Download Local and systemic autonomic nervous effects

J Appl Physiol 94: 469–475, 2003.
First published September 27, 2002; 10.1152/japplphysiol.00411.2002.
Local and systemic autonomic nervous effects on cell
migration to the spleen
HEINER ROGAUSCH, DETLEV ZWINGMANN, MIRJAM TRUDEWIND,
ADRIANA DEL REY, KARL-HEINZ VOIGT, AND HUGO BESEDOVSKY
Department of Immunophysiology, Institute of Physiology,
Philipps-University, 35039 Marburg, Germany
Submitted 10 May 2002; accepted in final form 26 September 2002
spleen is intensely interwoven with the detection of antigenic material circulating in blood, and splenectomy increases the risk of
overwhelming infections or septic complications (3, 23).
In contrast to other lymphoid organs, the spleen lacks
significant afferent lymphatic vessels and blood flow
represents the predominant route for the influx of cells
and antigens. Two characteristic findings indicate the
interdependence between splenic perfusion and cell
uptake: 1) the flow per gram of tissue is 10 times higher
than that of resting skeletal muscle and nearly as high
as the blood flow of heart muscle, and 2) the rate of
lymphocyte circulation into the spleen equals the total
number of cells flowing into all other lymphatic and
nonlymphatic organs (15, 16, 27).
It is remarkable that a strong interdependence between cell and blood supply coexists with a dense
splenic noradrenergic innervation. If the content of
norepinephrine (NE) is taken as reflection of the degree of noradrenergic sympathetic innervation, the
spleen belongs to the most densely sympathetically
innervated organs (5, 12), and the NE turnover rate is
four to six times higher than that of the liver or lung.
Most splenic noradrenergic nerve fibers have vasoconstrictor function and reduce blood flow. Therefore, the
high splenic perfusion rate observed under basal conditions and during immune responses is surprising,
but it can be explained by our laboratory’s previous
observations that locally released cytokines, such as
interleukin (IL)-1␤, exert a tonic inhibition on the
noradrenergic vasoconstrictor tonus (18). An increase
in splenic blood flow mediated by a cytokine-inhibited
NE release by sympathetic nerves may be a main
mechanism influencing lymphoid cell uptake from the
circulating cell pool (19). In addition, it is expected that
the special morphological structure of the spleen
guides cells and antigenic material from the circulation
into the resident pool. The analysis of how lymphocytes
enter into splenic cell compartments suggests local
regulatory influences either during the phase of cell
uptake or during homing into specific areas (2, 26).
However, these local mechanisms will still depend on
the supply of lymphoid cells by blood circulation and on
the particular structure of the spleen. Until now, the
role of hemodynamic forces that determine splenic perfusion and cell uptake in this organ was not systematically investigated in vivo, and it is not known whether
adhesion molecules can interfere or even override noradrenergic influences on splenic perfusion and blood
cell supply. The experiments reported here tested the
hypothesis that noradrenergic regulation of vascular
blood flow plays a significant, but not exclusive, role for
the splenic extraction of immune cells from the circulating pool.
We studied the influence of increased splenic vascular perfusion induced by either local denervation or
general sympathectomy on lymphoid cell uptake by the
Address for reprint requests and other correspondence: H. O.
Besedovsky, Philipps-Univ. Marburg, Institute of Physiology,
Deutschhausstrasse 2, D-35037 Marburg, Germany (E-mail:
[email protected])
The costs of publication of this article were defrayed in part by the
payment of page charges. The article must therefore be hereby
marked ‘‘advertisement’’ in accordance with 18 U.S.C. Section 1734
solely to indicate this fact.
norepinephrine; sympathetic innervation; heart minute volume
THE FUNCTION OF THE MAMMALIAN
http://www.jap.org
8750-7587/03 $5.00 Copyright © 2003 the American Physiological Society
469
Downloaded from http://jap.physiology.org/ by 10.220.33.4 on April 13, 2017
Rogausch, Heiner, Detlev Zwingmann, Mirjam
Trudewind, Adriana del Rey, Karl-Heinz Voigt, and
Hugo Besedovsky. Local and systemic autonomic nervous
effects on cell migration to the spleen. J Appl Physiol 94:
469–475, 2003. First published September 27, 2002; 10.1152/
japplphysiol.00411.2002.—This work is based on the hypothesis that sympathetic nerves regulate the uptake of circulating cells by the spleen by affecting splenic blood flow and that
the quantity of cells sequestered depends on whether
changes in noradrenergic transmission occur at local or systemic levels. Fluorescently labeled lymphoid cells were injected into rats, and organ blood flow was measured by the
microsphere method. Increased retention of cells in the
spleen paralleled by increased blood flow was detected after
local denervation of this organ or administration of bacterial
endotoxin. A comparable enhanced splenic blood flow was
observed after general sympathectomy. However, the redistribution of blood perfusion during general vasodilatation
resulted in deviation of leukocyte flow from the spleen, thus
resulting in reduced uptake of cells by this organ. These
results indicate that, although the uptake of cells by the
spleen depends on arterial blood supply, enhanced perfusion
does not always result in increased cell sequestration because general vasodilatation reduces cell uptake by this
organ and even overrides stimulatory effects of endotoxin.
470
BLOOD FLOW AND IMMUNE CELL TRAFFIC
spleen of normal and endotoxin-stimulated rats. The
results obtained show that, although both procedures
result in comparable increases in splenic blood perfusion, only local denervation produces a net increased
influx of lymphoid cells into the spleen.
METHODS
General Procedures
Male inbred Wistar-Koyoto rats (300–350 g body wt) were
housed in single cages and kept at a 12:12-h day-night cycle
with water and standard pellets available ad libitum. Surgery was performed while the animals were under general
anesthesia with pentobarbital sodium (60 mg/kg Narcoren).
Animals were placed during surgery and experiments on a
heating plate regulated to maintain core temperature between 36.5 and 37°C.
Determinations of Blood Flow
Q̇ ti ⫽ 100 䡠 Q̇ref共Iti/Iref 䡠 Wti兲
where Iti and Iref are number of spheres per tissue or reference probe, respectively, and Wti is weight of the tissue.
The heart minute volume was determined with an ultrasonic flow probe (model T206, blood flowmeter, Transonic
Systems, Ithaca, NY) in animals that were not used for organ
perfusion and cell uptake studies. The flow probe was placed
around the thoracic aorta, and, after closure of the the chest
and stabilization of blood gases and peripheral arterial resistance, blood flow was digitally registered at 80-kHz real-time
display throughput (Dataq Instruments, Akron, OH).
In all animals, arterial blood pressure was measured via a
catheter implanted in the femoral artery combined with a
Statham pressure transducer; arterial PO2 and PCO2 were
measured by electrochemical detection (Gas Check AVL, Bad
Homburg), and core temperature was monitored with a thermistor probe in the abdominal cavity.
A piece of spleen was used to evaluate the uptake of
labeled cells and the amount of trapped fluorescent spheres.
The number of labeled cells per 100,000 splenocytes was
determined from cytocentrifuge preparations of recipient
spleen cell suspensions by using a fluorescent microscope.
Because this step is critical for subsequent calculations,
determinations were performed in parallel by two independent observers. Twelve cytocentrifuge preparations were
evaluated from each cell suspension. Uptake of fluorescent
cells was expressed either as the percentage of the number of
injected cells, or per spleen or per 100,000 splenocytes.
Determination of IL-1␤ in the Spleen
Approximately one-third of the spleen was sonicated (20 s,
20 strokes, in ice-cold water), the homogenate was centrifuged (10 min, 20,000 g, 4°C), and IL-1␤ concentration in the
supernatant was determined by using a commercially available kit for determination of rat IL-1␤ (Endogen, Woburn,
MA).
Experiments
Control group. The uptake of fluorescent cells under baseline conditions was evaluated 15 min, 6 h, and 24 h after the
injection of labeled cells (6 animals per time point). Blood
flow was measured in the same animals 15 min or 6 h after
cell injection, and the infusion of microspheres started 3 min
before the spleen was removed.
Effect of vasodilatation induced by cutting the splenic
nerve. The effect of local vasodilatation on fluorescent cell
uptake into the spleen was evaluated by surgically interrupting splenic sympathetic nervous supply 5 days before the
experiments were started as described previously (18). At
this time, animals were recovered from the operation, as
shown by their normal weight gain and corticosterone blood
levels (see Table 1).
Successful denervation was documented by a 90% reduction of splenic NE content as evaluated by HPLC. The number of labeled cells that were sequestrated by the spleen and
flow values were determined at the same times mentioned
above. Sham-operated animals that were injected in parallel
Table 1. Cardiovascular parameters in animals
with intact splenic innervation (control),
surgically denervated spleen (local
denervation), or general depletion of vesicular
noradrenergic stores (NE depletion)
Control
Labeling Procedure for Splenocytes
Splenocyte suspensions were prepared from inbred donor
rats and labeled under sterile conditions with a PKH fluorescent cell linker kit (107 cells/ml in a 2 ␮M staining solution
for 2 min; excitation/emission of PKH 551/567 nm; Sigma
Chemical, St. Louis, MO). These cell suspensions were used
because they mainly consist of lymphocytes; they need a
minimum of purification steps. Before injection into the recipient, the number of dead cells was determined with trypan
blue; the suspensions normally contained 6–10% dead cells;
suspensions with ⬍15% dead cells were also used. Cells (108
per kg body wt) were infused into the left ventricle over a
period of 2 min in a volume of 1 ml, and the catheter was
subsequently rinsed with 0.5 ml saline.
J Appl Physiol • VOL
Arterial blood pressure,
mmHg
Heart minute volume,
ml/min
Arterial hemoglobin,
g/dl
Arterial PO2, Torr
Arterial PCO2, Torr
Splenic NE content,
ng/g
Renal NE content, ng/g
Corticosterone in blood
plasma, ␮g/dl
Local
Denervation
NE Depletion
103.5 ⫾ 3.0
103.0 ⫾ 3.2
90.2 ⫾ 2.8*
50.8 ⫾ 3.4
55.3 ⫾ 2.2
87.0 ⫾ 2.7†
14.21 ⫾ 0.33
13.92 ⫾ 0.33
13.2 ⫾ 0.24
95.8 ⫾ 1.3
40.3 ⫾ 0.8
412.7 ⫾ 26.2
96.5 ⫾ 0.6
40.2 ⫾ 0.8
13.8 ⫾ 6.6†
95.2 ⫾ 1.0
40.7 ⫾ 0.7
Not
detectable†
7.1 ⫾ 5.9†
1.9 ⫾ 1.0
127.7 ⫾ 9.1
2.4 ⫾ 0.9
128.0 ⫾ 8.7
2.6 ⫾ 0.4
Values are means ⫾ SE for 6 animals per group. NE, norepinephrine. * P ⬍ 0.05; † P ⬍ 0.001 compared with control.
94 • FEBRUARY 2003 •
www.jap.org
Downloaded from http://jap.physiology.org/ by 10.220.33.4 on April 13, 2017
The procedure for the determination of organ blood flow
with the microsphere technique is similar to standard, previously reported techniques (14, 21). Briefly, fluorescent-dye
labeled polystyrene microspheres [excitation/emission 450/
480 nm, diameter 15.5 ␮m (⫾ 2%); Molecular Probes (MoBiTec, Göttingen, Germany)] were injected at 0.2 ml/min into
the left ventricle at a dose of 450,000 spheres per animal.
Simultaneously, a reference probe (Q̇ref) was withdrawn from
the abdominal aorta at the same flow velocity through the
femoral artery, and blood flow in spleen, liver, heart, and
skeletal muscle (Q̇ti) was determined by using the following
equation
Evaluation of Splenic Cell Uptake
BLOOD FLOW AND IMMUNE CELL TRAFFIC
471
with aliquots of the same labeled splenocyte suspension
served as controls.
Effect of vasodilatation by depletion of sympathetic noradrenergic stores. Because, under clinical conditions, pharmacological interventions can affect not only splenic but also
sympathetic nerve endings in other organs, general noradrenergic transmission was interrupted by depleting noradrenergic stores with reserpine (10 mg/kg ip; 24 h before the
experiment). Six animals treated with reserpine and six
animals injected with the vehicle alone were studied in parallel at the same points of time as in the other studies.
Effect of endotoxin on splenic cell uptake. Endotoxin [lipoplysaccharide (LPS)] from Escherichia coli (026:B6; TCA
extract; Sigma Chemical; 10 ␮g/kg body wt) was dissolved in
isotonic sodium chloride solution and injected into the tail
artery in awake animals. Six hours later, labeled cells were
injected as described above, and cell uptake was determined
at the same time intervals as in the other experimental
groups.
Statistics
RESULTS
Relation Between Splenic Blood Flow
and Cell Uptake
Blood flow of the spleen was measured under basal
conditions and after interruption of noradrenergic innervation. Basal flow was 73.5 ⫾ 8.65 ml 䡠 100
g⫺1 䡠 min⫺1. This value approached myocardial perfusion (91 ⫾ 8.00 ml 䡠 100 g⫺1 䡠 min⫺1) and was nearly 20
times higher than the perfusion of skeletal muscle at
rest (3.8 ⫾ 0.6 ml 䡠 100 g⫺1 䡠 min⫺1), as measured in the
same animal and under the same experimental conditions. Because the spleen has a high basal blood flow
value, it was important to determine whether it can be
increased after interruption of the noradrenergic innervation. The results showed that despite high resting flow, the ablation of the noradrenergic innervation
led to a significant increase: 276.7 ⫾ 10.6 ml 䡠 100
g⫺1 䡠 min⫺1 after local surgical denervation (P ⬍ 0.001),
and 265 ⫾ 10.6 ml 䡠 100 g⫺1 䡠 min⫺1 after depletion of
noradrenergic presynaptic stores (P ⬍ 0.001; Fig. 1).
The increase in flow indicates that the noradrenergic
sympathetic innervation of the spleen has profound
influence on establishing the level of splenic perfusion.
Sham-operated and untreated animals exhibited no
significant differences of splenic blood flow.
Next, we wanted to know whether an increased local
perfusion favors the accumulation of injected cells in
the spleen. Both 15 min and 6 h after the injection of
labeled cells, a higher sequestration of injected cells
(Fig. 2A), higher number of cells retained per spleen
(Fig. 2B), and higher number of cells retained per
100,000 splenocytes (Fig. 2C) were observed in the
denervated spleen compared with the controls. Even
after 24 h, more cells accumulated in the denervated
spleen compared with the sympathetically innervated
organ. These results suggest the possibility of parallel
J Appl Physiol • VOL
changes between spleen perfusion and cell uptake,
which was corroborated by the experiments described
below.
Cell Sequestration into the Spleen Is Favored
by an Increased Splenic Perfusion Induced
by Bacterial Endotoxin
Studies in which LPS was used were included because this cytokine raises blood flow and can favor the
adhesion of leukocytes on endothelial cells. If adhesion
is additive to the effect of perfusion, a higher cell
uptake into the spleen, which exceeds the effect of
increased perfusion, would be expected. However, Fig.
3 indicates parallel changes between the level of
splenic perfusion and cell uptake. A positive correlation between local perfusion and the degree of cell
extraction from circulation is described by a secondorder polynomial equation (y ⫽ 1 ⫻ 10⫺5x2 ⫹ 0.008x ⫹
0.074, R2 ⫽ 0.74; Fig. 3).
Because, as our laboratory reported before (19),
IL-1␤ is the main mediator of the increase in splenic
blood flow induced by LPS, the production of this cytokine in the spleen of rats subject to either local or
general denervation was evaluated. LPS administration significantly increased IL-1␤ in the spleen, but the
increase was comparable in control or local and systemic denervated rats (IL-1␤ ng/spleen: none ⫹ vehicle ⫽ 2.57 0.35; none ⫹ LPS 28.9 6.32; sham operated ⫹ vehicle ⫽ 5.7 0.55; sham ⫹ LPS ⫽ 27.9 3.85;
local splenic denervated ⫹ LPS ⫽ 19.37 3.19; system-
94 • FEBRUARY 2003 •
www.jap.org
Downloaded from http://jap.physiology.org/ by 10.220.33.4 on April 13, 2017
The data were compared by ANOVA and Scheffé’s test. All
values are reported as means ⫾ SE from six rats, and statistical significance was set at P ⬍ 0.05.
Fig. 1. Interruption of sympathetic transmission in the spleen by
local denervation or by systemic depletion of noradrenergic stores is
followed by increased splenic perfusion. Splenic blood flow was determined in anesthetized rats with the microsphere method in the
normally innervated (ctrl) or sympathetically denervated spleen.
Local denervation of the spleen was performed surgically 5 days
before the measurements (local denerv). In another group of rats,
noradrenergic vesicular stores were depleted by reserpine administration 24 h before the experiment (reserp). Values are means ⫾ SE
for 6 animals per group. Both denervation procedures induced a
significant increase of splenic blood flow compared with controls
(*** P ⬍ 0.001); no significant differences were observed between the
denervation procedures.
472
BLOOD FLOW AND IMMUNE CELL TRAFFIC
ically denervated ⫹ LPS ⫽ 24.1 5.8; 4–6 rats per
group). The IL-1␤ content in the spleen of rats from all
groups that received LPS differed significantly from
the controls (P ⬍ 0.05); the different denervation procedures did not significantly affect LPS-induced IL-1␤
content.
Systemic Vasodilatation Interferes with Local
Cell Supply
Fig. 2. Increased splenic perfusion induced by local denervation is
followed by increased uptake of circulating leukocytes. Splenocytes
from an inbred donor animal were labeled with fluorescent dye (PKH)
and injected into the lumen of the left ventricle of the recipient to
guarantee an even distribution of cells with the heart minute volume. A:
splenic cell uptake expressed as percentage of the total number of
injected cells. B: cell uptake expressed per spleen. C: cell uptake expressed per 105 splenocytes. Open bars, values determined in control
animals with innervated spleen (control); solid bars, data obtained from
rats in which the spleen was surgically denervated (local denerv) 5 days
before the experiment. Values are means ⫾ SE from 6 experiments.
Statistical comparisons were made between denervated animals and
their respective controls: *P ⬍ 0.05, **P ⬍ 0.01, ***P ⬍ 0.001.
J Appl Physiol • VOL
The vasodilatation observed in the spleen after systemic depletion of noradrenergic stores was similar to
the effect of local denervation (Fig. 1). Both procedures
induced comparable reductions of the splenic NE content, but only systemic denervation affected the content of the neurotransmitter in other organs, such as
the kidney (Table 1). Accordingly, although local denervation increased blood flow only in the spleen, the
abrogation of general noradrenergic vasoconstrictor tonus induced higher blood flow also in other sympathetically controlled organs. This effect was followed by a
significant increase of the heart minute volume to
maintain arterial blood pressure at normal levels (Table 1).
The redistribution of the heart minute volume between peripheral organs included increased blood flow
in large parenchymatous organs. For example, liver
blood flow rose from 3.7 ⫾ 0.5 to 10.1 ⫾ 1.4 ml/min, i.e.,
by 6.4 ml/min (Fig. 4); skeletal muscle flow increased
from 3.8 ⫾ 0.6 to 45.0 ⫾ 11.4 ml 䡠 100 g⫺1 䡠 min⫺1. As
already shown in Fig. 1, systemic NE depletion also
resulted in an increase of splenic blood flow of ⬃200%,
but this increase was low (only 1.5 ml/min) compared
with that observed in other organs when expressed in
94 • FEBRUARY 2003 •
www.jap.org
Downloaded from http://jap.physiology.org/ by 10.220.33.4 on April 13, 2017
Fig. 3. Correlation between splenic perfusion and cell uptake. Open
symbols represent the values derived from each individual animal:
controls with intact splenic innervation (E), animals that received
endotoxin (10 ␮g/kg body wt) 6 h before cell injection (‚), and animals
in which the spleen was surgically denervated (䊐). Solid symbols
represent the mean of the respective experimental groups.
BLOOD FLOW AND IMMUNE CELL TRAFFIC
absolute terms (from 0.8 ⫾ 0.1 to 2.3 ⫾ 0.6 ml/min).
Thus the larger proportion of the heart minute volume
was deviated during general vasodilatation to the liver
and to other large parenchymatous organs. This rise
was particularly evident in the skeletal muscle because, when it is considered that 40% of the animal
body weight is represented by this tissue, the increment of muscle perfusion is 34.6 ml/min (⬃1,100%).
A plot of cell uptake vs. splenic perfusion (Fig. 5)
showed that increases of splenic cell sequestration are
observed in accordance with the rise of spleen perfusion induced by local denervation, as has been already
shown in Fig. 3 for individual experiments. However,
after general vasodilatation, fewer cells were retained,
although the flow was similarly high as in locally
denervated organs (P ⬍ 0.001). LPS-treated recipients
exhibited higher cell uptake into the spleen than shaminjected animals. Splenic cell uptake is in LPS-treated
rats further augmented by local denervation of the
spleen. In contrast, after general vasodilatation, fewer
cells were retained in the spleen of animals treated
with LPS. These results indicate that the cardiovascular system not only determines the random conditions
for the distribution of cells to the spleen but may even
interfere with the favoring effects of LPS on splenic cell
uptake.
induces opposite effects on immune cell uptake by the
spleen. With respect to LPS, it should be mentioned
that the increase in blood perfusion induced by doses of
the endotoxin that do not cause shock is restricted to
the spleen without affecting other lymphoid organs,
and this effect is mediated by the capacity of locally
released IL-1␤ to inhibit the sympathetic tonus (19).
Because cytokines released after endotoxin administration can promote the expression of adhesion molecules on endothelial cells, it can be expected that LPS
stimulates cell uptake by that mechanism (22, 28). The
spleen has no high endothelial venules, but adhesion
molecules, such as integrins and selectins, are expressed in the extracellular matrix of the splenic meshwork, and they may be upregulated by cytokines induced by LPS and contribute to the high capacity of the
spleen to sequestrate cells from the blood (15, 16).
IL-1␤ is equally produced in the spleen in locally and
systemically denervated or normal innervated spleens,
and differences in cell uptake cannot be explained by
differences in IL-1␤ production. However, the results
reported here, although corroborating that splenic cell
uptake is favored by LPS, indicate that hemodynamic
influences also play a relevant role in the capacity of
the spleen to uptake circulating cells.
We studied splenic cell uptake instead of determining the circulation half-time of injected cells. This approach was chosen because the number of circulating
lymphoid cells depends on a large number of factors,
including redistribution of cells between marginating,
adhering, or emigrating pools. Furthermore, the measurement of half-times of circulating cells provides no
information about where cells are homing, whereas the
determination of cells accumulating in the different
organs gives a better indication of local cell uptake.
Because it has been shown in vivo that removal of
sympathetic noradrenergic transmission has no measurable influence on the size of splenic cell compart-
DISCUSSION
Our results demonstrate that the uptake of circulating lymphocytes by the spleen is favored by a selective,
local increase in splenic perfusion induced by splenic
denervation and by the bacterial endotoxin LPS. Furthermore, we report here that general vasodilatation
J Appl Physiol • VOL
Fig. 5. Relation between blood flow and cell uptake after local
splenic denervation (local denerv), general depletion of norepinephrine stores (reserp), administration of endotoxin [lipolysaccharide
(LPS)], or their combination [(LPS ⫹ local denerv) and (LPS ⫹
reserp)]. Values are means ⫾ SE for 6 animals per point.
94 • FEBRUARY 2003 •
www.jap.org
Downloaded from http://jap.physiology.org/ by 10.220.33.4 on April 13, 2017
Fig. 4. Local organ denervation and general norepinephrine depletion differentially affect organ blood flow. After local splenic denervation (local denerv), blood flow increased significantly only in the
spleen, whereas the perfusion of other organs, such as the liver,
remained unchanged. After interruption of the noradrenergic transmission by reserpine (reserp), a higher portion of the heart minute
volume is directed to the liver than to the spleen. Control (ctrl), blood
flow in the organ of normal rats. Values are means ⫾ SE for 6
animals per group. *** P ⬍ 0.001.
473
474
BLOOD FLOW AND IMMUNE CELL TRAFFIC
J Appl Physiol • VOL
extrasplenic circulation might be also relevant for the
dynamics of uptake of particulate antigenic material.
The present data may have clinical implications.
Vasodilating drugs and physiological or pathophysiological conditions leading to general vasodilatation
would reduce splenic uptake of cells or circulating
antigenic material and therefore interfere with the
protective function of the spleen. For example, during
physical stress, blood flow decreases preferentially in
the spleen because it is one of the most densely noradrenergically innervated peripheral organs, whereas
other organs are not affected so much or are even more
perfused, such as skeletal muscle at work (7, 8). The
present results predict that, under this condition, the
contact of circulating cells with splenic tissue is reduced. Such interpretation is supported by recent experiments showing that after prolonged ␣-adrenergic
stimulation, the number of leukocytes increases in
blood circulation and decreases in the spleen (24).
It can be concluded from our results that after the
redistribution of the heart minute volume during intense muscular work, i.e., in a condition where NE
levels increase and skeletal muscle vasodilatation prevails, not only cell uptake, but also trapping of circulating antigenic material in the spleen, is reduced. This
condition may contribute to the impairment of immune
defense observed after exhausting physical training, a
situation during which the blood flow of the muscle is
increased because of high metabolic demands but
splenic blood flow is reduced because of sympathetic
activation (13).
In conclusion, our results stress the relevance of
hemodynamic forces controlled by the sympathetic nervous system for cell and antigen uptake by the spleen.
Immune processes that cause only a local increase in
blood flow would favor splenic cell uptake and immune
defense. On the contrary, general vasodilatation would
interfere with the capacity of the spleen to extract cells
from the circulation and thus interfere with splenic
immune functions.
We appreciate the expert technical assistance of Sigrid Petzoldt in
the performance of the studies.
This work was supported by a grant from the Deutsche Forschungsgemeinschaft (Sonderforschungsbereich 297, Project B2) to
H. Rogausch and A. del Rey.
REFERENCES
1. Besedovsky HO and del Rey A. Immune-neuro-endocrine
interactions: facts and hypotheses. Endocr Rev 17: 64–102, 1996.
2. Binns RM, Licence ST, and Pabst R. Homing of blood,
splenic, and lung emigrant lymphoblasts: comparison with the
behavior of lymphocytes from these sources. Int Immunol 4:
1011–1019, 1992.
3. Brigden ML, Pattullo A, and Brown B. Pneumococcal vaccine
administration associated with splenectomy: the need for improved education, documentation, and the use of a practical
checklist. Am J Hematol 65: 25–29, 2000.
4. Carlson SL, Fox S, and Abell KM. Catecholamine modulation
of lymphocyte homing to lymphoid tissue. Brain Behav Immun
11: 307–320, 1997.
5. Felten DL. Neural influence on immune responses: underlying
suppositions and basic principles of neural-immune signaling.
Prog Brain Res 122: 381–389, 2000.
94 • FEBRUARY 2003 •
www.jap.org
Downloaded from http://jap.physiology.org/ by 10.220.33.4 on April 13, 2017
ments, the rate of cell proliferation, or apoptosis (6), it
is unlikely that these processes would influence the
number of cells uptaken by the spleen. The interruption of the noradrenergic transmission is frequently
used to detect autonomic nervous influences on immune functions (for reviews, see Refs. 5, 16). The
nearly complete abrogation of the noradrenergic vasoconstrictor influence, most likely mediated by inhibition of NE release after administration of LPS to animals with an intact innervation, results in an increase
in blood flow that is close to that induced by sympathectomy (19, 20). Thus the effect of denervation can be
considered as a situation reflecting what would occur
under more physiological conditions. An increase in
blood flow similar to that caused by denervation is also
noticed during local inflammatory processes and in the
local hyperemia that precedes specific immune responses. Our results indicate that, under these conditions, the locally increased blood flow would direct the
cells to the sites of immune defense.
The fact that splenic blood flow increases in a comparable magnitude both after local and general interruption of noradrenergic sympathetic transmission
and after LPS, but that only local denervation or LPS
administration results in increased splenic cell uptake,
may be related to the particular function of the spleen,
which can be considered as a filter inserted in the
arterial circulation. This view is briefly discussed below.
There are still controversial results about adhesion
and recognition of circulating leukocytes at the endothelial lining of blood vessels and about the effects of
NE and sympathetic nerves on these processes (4, 10,
11, 17, 24, 26). An argument against cell sorting at the
site of entrance into the spleen is the finding that
memory cells and cytotoxic effector T lymphocytes are
migrating in comparable numbers into the spleen,
whereas they are differentially extracted from the circulating pool into other lymphoid organs (25).
Our results indicate that the absolute level of splenic
perfusion (ml/min) is not the only variable determining
the number of cells that are trapped by the spleen,
because it also depends on the distribution of blood flow
within peripheral organs. We showed that normal
spleen perfusion represents ⬃2% of the total heart
minute volume and that it increases to ⬃4% after local
denervation. After general interruption of sympathetic
noradrenergic transmission, the spleen receives ⬃2%
of the heart minute volume, i.e., not more than when
sympathetic innervation is intact. This may explain
why during general vasodilatation not more cells are
taken up by the spleen than under control conditions,
despite the fact that splenic perfusion is higher than
under basal conditions. The increased blood flow in
large parenchymatous organs may diminish the
chance of circulating lymphocytes to contact the
splenic meshwork. After contacting endothelial cells,
only 4–6 of 10 lymphocytes emigrate from the bloodstream (9), and prolonged circulation through extrasplenic pathways during general vasodilatation may
further reduce splenic cell uptake. Such a prolonged
BLOOD FLOW AND IMMUNE CELL TRAFFIC
J Appl Physiol • VOL
18. Rogausch H, del Rey A, Kabiersch A, and Besedovsky HO.
Interleukin-1 increases splenic blood flow by affecting the sympathetic vasoconstrictor tonus. Am J Physiol Regul Integr Comp
Physiol 268: R902–R908, 1995.
19. Rogausch H, del Rey A, Kabiersch A, Reschke W, Örtel J,
and Besedovsky HO. Endotoxin impedes vasoconstriction in
the spleen: role of endogenous interleukin-1 and sympathetic
innervation. Am J Physiol Regul Integr Comp Physiol 272:
R2048–R2054, 1997.
20. Rogausch H, del Rey A, Örtel J, and Besedovsky HO.
Norepinephrine stimulates lymphoid cell mobilization from the
perfused rat spleen via ␤-adrenergic receptors. Am J Physiol
Regul Integr Comp Physiol 276: R724–R730, 1999.
21. Sexton WL and Poole DC. Costal diaphragm blood flow heterogeneity at rest and during exercise. Respir Physiol 101: 171–
182, 1995.
22. Shephard RJ, Gannon G, Hay JB, and Shek PN. Adhesion
molecule expression in acute and chronic exercise. Crit Rev
Immunol 20: 245–266, 2000.
23. Splenectomy. Patient Care Committee of the Society for
Surgery of the Alimentary Tract (SSAT). Practice guideline.
J Gastrointest Surg 3: 218–219, 1999.
24. Stevenson JR, Westermann J, Liebmann PM, Hortner M,
Rinner I, Felsner P, Wofler A, and Schauenstein K. Prolonged alpha-adrenergic stimulation causes changes in leukocyte distribution and lymphocyte apoptosis in the rat. J Neuroimmunol 120: 50–57, 2001.
25. Weninger W, Crowley MA, Manjunath N, and von Adrian
UH. Migratory properties of naive, effector, and memory CD8⫹
T cells. J Exp Med 194: 953–966, 2001.
26. Westermann J and Pabst R. How organ-specific is the migration of “naive” and “memory” T lymphocytes? Immunol Today 17:
278–282, 1996.
27. Willführ KU, Westermann J, and Pabst R. Absolute numbers
of lymphocyte subsets migrating through the compartments of
the normal and transplanted rat spleen. Eur J Immunol 20:
903–911, 1990.
28. Young AJ. The physiology of lymphocyte migration through the
single lymph node in vivo. Semin Immunol 11: 73–83, 1999.
94 • FEBRUARY 2003 •
www.jap.org
Downloaded from http://jap.physiology.org/ by 10.220.33.4 on April 13, 2017
6. Herklotz J and Westermann J. Local immune responses in
the spleen: regulation by the nervous system and effects on
behavior (Abstract). Immunobiology 204: 380, 2001.
7. Laughlin MH, Armstrong RB, White J, and Rouk K. A
method for using microspheres to measure muscle blood flows in
exercising rats. J Appl Physiol 52: 1629–1635, 1982.
8. Laughlin MH, Korthuis RJ, Sexton WL, and Armstrong
RB. Regional muscle blood flow capacity and exercise hyperemia
in high-intensity trained rats. J Appl Physiol 64: 2420–2427,
1988.
9. MacDonald IC, Ragan DM, Schmidt EE, and Groom AC.
Kinetics of red cell passage through interendothelial slits into
venous sinuses in rat spleen, analysed by in vivo microscopy.
Microvasc Res 33: 118–134, 1987.
10. Mackay CR, Kimpton WG, Brandon MR, and Cahill RN.
Lymphocyte subsets show marked differences in their distribution between blood and the afferent and efferent lymph of peripheral lymph nodes. J Exp Med 167: 1755–1765, 1988.
11. Mackay CR, Marston WL, and Dudler L. Naive and memory
T cells show distinct pathways of lymphocyte recirculation. J
Exp Med 171: 801–817, 1990.
12. Madden KS, Sanders VM, and Felten DL. Catecholamine
influences and sympathetic neural modulation of immune responsiveness. Annu Rev Pharmacol Toxicol 35: 417–448, 1995.
13. Nagatomi R, Kaifu T, Okutsu M, Zhang X, Kanemi O, and
Ohmori H. Modulation of the immune system by autonomic
nervous system and its implication in immunological changes
after training. Exerc Immunol Rev 6: 54–74, 2000.
14. Naredi P, Mattson J, Hafström L, and Jacobsson L. Evaluation of blood flow measurements with microspheres and rubidium—an experimental study in rats. Int J Microcirc Clin Exp
9: 423–437, 1990.
15. Pabst R. The spleen in lymphocyte migration. Immunol Today
9: 43–45, 1988.
16. Pabst R and Binns RM. Heterogeneity of lymphocyte homing
physiology. Immunol Rev 108: 83–109, 1989.
17. Reynolds JD, Chin W, and Shmoorkoff J. T and B cells have
similar recirculation kinetics in sheep. Eur J Immunol 18: 835–
840, 1988.
475