidney, the presence of H3 muscimol, a specific marker of GABA-A receptors [27] was demonstrated. Thymic Microenvironment and its Innervation Comparing our experimental morphological and biochemical results with those reported by others [28, 29], confirms that the human thymus is a central lymphoid organ which plays a key role in cell-mediated immunity. It weighs about 15–20 gm at birth, increases to approximately 30 gm at puberty, and thereafter progressively atrophies to become a fibrofatty tissue of approximately 10 gm containing only small islands of recognizable thymic tissue in the adult. In elderly subjects the thymus becomes a fibrous pad containing only adipose cells with a few small residual islands of recognizable thymic tissue. Two lobes surrounded by a connectival capsule form the thymus. From this capsule originate the fibrous septa that divide each lobe into lobules. The lobules present a peripheral cortex and a central medulla. The cortex is composed of small closely packed lymphocytes, termed thymocytes, belonging to the T-cell systems. Scattered among these lymphocytes are larger, paler cells capable of active phagocytosis termed reticular cells. Due to the abundant cytoplasm and pale vesicular nuclei, the reticular cells are termed reticulo-epithelial or epithelial cells. On the contrary, the medulla is composed of numerous reticular cells with a few lymphocytes between them. In addition, the medulla contains some layered keratinizing epithelial cells known as corpuscles of Hassal. All of these structures form the thymic microenvironment [30, 31]. In particular, it should be noted that normal human thymus does not contain lymphoid
1077
follicles; their appearance is interpreted as thymic hyperplasia. Innervation of the thymic micro-environment arises from the parasympathetic nuclei localized in brain stem and in the spinal cord through the X cranial nerve [32]. Cholinergic innervation induces an increase of the immunoreactivity of thymic lymphocytes [33]. Cyclic nucleotides are also capable of increasing the activation and function of lymphocytes [34]. The thymus is innervated by AchE-positive fibers of the vagus, the recurrent laryngeal, and phrenic nerves. Catecholaminergic innervation derives from the stellate ganglia and other small ganglia of the thoracic sympathetic nervous system chain. Innervation of the thymic micro-environment also arises from the sympathetic system by means of trunks belonging to the cardiac plexuses surrounding the origin of the main vessels emerging from the heart wall [35]. The phrenic nerve also provides nerve fibers for the thymus. The thymic nerves run parallel to the vessels, forming periadventitial plexi and, with the vasculature, perforate the capsule of the thymus, spreading into the connective tissue and passing between lobules. Few thin nerve fibers penetrate the medulla when they end with varicosities. The thymic cortex does not present nerve fibers. Notwithstanding this, some specific coloration techniques appear to demonstrate that many types of peptidergic nerve fibers end their course in the thymic cortex [36]. Sympathetic innervation of the thymus is inhibitory to the immunoreactivity of T lymphocytes and to the proliferative response of the thymus [37– 40]. GABAergic innervation of the thymus (and in this experiment GABA-t is an enzyme linked to the metabolism of GABA) is also inhibitory to the immune response of the thymus [11], but further experiments are needed to clarify the role and the function of GABA-t activity in the human thymus. ACKNOWLEDGMENTS
The authors are greatly indebted to Drs. B. Nagar and J. Grasso for their suggestions and criticisms. The Medline, Internet and other informatic consulting services of Drs. M. Cameroni and F.M. Tranquilli Leali are gratefully acknowledged. The technical assistance of Mr. Dario Caporuscio, the photographic service of Mr. Giuseppe Leoncini, the excellent secretarial work of Ms. Silvana Casamento, and the kind help of Mrs. Sharon Hobby for revising the English version of the manuscript are also gratefully acknowledged.
REFERENCES 1. Crotti A: Thyroid and Thymus. Philadelphia, Lea & Febiger, 1918. 2. Hammer JA: Von Innervationsverhaltnisse der in Krelor-
1078
3.
4. 5. 6.
7.
8.
9. 10.
11.
12.
13.
14. 15.
16.
17.
18.
19.
gans der Thymus bis den 4 Fetal Monat. Zeitsch Mikrosk Anat Forsch 8:253, 1935. Fujiwara M, Muryobayasaky T, Shenamoto K: Histochemical demonstration of monoamines in the thymus of rat. Jap J Pharm 16:493, 1966. Fillenz M: The innervation of the cat spleen. Proc Roy Soc London 174:459, 1970. Sergeeva VE: Histotopography of catecholamines in the mammalian thymus. Bull Exp Biol Med 77:456, 1974. Reilly FD, McCuskey PA, Meineke HW: Studies of the hemopoietic microenvironment. VIII. Adrenergic and cholinergic innervation of the murine spleen. Anat Rec 185:109, 1976. Besedovsky HO, Ray A, Del Sorkin E, Prada MD, Keller HH: Immunoregulation mediated by the sympathetic nervous system. Cell Immunol 48:346, 1979. Williams JM, Peterson RJ, Shea PA, Schmedtje JF, Bauer DC, Felton DL: Sympathetic innervation of murine thymus and spleen. Evidence for a functional link between the nervous and immune systems. Brain Res Bull 6:83, 1981. Singh U: Sympathetic innervation of fetal mouse thymus. Eur J Immunol 14:757, 1984. Raymond F, Morel E, Bach JF: Evidence for the presence of immunoreactive acetylcholine receptors on human thymus cells. J Neuroimmunol 6:31, 1984. Cavallotti D, Artico M, D’Andrea V, Cavallotti C: Gabatransaminase activity in the human thymus during immunoresponse. Neuropeptides, in press. Cavallotti C, Artico M, De Santis S: Specific vascular localisation of Gabatransaminase in the thymus gland of juvenile and aged rats. Eur J Histochem, in press. Cavallotti D, Artico M, Cavallotti C, De Santis S, Tranquilli Leali FM: Interleukines and Gabatransaminase activity in rat thymus. Int J Immunopharmacol, in press. Townsend FM: Manual of Histologic and Special Staining Techniques. New York, McGraw-Hill, 1960. Bodian D: A new method for staining nerve fibers and nerve ending in mounted paraffin sections. Anat Rec 65:89, 1936. Van Gelder NM: The histochemical demonstration of ␥-aminobutyric acid metabolism by reduction of tetrazolium salts. J Neurochem 12:231, 1965. Cavallotti C, Jacopino L, Amenta F: Stimulatory effect of  estradiol treatment on Gaba degratative enzymes within rat cerebellar cortex. Neurosci Lett 39:205, 1983. Lowry OH, Rosebrough NJ, Farr AL, Randall RJ: Protein measurement with the Folin phenol reagent. J Biol Chem 193:265, 1951. Jung MJ, Lippert B, Metcalf BW, Schechter PJ, Bohlen P, Sjoerdsma A: The effect of 4-amino hex 5-ynoic acid (␥-acetylenic GABA y-ethynyl GABA) a catalytic inhibitor of GABA transaminase, on brain GABA metabolism in vivo. J Neurochem 28:717, 1977.
D. Cavallotti et al.
20. Serio A: Appunti Dalle Lezioni di Statistica Sanitaria. Rome, Kappa, 1986. 21. Castino M, Roletto E: Statistica Applicata. Padua, Piccin, 1992. 22. Zachmann M, Tocci P, Nyhan WL: The occurrence of gamma-aminobutyric acid in human tissues other than in brain. J Biol Chem 241:1355, 1966. 23. Wu JY, Lin H, Lin Y, Xu GW, Lin BH, Whang BH, Wei SP: Immunochemical characterization and visualization of glutamate decarboxylase and GABA-transaminase in peripheral tissues. In Erdo SL, Bowery NG (eds): GABAergic Mechanisms in the Mammalian Periphery. New York, Raven Press, 1986. 24. Amenta F: Autoradiographic localization of GABA receptor sites in peripheral tissues. In Erdo SL, Bowery NG (eds): GABAergic Mechanisms in the Mammalian Periphery. New York, Raven Press, 1986. 25. Gerber JC, Hare TA: Gamma-aminobutyric acid in peripheral tissue, with emphasis on the endocrine pancreas. Diabetes 28:1073, 1979. 26. Vincent SR, Kimura H, McGeer EG: The pharmacohistochemical demonstration of GABA-transaminase. Neurosci Lett 16:345, 1980. 27. Cavallotti C, Amenta F, Jacopino L, Erdo SL: Autoradiographic localization of the GABAA receptor against 3Hmuscimol within rat kidney. Pharmacology 36:390, 1988. 28. Simpson JG, Gray ES, Beck JS: Age involution in the normal human adult thymus. Clin Exp Immunol 19:261, 1975. 29. Singh J, Singh AK: Age related changes in human thymus. Clin Exp Immunol 37:507, 1979. 30. Haynes BF: The human thymic microenvironment. Adv Immunol 36:87, 1984. 31. Kendall MD: Functional anatomy of the thymic microenvironment. J Anat 117:1, 1991. 32. Bulloch K, Moore RY: Innervation of the thymus gland by brain stem and spinal cord in mouse and rat. Am J Anat 162:157, 1981. 33. Strom TB, Stykowski AJ, Carpenter CB, Merril JP: Cholinergic augmentation of lymphocyte mediated cytotoxicity. A study of the cholinergic receptors on cytotoxic lymphocytes. Proc Natl Acad Sci USA 71:1330, 1974. 34. Strom TB, Lundin AP, Carpenter CB: The role of cyclic nucleotides in lymphocyte activation and function. Proc Soc Clin Immunol 3:115, 1977. 35. Bulloch K, Pomerantz W: Autonomic nervous system innervation of thymic-related lymphoid tissue in wildtype and nude mice. J Comp Neurol 228:157, 1984. 36. Felten DL, Felten SY, Carlson SL, Olshowka JA, Livnat S: Noradrenergic and peptidergic innervation of lymphoid tissue. J Immunol 135(2 Suppl):755, 1985. 37. Singh U: Effect of catecholamines on lymphopoiesis in fetal mouse thymic explants. J Anat 129:279, 1979.
GABA-t in Human Thymus
38. Singh U: Effect of sympathectomy on the maturation of fetal thymocytes grown within the anterior eye chambers in mice. Proceedings of the VIII International Congress of Lymphatic tissue Germinal Centres, Cambridge. Adv Exp Med Biol 186:349, 1985.
1079
39. Singh U: Lymphopoiesis in the nude fetal thymus following sympathectomy. J Cell Immunol 93:222, 1958. 40. Singh U, Millson DS, Smith PA, Owen JJT: Identification of adrenoreceptors during thymocyte ontogeny in mice. Eur J Immunol 9:31, 1979.