| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Published online before print November 21, 2003
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Departments of
* Ophthalmology and
Biomedical Physics and Bioengineering, Aberdeen University Medical School, Scotland, United Kingdom
1 Correspondence: Department of Ophthalmology, Aberdeen University Institute of Medical Sciences, Foresterhill, Aberdeen AB25 2ZD, UK. E-mail: h.xu{at}abdn.ac.uk (H. X.) or i.j.crane{at}abdn.ac.uk (I. J. C.).
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Key Words: inflammation • leukocyte trafficking • blood flow • experimental autoimmune uveoretinitis • scanning laser ophthalmoscopy
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
A number of molecular and biophysical parameters are involved in the multistep process of leukocyte infiltration in inflammation [1 2 3 ]. Over the years, many studies have focused on adhesion-molecule regulation during the leukocyte-endothelium interactions. However, little is known about hydrodynamic factors, such as shear stress, which occur during this process. Hydrodynamic factors are thought to be important for leukocyte rolling and extravasation [4 , 5 ]. The rolling process of leukocytes involves a complex balance of forces arising from hydrodynamic shearing effects and the strength of the adhesive bond between leukocyte and endothelium [4 ]. Hydrodynamic force allows cell rolling to occur by driving the cells forward, and the specific adhesion force holds them at the margin of the vessel. The kinetics and mechanics of cell rolling are coupled, as force can influence the formation and dissociation of receptor-ligand bonds. The kinetic rates and their force dependence determine how likely, how rapidly, and how strongly cells bind as well as how long they remain bound.
Previous studies in mesenteric vessels have shown that shear stress cannot account for the difference in leukocyte rolling and adhesion between arterioles and venules [6 , 7 ], but it is negatively correlated with leukocyte rolling flux [8 ]. However, in the central nervous system (CNS), the circumstances are totally different, and shear stress can be tenfold higher. Blood flow in the CNS is largely independent of perfusion pressure and is autoregulated mainly by changes in vascular resistance. These changes are brought about by local chemical and endothelial factors as a result of trauma, ischaemia, seizures, and inflammation [9 ]. In addition, the blood-brain barrier in vascular endothelial cells of the CNS usually prevents transmigration of leukocytes under normal circumstances [10 , 11 ]. As the brain vasculature is difficult to study in vivo, little is known about the hydrodynamic changes and their influence on leukocyte trafficking during inflammation in the CNS.
The retina is an extension of the CNS, and its vessels possess barrier properties (BRB) identical to those of vessels elsewhere in the CNS. It is an ideal tissue in which to characterize the hemodynamic changes of the CNS during inflammation, as it is possible to visualize the entire retinal vasculature without causing any physical injury to the studied animal [12 ]. We have therefore examined the changes in hemodynamic factors and their relationship to leukocyte rolling, adherence, and infiltration in the retina as inflammation develops, using a noninvasive, in vivo leukocyte-tracking method, which uses scanning laser ophthalmoscopy (SLO). We initially developed this system in rats [13 ] and have adapted it more recently for mice [14 , 15 ]. Individual, labeled cells can be visualized in retinal vessels under physiological flow conditions, and the velocity of the leukocytes and shear stress can be accurately measured. To further enhance the physiological relevance of the findings at an early stage of the inflammatory process, we have tracked normal and in vivo antigen-primed leukocytes. We could therefore investigate how antigen priming relates to the ability of cells to adhere to and infiltrate the retina. Confocal microscopy of retinal flatmounts was used to determine whether labeled cells adhered to vessels or had migrated into the retina.
Experimental autoimmune uveoretinitis (EAU), a model for the sight-threatening human disease, endogenous posterior uveitis [16 ], was used in this study. EAU is a T cell-mediated autoimmune disease that can be induced in rodents by challenge with retinal antigen or its partial polypeptide [17 , 18 ] or by systemic administration of in vitro-activated T cells specific for a retinal antigen [19 , 20 ]. Disease is initiated as a result of sensitization of lymphocytes to retinal antigens, leading to the emigration of lymphocytes and macrophages, from the circulation into the neuroretina and structural damage to the retina [16 ].
Using this model and the physiological tracking of labeled leukocytes and determination of their subsequent extravasation, we have been able to show how the ability of leukocytes to cross the BRB is related to shear stress, activation of vascular endothelial cells, and leukocyte priming.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
EAU was induced in B10.RIII mice as described previously [21 22 23 ]. Briefly, mice were immunized subcutaneously with 50 µg human interphotoreceptor retinal-binding protein (IRBP) peptide 161–180 (SGIPYIISYLHPGNTILHVD; purity, >85%; Sigma, Cambridge, UK), emulsified with 50 µl Freund’s complete adjuvant (CFA; H37Ra; Difco Laboratories, Detroit, MI) in a total volume of 100 µl. Naive mice were untreated. Immunization with the same volume of phosphate-buffered saline (PBS) instead of IRBP peptide in CFA had no detectable effect clinically or histologically in the retina, and trafficking in these mice or using cells from these mice did not differ from that where animals were untreated (data not shown).
Animals were observed using an ophthalmic operating microscope and slit-lamp for clinical evaluation (grading) of the eye at day 3 postimmunization (p.i.) and daily from day 7 to day 21. Three mice at each time point (days 7, 9, 12, 15, and 20 p.i.) were killed, and their eyes were fixed in 2.5%-buffered glutaraldehyde (Fisher Chemicals, Leics, UK) and embedded in resin for standard hematoxylin and eosin (H&E) staining. The intensity of uveoretinitis was evaluated histologically and graded using a modified version of the customized histologic grading system [23 , 24 ] by independent observers.
Experimental design
Three groups of experiments were performed. Group 1: Spleen cells from naïve mice were injected into IRBP peptide-induced mice at days 0, 3, 7, 9, and 16 p.i. (nonactivated leukocyte group). Group 2: Spleen cells from IRBP peptide-immunized mice at days 3, 7, 9, and 16 p.i. were injected into peptide-immunized mice at days 3, 7, 9, and 16 p.i., respectively (primed leukocyte group). Group 3: Spleen cells from IRBP peptide-immunized mice at days 3, 7, 9, 12, and 16 p.i. were injected into naïve mice (nonactivated endothelium group). At each time point, three mice were used in the SLO study (except Group 2–day-9 p.i.-recipient mice, where n=6), and five mice were used in the confocal microscopy study.
SLO and image analysis
The technique of digital SLO in mice is described in detail elsewhere [14
, 15
]. Briefly, splenocytes were fluorescently labeled with calcein-AM (C-AM; Molecular Probes Europe BV, Leiden, The Netherlands) in vitro [15
]. Animals were anesthetized with intramuscular injection of 0.4 ml/kg Hypnorm (Janssen-Cilag Ltd., Belgium) and 1 ml/kg Diazepam (Phoenix Pharmaceuticals Ltd., Gloucester, UK) intraperitoneally. Pupils were dilated with Tropicamide (1 drop, 0.5% w/v, Chauvin Pharmaceuticals Ltd., Essex, UK) and a hard contact lens [15
] placed on the cornea to obtain a clear view of the fundus. Sodium fluorescein [Sigma; 100 µl 0.05% (v/v)], which has no effect on endothelial cell function, was injected via the tail vein to outline the vessels, followed by 1 x 107 C-AM-labeled cells in 150 µl PBS, and the animals were examined by SLO. SLO images were recorded simultaneously on videotape (S-VHS) and digitally at 25 frames per second. In each animal, three different areas adjacent to the optic disc were examined. At least 900 digital frames, within the first 30 min after injection of labeled cells, were intermittently recorded for image analysis.
Vessel diameters of retinal arteries and veins were measured by confocal microscopy within 1 mm of the optic disc [15
]. The velocities of individual fluorescent cells in each SLO image frame were calculated by computer program [15
]. The shear stresses of retinal arteries and veins were measured and calculated as described earlier [15
]. In brief, mean red blood cell velocity was calculated as Vmean = Vwhite blood cell/1.6. Wall shear rate (
) = 8(Vmean/diameter), and shear stress was x blood viscosity, where blood viscosity was assumed to be 0.025 poise. Identification of vessel types was based on direction of blood flow, size, and position [15
].
Leukocytes were classified according to their interaction with the endothelial lining as rolling cells and as adherent cells (sticking cells). Rolling cells were defined as those cells with a velocity below the critical velocity, as calculated by Vcrit = 
x 
x(2–), where = Vmax/(2–2), and is the ratio of the leukocyte diameter to vessel diameter [7
, 25
]. Sticking cells were defined as cells that remained firmly adherent to the endothelium for 20 s or longer. The rolling efficiency was calculated as the percentage of labeled, rolling cells among the total number of labeled leukocytes that entered a venule. The sticking efficiency was determined as the percentage of labeled leukocytes becoming firmly adherent compared with the total number of labeled leukocytes that rolled in a venule during the same time interval.
Instantaneous rolling velocity of individual cells was measured frame-by-frame, every 0.04 s, using MetaMorph image analysis system (Universal Imaging Corp., Downingtown, PA). Individual rolling velocity was measured by following cell displacements over 2- to 5-s intervals. Rolling velocities for cells within different vessels were obtained by measuring 10–20 leukocytes rolling along the same venule segment. These data were collected at a variety of shear stresses within different venules.
Retinal flatmounts for confocal microscopy
Retinal whole flat mounts were prepared according to the method of Chan-Ling [26
]. In brief, 50 min or 24 h after injection of labeled cells, animals were injected via the tail vein with 100 µl 2% (w/v) Evans Blue (Sigma). Animals were killed by CO2 inhalation 10 min later. The eyes were removed and were immediately immersed in 2% (w/v) paraformaldehyde (Agar Scientific Ltd., Cambridge, UK) for 1 h. The anterior segment of the globe was removed, and the retina was peeled from the choroid. Retinas were then spread on clean glass slides and mounted vitreous side-up under coverslips with Vectashield (Vector Laboratories, Burlingame, CA). Retinal flatmounts were analyzed by confocal scanning laser microscopy (Bio-Rad Microsciences MRC 1024, Hemel Hempstead, UK).
Flow cytometric analysis of cell-surface adhesion molecules
Splenocytes were stained with specific antibodies to mouse CD62L [L-selectin, rat immunoblobulin G (IgG)2a], CD44 (rat IgG2b), CD11a [lymphocyte functional antigen-1 (LFA-1), rat IgG2a], and CD162 [P-selectin glycoprotein ligand-1 (PSGL-1), rat IgG1] and their isotype controls (all were purchased from BD PharMingen, Oxford, UK) in 1% (w/v) bovine serum albumin/PBS at 4°C for 20 min. Flow cytometry was performed on a FACS Caliber (BD Biosciences, San Jose, CA) and was analyzed using CellQuest software (BD Biosciences).
Data analysis
Vessel diameters were quantified off-line at least three times for each vessel, and the means were used for further calculations. Vessels and cells were considered as independent variables, and mean ± SEM was calculated at each time point. When control and EAU mice were compared at each time point, probability values were calculated using Dunnett’s multiple comparison test. For rolling cells and sticking cells, we used the 
2 test. Probability values of P < 0.05 were considered significant.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
|
Vessel diameter and shear stress
In naïve mice, the average diameter of retinal vessels in SLO images was 57.30 ± 2.90 µm for retinal arteries and 74.06 ± 1.95 µm for retinal veins. In IRBP peptide-induced mice, at day 9 p.i., retinal veins were significantly dilated (diameter=99.69±6.16 µm, P<0.001 compared with normal mice). By day 16 p.i., the diameter of retinal veins had returned to normal size (73.01±6.96 µm). The diameter of retinal arteries did not change during the progress of EAU.
Shear stress in retinal veins was reduced in peptide-immunized mice from day 7 to day 16 p.i. (Fig. 3A ). However, at day 9 p.i., shear stress was significantly lower than at day 7 and day 16 p.i. (Fig. 3A , P<0.05). Again, in retinal arteries, there was no significant change in shear stress during the progress of EAU (Fig. 3A) .
|
The effect of shear stress on leukocyte rolling and sticking
Rolling cells were observed in retinal veins and PCVs of days 9 and 16 p.i.-recipient mice (to view cell rolling in retinal vessels, see our video report [14
], or visit our website at http://www.biomed.abdn.ac.uk/Abstracts/A00665/). No rolling leukocytes were seen before day 9 p.i. or at day 21 p.i. when the disease was resolving (data not shown). There was more cell rolling at day 9 p.i. than day 16 p.i. (Fig. 4A
and 4B
). To evaluate the effect of wall shear stress on cell rolling during inflammation, data for shear stress, cell rolling efficiency, and rolling velocity were collected from 23 venules in six day-9 p.i.-recipient mice. Linear regression analysis shows that rolling efficiency is negatively correlated with wall shear stress (Fig. 4C) . However, the mean rolling velocity shows no significant negative or positive correlation over this range of shear stresses (Fig. 4D)
.
|
The effect of leukocyte activation on cell rolling and infiltration
Change of velocities—comparison of naïve and IRBP peptide-primed leukocytes
When cells from peptide-immunized mice were injected into naïve mice, no rolling cells were observed, indicating the importance of an activated endothelium. There was no significant difference in velocity between naïve cells and cells from mice at any of the time points p.i. in any of the vessel types (Fig. 5A
). However, in PCVs in IRBP peptide-immunized recipient mice at day 7 p.i., IRBP peptide-primed cells moved significantly slower than naïve cells (Fig. 5B)
, although in other vessels (Fig. 5B)
and at days 3, 9, and 16 p.i. (data not shown), these two cell populations moved at similar speeds.
|
|
Infiltration—comparison of naïve and IRBP peptide-primed leukocytes
Cells which had entered the retinal parenchyma by transendothelial migration (TEM) were studied using confocal microscopy 1 h and 24 h after cell injection, and the relationship of TEM to antigen priming of leukocytes and endothelial cell activation was determined. Retinas were examined at 1 h and 24 h after infusion of cells, as it had been reported that TEM could occur at 24 h [27
].
One hour after adoptive transfer of naïve cells into IRBP peptide-immunized mice at days 7 and 9 p.i., all of the fluorescently labeled cells were located intravascularly (mostly, in the venules and PCVs). Even in day-9 p.i. recipients, where the mice showed severe disease in the retina (Fig. 7A ), no leukocyte trafficking of nonprimed cells into retinal tissues could be detected 1 h after cell infusion. In contrast, adoptively transferred IRBP peptide-primed cells from mice with EAU (day 9 p.i.) marginated along the vessel wall and even showed early TEM into retinal tissue 1 h after cell injection into day-9 recipients (Fig. 7B) . This was identified by a marked shape change in the cell to a polarized morphology, which appeared also to develop pseudopodial extensions (see Fig. 7B inset). However, by day 16 p.i., even naïve cells could be seen to move freely into retina tissue from retinal venules just 1 h after injection (Fig. 7C) .
|
|
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Shear stress in retinal veins and venules was already reduced 24 h before leukocyte infiltration, which occurred at day 8 p.i., and this reduction was more marked with increased severity of disease (day 9 p.i.). In the veins and PCVs, the shear stress reduction was accompanied by a reduction in leukocyte velocity at day 7. Shear stress reduction was not a result of dilation of the retinal veins, as at day 7 p.i., these were not significantly dilated. Factors such as up-regulation of endothelial cell adhesion molecules and/or a change in blood components, fibrinogen and thrombin, which are part of the acute-phase response at this time point, may well be involved.
Walpola et al. [5 , 28 ] showed that low shear stress was an adequate stimulus to induce monocyte adhesion and migration and up-regulate vascular cell adhesion molecule-1 expression on endothelial cells and endothelial cell desquamation in rabbit carotid arteries. Previously, we have shown that in EAU at day 7 p.i., adhesion molecules, intercellular adhesion molecule-1 (ICAM-1) and P-selectin, were significantly increased on the endothelial cells of retinal veins and venules [22 ]. This correlates with the reduction of shear stress in retinal veins in the current study, and whether they are directly linked needs further investigation.
Previous studies in mesenteric vessels, where very different conditions to the CNS prevail and where, in particular, shear stresses are much lower have shown that leukocyte rolling is inversely related to the wall shear stress [8 ]. In this study in the retinal circulation, we found that not only rolling but also firm adhesion of leukocytes were negatively correlated with wall shear stress, indicating that shear stress influences leukocyte behavior at the BRB and is biologically significant during inflammation. Shear stress, however, did not correlate with leukocyte velocity at this stage (day 9 p.i.) of the disease. The cell velocity is already reduced by this time, and an appropriate shear stress, which enables the leukocyte to roll and sample the endothelial, is critical.
However, reduction of shear stress is only one of the factors influencing leukocyte trafficking at the BRB. Leukocytes rolled in day-16 p.i. IRBP-immunized mice but not in day-7 p.i. mice, although the shear stresses were almost the same at these two time points, indicating that other factors such as adhesion molecules are also important to initiate leukocyte rolling. We have shown before that at day 16 p.i. in IRBP-immunized mice, retinal vessels expressed higher ICAM-1, P-selectin, and E-selectin than at day 7 p.i. [22 ].
Leukocyte rolling was not observed in the retinal vessels of naïve mice but only in the mice with EAU. In these mice with EAU (day 9 p.i.), naïve and IRBP-primed cells displayed rolling behavior in retinal veins and venules, suggesting that activation of endothelial cells is important and a prerequisite for inflammatory cell recruitment in the retina. This is in agreement with a previous report [29 ]. However, the rolling velocity of IRBP-primed cells was much less than for naïve cells. Furthermore, time-dependent reduction of rolling velocity was observed in IRBP-primed cells but not in naïve cells during the rolling process, indicating that naïve and primed cells differ in their rolling behavior. The velocity of rolling leukocytes is thought to be determined by the expression of adhesion molecules as well as the prevailing shear stress [4 ]. However, in this case, the shear stress and endothelial cell adhesion molecule expression at day 9 p.i. EAU were the same for primed and naïve cells. The increased expression of adhesion molecules such as CD44, PSGL-1, and LFA-1 (Table 2) on the IRBP-primed cells is more likely to account for the reduced rolling velocity of these cells.
A decreased rolling velocity increases the transit time of rolling leukocytes, which has recently been shown to be an important determinant of leukocyte recruitment in vivo [30 ]. When cells roll more slowly, they have more time to be further activated by chemokines secreted or presented by the endothelium and to become firmly adhesive and finally migrate into retinal tissues. Leukocyte chemokines and cytokines and their receptors will also enhance these interactions. The effect of reduced rolling velocity on adherence of the leukocytes to activated endothelium is shown in our study. Far fewer naïve cells than primed cells arrested on retinal vessels when injected into a day-9 p.i. recipient, although the rolling cell number was the same as that of primed cells, as their rolling velocity was significantly higher than that of the primed cells.
When the number of cells that had migrated into the retina was assessed after 24 h, although a small number of naïve cells had migrated into the retinal parenchyma of day-8 p.i. mice, and this number had increased at day 9 p.i., the number of IRBP-primed cells that had migrated was much greater at both time points, further supporting the notion that priming of leukocytes is important for leukocyte recruitment. It is possible that selective retention rather than selective recruitment accounts for some of the increase in antigen-primed cells found in the retinal parenchyma 24 h after cell infusion, but the increase in these cells in the vessels only 1 h after cell infusion indicates that leukocyte retention cannot be the whole explanation. Potential mechanisms for homing of antigen-primed cells to the eye are currently under investigation in our laboratory.
In conclusion, this in vivo pathophysiological study has shown how reduction in shear stress in combination with activation of the endothelium and priming of leukocytes are prerequisites for leukocyte passage across the BRB into the retina as an inflammatory reaction develops.
|
ACKNOWLEDGEMENTS |
|---|
|
FOOTNOTES |
|---|
Received October 7, 2002; revised July 31, 2003; accepted October 7, 2003.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  H. Xu, A. Manivannan, I. Crane, R. Dawson, and J. Liversidge Critical but divergent roles for CD62L and CD44 in directing blood monocyte trafficking in vivo during inflammation Blood, August 15, 2008; 112(4): 1166 - 1174. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. Xu, C. L. Cheng, M. Chen, A. Manivannan, L. Cabay, R. G. Pertwee, A. Coutts, and J. V. Forrester Anti-inflammatory property of the cannabinoid receptor-2-selective agonist JWH-133 in a rodent model of autoimmune uveoretinitis J. Leukoc. Biol., September 1, 2007; 82(3): 532 - 541. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
I. J. Crane, H. Xu, C. Wallace, A. Manivannan, M. Mack, J. Liversidge, G. Marquez, P. F. Sharp, and J. V. Forrester Involvement of CCR5 in the passage of Th1-type cells across the blood-retina barrier in experimental autoimmune uveitis J. Leukoc. Biol., March 1, 2006; 79(3): 435 - 443. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 O. O. Ogunshola, V. Djonov, R. Staudt, J. Vogel, and M. Gassmann Chronic excessive erythrocytosis induces endothelial activation and damage in mouse brain Am J Physiol Regulatory Integrative Comp Physiol, March 1, 2006; 290(3): R678 - R684. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. Xu, A. Manivannan, R. Dawson, I. J. Crane, M. Mack, P. Sharp, and J. Liversidge Differentiation to the CCR2+ Inflammatory Phenotype In Vivo Is a Constitutive, Time-Limited Property of Blood Monocytes and Is Independent of Local Inflammatory Mediators J. Immunol., November 15, 2005; 175(10): 6915 - 6923. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. R. Ritter, E. Aguilar, E. Banin, L. Scheppke, H. Uusitalo-Jarvinen, and M. Friedlander Three-Dimensional In Vivo Imaging of the Mouse Intraocular Vasculature during Development and Disease Invest. Ophthalmol. Vis. Sci., September 1, 2005; 46(9): 3021 - 3026. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. Xu, R. Dawson, I. J. Crane, and J. Liversidge Leukocyte Diapedesis In Vivo Induces Transient Loss of Tight Junction Protein at the Blood-Retina Barrier Invest. Ophthalmol. Vis. Sci., July 1, 2005; 46(7): 2487 - 2494. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
