Molecular and Cellular Biochemistry 2006.DOI: 10.1007/s11010-005-9059-5 c�Springer 2006

Characterization of thymocyte phenotypicalterations induced by long-lastingβ-adrenoceptor blockade in vivo and its effectson thymocyte proliferation and apoptosis

G. Leposavic,2,3 N. Arsenovic-Ranin,1 K. Radojevic,2

D. Kosec,2 V. Pesic,3 B. Vidic-Dankovic,2 B. Plecas-Solarovic3

and I. Pilipovic2

1Department of Immunology and Microbiology, Faculty of Pharmacy, Belgrade, Serbia and Montenegro; 2ImmunologyResearch Center “Branislav Jankovic”, Institute of Immunology and Virology “Torlak”, Belgrade, Serbia and Montenegro;3Department of Physiology, Faculty of Pharmacy, Belgrade, Serbia and Montenegro

Received 22 July 2005; accepted 18 October 2005

Abstract

Adult male Wistar rats were subjected to propranolol (P, 0.40 mg/100 g/day) or saline (S) administration (controls) over 14days. The expression of major differentiation molecules on thymocytes and Thy-1 (CD90) molecules, which are shown to adjustthymocyte sensitivity to TCRαβ signaling, was studied. In addition, the sensitivity of thymocytes to induction of apoptosis andconcanavalin A (Con A) signaling was estimated. The thymocytes from P-treated (PT) rats exhibited an increased sensitivityto induction of apoptosis, as well as to Con A stimulation. Furthermore, P treatment produced changes in the distributionof thymocyte subsets suggesting that more cells passed positive selection and further differentiated into mature CD4+ orCD8+ single positive (SP) TCRαβhigh cells. These changes may, at least partly, be related to the markedly increased density ofThy-1 surface expression on TCRαβ low thymocytes from these rats. The increased frequency of cells expressing the CD4+25+phenotype, which has been shown to be characteristic for regulatory cells in the thymus, may also indicate alterations in thymocyteselection following P treatment. Inasmuch as positive and negative selections play an important role in continuously reshapingthe T-cell repertoire and maintaining tolerance, the hereby presented study suggests that pharmacological manipulations withβ-AR signaling, or chemically evoked alterations in catecholamine release, may interfere with the regulation of thymocyteselection, and consequently with the immune response. (Mol Cell Biochem xxx: 1–13, 2005)

Key words: β-adrenoceptor blockade, CD90 expression, Con A, T-cell differentiation, thymocyte apoptosis, thymocyteproliferation

Introduction

In the thymus T lymphocytes develop from hematopoeticstem cells via phenotypically identifiable stages, includingsubsequently CD4−8− double negative (DN) pre-T cells,

Address for offprints: G. Leposavic, Department of Physiology, Faculty of Pharmacy, 450 Vojvode Stepe, 11521 Belgrade (Kumodraz), Serbia and Montenegro(E-mail: Gordana.Leposavic@pharmacy.bg.ac.yu)

CD4+CD8+ double positive (DP) cells, and finally ma-ture CD4+ or CD8+ single positive (SP) cells. This chainof events includes different selection steps. The first iscalled β-selection. It is driven by pre-T-cell receptor (TCR)i.e. CD3/β/invariant pre-TCRα complex and results in the

conversion of DN cells into DP cells. The cells that pass thisselection step undergo the selection events driven by the ma-ture TCRαβ [1–4]. The nonlymphoid thymic cells present anevolutionary optimized set of thymic peptides bound to MHCmolecules to the developing thymocytes expressing TCR. Ifthe resulting avidity of MHC:peptide-TCR reaction reaches athreshold level the cell receives a survival signal and, depend-ing on the MHC class recognized by TCR, downregulates theexpression of either the CD4 or CD8 molecule and furtherincreases the expression of the CD3/TCRαβ molecular com-plex, that is termed positive selection. The high avidity TCRligation induces negative selection and apoptotic cell death[4, 5], while very low avidity leads to apoptotic death dueto ”neglect” [4, 5]. The selection events are responsible forthe generation of the MHC-restricted self-tolerating T-cellrepertoire of SP mature T lymphocytes.

Under physiological circumstances, some T cells that ex-press self-reactive antigen receptors fail to be eliminated dur-ing development. These cells escape into the periphery form-ing a part of the normal T-cell repertoire. An accumulatingbody of evidence suggests that these T cells are under contin-uous control of a dominant T-cell mediated protective mech-anism. Namely, it is believed that the thymus, in addition tomediating positive and negative selection of the conventionalT cells, has an essential third function providing generationof regulatory T-cell subpopulations that play a critical rolein the maintenance of tolerance and in the prevention of au-toimmunity [6, 7].

It has been well documented that the sympathetic nervoussystem, a major component of the autonomous nervous sys-tem, innervates primary lymphoid organs (thymus and bonemarrow) [8–10]. Light microscopic studies have revealed thatpostganglionic sympathetic nerve fibers: (i) enter the thy-mus along with blood vessels, (ii) distribute to the capsularand septal system and (iii) further enter thymus parenchymaforming varicose plexuses in the subcapsular cortex and cor-ticomedullary junction. Extremely sparse sympathetic nervefibers have been found in the medulla, as well [12, 13]. Atthe ultrastructural level sympathetic varicosities are evidentin close proximity to both thymocytes and thymic nonlym-phoid cells [13–15], but classical synapses between the neu-ronal elements and the thymic cells have not been observed[13]. Therefore, it is reckoned that the sympathetic neuro-transmitters are released nonsynaptically i.e. from varicoseaxon terminals [13].

The expression of β-ARs has been revealed on the surfaceof both thymocytes [16, 17] and thymic nonlymphoid cells [9,18–20]. Accordingly, it has been suggested that noradrenaline(NA), the principal neurotransmitter released from sympa-thetic nerve terminals, influences T-cell maturation, not onlydirectly via β-ARs on the developing T cells, but also indi-rectly, by acting on the thymic nonlymphoid cells via β-ARs[20]. However, knowledge about the role of β-AR-mediated

signaling in the modulation of intrathymic T-cell develop-ment is still extremely limited. Since it has been shown thatthymocytes express a significantly lower number of β-ARson their surface in comparison with circulating peripheral Tcells [21, 22], it is assumed that the surface expression ofβ-ARs increases during T-cell maturation. Furthermore, ithas been speculated that stimulation of β-ARs may impedeT-cell development, and that lowering of β-AR expression indeveloping T cells is a beneficial phenomenon [13].

In vitro studies revealed that NA influences the expres-sion of Thy-1 antigens on fetal mouse thymic stem cells[16, 23]. After analyzing the expression of TCRαβ (by one-color flow cytometry) and CD4/CD8 coreceptor molecules(by two-color flow cytometry) in adult male Wistar rats sub-jected to a long-lasting P treatment, we suggested that β-ARblockade influences T-cell differentiation and consequentlyT-cell dependent functions [24].

The understanding the influence of β-AR-mediated inputson T-cell maturation, and consequently T-cell dependant re-sponse, is important for several reasons. Firstly, many con-ditions in healthy individuals (e.g., chronic exercise training,psychologically induced stress) are associated with increasedsympathetic tone and catecholamine release. Moreover, it hasbeen shown that exposure to stressors or agents that acti-vate sympathetic nervous system reduces T-cell responses,anti-viral immune reactivity, and NK cell activity, as wellas that these immunosuppressive effects may be preventedby pretreatment with β-AR blockers [25]. Secondly, alter-ations in immune cell number and function are observed inpatients with congestive heart failure and other cardiovas-cular clinical diseases linked to prolonged elevation in cate-cholamines [26]. Thirdly, β-AR agonists and antagonists, aswell as some drugs (e.g. cyclophosphamide) and chemicals(e.g. pyrethroids) causing catecholamine release, are in ex-tensive current use. Cyclophosphamide, a medicament usedto inhibit the growth of some tumors, has been shown to in-duce dose- and time-dependent elevations in NA concentra-tion in the thymus [27]. In light of these findings, it has beenspeculated that the greater susceptibility of subjects receiv-ing cyclophosphamide to opportunistic infections is induced,not only by the anti-neoplastic action of the drug, but also bythe drug-evoked changes in nervous system – immune sys-tem communications [27]. Pyrethroids, insecticides widelyutilized for agricultural and environmental applications [28],are shown to be strong inducers of adrenaline and NA release[29] and their effects on the thymus cellularity and distribu-tion of thymocyte subsets has been related to the induction ofcatecholamine release [30–32]. Finally, identification of cat-echolamines as potential modulators of T-cell maturation andunderstanding underlying mechanisms of their action may beof practical value since it has been demonstrated that: (i) inspite of a decline in thymic function with age, considerableT-cell output is maintained into late adulthood [33, 34] in

order to provide optimal diversity of the T-cell repertoire, onthe one hand, and to maintain tolerance to self-antigens, onthe other hand and (ii) adult thymus can contribute to immunereconstruction in cases of pathogenically- or therapeutically-induced T-cell deficiency [33].

Having all that in mind, we undertook the present study inorder to identify thymocyte developmental stages sensitive tolack of β-AR-mediated inputs and underlying mechanisms oftheir action. To this end effects of 14-day-long P treatment onthe relative and absolute numbers of subtle thymocyte sub-populations that may be delineated by three-colour flow cy-tometric analysis (FCA) of CD4/8/TCRαβ expression wereexamined. Furthermore, since it has been demonstrated that:(i) exogenous cAMP and NA can induce a decrease in steadystate Thy-1 mRNA levels in T-lineage cells and murine thy-mocytes through posttranscriptional destabilization of Thy-1mRNA [35], and (ii) this molecule is involved in the regula-tion of TCR signaling, and thereby TCR-dependent selectionprocesses [36], we hypothesized that β-AR blockade mayaffect thymocyte expression of Thy-1 molecules, and con-sequently thymocyte selection and distribution of thymocytesubsets. To test this hypothesis we analyzed the expression ofThy-1 on the surface of thymocyte subsets delineated by thelevel of TCRαβ surface expression. Moreover, since it hasbeen shown that intrathymic T-cell differentiation and matu-ration encompasses proliferation or apoptosis of thymocytesat distinct stages of development [4, 5], we assumed thatthymocyte sensitivity to induction of apoptosis and prolifer-ative signals is altered in rats subjected to β-AR blockade. Toprove this assumption in PT rats we analyzed spontaneous anddexamethasone-induced thymocyte apoptosis and thymocyteresponse to ConA stimulation. Finally, since substantial evi-dence indicates the importance of dysfunctional communica-tion between the sympathetic nervous system and the immunesystem in the pathogenesis of autoimmune diseases, such asrheumatoid arthritis and multiple sclerosis [37], the relativeproportion and absolute number of cells expressing the regu-latory (CD4+25+) phenotype in the thymus of rats subjectedto P treatment were studied to assess whether long-term β-AR blockade affects the development of regulatory T cells inthe thymus.

Methods

Animals

The experiments were performed on male Wistar rats, 75 daysold at the beginning of the experiment. The rats were pur-chased from the vivarium of the Medical Military Academy,Belgrade, and bred under conventional laboratory condi-tions. They were randomly assigned to P or S administra-tion (controls). Each group consisted of at least 7 animals.

Accordingly, over 14 consecutive days the animals receiveds.c. injections of P (0.40 mg/100 g/day) or the S vehicle(100 μl/100 g/day). The protocol was based on our previousfindings that P at this dose, administered to rats of the sameage and strain, for 14 days, induced alterations in distribu-tion of the main thymocyte subsets delineated by CD4/8 andTCRαβ expression, as determined by two- and one-colourFCA, respectively [24].

All animals were killed under ether anesthesia by exsan-guinations. Their thymuses were removed, dissected free ofparathymic lymph nodes and adherent membranous tissue,and weighed.

The all experiments were approved by the Animal Use andCare Advisory Committee of the Faculty of Pharmacy.

Chemicals, antibodies and immunoconjugates

Propranolol HCl, sodium azide, merocyanine (MC) 540and Concanavalin A (Con A) were purchased from Sigma-Aldrich Chemie GmbH, Taufkirchen, Germany. Dexametha-sone (Dexason, Dex) was obtained from Galenika A.D.,Zemun, Serbia and Montenegro.

RPMI 1640 powdered medium (Sigma-Aldrich ChemieGmbH) was dissolved in redistilled water according tothe manufacturer’s instructions. To prepare complete RPMImedium, 2 mM L-glutamine (Serva, Heidelberg, Germany),1 mM sodium pyruvate (Serva), 100 units/ml penicillin (ICN,Costa Mesa, CA, USA), 100 μg/ml streptomycin (ICN) and10% fetal calf serum (FCS) (Gibco, Grand Island, NY, USA)were added. FCS was inactivated by heating the serum at56 ◦C for 30 min.

For immunolabeling the following monoclonal antibod-ies (mAbs) were used as the first-step reagents: phyco-erythrin (PE)-conjugated anti-CD4 (clone W3/25, Serotec,Oxford, UK), fluorescein-isothiocyanate (FITC)-conjugatedanti-CD8 (clone MRC OX-8, Serotec), peridinin chlorophyllprotein (PerCP)-conjugated anti-TCRαβ (clone R73, BDBiosciences Pharmingen, Mountain View, CA, USA), biotin-conjugated anti-CD25 (clone MRC OX-39, Serotec), FITC-conjugated anti-CD90 (Thy-1.1) (clone HIS 51, BD Bio-sciences Pharmingen). Streptavidin-PerCP, purchased fromBD Biosciences Pharmingen, was applied as the second-stepreagent. The appropriate IgG isotype controls were obtainedfrom BD Biosciences Pharmingen.

Preparation of thymic cell suspensions

The thymic lobes were excised and placed in individual Petridishes containing ice-cold phosphate-buffered saline (PBS,pH 7.3). The thymocyte suspension was prepared by grind-ing the thymic tissue between the frosted ends of microscope

slides and passing the resultant suspension through a fine ny-lon mesh. The single-cell suspension obtained was washedthree times in ice-cold PBS containing 2% FCS and 0.01%sodium azide (PS medium). The cells were then counted inan improved Neubauer hemacytometer and cell density wasadjusted to 107 cells/ml by addition of PS medium. The via-bility of such cell preparations, as determined by Trypan blueexclusion, was routinely greater than 95%.

Flow cytometry analysis of surface antigen expressionon thymocytes

The immunolabeling was performed as previously described[38]. Twenty thousand cells per sample were analyzed us-ing a FACScan flow cytometer (Becton Dickinson, MountainView, CA, USA). Non-specific IgG isotype-matched controlswere used for each fluorochrome type to define backgroundstaining, while dead cells and debris were excluded fromanalysis by selective gating based on anterior and right-anglescatter. The percentage of positive cells for each labeling wasdetermined using the CellQuest software (Becton Dickin-son). In addition, the analysis of CD90 expression includedestimation of the mean channel number, which represents thedensity of expression of surface marker. The mean channelnumber was determined for the thymocytes from both P- andS-administered rats, and the percentage change in the meanfluorescence intensity (MFI) was calculated as it has beensuggested by Kamath et al. [39]:

Percentage change in MFI

= MFI for histogram of PT rats – MFI for histogram of S-injected rats

MFI for histogram of S-injected rats×100

The data from three experiments were pooled and depictedas mean percentage change in MFI ± standard error of themean (SEM).

Detection of apoptotic thymocytes

It has been demonstrated that thymocyte apoptosis inducedby chemicals in vivo is difficult to detect because of rapidclearance of apoptotic cells by phagocytes [39, 40]. To over-come this problem we used the previously reported strat-egy and cultured the thymocytes exposed in vivo to P or Sin vitro for an additional 18 h [39, 40]. Thymocytes (2 ×105 cells/well) were cultivated in 96-well flat-bottom plates(Nunc A/S, Roskilde, Denmark) in RPMI 1640 without orwith 100 nM Dex for 18 h, and the relative numbers of apop-totic cells were estimated using two methods of apoptoticcell staining. This dose of Dex was shown to be effective

in inducing apoptosis of thymocytes from male rats in vitro[41]. After the estimation of cell viability by Trypan bluedye exclusion, the apoptotic cells were labeled with MC 540and FITC-conjugated Annexin V/propidium iodide (PI) solu-tion (Apoptosis detection kit, BD Biosciences Pharmingen),respectively.

Detection of apoptotic thymocytes using lypophilicdye MC 540M540 is believed to detect a decreased packing order of phos-pholipids in the outer leaflet of the apoptotic cell plasmamembrane. The binding of MC 540 was measured accordingto the procedure of Mower et al. [42]. A 1 mg/ml stock solu-tion of MC540 in double-distilled H2O was filtered througha 0.22 nm filter, then stored in the dark at 4 ◦C no longerthan 1 month. Just before FACS analysis, 5 μl of stockMC540 was added to the 1 ml thymocyte suspension con-taining 106 to 107 cells. All samples were analyzed on aFACScan flow cytometer by CellQuest Software (BectonDickinson).

Detection of apoptotic thymocytes using Annexin Vand propidium iodideIn the same thymocyte suspensions thymocyte apoptosis wasalso examined using Annexin V and PI staining. AnnexinV binds to phosphatidyl serine, which translocates from theinner to the outer leaflet of the plasma membrane during apop-tosis [43]. Conjugation of Annexin V to detectable prostheticgroups, like fluorescein and biotin, thus allows the detectionof cells, which expose phosphatidyl serine. PI is a nonspe-cific DNA dye that is excluded from living cells with intactplasma membranes but incorporated into nonviable cells [44].The staining was performed according to the producer’s man-ual. Twenty thousand cells per sample were analyzed using aFACScan flow cytometer (Becton Dickinson) and CellQuestsoftware (Becton Dickinson).

Detection of BrdU+ cells

Bromodeoxyuridine (BrdU) incorporation was utilized toidentify DNA-synthesizing cells in vitro. A total of 2 × 105

cells/well (100 μl) of thymocytes from PT or S-administeredrats were dispersed into plastic 96-well plates (Nunc A/S)and cultured for 48 h at 37 ◦C in 5% CO2 humified airatmosphere without or with 2.5 μg/ml ConA in a totalvolume of 200 μl of RPMI 1640 culture medium. All cul-tures were run in triplicate. During the last 18 h of culturethe cells were pulsed by 1 μM BrdU/well. The BrdU/7-AAD Flow kit (BD Biosciences Pharmingen) was used todetect BrdU incorporation into cells. The 7-AAD labelingenabled detection of the apoptotic cells. To determine the

phenotypic characteristics of the BrdU+ cells, the stainingwith FITC-conjugated anti-BrdU Ab and 7-AAD was com-bined with immunolabeling with biotin-conjugated anti-CD3mAb (clone G4.18, BD Biosciences Pharmingen), as the firststep-reagent, and streptavidin-PE as the second-step reagent.The staining was performed according to the BrdU/7-AADFlow kit producer’s manual. All samples were analyzed ona FACScan flow cytometer by CellQuest software (BectonDickinson).

Statistical analysis

Data are expressed as X mean ± SEM. Student’s t-test wasused to assess the significance of differences between means.P < 0.05 was considered significant.

Results

Body weight, thymus weight and thymocyte yield

Fourteen-day long P treatment did not produce signifi-cant changes in the body weight of the adult Wistar rats(413.57 g ± 14.26 g in PT vs. 407.14 g ± 5.21 g in controls).Furthermore, this treatment affected significantly neither thethymus weight (0.63 g±0.03 g in PT rats vs. 0.62 g ± 0.05 gin controls) nor the thymocyte yield (114.20 × 107 ± 5.70 ×107 cells/thymus in PT rats vs. 116.00 × 107 ± 9.6 ×107 cells/thymus in controls).

Thymocyte apoptosis

MC 540 staining demonstrated that, in the absence of Dex,the relative number of apoptotic cells was significantly (p <

0.05) greater in the cultures of thymocytes from PT rats thanin those from S-administered rats. This increase reflected aconsiderable (p < 0.01) rise in the frequency of cells in earlyapoptosis, as the frequency of cells in late apoptosis/necrosisdid not significantly differ between the cultures of thymo-cytes from the two groups of rats (Fig. 1). According tothe intensity of MC540 fluorescence and forward scatter (asthe apoptotic cell shrinks and subsequently condenses), twosubsets of apoptotic cells can be distinguished: (i) cells inearly apoptosis (showing high binding of MC 540 and re-duced forward scatter) and (ii) cells in late apoptosis/necrosis(showing lower MC 540 binding and reduced forward scat-ter) [42]. It has been reported that apoptotic cells remainMC540 bright until their DNA content reaches a certain pointand then display decreasing levels of MC540 staining. Typ-ically, these cells also exhibit low forward and side light

Fig. 1. In panel A are shown representative dot-plots of the MC540 bindingto thymocytes isolated from propranolol-treated rats (lower dot-plots) andsaline-injected (upper dot-plots) and cultured for 18h without (−Dex) orwith dexamethasone (+Dex). The total number of apoptotic cells = cellsin region R1 (cells in early apoptosis) + cells in region R2 (cells in lateapoptosis/ necrosis). The total number of alive nonapoptotic cells = cells inregion R3. In the histograms are shown the relative proportions of (B) allapoptotic cells, and (C) cells in different phases of apoptosis in the cultures ofthymocytes from saline-injected and propranolol-treated rats, in the absence(−Dex) and presence of Dex (+Dex).The data are expressed as mean ± SEM(n = 7). ∗ p < 0.05; ∗∗p< 0.01. aSaline –Dex vs. Propranolol –Dex. bSaline+Dex vs. Propranolol +Dex. cSaline –Dex vs. Saline +Dex. dPropranolol–Dex vs. Propranolol +Dex.

scatter and, most likely, they are mainly late apoptotic cells[45].

Similar data were provided by Annexin V/PI staining(Fig. 2). Namely, when normal thymocytes are cultured invitro they undergo spontaneous apoptosis and appear ini-tially as Annexin V+/PI– cells. However, with increasing

Fig. 2. In panel A are shown representative dot-plots of double Annexin V/PIstained thymocytes isolated from (upper dot-plots) saline-injected and (lowerdot-plots) propranolol-treated rats and cultured for 18h without (−Dex) orwith dexamethasone (+Dex). The total number of apoptotic cells = cells inthe lower right quadrant (cells in early apoptosis) + cells in the upper rightquadrant (cells in late apoptosis/necrosis).The cells in the upper left quadrant= necrotic cells. In the histograms are shown the relative proportions of (B)all apoptotic cells and (C) cells in different phases of apoptosis in −Dex and+Dex cultures of thymocytes from saline-injected and propranolol-treatedrats. The data are expressed as mean ± SEM (n = 7). ∗ p < 0.05; ∗∗ p <

0.01. aSaline –Dex vs. Propranolol –Dex. bSaline +Dex vs. Propranolol+Dex. cSaline –Dex vs. Saline +Dex. dPropranolol –Dex vs. Propranolol+Dex.

culture duration, the same cells appear as Annexin V+/PI+cells [40]. Thus, single positive populations are consideredearly apoptotic (Annexin V+/PI−) or dead cells (AnnexinV−/PI+), whereas double positive (Annexin V+/PI+) cellsare thought to be in a late stage of apoptosis rather than innecrosis [40, 43, 46]. The data shown in Fig. 2 indicate a

significant (p < 0.01) increase in the percentage of AnnexinV+/PI-cell in Dex free cultures of thymocytes from PT ratswhen compared to the corresponding cultures of thymocytesfrom S-injected rats. The percentage of Annexin V+/PI+cells did not significantly differ between the cultures of thy-mocytes from the two groups of rats. Thus, in the absence ofDex the overall frequency of apoptotic i.e. Annexin V+/PI–and Annexin V+/PI+ cells in the thymocyte cultures fromrats treated with P was slightly, but significantly (p < 0.05),greater than that in the thymocyte cultures from controls.The relative number of Annexin V−/PI+ cells did not sig-nificantly differ between the cultures of thymocytes from thetwo groups of rats.

Having in mind that: (i) sensitivity to glucocorticoid-induced apoptosis is dependent on cAMP level [47], (ii)β-ARs are adenylate cyclase-linked receptors and (iii) NA is,most likely, produced by the thymocytes themselves [48, 49],we attempted to assess whether the long-lasting P treatmentaffects thymocyte sensitivity to Dex-induced apoptosisin vitro. In the cultures of thymocytes from both P- andS-administered rats Dex caused a significant (p < 0.01) risein the overall relative number of apoptotic cells, increasingthe frequency of thymocytes in both the early (p < 0.01) andlate phase of apoptosis /necrosis (p < 0.01), as determined byboth methods of staining (Figs. 1 and 2). However, the Dex-induced increase in the percentage of apoptotic cells was lesspronounced in the cultures of thymocytes from PT rats thanin the control cultures of thymocytes from S-administeredrats. Namely, the mean values for the former were 72%and 90% greater than in corresponding Dex free culturesas determined by MC 540 and Annexin V/PI staining,respectively, while for the latter the mean values were 103%and 116% greater than in corresponding Dex free culturesfor each staining method, respectively. The percentage ofcells in early apoptosis was significantly less (p < 0.01) inthe cultures of thymocytes from rats treated with β-blockerthan in the thymocyte cultures from control rats, while that ofcells in late apoptosis/necrosis was slightly, but significantly,greater, as determined by both methods of staining (Figs. 1and 2).

The relative numbers of Annexin V−/PI+ cells were sig-nificantly greater (p < 0.01) in the Dex free cultures fromboth PT and S-injected rats when compared to the corre-sponding Dex free cultures. However, the relative number ofthese cells was significantly increased in the Dex+ culturesof thymocytes from PT rats over that in the correspondingcontrol cultures (Fig. 2).

BrdU+thymocytes

The sensitivity of thymocytes from PT rats to Con A stim-ulation was examined in vitro using BrdU/7-AAD staining.

Fig. 3. In the histograms are shown (A) the relative proportions of BrdU+ alive cells in the 48 h cultures of thymocytes from propranolol-treated and saline-injected rats in the presence and absence of Con A and (B) the relative proportions of cells expressing different levels of CD3 (CD3−, CD3low and CD3high)within the population of BrdU+ thymocytes. The data are expressed as mean ± SEM (n = 7). ∗∗p< 0.01; ∗p < 0.05. aSaline +ConA vs. Propranolol + ConA.bSaline −ConA vs. Saline + ConA. cPropranolol −ConA vs. Propranolol + ConA.

It was shown that, in the absence of Con A, the percent-age of alive cells was significantly (p < 0.01) lower inthe thymocyte cultures from PT rats (71.17% ± 0.20%)than in the thymocyte cultures from control rats (75.30% ±0.11%). In these cultures the frequency of BrdU+ alivecells did not significantly differ between the two groups ofrats (Fig. 3).

In the presence of Con A, the difference in mean per-centage of alive cells between the cultures of thymo-cytes from P- (78.18% ± 0.47%) and S-administered rats(76.96% ± 0.27%) was not significant. However, the per-centage of BrdU+ alive cells increased considerably (p <

0.01) in the cultures of thymocytes from both PT and S-injected rats when compared with the corresponding ConA free cultures (Fig. 3). Furthermore, in the presence ofCon A, the percentage of BrdU+ alive cells was signifi-cantly (p < 0.01) greater in the thymocyte cultures fromPT rats than in the corresponding cultures from control rats(Fig. 3).

Analysis of the BrdU+ cells for the level of CD3 surfaceexpression in the presence of Con A, showed a significantlydifferent phenotypic profile of these cells in the thymocytecultures from PT rats from that in the corresponding culturesfrom the control rats. There was a significant (p < 0.01)increase in the relative proportion of CD3high within BrdU+subset followed by a proportional decrease (p < 0.01) ofCD3low cells in the former. The relative proportion of CD3−

cells within the BrdU+ subset did not differ significantlybetween the cultures of thymocytes from the two groups ofrats (Fig. 3).

Expression of TCRαβCD4/CD8 on thymocytes

For the analysis of TCRαβ expression mAbs, most likely, di-rected at a constant determinant of the rat αβ heterodimeric

TCR were used [50]. As it has been already shown [24,51] the thymocytes from the both groups of rats displayeda triphasic distribution of staining with these mAbs, thusthat subsets with: (i) a high level of TCRαβ expression(TCRαβhigh); (ii) a lower, but clearly measurable level of anti-gen expression (TCRαβ low) and (iii) the absence of measur-able binding of these mAbs (TCRαβ−) were distinguished.In respect to the expression of CD4 and CD8 molecules andthe level of TCRαβ expression twelve subsets of thymocyteswere delineated and their relative and absolute numbers werecalculated.

TCRαβ− thymocytesThe phenotype analysis of TCRαβ− cells for CD4 and CD8expression, showed that both the relative and absolute num-bers of CD4+8− cells SP were significantly (p < 0.01) in-creased, while those of CD4−8+ SP cells were significantly(p < 0.05) reduced in PT rats compared to correspondingvalues in the controls. Neither the relative numbers nor abso-lute numbers of CD4−8− DN and CD4+8+ DP cells wereaffected by P treatment (Fig. 4).

TCRαβ low thymocytesThe analysis of TCRαβ low thymocyte subset showed that thepercentage of CD4+8− SP cells was significantly (p < 0.05)greater in PT rats than in controls. By contrast, the percent-ages of CD4−8+ SP and CD4+CD8+ DP cells were sig-nificantly (p < 0.01) decreased in PT rats when comparedwith the controls. The relative number of CD4-8- cells wasnot affected by the treatment (Fig. 3). The absolute numberof CD4+CD8+ DP cells was significantly (p < 0.05) dimin-ished, while those of CD4+8− SP (p < 0.05) and CD4−8−DN (p < 0.01) cells were significantly increased in PT ratscompared with the controls (Fig. 4).

Fig. 4. Relative proportions (A) and absolute numbers (B) of the (a)TCRαβ− and (b) TCRαβ low and (c) TCRαβhigh thymocyte subsets delin-eated by expression of CD4/8 from propranolol-treated and saline-injectedadult male rats. The data are expressed as mean ± SEM (n = 7). ∗∗p< 0.01;∗p < 0.05.

TCRαβhigh thymocytesThe blockade of β-ARs significantly (p < 0.01) increasedthe frequency of cells in all TCRαβhigh thymocyte subsets,except for the CD4−8− DN subset, where the percentageremained unaltered (Fig. 4). The absolute numbers of cellswithin these subsets showed the same pattern of changes inPT rats (Fig. 4).

Expression of CD4/CD25 on thymocytes

CD25 has been shown to be a useful marker for regulatoryCD4+T cells present in the rat thymus, which are likely tobe the precursors of peripheral regulatory T cells engaged inthe maintenance of self-tolerance [52]. These cells seem tobe members of a unique lineage of T cells that are selectedduring the process of T-cell differentiation in the thymus [7].As it has been suggested that the process of T-cell differenti-ation is affected by P treatment [24], the influence of β-ARblockade on the generation of CD4+CD25+ regulatory Tcells in thymus was also analyzed. In the rats exposed to Ptreatment the frequency and absolute number of thymocytesexpressing this regulatory phenotype were significantly in-creased (Fig. 5).

Expression of CD90 on thymocytes

Long-lasting P treatment did not significantly affect the per-centage of CD90+ thymocytes (98.78 ± 0.13% in PT rats vs.98.23 ± 0.15 in S-injected rats). In addition, the comparisonswere also made in MFI for CD90+. MFI indicates the den-sity of expression of surface marker. As previously reported[39], a 100% increase in MFI represents a twofold increasein density of the cell surface marker. The long-lasting β-ARblockade produced a marked enhancement in MFI suggest-ing an increase in the density of CD90 expression (Fig. 5).Having in mind that this molecule is involved in the reg-ulation of TCR signaling, and thereby TCR-dependent se-lection processes [36], on the one hand, and that TCRαβ low

thymocytes are supposed to be mainly the cells undergoingselection processes [1, 5], on the other hand, we analyzed theexpression of CD90 by TCRαβ low cell subset. The resultsdemonstrated that MFI for CD90, and therefore the densityof CD90 expression, was also markedly increased on the sur-face of TCRαβ low thymocytes (Fig. 6).

Discussion

This study has confirmed our previous observation thatlong-lasting β-AR blockade in adult Wistar rats, affects

Fig. 5. The relative proportions (A) and the absolute numbers (B) of CD4+/25+ cells in the thymus of propranolol-treated and saline-administred adult malerats. The data are expressed as mean ± SEM (n = 7). ∗ p < 0.025, ∗∗p < 0.01.

intrathymic T-cell maturation without influencing thymuscellularity [24]. The finding was extended by demonstrating:(i) the thymocyte differentiation steps that are particularlysensitive to P action, and (ii) the changes in surface Thy-1 ex-pression that may indicate possible underlying mechanisms.Furthermore, as an increase in the percentage and absolutenumber of CD4+25+ cells was found in the thymus of PTrats, the study may also suggest that β-AR blockade influ-ences the maturation of not only conventional, but also CD4+T regulatory cells.

This is supported by data that long-lasting treatment ofaged mice by nadolol, a non-selective β-AR blocker, doesnot affect the total thymus cellularity, while it significantlymodifies thymocyte differentiation, as shown by changes inthe relative distribution of thymocyte subsets delineated byexpression of CD4/8 molecules [53]. Namely, even if theoverall number of thymocytes is unaltered, there might bespecific points in the differentiative pathway where progres-sion to more mature stages is enhanced. They might be iden-tified by depletion of cells at one point, with a proportionalaccumulation of cells in the downstream developmental step.

However, 16-day-long P treatment in inbred adult DA ratswas found to diminish both thymus cellularity and its weight[38]. This discrepancy may be reconciled by data indicat-ing genetic differences in the expression of cytochrome P450(CYP) isoforms involved in drug metabolism between DAand Wistar rats [54]. Namely, it has been demonstrated thatDA rats exhibit quite low mRNA levels of CYP2D2 (majorisoform responsible for P metabolism), and that liver micro-somes from DA rats have low activities toward the biotrans-formation of some typical CYP2D6 substrates, compared toWistar rats [54]. Accordingly, substantial differences in Pmetabolism between DA and Wistar rats have been observed.It has been reported that P exhibits 35-fold higher exposuresin DA rats than in Wistar rats [54]. Thus, differential effect

of P treatment in these two strains of rats may be related todifferences in the effective dose of drug.

Our investigation of the sensitivity of thymocytes fromPT to induction of apoptosis in culture revealed a small,but significant, increase in the frequency of cells undergo-ing apoptosis spontaneously. This finding is consistent withprevious data showing that 6-hydroxydopamine, an agent thatcauses degeneration of noradrenergic nerves, induces mousethymocytes to undergo apoptosis both in vivo and in vitro.Moreover, it has been shown that this effect may be inhibitedby catecholamine uptake blocker [55]. The data obtained inthe culture of thymocytes from adult DA rats subjected to Ptreatment provide further support [38]. The relatively smallmagnitude of increase in the percentage of apoptotic cells inthe cultures of thymocytes from PT rats compared with thatfrom S-injected animals may be related to the fact that thefrequency of apoptotic cells was estimated within an unfrac-tionated thymocyte population. Namely, the changes withinselected thymocyte subpopulation/s in culture may be min-imized or even completely counteracted by changes in theopposite direction within other thymocyte subpopulation/s,and thus they may be underestimated or even completelyoverlooked when an unfractionated population is analyzed.In this context it should be pointed out that this study in-dicates that thymocytes from PT rats exhibited decreasedsusceptibility to Dex-induced apoptosis in vitro comparedwith control cells. This is consistent with data showing thatglucocorticoid-dependent apoptosis in mouse thymocytes in-creases with the cAMP level elevation that may be inducedby β-AR stimulation [47]. On the other hand, it has beenshown that: (i) P prevents the exogenous cAMP- or NA-induced decrease in Thy-1 mRNA levels in T-cell lines [35]and (ii) mAbs recognizing determinants in a defined Thy-1 structural domain induce marked apoptosis of immaturethymocytes [56]. Accordingly, as P treatment increased the

Fig. 6. Representative histograms of the expression of CD90 on (A) allthymocytes and (B) TCRαβ low thymocytes from adult male rats treatedwith propranolol (thick solid black line) and injected with saline (dottedgray line) over 14 days. The thin black line represents background stainingby the appropriate PE labeled isotype control. In the histogram (C) are shownthe relative changes in MIF for CD90 expression on all thymocytes, and onTCRαβ low thymocytes in propranolol-treated rats relative to saline-injectedcontrols (n = 21).

surface density of CD90 expression, an increased sensitivityof some thymocyte subsets to induction of apoptosis may beexpected. Collectively, all these data suggest that, dependingon the triggering signal, P may modulate the susceptibility ofdistinct thymocytes to induction of apoptosis in the oppositedirection.

In the presence of Con A, the frequency of BrdU+ cellswas also slightly, but statistically significantly, increased incultures of thymocytes from PT rats suggesting an increasedsensitivity of these cells to Con A signaling. Moreover, theanalysis of phenotypic characteristics of BrdU+ cells indi-cated that this increase reflected a rise in the CD3high cells.In the same line are data that both adrenaline and NA inhibit

ConA-induced proliferation of murine thymus cells in vitro[57]. In addition, it has been demonstrated that thymocytesfrom rats perinatally exposed to pyrethroids, insecticides thatare strong inducers of adrenaline and NA release [29, 58], ex-hibit an impaired ability to proliferate in response to differentdoses of Con A [29].

Our results on the effects of the long-lasting β-AR block-ade on the susceptibility of the overall thymocyte populationto induction of apoptosis and sensitivity to Con A in vitro, areconsistent with data showing that chemical sympathectomyin adult CSE rats, affecting distinct thymic subsets, increasesthe percentage of both apoptotic and proliferating cells in situin the thymus [59]. All these findings taken together suggestthat in the present study P treatment led to the establishmentof a new balance between thymocyte apoptosis and prolifer-ation, so that organ cellularity remained unchanged.

The blockade of the β-AR in Wistar rats also produced analteration in the relative distribution of the thymocyte sub-sets delineated by the expression of CD4/8/TCRαβ. Alongthe developmental route from CD4−8−TCRαβ− triple neg-ative to CD4+CD8+DP TCRαβ low thymocytes, small im-mature SP (ISP) TCRαβ−/low subpopulations could be dis-tinguished in both mice and rats [51, 60]. The subtle changesdemonstrated in relative and absolute numbers of ISP thymo-cytes in the experimental group of rats may suggest that earlyphases of thymocyte development are sensitive to P treatment.Furthermore, in PT rats both the relative and absolute num-bers of CD4+CD8+DP TCRαβ low cells were reduced. Thecells expressing this phenotype are believed to enter TCRαβ-dependent selection processes [1, 4, 5]. In the light of the in-creased relative proportions of CD4+CD8+ DP TCRαβhigh

cells, which are supposed to be cells that have just passed pos-itive selection [1, 2], and their SP descendents (CD4+CD8−TCRαβhigh and CD4−CD8+TCRαβhigh cells), this findingmay suggest an enhanced positive selection and further matu-ration of the positively selected cells. The P-induced changesin the phenotype of BrdU+ cells, which were characterizedby an increase in the relative proportion of CD3high cells fol-lowed by a proportional decrease in the CD3low cells in thecultures with Con A, seem to favor this option. Namely, Pon-gracz et al. [61], showed that Con A has differential effects onmature and immature T cells, causing activation and prolifer-ation of the former cells but positive selection and thereforesurvival and further maturation of immature DP cells. Thehereby demonstrated P-induced increase in the density ofCD90 surface expression on TCRαβ low thymocytes is alsoconsistent with the previous assumption. Namely, it has beenshown that a lack of Thy-1-mediated regulation of TCR-αβ signaling in developing T cells of Thy-1−/− rats pro-duces hyperresponsiveness to TCRαβ, triggering exagger-ated negative selection and consequently markedly reducedde novo production of CD4+8− and CD4−8+ SP thymo-cytes from their CD4+8+ DP precursors [62]. However, it

should be mentioned that the increase in percentage of SPTCRαβhigh cells may also be related to a decelerated egres-sion rate of mature cells into the periphery and/or an increasedrate of pre-migrational proliferation. The P-induced increasein the frequency of BrdU+ cells of mature CD3high phenotypein the cultures of thymocytes from PT rats in the presence ofCon A, is consistent with the latter option. Finally, the sizeof the most mature SP pool may be affected by an increasedentry of recirculating peripheral T cells [63]. However, as thefrequency of RT6.1+ mature T cells in the thymuses of PTrats did not differ significantly from that in thymuses of con-trols (data not shown), this last option does not seem likely.

The study also demonstrated that β-AR blockade in thethymus produced an increase in the relative and absolutenumber of cells showing the CD4+25+ phenotype. Stud-ies with the K14 transgenic mouse have indicated that thesecells acquire CD25 expression and suppressor function inthe thymic cortex during the process of positive selection oncortical epithelial cells. Some of these CD25+ cells then un-dergo negative selection on bone-marrow-derived cells in themedulla and die by apoptosis [64]. Therefore, the increasedpercentage of CD4+25+ cells in the thymus of PT rats mayalso indicate modified regulation of the selection events. Theincreased production of these cells further supports the hy-pothesis that stress- or drug-induced altered communicationbetween the sympathetic nervous system and the immunesystem may be important in the pathogenesis of autoimmunediseases [37].

It should be pointed out that the P-induced effects on T-cellmaturation may reflect, not only direct P action on the devel-oping thymocytes, but also its action on the thymic epithelialcells [65, 66]. Namely, it has been reported that cateholaminesmay affect cytokine secretory capacity of the cultivated ratthymic epithelial cells that show a preferential expressionof subcapsular/perivascular and medullary markers [65, 66].Only a small subset (5–10%) of these cells was shown to bepositive for cortical markers [65, 66].

Furthermore, as P crosses the blood-brain barrier and ac-cumulates in the central nervous system, an influence onT-cell development via central neuroendocrine regulatorymechanisms may not be excluded [67]. Namely, it has beendemonstrated that P, affecting the release of the hypothala-mic releasing/inhibiting hormones, may influence secretionof several pituitary hormones including adrenocorticotropin,thyroid-stimulating hormone, growth hormone and prolactin[68–70]. An accumulating body of evidence suggests thatthese hormones can directly and/or indirectly, via inducingrelease of the hormones from the target glands, affect T-cell maturation acting on the developing T cells and/or onthe nonlymphoid thymic cells supporting T-cell development[71, 72].

In conclusion, the results clearly showed that in adult ratsin vivo administration of P: (i) increases the density of Thy-1

expression on TCRαβ low cells and (ii) produces substantialchanges in the distribution of thymocyte subsets that suggestan enhanced thymocyte positive selection and further matu-ration into SP cells. Furthermore, the study indicates that β-AR blockade enhances maturation of the CD4+ T regulatorycells. Collectively, the results support the hypothesis that cate-cholamines, via β-AR, deliver signals that, under physiologi-cal conditions, provide fine-tuning of TCR-dependent signal-ing, and thus regulate the selection thresholds. Inasmuch aspositive and negative selection plays an important role in con-tinuous reshaping of the T cell repertoire, and maintenanceof tolerance, the current study suggests that pharmacologicalmanipulations with β-AR signaling, as well as chemically-induced alterations in catecholamine release may interferewith these regulatory mechanisms, thereby affecting the pe-ripheral T-cell compartment and immune response.

Acknowledgments

This work was supported by research grant No 1239 from theMNZZS of the Republic of Serbia. The authors are gratefulto Ms. Milica Perisic for excellent technical assistance inpreparing the manuscript.

References

1. Jameson SC, Hogquist KA, Bevan MJ: Positive selection of thymocytes.Annu Rev Immunol 13: 93–126, 1995

2. Guidos CJ: Positive selection of CD4+ and CD8+ cells. Curr OpinImmunol 8: 225–232, 1996

3. Ellmeier W, Sawada S, Littman DR: The regulation of CD4 and CD8coreceptor gene expression during T cell development. Annu Rev Im-munol 17: 523–554, 1999

4. Cantrell DA: Transgenic analysis of thymocyte signal transduction.Nature Rev Immunol 2: 20–27, 2002

5. Chan SH, Cosgrove D, Waltzinger C, Benoist C, Mathis D: Anotherview of the selective model of thymocyte selection. Cell 73: 225–236,1993

6. Seddon B, Mason D: The third function of the thymus. Immunol Today21: 95–99, 2000

7. Shevach EM: CD4+CD25+ suppressor T cells: more questions thananswers. Nature Rev Immunol 2: 389–400, 2002

8. Williams JM, Felten DL: Sympathetic innervation of murine thymus andspleen: a comparative histofluorescence and biochemical study. AnatRes 199: 531–542, 1981

9. Madden KS, Felten DL: Experimental basis for neural-immune inter-actions. Physiol Rev 75: 77–106, 1995

10. Friedman EM, Irwin MR: Modulation of immune cell function by au-tonomic nervous system. Pharmacol Therapeut 74: 27–38, 1997

11. Nance MD, Hopkins HD, Bieger D: Re-investigation of the innervationof the thymus gland in mice and rats. Brain Behav Immun 1: 134–147,1987.

12. Leposavic G, Micic M, Ugresic N, Bogojevic M, Isakovic K: Com-ponents of sympathetic innervation of the rat thymus during late fetaland postnatal development: histofluorescence and biochemical study.Thymus 19: 77–87, 1992

13. Elenkov IJ, Wilder RL, Chrousos GP, Vizi ES: The sympathetic nerve –an integrative interface between two supersystems: the brain and im-mune system. Pharmacol Rev 52: 595–638, 2000

14. Novotny GEK, Sommerfeld H, Zirbes T: Thymic innervation of rat: alight and electron microscopical study. Comp Neurology 302: 552–556,1990

15. Vizi ES, Orso E, Osipenko ON, Hasko G, Elenkov IJ: Neurochemical,electrophysiological and immunocytochemical evidence for a noradren-ergic link between the sympathetic nervous system and thymocytes.Neurosci 68: 1263–1276, 1995

16. Singh U: Effects of catecholamines in lymphopoesis in fetal mousethymic explants. J Anat 129: 279–285, 1979

17. Marchetti B, Morale MC, Paradis P, Bouvier M: Characterization, ex-pression, and hormonal control of a thymic beta(2)-adrenergic receptor.Am J Physiol 267: E18–31, 1994

18. Borne HR, Lichtenstein TM, Mean KL: Pharmacological control ofallergic histamine release in vitro: evidence for an inhibitory role of3′,5′-adenine monophosphate in human leukocytes. J Immunol 142:695–705, 1989

19. Kurz B, Feindt J, von Gaudecker B, Kranz A, Loppnow H, Mentlein R:β-Adrenoceptor-mediated effects in rat cultured thymic epithelial cells.B J Pharmacol 120: 1401–1408, 1997

20. Sanders VM, Kasprowicz JD, Swanson-Mungerson MA, Podojil JR,Kohm PA: Adaptive immunity in mice lacking the β2-adrenergic recep-tors. Brain Behav Immun 17: 55–67, 2003

21. Pochet R, Delespresse G: Beta-adrenoceptors display different effi-ciency on lymphocyte subpopulations. Biochem Pharmacol 32: 1651–1655, 1983

22. van de Griend RJ, Astraldi A, Wijermans P, van Doorn R, Ross D: Lowbeta-adrenergic receptor concentration on human thymocytes. Clin ExpImmunol 51: 55–63, 1983

23. Singh U, Owen JJ: Studies on the maturation of thymus stem cell. Theeffects of catecholamines, histamine and peptide hormones on the ex-pression of T cell alloantigens. Eur J Immunol 6: 59–62, 1976

24. Leposavic G, Plecas-Solarovic B, Kosec D: Differential effects ofchronic propranolol treatment on the phenotypic profile of thymocytesfrom immature and adult rats. Immunopharmacol 46: 79–87, 2000

25. Madden KS: Catecholamines, sympathetic innervation, and immunity.Brain Behav Immun 17: S5–S10, 2003

26. Sanders VM, Straub RH: Norepinephrine, the β-adrenergic receptor,and immunity. Brain Behav Immun 16: 290–333, 2002

27. Karp JD, Szczytkawski JL, Gentile CF: Noradrenergic responses ofperipheral organs to cyclophosphamide in mice. Life Sci 75: 2077–2089, 2004

28. Aldridge WN: An assessment of the toxicological properties ofpyrethroids and their neurotoxicity. CRC Crit Rev Toxicol 21: 89–104,1990

29. Ray DE, Cramer JE: The action of decamethrin (a synthetic pyrethroid)on the rat. Pestic Bioch Physiol 10: 33–340, 1979

30. Desi I, Varga L, Dobrony I, Saklenerik GY: Immunotoxicological inves-tigations of the effect of pesticide cypermethrin. Arch Toxicol 8 (Suppl.):300–304, 1985

31. Madsen C, Claesson MH, Ropke C: Immunotoxicity of the pyrethroidinsecticides deltametrin and α-cypermetrin. Toxicology 107: 219–227,1996

32. Santoni G, Cantalamessa F, Cavagna R, Romagnoli S, Spreghini E,Piccoli M: Cypermethrin-induced alteration of thymocyte distribu-tion and functions in perinatally-exposed rats. Toxicology 125: 67–68,1998

33. Douek DC, McFarland RD, Keiser PH, Gage EA, Massey JM, HaynesBF, Polis MA, Haase AT, Feinberg MB: Changes in thymic function withage and during the treatment of HIV infection. Nature 396: 690–695,1998

34. McFarland RD, Douek DC, Koup RA, Picker LJ: Identification of ahuman recent thymic emigrant phenotype. Proc Natl Acad Sci USA 97:4215–4220, 2000

35. Wajeman-Chao SA, Lancaster SA, Graf LH Jr, Chambers D: Mechanismof catecholamine-mediated destabilization of messenger RNA encodingThy-1 protein in T-lineage cells. J Immunol 161: 4825–4833, 1998

36. Killeen N: Thy-1-hiding in full view. Curr Biol 7: R774–R777, 199737. Harbuz, M: Neuroendocrinology of autoimmunity. Inter Rev Neurobiol

52: 133–161, 200238. Rauski A, Kosec D, Vidic-Dankovic B, Radojevic K, Plecas-Solarovic

B: Thymopoiesis following chronic blockade of beta-adrenoceptors.Immunopharm Immunotoxicol 25: 515–528, 2003

39. Kamath AB, Nagarkatti PS, Nagarkatti M: Characterization of phe-notypic alterations induced by 2,3,7,8-tetrachlorodibenzo-p-dioxin onthymocytes in vivo and its effect on apoptosis. Toxicol Appl Pharmacol150: 117–124, 1998

40. Okasha SA, Ryu S, Do Y, McKallip RJ, Nagarkatti M, Nagarkatti PS:Evidence for estradiol-induced apoptosis and dysregulated T cell mat-uration in the thymus. Toxicology 163: 49–62, 2001

41. Brown DG, Sun X-M, Cohen G: Dexamethasone-induced apoptosisinvolves cleavage of DNA to large fragments prior to internucleosomalfragmentation. J Biol Chem 268: 3037–3039, 1993

42. Mower DA Jr, Peckham DW, Illera VA, Fishbaugh JK, Stunz LL, Ash-man RF: Decreased membrane phospholipid packing and decreased sizeprecede DNA cleavage in mature mouse B cell apoptosis. J Immunol152: 4832–4842, 1994

43. Vermes I, Haanen C, Steffens-Nakken H, Reutelingsperger C: A novelassay for apoptosis. Flow cytometric detection of phosphatidylserineexpression on early apoptotic cells using fluorescein labelled AnnexinV. J Immunol Meth 184: 39–51, 1995

44. Eray M, Matto M, Kaartinen M, Andersson L, Pelkonen J: Flow cy-tometric analysis of apoptotic subpopulations with a combination ofannexin V-FITC, propidium iodide, and SYTO 17. Cytometry 43: 134–142, 2001

45. Reid S, Cross R, Snow ER: Combined Hoechst 33342 and merocya-nine 540 staining to examine murine B cell cycle stage, viability andapoptosis. J Immunol Method 192: 43–54, 1996

46. Lecoeur H, Prevost MC, Gougeon ML: Oncosis is associated withexposure of phosphatidylserine residues on the outside layer of theplasma membrane: a reconsideration of the specificity of the annexinV/propidium iodide assay. Cytometry 44, 65–72, 2001

47. McConkey DJ, Orrenius S, Okret S, Jondal M: Cyclic AMP potentiatesglucocorticoid-induced endogenous endonuclease activation in thymo-cytes. FASEB J 7: 580–585, 1993

48. Josefsson E, Bergquist R, Ekman R, Tarakowski A: Catecholamines aresynthesized by mouse lymphocytes and regulate function of these cellsby induction of apoptosis. Immunology 88: 140–146, 1996

49. Vidic-Dankovic B, Radojevic K, Leposavic G: Orchidectomy affectsthe monoamine content in rat thymocytes. J Neuroimmunol 154: 132,2004

50. Hunig T, Wallny HJ, Hartley JK, Lawetzky A, Tiefenthaler G: A mon-oclonal antibody to a constant determinant of the rat T cell antigenreceptor that induces T cell activation. J Exp Med 169: 73–86, 1989

51. Tsuchida M, Konishi M, Jojima K, Naito K, Fujikura Y, FukumotoT: Analysis of cell surface antigens on glucocorticoid-treated rat thy-mocytes with monoclonal antibodies. Immunol Lett 39: 209–217,1994

52. Stephens LA, Mason D: CD25 is a marker for CD4 thymocytes thatprevent autoimmune diabetes in rats, but peripheral T cells with thisfunction are found in both CD251 and CD252 subpopulations. J Im-munol1 65: 3105–3110, 2000

53. Madden KS, Felten DL: β-Adrenoceptor blockade alters thymocytedifferentiation in aged mice. Cell Mol Biol 47: 189–196, 2001

54. Komura H, Iwaki M: Pharmacokinetics and metabolism of metoprololand propranolol in the female DA and female Wistar rat: the female DArat is not always an animal model for poor metabolizers of CYP2D6. JPharmaceutical Sci 94: 397–408, 2005

55. TsaoC-W, Cheng J-T, Shen C-L, Lin Y-S: 6-Hydroxydopamine inducesthymocyte apoptosis in mice. J Neuroimmunol 65: 91–95, 1995

56. Hueber AO, Roposo G, Pierres M, He HT: Thy-1 triggers mouse thy-mocyte apoptosis through a bcl-2 resistant mechanism. J Exp Med 179:785–792, 1994

57. Cook-Milles MJ, Cohen RL, Perlan RL, Chambers DA: Inhibition oflymphocyte activation by catecholamines: evidence for non-classicalmechanisms of catecholamine action. Immunology 85: 544–549, 1995

58. Cremer JE, Seville MP: Comparative effects of two pyrethroidsdeltamethrin and cismethrin on plasma catecholamines and on bloodglucose and lactate. Toxicol Appl Pharmacol 66: 124–133, 1982

59. Al-Shawaf AA, Kendal DM, Cowen T: Identification of neural profilescontaining vasoactive intestinal polypeptide, acetylcholinesterase andcatecholamines in the rat thymus. J Anat 174: 131–143, 1991

60. Petrie HT, Hugo P, Scollay R, Shortman K: Lineage relationships anddevelopmental kinetics of immature thymocytes: CD3, CD4 and CD8acquisition in vivo and in vitro. J Exp Med 172: 1583–1588, 1990

61. Pongracz J, Parnell S, Anderson G, Jaffrezou J-P, Jenkinson E: Con Aactivates an Akt/PKB dependent survival mechanism to modulate TCRinduced cell death in double positive thymocytes. Mol Immunol 39:1013–1023, 2003

62. Hueber AO, Bernard AM, Battari CL, Marguet D, Massol P, Foa C,Brun, N, Garsia S, Steward C, Pierres M, He HT: Thymocytes in Thy-1-/- mice show augmented TCR signaling and impaired differentiation.Curr Biol 7: 705–708, 1997

63. Westermann J, Smith T, Peters U, Tschernig T, Pabst R, Steinhoff G,Sparshott SM, Bell EB: Both activated and non-activated leukocytesfrom the periphery continuously enter the thymic medulla of adult rats:

phenotypes, source and magnitude of traffic. Eur J Immunol 26: 1866–1874, 1996

64. Bensinger SJ, Bandeira A, Jordan MS, Caton AJ, Laufer TM: Majorhistocompatibility complex class II-positive cortical epithelium medi-ates the selection of CD4(+)25(+) immunoregulatory T cells. J ExpMed 194: 427–438, 2001

65. von Patay B, Loppnow H, Feindt J, Kurz B, Mentlein R: Catecholaminesand lipopolysaccharide synergistically induce the release of interleukin-6 from thymic epithelial cells. J Neuroimmunol 86: 182–189, 1998

66. von Patay B, Kurz B, Mentlein R: Effects of transmitters and co-transmitters of the sympathetic nervous system on interleukin-6 syn-thesis in thymic epithelial cells. Neuroimmunomodulation 6: 45–50,1999

67. Neil-Dwyer G, Bartlett J, McAinsh J, Cruickshank JM: β-Adrenoceptorblockers and the blood-brain barrier. Br J Clin Pharmacol 11: 549–553,1981

68. Oberbeck R, Schurmeyer T, Jacobs R, Benschop RJ, Sommer B,Schmidt RE, Schedlowski M: Effects of beta-adrenoceptor-blockadeon stress-induced adrenocorticotrophin release in humans. Neuroen-docrinology 69: 309–315, 1999

69. Cheolyoung P, Inmyung Y, Jeongtaek W, Sungwoon K, Jinwoo K,Youngseol K, Seungjoon P: Acute hyperglycemia and activation of thebeta-adrenergic system exhibit synergistic inhibitory actions on growthhormone (GH) releasing hormone-induced GH release. Eur J Endocrinol148: 635–640, 2003

70. Daftary SS, Boudaba C, Tasker JG: Noradrenergic regulation of parvo-cellular neurons in the rat hypothalamic paraventricular nucleus. Neu-rosci 96: 743–51, 2000

71. Blalock JE: A molecular basis for communication berween immune andneuroendocrine system. Physiol Rev 69: 1–32, 1989

72. Savino W, Dardenne M: Neuroendocrine control of thymus physiology.Endocrine Rev 21: 412–443, 2000