Skip Navigation

Cardiovascular Research 1999 41(1):246-254; doi:10.1016/S0008-6363(98)00231-4
© 1999 by European Society of Cardiology
This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow E-letters: Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when E-letters are posted
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrowRequest Permissions
Right arrow Disclaimer
Google Scholar
Right arrow Articles by Hart-Favaloro, J. L.
Right arrow Articles by Woodman, O. L.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Hart-Favaloro, J. L.
Right arrow Articles by Woodman, O. L.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

Copyright © 1999, European Society of Cardiology

Rabbit mononuclear leukocytes cause contraction of isolated aorta by the release of serotonin

Joanne L. Hart-Favaloro and Owen L. Woodman*

Department of Pharmacology, University of Melbourne, Parkville, Victoria 3052, Australia

* Corresponding author. Fax: +61-3-9347-1452; e-mail: o.woodman@pharmacology.unimelb.edu.au

Received 7 January 1998; accepted 30 June 1998


   Abstract
Top
 Abstract
1 Introduction
 2 Methods
 3 Results
 4 Discussion
 References
 
Objective: The aim of this study was to examine the vasoactive effects of rabbit isolated mononuclear leukocytes, and to identify the mediators responsible for those vasoactive effects. Methods: Mononuclear leukocytes (MNLs) were isolated from New Zealand White rabbit whole blood, suspended at 5x107 cells/ml and incubated for 30 min at 37°C. This cell suspension, or the cell-free supernatant from this suspension, were then examined for vasoactive effects in rabbit isolated thoracic aorta. Results: Both the MNL suspension and the cell-free supernatant from this suspension caused endothelium-independent contraction of aortic rings, both from resting tension and when pre-contracted. The MNL suspension caused a significantly greater contraction than the MNL supernatant under all conditions. The contractions to the MNL product were significantly inhibited by the 5-HT2 receptor antagonist ketanserin (0.1 µM), but not by the {alpha}1-adrenoceptor antagonist prazosin (10 µM). High-performance liquid chromatography (HPLC) analysis showed that the MNL supernatant contained serotonin (5-hydroxytryptamine, 5-HT) at an average concentration of 5 µM. Conclusions: We conclude that MNLs cause contraction of rabbit isolated aortic rings by the release of 5-HT.

KEYWORDS Leukocyte; Monocyte; Lymphocyte; Vasocontraction; Rabbit, aorta; Serotonin


   1 Introduction
Top
 Abstract
 1 Introduction
2 Methods
 3 Results
 4 Discussion
 References
 
Mononuclear leukocytes (MNLs) accumulate at sites of inflammation and are commonly found in atherosclerotic lesions. They infiltrate the subendothelial space and are proposed to contribute to vascular lesion development by releasing cytokines [1], which promote smooth muscle cell proliferation [2]. Cytokines also promote further recruitment of leukocytes [3], which may lead to endothelial cell damage [2]and the release of vasoactive substances, including arachidonic acid metabolites such as prostaglandin E2 (PGE2) and thromboxane A2 (TXA2) [4]and oxygen free radical species [5]. Furthermore, human mononuclear cells have recently been shown to contain both angiotensin I and angiotensin II [6]and human macrophages are reported to synthesize and secrete endothelin-1 and endothelin-3 [7]. Thus, MNL-derived products have the potential to cause significant effects on vascular tone, particularly in areas of inflammation.

There have been reports of vasoactive responses to MNLs, however, none of the studies identified the active factor. One study indicated that activated human MNLs caused increased vascular resistance in cynomolgus monkey femoral artery, by releasing a stable factor, which was <1000 Da in size, but was not further characterized [8]. Other studies indicate that MNLs can increase the potency of endothelin-1 [9], inhibit endothelium-dependent relaxation [10]and that migration of MNLs into vascular lesions may increase the sensitivity of the rabbit carotid artery to the contractile effects of serotonin (5-HT) [11]. Monocyte/macrophages, but not lymphocytes, are also implicated as a source of vasoactive mediators in cerebral ischaemia and stroke [12]but, as noted above, the identity of the vasoconstrictor(s) remains unknown.

The aims of this study were to characterize the vasoactive effects of MNLs or products released from MNLs on rabbit isolated aorta, and to identify the mediators involved. This report provides evidence for the release of 5-HT from MNLs. This finding may be of particular significance in atherosclerosis, as MNLs are known to accumulate, and are probably activated, in atherosclerotic vascular lesions [13].


   2 Methods
Top
 Abstract
 1 Introduction
 2 Methods
3 Results
 4 Discussion
 References
 
All experimental procedures were carried out within the guidelines of the National Health and Medical Research Council of Australia, and were approved by the University of Melbourne Animal Experimentation Ethics Committee.

2.1 General procedures
New Zealand White rabbits of either sex were anaesthetised by intravenous administration of Saffan® (alphaxalone, 9 mg/ml, and alphadolone acetate, 3 mg/ml, diluted 1:2 with sterile normal saline). The carotid artery was isolated and cannulated and blood was collected into syringes containing sodium citrate (final concentration, 3.8%, w/v). When all of the blood was removed (usually 100–120 ml) the thorax was quickly opened and the thoracic aorta was removed and stored in Krebs–bicarbonate solution [Krebs; composition (in mM): NaCl 118, KCl 4.7, KH2PO4 1.2, MgSO4·7H2O 1.2, (+)-glucose 5.0, NaHCO3 25.0, CaCl2·2H2O 2.5].

2.2 Isolation of rabbit mononuclear leukocytes
MNLs were isolated from whole blood under sterile conditions using cell-culture grade, endotoxin-tested reagents and solutions. Plasma was removed by centrifugation (15 min, 750xg) and then centrifuged again to obtain platelet-poor plasma (10 min, 1000xg). Most of the erythrocytes were removed from the remaining blood by sedimentation with Dextran T500 (5 ml; 6%, w/v). The leukocyte-rich top layer was removed and centrifuged (10 min, 750xg). The resulting pellet was resuspended in a solution of platelet-poor plasma (2 ml) and normal saline (8 ml), then underlayered with Lymphoprep® (3 ml) and centrifuged (20 min, 850xg) to isolate the mononuclear cells, which are concentrated in the middle layer. The MNLs were incubated in platelet-poor plasma (10 min, 37°C) then washed (10 min, 500xg) in Tyrode's solution [composition (in mM): N-2-hydroxyethylpiperazine-N''-2-ethanesulphonic acid (HEPES) 10, NaCl 137, NaHCO3 11.9, NaH2PO4 0.4, KCl 2.7, MgCl2 0.26, D-glucose 11 and 0.25% (w/v) bovine serum albumin (BSA); pH 7.4). The cells were resuspended at a final concentration of 5x107 cells/ml and incubated for 30 min at 37°C. The cell suspensions were then used in isolated aorta experiments, or the incubation medium was obtained as the supernatant from the cell suspension following centrifugation (5 min, 1000xg). This supernatant was used immediately, or stored for up to four months at –20°C.

2.3 Composition and viability of MNL suspensions
A sample (50 µl) of each of the isolated mononuclear or polymorphonuclear cells was stained with gentian violet and trypan blue to assess cell number and viability. Cell viability was always >95%. In addition, the number of platelets was also counted in some of the preparations. In some experiments, a further sample of the MNL suspension was applied to a microscope slide using a cytospin centrifuge. These slides were fixed with ethanol, then stained with Giemsa's stain and the MNL subtypes were counted.

2.4 Preparation of platelet supernatant
Platelets isolated from the plasma were washed twice in Tyrode's solution (5 min, 1000xg) and then resuspended at 5x107 platelets/ml in Tyrode's solution, the concentration found to be contaminating the MNL suspension. The platelet suspension was incubated at 37°C for 30 min. The supernatant was then obtained by centrifuging the suspension again (5 min, 1000xg) and discarding the platelets.

2.5 Isolated aorta preparation
The thoracic aorta was cleared of connective tissue and cut into rings that were approximately 4 mm in length. The rings were mounted in siliconised, 10 ml organ baths on a fixed hook and a second hook, which was connected to a force transducer (model # FT03, Grass Medical Instruments, USA) bathed in Krebs–bicarbonate solution maintained at 37°C with a water jacket and continuously bubbled with a mixture of 95% O2 and 5% CO2. The rings were allowed to equilibrate for at least 40 min, during which time the passive tension was adjusted to 2 g. Changes in tension were recorded on a chart recorder (model # R-02, Rikadenki Kogyo, Japan), which was calibrated immediately prior to each experiment. In some experiments, the endothelium was deliberately removed by rubbing the luminal surface gently with a wooden stick, to examine the role of the endothelium in the responses.

2.6 High-performance liquid chromatography (HPLC) analysis of amine concentrations
The concentration of noradrenaline, adrenaline and serotonin in samples of the MNL supernatant and the vehicle were measured by HPLC analysis. These samples were first treated with 0.4 M perchloric acid, to preserve the monoamines and precipitate any proteins that were present. Samples were then centrifuged to remove the protein and then were diluted with water prior to analysis. The HPLC analysis equipment consisted of a reversed-phase micro-guard cartridge (BioRad) followed by a clinical HPLC reversed-phase column (BioRad). Noradrenaline, adrenaline and serotonin in the eluent from the column were detected by an electrochemical cell (BAS LC4B amperometric detector) set at 650 mV. The mobile phase employed consisted of 70 mM KH2PO4, 8 µM sodium octane sulphonate, 0.5 mM Na2EDTA, with the pH adjusted to 3.0 with orthophosphoric acid. Methanol (10%, v/v) was added to the mobile phase for the detection of serotonin. Using this method, the detection thresholds for these analyses were determined to be 0.02 µM for noradrenaline, 0.04 µM for adrenaline and 0.03 µM for 5-HT.

2.7 Experimental protocols
Following a 45-min equilibration period, all aortic rings were assessed for endothelial integrity by recording the endothelium-dependent relaxation elicited by acetylcholine (ACh, 3 µM), after sub-maximal contraction with phenylephrine (PE, 0.03–0.1 µM). The pre-contractions were always matched between aortic rings and the mean relaxation was 78±1% of the PE-induced pre-contraction, and it was accepted that the endothelium was functionally intact if this relaxation was greater than 50% of the pre-contraction. Any unrubbed rings with less than 50% relaxation were discarded. Following washout and a further equilibration period, the effects of the MNL supernatant or suspension were examined in aortic rings that were either pre-contracted to similar, submaximal levels with PE, which would enable observation of any relaxant effects, or the effects of the MNL supernatant or suspension were examined in rings that were left at resting tension. MNL supernatant, MNL suspension or vehicle (Tyrode's solution with 0.25% BSA) were added to the 10 ml organ bath in successive volumes of 3, 10, 30, 100 and 300 µl (total 443 µl). In further experiments, the response to MNL supernatant was examined in aorta from resting tension in the presence or absence of an enzyme inhibitor or receptor antagonist, which was always added at least 10 min prior to commencing the MNL sample dose–response curve. Only one dose–response curve to the MNL supernatant or suspension was constructed in any single aortic ring. The concentrations of prazosin (10 µM), indomethacin (3 µM) and REV-5901 (0.1 µM) used in these experiments are equal to or greater than concentrations have been shown to effectively inhibit {alpha}1-adrenoceptor-mediated contraction [14], prostaglandin synthesis [15]or leukotriene receptor-mediated contraction [16], respectively. The effectiveness of ketanserin (0.1 µM), bosentan (10 µM) and saralasin (0.1 µM) was confirmed by their ability to inhibit contractile responses to 5-HT, endothelin-1 and angiotensin II, respectively. In these experiments, the same MNL sample was used to obtain results with and without the inhibitor. The response to platelet-released products (platelet supernatant) was also examined in aorta from resting tension, with intact endothelium, using the same cumulative dose–response method as described above.

2.8 Drugs and reagents used
For the collection and preparation of isolated leukocytes, Saffan® was obtained from Pitman Moore (Australia) and diluted two-fold in sterile 0.9% NaCl prior to use. Dextran T500 was from Pharmacia (Sweden) and prepared as a 6% solution in sterile 0.9% NaCl. Lymphoprep® was obtained from Nycomed (Norway). Tyrode's salts, 1 M HEPES buffer and 0.25% BSA were all endotoxin-tested reagents purchased from Sigma (St. Louis, MO, USA). Crystal violet and trypan blue were also from Sigma, and the Diff-Quick® kit (modified Wright stain) was from Lab-Aids (Australia). L-Phenylephrine hydrochloride (PE), indomethacin, 5-hydroxytryptamine creatinine sulphate (5-HT), angiotensin II acetate salt and [Sar1, Ala8]-angiotensin II acetate salt (saralasin) were all obtained from Sigma. O-Acetylcholine perchlorate was obtained from BDH Chemicals (UK), ketanserin tartrate was from Jannsen (Belgium), benzenemethanol,a-pentyl-3-(2-quinolinylmethoxy)-C22H25NO2 (REV 5901) was from Cayman Chemicals (USA) and prazosin hydrochloride was obtained from Pfizer (USA). PE, ACh, bosentan, 5-HT, ketanserin, prazosin and saralasin were all dissolved in water, indomethacin was dissolved in Na2CO3 (0.1 M), Angiotensin II was dissolved in acetate (0.05 M) and REV-5901 was dissolved in ethanol. All drugs were further diluted in Krebs solution. For the HPLC analysis, standards were prepared from noradrenaline hydrochloride, adrenaline bitartrate and 5-hydroxytryptamine hydrochloride (5-HT), all of which were obtained from Sigma and dissolved in 0.4 M perchloric acid.

2.9 Statistical analyses
Results are expressed as mean±standard error of the mean. Volume–response curves were compared by two-way analysis of variance (ANOVA) with post-hoc comparisons by the Newmann-Keuls test. For paired data, where pEC50 values were compared in the absence and presence of an antagonist, Student's t-test was used. Statistical significance was accepted when P<0.05.


   3 Results
Top
 Abstract
 1 Introduction
 2 Methods
 3 Results
4 Discussion
 References
 
3.1 Composition of the MNL suspensions
The MNL suspensions contained (in%): lymphocytes 82.6±2.8, monocytes 13.3±2.1, neutrophils 4.3±1.8 and basophils 0.5±0.1 (n=14). Platelets were present in the MNL suspensions at an average concentration of 5.8±1.7x107/ml (n=11).

3.2 Effect of MNL suspension on aorta
Original traces of the responses to MNL suspension in rabbit aortic rings from resting tension, or submaximally pre-contracted, with intact endothelium are reproduced in Fig. 1. MNL suspension caused a dose-dependent contraction of rings from resting tension, and a dose-dependent further contraction of rings that had been pre-contracted with PE. The group data is shown in Fig. 2. Contractions to the MNL suspension were not dependent upon a functioning endothelium, and the threshold of the response to MNL suspension in pre-contracted rings is lower than the threshold in the rings from resting tension (Fig. 2).


Figure 1
View larger version (8K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 1 Reproduction of traces showing the response to MNL suspension. MNL suspension (5x107 MNLs/ml) caused a dose-dependent contraction in aorta from resting tension (A) and in aorta that had been submaximally pre-contracted with phenylephrine (PE, 0.03 µM) (B). Doses added are in µl.

 

Figure 2
View larger version (13K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 2 Effect of MNL suspension on aorta. MNL suspension (5x107 MNLs/ml) caused a dose-dependent and endothelium-independent contraction of aorta from resting tension [A, {blacksquare}=endothelium intact (EI), {square}=endothelium denuded (EX), n=4] and in pre-contracted aorta [B, bullet=EI, {circ}=EX; pre-contraction with phenylephrine (0.03–0.1 µM): EI; 2.7±0.2 g, EX; 3.7±0.4 g, n=6].

 
3.3 Effect of MNL supernatant on aorta
Original traces of the response to MNL supernatant in aorta from resting tension, or submaximally pre-contracted, with intact endothelium are reproduced in Fig. 3. MNL supernatant caused a dose-dependent contraction of rings from resting tension, and a dose-dependent further contraction of aorta that had been pre-contracted with PE. The group data is shown in Fig. 4. The maximum response to MNL supernatant from resting tension was not different to the response in pre-contracted aorta, nor was the level of the response influenced by the presence or absence of the endothelium (Fig. 4). The threshold of the response to MNL supernatant in pre-contracted rings is lower than the threshold in the rings from resting tension (Fig. 4). The contraction response to MNL suspension was greater than the response to MNL supernatant in both aorta from resting tension and when pre-contracted (ANOVA, P<0.05).


Figure 3
View larger version (7K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 3 Reproduction of traces showing the response to MNL supernatant. MNL supernatant (derived from 5x107 MNLs/ml) caused a dose-dependent contraction in aorta from resting tension (A) and in aorta that had been submaximally pre-contracted with phenylephrine (0.03 µM) (B). Doses added are in µl.

 

Figure 4
View larger version (12K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 4 Effect of MNL supernatant on aorta. MNL supernatant (derived from 5x107 MNLs/ml) caused a dose-dependent and endothelium-independent contraction of aorta from resting tension (A, {blacksquare}=EI, {square}=EX, n=10) and in pre-contracted aorta [B, bullet=EI, {circ}=EX; pre-contraction with phenylephrine (0.03–0.1 µM): EI: 3.5±0.2 g, EX: 3.8±0.2 g, n=5].

 
3.4 Effect of inhibitors on the response to MNL-derived products
3.4.1 Indomethacin
The contraction response to MNL supernatant was unaffected by the cyclo-oxygenase inhibitor indomethacin (3 µM) [maximum contraction (g): control 3.5±1.1; indomethacin-treated 3.4±1.1, n=6].

3.4.2 REV 5901
The leukotriene receptor antagonist and 5-lipoxygenase inhibitor REV 5901 (0.1 µM) had no effect on the contraction response to MNL supernatant [maximum contraction (g): control 3.5±1; REV 5901-treated 3.7±1.0, n=5].

3.4.3 Bosentan
The contraction response to MNL supernatant was unaffected by the endothelin receptor antagonist bosentan (10 µM). [Maximum contraction (g): control 4.9±0.7; bosentan-treated 4.6±1.5, n=4]. This concentration of bosentan abolished the response to 0.1 µM endothelin-1 (data not shown).

3.4.4 Saralasin
The contractile response to MNL supernatant was unaffected by the angiotensin II receptor antagonist saralasin (0.1 µM) [maximum contraction (g): control 2.9±0.4; saralasin treated 4.2±0.5, n=4]. This concentration caused approximately a ten-fold shift in the dose–response curve to angiotensin II (data not shown). The contractile response to MNL suspension was not affected by the presence of saralasin [maximum contraction (g): control 5.2±1.5; saralasin-treated 6.0±1.5, n=5).

3.4.5 Ketanserin
Both the contractile response to MNL supernatant (n=6) and MNL suspension (n=4) were significantly inhibited by 0.1 µM ketanserin, (P<0.05, ANOVA, Fig. 5). The 5-HT2 receptor antagonist ketanserin (0.1 µM) caused an approximately 14-fold shift in the 5-HT dose–response curve, pEC50: control 7.12±0.15; ketanserin-treated 5.97±0.14 (P<0.05, Student's t-test for paired data) but the maximum contraction to 5-HT was unchanged: control 7.52±0.9; ketanserin-treated 7.68±1.08, n=5 (Fig. 6).


Figure 5
View larger version (12K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 5 Effect of ketanserin on the response to MNL supernatant. (A) In aorta from resting tension, ketanserin (0.1 µM) significantly reduced the contraction response to MNL supernatant (derived from 5x107 MNLs/ml) (A, control {blacksquare}, ketanserin {square}, n=6) and MNL suspension (5x107 MNLs/ml) (B, control bullet, ketanserin {circ}, n=4). (*=P<0.05, ANOVA).

 

Figure 6
View larger version (15K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 6 5-HT dose–response curve. The dose–response curve to 5-HT in aortic rings from resting tension (bullet) is significantly antagonised by ketanserin (0.1 µM, {circ}). (*=P<0.05, Student's t-test on EC50 calculation, control vs. ketanserin-treated). The equivalent response to 443 µl of 5 µM 5-HT is shown by the arrow/dashed line to indicate that 443 µl MNL supernatant (which was determined to contain 5 µM 5-HT by HPLC analysis) can elicit a similar response (see Fig. 4).

 
3.4.6 Prazosin
The contractile response to MNL supernatant was unaffected by the {alpha}1-adrenoceptor antagonist prazosin (10 µM) [maximum contraction (g): control 3.2±1.1, prazosin-treated 2.9±0.9, n=6].

3.5 HPLC measurement of amines in the MNL supernatant
Neither the vehicle nor samples of the MNL supernatant contained any detectable adrenaline or noradrenaline. However, the MNL supernatant contained 5.0±1.0 µM serotonin (n=4). There was no detectable 5-HT in the vehicle.

3.6 Effect of platelet contamination of the MNL preparation on the contractile responses
In some experiments, the MNL were given an extra wash in an attempt to reduce the platelet contamination of the MNL isolates. The extra wash resulted in approximately a 30% reduction in the number of platelets in the suspension [platelet count (x106/ml): normal wash 60±12; extra wash 39±7, n=4], but had no effect on the contractile response to the MNL supernatant (Fig. 7). Platelet supernatant obtained from a suspension of 5x107 platelets/ml (equivalent to the average platelet contamination of the MNL preparations) had no significant contractile effect (Fig. 7).


Figure 7
View larger version (14K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 7 Effect of platelet contamination on the response to MNL supernatant. The contractile response to MNL supernatant (derived from 5x107 MNLs/ml) (bullet, n=10) was unaffected by adding an extra wash step ({circ}, n=6) that reduced the contaminating platelet number by 40%. Supernatant prepared from 5x107 platelets/ml, the average platelet contamination, had no vasoactive effect ({blacktriangleup}, n=6). (*=P<0.05 compared to platelet supernatant).

 

   4 Discussion
Top
 Abstract
 1 Introduction
 2 Methods
 3 Results
 4 Discussion
References
 
These results show that mononuclear cells release a factor that causes contraction of rabbit thoracic aorta. MNL-derived products caused a dose-dependent contraction of rabbit aorta from resting tension, and a further contraction of pre-contracted aorta. The contractile responses were unaffected by the removal of the endothelium. Experiments carried out in the presence of a range of inhibitors indicate that the factor is not endothelin, an in situ product of cyclo-oxygenase, a lipoxygenase product or angiotensin II. However, the response to MNL products was significantly inhibited by the 5-HT2 receptor antagonist ketanserin, but not by the {alpha}1-adrenoceptor antagonist prazosin. The lack of effect of prazosin is important, as ketanserin is known to have some {alpha}-adrenoceptor antagonist properties [17], so this result indicates that the factor responsible for the contraction was 5-HT and not another amine acting on {alpha}-adrenoceptors. Further to this, supernatant from the MNL suspension was found to contain an average of 5 µM 5-HT when analysed by HPLC, whereas there was no detectable adrenaline or noradrenaline. All of the data presented indicate that the contractile factor released by MNLs is 5-HT.

MNLs have previously been shown to have vasoactive effects. Studies by Mügge et al. [8, 18]demonstrated that activated human mononuclear cells caused vasoconstriction of the cynomolgus monkey femoral artery in vitro. Interestingly, the results of these studies showed that human MNLs released a stable factor that caused an endothelium-independent contraction. The responsible factor was neither a metabolite of the cyclo-oxygenase or lipoxygenase pathways, nor a free radical, but a heat-stable molecule <1000 kDa in size. It was suggested that the factor may be angiotensin II, but no data was provided to support that hypothesis [4]. Since the observations of Mügge et al. [8]are consistent with our report, we suggest that the factor responsible for the contraction in their study may also have been 5-HT.

Further studies in cynomolgus monkeys showed that in vivo activation of leukocytes with the tri-peptide N-formyl–Met–Leu–Phe (f-MLP) caused a marked constriction of hindlimb large arteries [19]and a decrease in ocular and cerebral blood flow [20]in atherosclerotic, but not control, animals. Unfortunately, these in vivo studies are limited in that there can be no differentiation between the effects of activating polymorphonuclear leukocytes or MNLs, however, it is possible that 5-HT released from MNLs may have contributed to the constriction observed in those studies. It is also important to note that 5-HT-induced constriction is more likely to occur in atherosclerotic vessels [21, 22]and that MNLs accumulate at sites of inflammation, and are probably activated in atherosclerotic plaques [13]. There has also been a report suggesting that early migration of mononuclear cells into vascular lesions increases the sensitivity of the rabbit carotid artery to the contractile actions of 5-HT [11]. The results of the present study suggest that this may be due to the additive effects of the 5-HT released by the mononuclear cells to the exogenously applied 5-HT used to construct the dose–response curves.

Other reports of vasoactive effects of monocytes include a study which shows that endothelin-1-induced vasoconstriction is increased in the presence of monocytes and suggests that this effect is mediated by the stimulation of monocyte endothelin receptors. Although the mechanism of this potentiation was not determined [9], the effect is consistent with the ability of 5-HT to enhance contractile responses to endothelin [23]. Another group reported that a mediator derived from human MNLs, and also from other human cell types, causes inhibition of endothelium-dependent relaxation in rat aortic rings, together with small endothelium-dependent contraction of pre-contracted rat aorta [10]. Clearly, this factor was not 5-HT and suggests that there may be differences in mediator release across species. The literature cited here supports the hypothesis that leukocytes produce vasoactive factors, and we propose that 5-HT may be involved in some of the previous reports of MNL-derived constriction.

MNL products caused contraction of aorta from both resting tension and when precontracted. It is notable that the threshold dose of the MNL response in the precontracted rings is lower than in rings from resting tension. This lower threshold may be explained by 5-HT-mediated potentiation of the phenylephrine-induced contraction and is consistent with observations that sub-contractile concentrations of 5-HT enhance responses to other vasoconstrictors [24]. Both MNL suspension and supernatant caused contractions that were independent of the endothelium, as there was no statistically significant difference in the responses in rings with endothelium intact and endothelium that was denuded. Typically, 5-HT causes a greater contraction in aortic rings without endothelium, where the modulatory effect of the endothelium is removed [21], and there was a tendency for this to occur with the MNL suspension, but not the supernatant.

The contractions caused by the MNL suspension were significantly greater than those caused by the supernatant alone, which may be due to an interaction between the aorta and the MNLs. This interaction may result in activation of the leukocytes and a greater release of their stored substances, as it is reported that MNLs activated with interferon {gamma} [25]or cisplatin [26]release larger amounts of serotonin. The maximum contractions caused by the MNL products were smaller when the aortic rings were precontracted, probably because the rings were already approaching the maximum attainable contraction. Although the MNLs were not specifically stimulated by any agent, the activation status of cells was not assessed, so it is possible that the MNLs were activated by the isolation and incubation procedures. The MNL suspension used in this study contained mostly lymphocytes (83%) and some monocytes (13%). It is not known which of these cell types is responsible for the 5-HT release, or indeed if both cell types are involved.

The suspension was also contaminated with basophils and platelets. Platelet contamination was minimized as far as possible in the preparation procedure by removing the platelet-rich plasma prior to dextran sedimentation, and by further washing after separating the MNLs from the polymorphonuclear cells. However, platelet contamination was still evident (average of 5.8x107 platelets/ml in each MNL suspension). To control for this, a suspension of platelets was prepared at a similar concentration (5x107 platelets/ml), incubated and the supernatant collected as for the MNL supernatant. This platelet supernatant had no significant contractile effect, indicating that the platelet contamination had little role in the responses shown. In addition, a thorough study of the contents of platelets, undertaken by Meyers et al. [27], showed that 5x107 rabbit platelets contain only 1.7 nmol 5-HT, more than 2000-fold less than was measured in the MNL supernatant, supporting our result. It is important to note that platelets and leukocytes are known to interact and modulate each other's function. In particular, leukocytes are reported to inhibit platelet-derived vasorelaxation [4, 28]and, although it is unlikely in the present study given the relatively low level of platelets, a synergistic interaction between the MNLs and the platelets, which may enhance vasoconstriction, cannot be fully discounted. However, it should also be noted that reducing the level of platelet contamination by 30% did not change the response to the MNL supernatant. Basophils are also known to release 5-HT [29], however, the number of basophils present in the final MNL suspension (<1%) would be unlikely to release such large amounts of 5-HT. The final concentration of MNL in the organ bath (equivalent to 2.2x106 cells/ml) was comparable to the circulating concentration of MNLs, but conservative for an inflammatory situation, such as would occur in an atherosclerotic lesion, where MNLs accumulate and are activated.

The observations presented here are consistent with other reports of an association between MNLs and 5-HT. 5-HT is known to bind in a specific manner to high affinity binding sites on both human lymphocytes and monocytes [30]and, indeed, human lymphocytes have been shown to have a high affinity uptake mechanism for 5-HT [31]. In addition, human lymphocytes and macrophages are reported to have the ability to metabolize tryptophan to 5-HT [32]. 5-HT plays an immunomodulatory role in mononuclear leukocyte function [33, 34]and it has been shown to activate monocyte phagocytosis [35]and to have a role in macrophage function [36]. In addition, regulation of the interplay between monocytes and a subgroup of lymphocytes, the NK (natural killer) cells, is transduced by 5-HT receptors [37, 38].

In conclusion, this study provides pharmacological and chromatographic evidence that MNLs can release 5-HT, and that MNL-derived 5-HT causes significant contraction of rabbit aortic rings. MNLs are known to play a role in vasoconstriction, particularly in atherosclerosis, and MNL-derived 5-HT may be the mediator responsible for the vasoactive effects observed in several studies. 5-HT is not only specifically taken up by MNLs, but there is evidence that these cells may also have the capacity for 5-HT synthesis and, in addition, 5-HT has a role in MNL immunomodulation. The data presented suggests that MNL-derived 5-HT may be an important modulator of vessel tone.

Time for primary review 27 days.


   Acknowledgements
 
We wish to thank Dr. James Ziogas for supplying [Sar1, Ala8]-angiotensin II, and the Roche Co. for their generous gift of bosentan. We also thank Ms. Melanie O'Farrell and Dr. Philip Marley for carrying out the HPLC analyses. Thanks also to Ms. Vitina Sozzi and Ms. Sarah Emselle for their expert technical assistance. This work was supported by the Smoking and Health Research Foundation of Australia.


   References
Top
 Abstract
 1 Introduction
 2 Methods
 3 Results
 4 Discussion
 References
 

  1. Clinton S.K., Libby P. Cytokines and growth factors in atherogenesis. Arch Pathol Lab Med (1992) 116:1292–1300.[Web of Science][Medline]
  2. Lopes-Virella MIF. Atherosclerosis and autoimmunity. Clin Immunol Immunopathol 1994;73:155–167.
  3. Zimmennan G.A., McIntyre T.M., Prescott S.M., Otsuka K. Brief review: Molecular mechanisms of neutrophil binding to endothelium involving platelet activating factor and cytokines. J Lipid Mediators (1990) 2:S31–S43.[Web of Science][Medline]
  4. Kaul S., Padgett R.C., Heistad D.D. Role of platelets and leukocytes in modulation of vascular tone. Ann NY Acad Sci (1994) 714:122–135.[Web of Science][Medline]
  5. Dale M, Foreman JC. Textbook of immunopharmacology. London: Blackwell Scientific, 1984.
  6. Kitazono T., Padgett R.C., Armstrong M.L., Tompkins P.K., Heistad D.D. Evidence that angiotensin II is present in human monocytes. Circulation (1995) 91:1129–1134.[Abstract/Free Full Text]
  7. Ehrenreich H., Anderson R.W., Fox C.H., et al. Endothelins, peptides with potent vasoactive properties, are produced by human macrophages. J Exp Med (1990) 172:1741–1748.[Abstract/Free Full Text]
  8. Mugge A., Heistad D.D., Padgett R.C., et al. Mechanisms of contraction induced by human leukocytes in normal and atherosclerotic arteries. Circ Res (1991) 69:871–880.[Abstract/Free Full Text]
  9. Magazine H.I., Andersen T.T., Bruner C.A., Malik A.B. Vascular contractile potency of endothelin-1 is increased in the presence of monocytes or macrophages. Am J Physiol (1994) 266:H1620–H1625.[Web of Science][Medline]
  10. Liu J.J., Xie B., Thurlow P.J., Wiley J.S., Chen J.R. Various cells release a stable small molecule that inhibits endothelium-dependent relaxation. Am J Physiol (1995) 269:H1303–H1311.[Web of Science][Medline]
  11. Dusting G.J., Curcio A., Harris P.J., et al. Supersensitivity to vasoconstrictor action of serotonin precedes the development of atheroma-like lesions in the rabbit. J Cardiovasc Pharmacol (1990) 16:667–674.[CrossRef][Web of Science][Medline]
  12. Kochanek P.M., Hallenbeck I.M. Polymorphonuclear leukocytes and monocytes/macrophages in the pathogenesis of cerebral ischaemia and stroke. Stroke (1992) 23:1367–1379.[Abstract/Free Full Text]
  13. Ross R. The pathogenesis of atherosclerosis: A perspective for the 1990s. Nature (1993) 362:801–809.[CrossRef][Medline]
  14. Docherty I., Constantine I., Starke K. Smooth muscle of rabbit contains alpha 1- but not alpha 2-adrenoceptors. Naunyn-Schmiedeberg's Arch Pharmacol (1981) 317:5–7.[CrossRef][Web of Science][Medline]
  15. Levesque L., Drapeau G., Grose I., Rioux F., Marceau F. Vascular mode of action of kinin B 1 receptors and development of a cellular model for the investigation of these receptors. Br J Pharmacol (1993) 109:1254–1262.[Web of Science][Medline]
  16. Mullane K., Hatala M., Kraemer R., Sessa W., Westlin W. Myocardial salvage induced by REV-5901: an inhibitor and antagonist of the leukotrienes. J Cardiovasc Pharmacol (1987) 10:398–406.[Web of Science][Medline]
  17. Marwood J.I.F., Stokes G.S. Vascular smooth muscle serotonin and {alpha}-adrenergic receptors. Clin Exp Pharmacol Physiol (1983) 10:265–267.[CrossRef][Web of Science][Medline]
  18. Mugge A., Lopez I.A., Heistad D.D., Lichtlen P.R. Vasoconstriction in response to activated leukocytes: implications for vasospasm. Eur Heart J (1993) 14:87–92.[Web of Science][Medline]
  19. Lopez J.A.G., Armstrong M.L., Harrison D.G., Piegors D.I., Heistad D.D. Vascular responses to leukocyte products in atherosclerotic primates. Circ Res (1989) 65:1078–1086.[Abstract/Free Full Text]
  20. Faraci F.M., Lopez J.A.G., Breese K., Armstrong M.L., Heistad D.D. Effect of atherosclerosis on cerebral vascular responses to activation of leukocytes and platelets in monkeys. Stroke (1991) 22:790–796.[Abstract/Free Full Text]
  21. Vanhoutte P.M. Vascular effects of serotonin and ischemia. J Cardiovasc Pharmacol (1990) 16:S15–S19.
  22. Heistad D.D., Armstrong M.L. Sick vessel syndrome; Can atherosclerotic arteries recover? Circulation (1994) 89:2447–2450.[Free Full Text]
  23. Nakayama K., Ishigai Y., Uchida H., Tanaka Y. Potentiation by endothelin-1 of 5-hydroxytryptamine-induced contraction in coronary artery of the pig. Br J Pharmacol (1991) 104:978–986.[Web of Science][Medline]
  24. Vanhoutte P.M., Cohen R.A., Van Neuten I.M. Serotonin and arterial vessels. J Cardiovasc Pharmacol (1984) 6:S421–S428.[Web of Science][Medline]
  25. Finocchiaro LME, Nahmod VE, Launay JM. Melatonin biosynthesis and metabolism in peripheral blood mononuclear leukocytes. Biochem J 1991;280:727–731.
  26. Mantovani G., Maccio A., Esu S., Lai P. Evidence that cisplatin induces serotonin release from peripheral blood mononuclear cells and that methylprednisolone inhibits this effect. Eur J Cancer (1996) 32A:1983–1985.[CrossRef]
  27. Meyers K.M., Holmsen H., Seachford C.L. Comparative study of platelet dense granule constituents. Am J Physiol (1982) 243:R454–R461.[Web of Science][Medline]
  28. Kaul S., Waack B.I., Padgett R.C., Brooks R.M., Heistad D.D. Interaction of human platelets and leukocytes in modulation of vascular tone. Am J Physiol (1994) 266:H1706–H1714.[Web of Science][Medline]
  29. Conti P., Reale M., Barbacane R.C., et al. Role of lipoxins A4 and B4 in the generation of arachidonic acid metabolites by rat mast cells and their effect on [3H]serotonin release. Immunol Lett (1992) 32:117–124.[CrossRef][Web of Science][Medline]
  30. Giacovazzo M., Bernoni R.M., Di Sabato F., Martelletti P. Impairment of 5-HT binding to lymphocytes and monocytes from tension-type headache patients. Headache (1990) 30:220–223.[CrossRef][Web of Science][Medline]
  31. Faraj B.A., Olkowski Z.L., Iackson R.T. Expression of a high-affinity serotonin transporter in human lymphocytes. Int J Immunopharmacol (1994) 16:561–567.[CrossRef][Web of Science][Medline]
  32. Finocchiaro L.M.E., Arzt E.S., Femandez-Castello S., et al. Serotonin and melatonin synthesis in peripheral blood mononuclear cells: Stimulation by interferon-{gamma} as part of an immunomodulatory pathway. J Interferon Res (1988) 8:705–716.[Web of Science][Medline]
  33. Hellstrand K., Hermodsson S. Enhancement of human natural killer cell cytotoxicity by serotonin: role of non-T/CD16+ NK cells, accessory monocytes, and 5-HT1A receptors. Cell Immunol (1990) 127:199–214.[CrossRef][Web of Science][Medline]
  34. Hellstrand K., Hermodsson S. Monocyte-mediated suppression of LL-2-induced NK-cell activation. Regulation by 5-HT1A-type serotonin receptors. Scand J Immunol (1990) 32:183–192.[CrossRef][Web of Science][Medline]
  35. Tokamov A.A., Kykhova M.I.P., Koshkina O.V., Vasilyev V.I. The effect of serotonin on the functional activity of monocytes. Tsitologia (1991) 33:81–87.[Medline]
  36. Arzt E., Costas M., Finkielman S., Nahmod V.E. Serotonin inhibition of tumour necrosis factor-alpha synthesis by human monocytes. Life Sci (1991) 48:2557–2562.[CrossRef][Web of Science][Medline]
  37. Hellstrand K., Hermodsson S. Serotonergic 5-HT1A receptors regulate a cell contact-mediated interaction between natural killer cells and monocytes. Scand J Immunol. (1993) 37:7–18.[CrossRef][Web of Science][Medline]
  38. Hellstrand K., Czerkinsky C., Ricksten A., et al. Role of serotonin in the regulation of interferon-gamma production by human natural killer cells. J Interferon Res (1993) 13:33–38.[Web of Science][Medline]

Add to CiteULike CiteULike   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us    What's this?


This article has been cited by other articles:


Home page
 Cardiovasc ResHome page
J.Y. Jeremy, G.D. Angelini, M. Khan, D.P. Mikhailidis, R.J. Morgan, C.S. Thompson, K.R. Bruckdorfer, and K.M. Naseem
Platelets, oxidant stress and erectile dysfunction: an hypothesis
Cardiovasc Res, April 1, 2000; 46(1): 50 - 54.
[Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow E-letters: Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when E-letters are posted
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrowRequest Permissions
Right arrow Disclaimer
Google Scholar
Right arrow Articles by Hart-Favaloro, J. L.
Right arrow Articles by Woodman, O. L.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Hart-Favaloro, J. L.
Right arrow Articles by Woodman, O. L.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?