| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
|
FJ
EXPRESS SUMMARY ARTICLE The Full-length version of this article is also available, published online May 20, 2003 as doi:10.1096/fj.02-0577fje. |
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
* Neurosciences Research Unit, The Canberra Hospital;
The Australian National University Medical School, and
The John Curtin School of Medical Research, The Australian National University, Canberra, Australia
2Correspondence: Neurosciences Research Unit, Bldg. 10, Level 6, Yamba Dr., Garran, Australian Capital Territory 2606 Australia. E-mail: maria.staykova{at}anu.edu.au
SPECIFIC AIM
The ability of nitric oxide (NO) to down-regulate both actively induced and transferred autoimmune encephalomyelitis (EAE) suggests a number of inhibitory mechanisms for this highly reactive molecule. One mechanism examined is the possible effect of NO on T lymphocyte migration.
PRINCIPAL FINDINGS
1. Inhibition of T lymphoblast migration by NO
Antigen-activated T lymphocytes (lymphoblasts) readily migrate through collagen-coated transwell membranes and rat brain endothelial cell monolayers (RBEC) grown on collagen-coated transwells. During a 4 h migration assay, the NO donor 1-hydroxy-2-oxo-3, 3-bis (2-aminoethyl)-1-triazene (HOBAT), an amine/NO adduct that releases two equivalents of NO per mole, inhibited migration of 51Cr-labeled T cells across collagen-coated transwells in a dose-dependent manner by as much as 60% (Fig. 1
A). The migration of T lymphoblasts through confluent layers of RBEC was also inhibited (45%) by HOBAT (Fig. 1B
) as was migration through "inflamed," IFN-
-treated endothelium. To determine whether NO was in fact the active molecule inhibiting migration, hemoglobin was added as a NO scavenger, which prevented the HOBAT-induced inhibition (Fig. 1C
).
|
2. NO does not inhibit migration of T lymphoblasts by inducing cell death
NO at high concentrations can induce apoptosis and/or necrosis in many cell types, including myelin reactive T cells. A 4 h exposure of T lymphoblasts to HOBAT concentrations of 50, 100, and 200 µM increased the percentage of annexin V positive cells by 6, 8, and 10%, respectively, compared with untreated cells. Necrotic cells, as determined by propidium iodide staining, increased by 1–2% at these concentrations. At HOBAT concentrations of 400 µM and above, there was an increase of 14–30% in annexin V positive apoptotic cells. Thus, NO-induced apoptosis of T cells occurred only at high HOBAT concentrations and thus contributed to migration inhibition. At lower concentrations, however, the observed in vitro effect on T lymphoblast migration cannot be attributed merely to cell death.
3. NO has no effect on adhesion molecule expression, tight junction integrity, or stress fiber formation in endothelial cells
NO must be acting on the T lymphoblasts in the migration across collagen membranes. Migration across RBEC, however, may be influenced by an effect of NO on the endothelial cells as well as the T cells. We therefore examined the effect of NO on three endothelial cell phenotypic markers that may influence cell migration: expression of the adhesion molecules ICAM-1 and VCAM-1, tight junction integrity, and stress fiber formation. Exposure of resting or IFN--treated RBEC to HOBAT concentrations of 100 and 200 µM for 4 h resulted in no change in the expression of ICAM-1 or VCAM-1. The tight junctions in confluent layers of RBEC and Madin-Darby canine kidney cells (based on the immunofluorescent pattern of the tight junctional protein ZO-1) were not altered by exposure to NO. Phalloidin staining of these cells treated with 100 or 200 µM HOBAT showed no change in F-actin, but at higher concentrations (400 µM) the development of numerous stress fibers was seen.
HOBAT pretreatment of T cells inhibited their migration through untreated endothelial cells whereas pretreatment of endothelial cells did not inhibit migration of untreated T cells. Thus, the migration inhibitory action of NO appears to be mediated solely through its effect on T cells and not endothelial cells.
4. NO alters T lymphoblast morphology
T lymphoblasts were either untreated or treated with 50, 100, or 200 µM HOBAT for 3 h, fixed, stained for F-actin with phalloidin, and examined using confocal microscopy. Figure 2
A depicts the typical appearance of activated T lymphoblasts, showing an amoeboid shape and the extension of variable numbers of membrane processes such as lamellipodia and filopodia containing polymerized actin. Approximately 90% of activated cells have this morphology. After treatment with HOBAT, the F-actin becomes concentrated in one or two particular regions and the cells take on a distinctly polarized shape with a leading edge and trailing uropod (Fig. 2B, C
). There was a dose response in the ability of HOBAT to cause these morphological changes (Fig. 2)
. The NO dependence of this effect was shown when the change in the morphology was inhibited by the presence of hemoglobin in the culture media (Fig. 2)
. A similar change in F-actin in response to chemoattractants has been described during the migration of neutrophils but not to our knowledge in T lymphocytes.
|
5. p70S6K activity is not required for the NO-induced morphological changes in T cells or for migration inhibition by NO
Several studies from one of our labs have shown a role for p70S6K in cytoskeleton regulation and possibly cell migration at least with regard to Swiss 3T3 fibroblasts. We therefore examined the activity of p70S6K in untreated vs. NO-treated T lymphoblasts. We measured enzyme activity in antigen-activated (migratory) T cells and postactivated (nonmigratory) T cells, and there was no difference in p70S6K activity. When HOBAT was added to antigen-activated T cells at concentrations that inhibited migration, again there was no alteration in p70S6K activity. To further examine a role for p70S6K in NO-induced migration inhibition, we cultured activated T cells with 100 nM rapamycin, a p70S6K specific inhibitor, for 4 h with or without concomitant exposure to HOBAT. Rapamycin treatment did not prevent the NO-induced polarization of the T cells nor did it reverse the NO-induced migration inhibition.
DISCUSSION
When injected intravenously into susceptible animals, antigen-activated encephalitogenic T cells migrate to the CNS and transmigrate through or between CNS vascular endothelial cells into brain parenchyma, where they recognize antigen, recruit other leukocytes, and set up inflammatory foci leading to neurological dysfunction. We had previously reported that intravenous injection of activated T cells on a background of high NO levels in the recipients results in greatly diminished inflammation and subsequently reduced clinical signs of EAE. Here we report for the first time that in vitro NO inhibits T lymphoblast migration through collagen matrices and RBEC monolayers. These data suggest that in vivo inhibition of passive EAE may be due at least in part to a NO-induced inhibition of T cell migration.
NO treatment appears to act solely on the T cells in our trans-endothelial migration assay, as treatment of endothelial cells had no effect on the expression of the major adhesion molecules ICAM-1 and VCAM-1 or on tight junction integrity.
An essential requirement in lymphocyte migration is the acquisition of a polarized morphology that allows the cell to use intracellularly generated forces to produce motility. An early event in polarization is a change in F-actin distribution from radial symmetry to its concentration in a particular region. In our model the lymphocytes are essentially in a solution exposed to NO as they settle onto the endothelial surface. We suggest that exposure to NO leads to a rapid polarization of the cells, perhaps occurring in seconds to minutes, as described for the onset of migration after chemoattractant-induced migration of T cells. The result of such polarization into what appear to be migrating cells may therefore predefine a limited number of directions in which the cell may migrate whereas the untreated activated cells maintain a variable number of lamellae around the perimeter of the cell, any one of which can become dominant and direct migration when stimulated by contact, chemoattractants, or NO produced locally by endothelial cells (Fig. 3
). Polarization of lymphocytes has also been shown to dramatically alter the distribution of adhesion molecules and chemoreceptors with adhesion molecules localized to the uropod and chemoreceptors concentrated in the leading edge. This could contribute to decreased migration by decreasing the chance of adhesive surfaces on the cells contacting endothelium.
|
A similar process may occur in vivo. Activated encephalitogenic T cells injected into a recipient with high NO levels would rapidly polarize into a migratory phenotype and decrease the chances of their randomly contacting endothelial cells and beginning directional migration. Where this polarization occurs in vivo is not known. We do know that in the PVG rat the spleen has significantly more NO producing macrophages than do other strains of rats and that these macrophages contribute to the high RNI levels after CFA treatment. Intravenously injected activated T cells circulate through the spleen before migrating to the CNS; it is possible that in a milieu of high local concentrations of NO they become polarized in the spleen, then re-enter the circulation. Two experiments were done in which HOBAT was removed after the 3 h incubation and the polarized cells did not revert to their prior morphology at least during the next 5 h. Thus the NO-induced polarization of the T lymphoblasts appears to be more than transient.
On a molecular level, the biochemical basis for the majority of NO's physiological actions is the activation of guanylate cyclase, which increases synthesis of cGMP. In earlier studies from one of our labs (M.F.C.) we showed that NO enhanced p70S6K activity and stress fiber formation in Swiss 3T3 fibroblasts by induction of cGMP-activated protein kinase. p70S6K colocalized with the stress fibers, suggesting that fiber formation may be regulated at least in part by translocation of these proteins to the cytoskeleton. In the present work we found no change in the activity of p70S6K in NO-treated T lymphoblasts nor did the enzyme inhibition by rapamycin inhibit F-actin polarization in the NO-treated cells. From these data we conclude that NO-provoked cytoskeletal changes in T lymphoblasts are mediated by a p70S6K-independent mechanism.
In conclusion, we have shown that NO induces F-actin driven polarization of T lymphoblasts and significantly reduces the migration of such cells. We suggest that the morphological changes result in cells with predefined migratory directionality, limiting the cells' ability to respond to random migratory signals.
FOOTNOTES
1 To read the full text of this article, go to http://www.fasebj.org/cgi/doi/10.1096/fj.02-0577fje; doi: 10.1096/fj.02-0577fje 
This article has been cited by other articles:
|
|
 |
|
  M. A. Staykova, J. T. Paridaen, W. B. Cowden, and D. O. Willenborg Nitric Oxide Contributes to Resistance of the Brown Norway Rat to Experimental Autoimmune Encephalomyelitis Am. J. Pathol., January 1, 2005; 166(1): 147 - 157. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
