- Email: [email protected]
Upregulation of intercellular adhesion molecule-1 (ICAM-1) expression in primary cultures of human brain microvessel endothelial cells by cytokines and lipopolysaccharide
Journal of Neuroimmunology, 39 (1992) 11-22 © 1992 Elsevier Science Publishers B.V. All rights reserved 0165-5728/92/$05.00
11
JNI 02193
Upregulation of intercellular adhesion molecule-1 (ICAM-1) expression in primary cultures of human brain microvessel endothelial cells by cytokines and lipopolysaccharide D ona l d W ong and Katerina Dorovini-Zis Department of Pathology, Section of Neuropathology, the University of British Columbia and VancouL,er General Hospital, Vancouver, B.C., Canada (Received 11 November 1991) (Revised received 3 January and 10 February 1992) (Accepted 10 February 1992)
Key words: Cerebral endothelium; Intercellular adhesion molecule-I; Cytokine; Lipopolysaccharide
Summary The expression of intercellular adhesion molecule-1 (ICAM-1) by human cerebral endothelium was studied in primary cultures of human brain microvessel endothelial cells following treatment with bacterial lipopolysaccharide (LPS), tumor necrosis factor-a (TNF-a), interleukin-1/3 (IL-1/3) and interferon- 7 (IFN-y). Surface expression of ICAM-1 was examined with the immunogold silver staining technique. Intact cerebral endothelial cells constitutively express low levels of ICAM-1. Stimulation with LPS and cytokines induces upregulation of ICAM-1 which is minimal with IFN-7 and maximal with LPS or a combination of IFN-7 and TNF-a. Upregulation of ICAM-1 expression is concentration- and time-dependent, is observed as early as 4 h following incubation and persists for up to 72 h in the continuous presence of LPS or cytokines. The ICAM-I expression is not reversed by 3 days after removal of the LPS or cytokines. These findings may be relevant to the interactions between leukocytes and brain microvessel endothelial cells in inflammatory and demyelinating diseases of the CNS.
Introduction Several disorders of the central nervous system (CNS) including infectious, inflammatory and autoimmune demyelinating diseases are characterized by the migration of acute and chronic inflammatory cells from blood into brain with inva-
Correspondence to: K. Dorovini-Zis, Department of Pathology, Section of Neuropathology, Vancouver General Hospital, 855 West 12th Avenue, Vancouver, B.C., Canada V5Z 1M9.
sion of the extravascular tissue. The factors responsible for the regulation of leukocyte traffic across the cerebral endothelial barrier that normally excludes white blood cells from entering the brain, remain largely unknown. Recent studies on in vitro interactions between leukocytes and extracerebral endothelium suggest that de novo expression or upregulation of certain surface glycoproteins - adhesion molecules - - by endothelial cells mediates leukocyte adhesion to endothelium. Expression of intercellular adhesion molecules has been induced in vitro on extracere-
12 bral small and large vessel endothelial cells by cytokines and bacterial lipopolysaccharide (Albelda and Buck, 1990; Pober and Cotran, 1990, 1991) and in vivo following intradermal injection of endotoxin (Munro et al., 1991). A number of endothelial adhesion molecules have been described and characterized, including intercellular adhesion molecule-1 (ICAM-1), endothelial leukocyte adhesion molecule-1 (ELAM-1) and vascular cell adhesion molecule-1 (VCAM-1). ICAM-1 is a member of the immunoglobulin gene superfamily. It binds to leukocyte functionassociated molecule 1 (LFA-1), an integrin on the surface of neutrophils, lymphocytes and monocytes, and has been shown to play a role in the adhesion of polymorphonuclear leukocytes and lymphocytes to cultured endothelium (Albelda, 1991; Pober and Cotran, 1990, 1991). ICAM-1 is expressed at low levels basally by human umbilical vein, dermal and retinal microvessel endothelial cells in culture (Pober et al., 1986, 1987; Detmer et al., 1990; Liversidge et al., 1990; Ruszczak et al., 1990; Wellicome et al., 1990; Swerlick et al., 1991). Following activation by interleukin-1 (IL-1), tumor necrosis factor-a (TNF-a), lipopolysaccharide (LPS) or interferon-7 (IFN-7), ICAM-1 expression reaches maximal levels after 24 h (Albelda and Buck, 1990; Pober and Cotran, 1990; Springer, 1990). Differences in its expression have been observed among endothelial cells from different vascular beds. In the nervous system, the role of ICAM-1 in the initiation of the inflammatory response has not yet been fully investigated. Expression of ICAM-1 and the addressin MECA-325 by CNS endothelium has been implicated in the adhesion of lymphocytes to endothelial cells during the acute and chronic relapsing phases of experimental allergic encephalomyelitis, the animal model of multiple sclerosis (Raine et al., 1990; Wilcox et al., 1990; Cannella et al., 1991). In addition, immunolocalization of ICAM-1 in human cerebrovascular endothelium has been reported in necrotic and inflammatory CNS lesions (Sobel et al., 1990). The results of these studies suggest that the cerebral endothelium, through its ability to express certain adhesion proteins that interact with corresponding leukocyte membrane molecules, may play an important role in modu-
lating leukocyte traffic across the blood-brain barrier. In the present study, we investigated the expression of ICAM-1 by human brain microvessel endothelial cells (HBMEC) in primary culture following stimulation with LPS, IFN-7, TNF-a and IL-1/3. Previous studies have shown that HBMEC in culture form confluent monolayers that retain important endothelial and blood-brain barrier characteristics including paucity of cytoplasmic vesicles and presence of tight junctions that restrict the intercellular movement of macromolecules (Dorovini-Zis et al, 1991). Intact HBMEC showed minimal basal expression of ICAM1. Incubation with LPS and cytokines led to upregulation of ICAM-1 expression, which was minimal following IFN-7 stimulation and maximal after treatment with LPS or a combination of TNF-a and IFN-y. This expression was sustained in the continuous presence of LPS or cytokines in culture. Our results indicate that human cerebral endothelial cells are capable, upon activation, of expressing intercellular adhesion molecules for circulating leukocytes and may thus play an important role in the initiation of CNS inflammation.
Materials and methods
Isolation and culture of human brain microL'essel endothelial cells (HBMEC) Primary cultures of microvessel endothelial cells were established from human brains at autopsy and from temporal lobectomy specimens by methods previously described (Dorovini-Zis et al., 1991). The endothelial origin of the cells was determined by their strongly positive, granular, perinuclear immunofluorescence for Factor VIII antigen and binding of Ulex europaeus agglutinin. The isolated clumps of endothelial cells were seeded onto fibronectin-coated 24-well plates (Corning Plastics, Corning, NY) and maintained in culture in minimum essential medium alpha (aMEM, Gibco, Burlington, ON) supplemented with 10% horse plasma-derived serum (PDS, Hyclone Laboratories, Logan, UT), 25 mM Hepes, 10 mM sodium bicarbonate, 100 /zg/ml heparin, 20/xg/ml endothelial cell growth supplement (all
13 from Sigma, St. Louis, MO), 100 Izg/ml penicillin, 100 /zg/ml streptomycin, and 2.5 /xg/ml amphotericin B (Gibeo) at 37°C in a humidified 2.5% CO2/97.5% air atmosphere. Culture media were changed every other day. Endothelial cells reached confluency 7-9 days after plating and were used for studies on or after the 10th day.
ICAM-1 induction by cytokines and LPS Bacterial lipopolysaccharide (LPS from Escherichia coli O55:B5, Sigma), human recombinant interferon-y (IFN-7, Collaborative Research Inc., Bedford, MA), human recombinant tumor necrosis factor-a (TNF-a, Sigma) or human recombinant interleukin-1/3 (IL-1/3, Boehringer Mannheim, Laval, PQ) were dissolved in complete media at a final concentration of 0.001, 0.01, 0.1, 1, 5 /zg LPS/ml; 10, 50, 100, 500 U IFN-y/ml; 1, 10, 100 U TNF-a/ml; or 0.1, 1, 10 U IL-1/3/ml. 10-11-day-old cultures were incubated with different concentrations of LPS and cytokines for 4, 12, 2 4 , 4 8 and 72 h. In addition, separate wells were incubated with various concentrations of TNF-a and IFN-y (100 U IFN-y + 100 U TNF-ce/ml; 500 U IFN-7 + 1 U TNF-a/ml; 500 U IFN-7 + 10 U TNF-a/ml; or 500 U IFN-y + 100 U TNF-a/ml) for 24 h to examine the combined effects of these two cytokines. Cytokines or LPS were replenished after media change every other day. Treatments were given to duplicate or triplicate wells. Several primary cultures derived from different autopsy brains and surgical specimens were utilized in these experiments. Reversibility of ICAM-1 expression The reversibility of ICAM-1 expression was studied by incubating confluent, 8-10-day-old monolayers with LPS (0.001, 0.01, 0.1, 1, 5 p~g/ml); IFN-y (100, 500 U/ml); TNF-a (1, 10, 100 U / m l ) or IL-1/3 (0.1, 1, 10 U/ml) for 48 h. Following incubation, the cultures were washed three times with medium 199 (M199, Gibco) to remove the cytokine and LPS, placed in complete media and then returned to the incubator for 3 days prior to ICAM-1 detection. Control monolayers of the same age, originating from the same culture, were treated with identical concentrations of cytokines or LPS for 48 h. The unstimu-
lated expression of ICAM-1 was also studied in identical cultures grown in separate wells.
Antibodies Mouse anti-human ICAM-1 antibody OKT27, a gift from Dr. P. Rao (Ortho Diagnostics, Raritan, NJ), and RR1/1, a gift from Dr. T. Springer (Harvard Medical School, Boston, MA), were used as the primary antibodies. Goat anti-mouse IgG coupled to 5-nm gold particles (Auroprobe T M LM GAM IgG, Janssen/Cedarlane Labs Ltd., Hornby, ON) was used as the secondary antibody. Immunocytochemistry Following incubation with cytokines or LPS, cultures were washed briefly three times with buffer containing phosphate-buffered saline (PBS) with 1% bovine albumin and 1% normal goat serum (PBS/BSA/NGS) (both from Sigma), then incubated for 40 min at room temperature with anti-ICAM-i antibody at a final concentration of 2 / z g / m l in carrier buffer containing PBS with 5% BSA and 4% NGS. At the end of the incubation period, the monolayers were washed for 3 rain twice with PBS/BSA/NGS, and then incubated with the secondary antibody (Auroprobe T M LM GAM IgG) at a 1:40 dilution in carrier buffer for 1 h at room temperature. Subsequently, cultures were washed with PBS/BSA/NGS, fixed in 9.25% formaldehyde and 45% acetone in PBS for 30 s, washed with distilled deionized water and incubated in silver enhancing solution (IntenSE MTM, Janssen/Cedarlane) for 22-26 min. Cultures were washed again with distilled deionized water, counterstained with Giemsa and coverslipped using JB-4 plus ( P o l y s c i e n c e s / A n a l y c h e m , Markham, Ont.) as a mountant. Controls included cultures incubated in growth media in the absence of cytokines and LPS and monolayers incubated with (1) normal mouse IgG (Cedarlane) at the same concentration as the primary antibody, (2) irrelevant antibody (antihuman pituitary follicle stimulating hormone IgG, Biogenex Lab, CA), and (3) carrier buffer instead of the primary antibody. Stained cells were examined under a Nikon Labophot light microscope. Quantitation of ICAM-1 expression was performed by counting
14
O A
Fig. 1. A. HBMEC expression of ICAM-1 without LPS or cytokine, detected by immunogold silver staining, expression of ICAM-1 after incubation with 0.1 /xg LPS/ml for 24 h, detected by immunogold silver staining, expression of ICAM-1 after incubation with 500 U IFN-y/ml for 12 h, detected by immunogold silver staining, expression of ICAM-1 after incubation with 10 U TNF-a/ml for 24 h, detected by immunogold silver staining, expression of ICAM-1 after incubation with 10 U IL-1/3/ml for 24 h, detected by immunogold silver staining. expression of ICAM-1 after incubation with 100 U TNF-a/ml and 500 U IFN-y/ml for 24 h, detected by staining. × 560. (Publisher's magnification is 0.97.)
x560. B. HBMEC x 560. C. HBMEC x 560. D. I-tBMEC x560. E. HBMEC ×560. F. HBMEC immunogold silver
15
cells as labelled or unlabelled in four peripheral and one central randomly selected fields with an ocular grid under 20 × magnification. Prior to counting, the identity of all wells was masked, so that all measurements were performed blindly. Student's t-test was applied to the data to determine the significance of the differences.
Results
so
1
I"
2O
ICAM-1 expression In control cultures incubated in growth media in the absence of cytokines and LPS, 20-40% of endothelial cells expressed ICAM-1 (Figs. 1-5). The degree of basal expression varied among primary cultures derived from different individuals, but remained constant in a given culture. Treatment of HBMEC with LPS, IFN-7, TNF-a, and IL-1/3 led to significant upregulation of 1CAM-1 expression. Positively stained cells displayed diffuse surface staining in the form of finely granular, dark-brown to black deposits (Fig. 1). There was some variability in the intensity of labelling among individual endothelial cells, the plumper, older ceils being decorated with the most dense reaction product. The overall intensity of staining was higher in experimental than in control cultures (Fig. la). LPS was the most potent inducer of ICAM-1 upregulation (Figs. 1A, B, 2). 5 ~g LPS/ml resulted in expression of ICAM-1 by 95% of the cells within 12 h (P < 0.05), and this level remained constant up to 72 h in the continuous presence of LPS. The high level of ICAM-1 induced by 1 /~g and 0.1 /.~g LPS/ml seemed to decline slightly after 24 h. 0.01 /~g LPS/ml produced an intermediate rise in ICAM1 which dropped slowly after 12 h. A statistically significant (P < 0.05) rise in ICAM-1 (40%) was not observed with 0.001/~g LPS/ml until after 48 h. In contrast to the large increase in ICAM-1 expression observed with LPS, treatment with IFN-y induced only a slight increase (Fig. 1C, 3), with maximal levels of expression reaching only 58%. By 12 h, a statistically significant level of upregulation to 47% was observed with all concentrations (P < 0.05), but at longer incubation times, the relatively larger variance made any
• ,
o 0
10 20 30 40 50 60 70
80
Time Fig. 2. HBMEC expression of ICAM-1 after incubation with 5 txg/ml (©), 1/xg/ml (~), 0.1 ,~g/ml (D), 0.01 txg/ml (zx), or 0.001 ~ g / m l (©) LPS. Values represent mean+SE of duplicate wells. * P < 0.05 compared to expression at 0 h.
increases in expression not to be statistically significant. The extent of ICAM-1 upregulation by both TNF-a (Figs. 1D, 4) and IL-1/3 (Figs. 1E, 5) was intermediate between that of LPS and IFN-y. By
100
N
o 0
10 20 30 4O 5O 60 70
80
Fig. 3. HBMEC expression of 1CAM-1 after incubation with 500 U / m l (a), 100 U / m l (•), 50 U / m l (zx), or 10 U / m l (©) IFN-y. Values represent mean + SE of triplicate wells. * P < 0.05 compared to expression at 0 h.
16
100
20
~,
0
,
0
,
,
,
,
,
,
10 20 30 40 50 60 70
80
T i m e CO)
nificant. The increase caused by 1 U TNF-a/ml was not statistically significant until 12 h, and after incubation for 72 h, the increase declined to non-significant levels. The upregulation of ICAM-1 elicited by IL-1/3 reached maximal levels at 24 h (84% for 10 U, 76% for 1 U, and 67% for 0.1 U, all P < 0.05), and then declined. Incubation of endothelial cells with TNF-a and IFN-y combined resulted in marked increase in ICAM-1 expression that was even greater than that induced by LPS (Fig. 1F, 6). The number of immunostained cells ranged from 99 to 100% with 500 U IFN-T/ml plus 1 U to 100 U TNFa / m l and reached 97% with 100 U IFN-y/ml plus 100 U TNF-a/ml.
Reversibility
Fig. 4. H B M E C expression of ICAM-1 after incubation with 100 U / m l (El), 10 U / m l (z~), or 1 U / m l ((3) TNF-a. Values represent mean + SE of triplicate wells. * P < 0.05 compared to expression at 0 h.
24 h, 85% of cells showed positive staining (P < 0.05), after which there was a general trend of slight decline. The increased staining induced by 10 U and 100 U TNF-a/ml was statistically sig-
11111
Stimulation of HBMEC with LPS or cytokines for 2 days followed by withdrawal and culture of cells in regular growth media for 3 days was associated with variable degrees of reversal of ICAM-1 upregulation following LPS and TNF-a incubation, and lack of reversal following IL-1/3 and IFN-y treatment. These differences were due to cytokine and LPS removal since cultures incubated with cytokine or LPS for 5 days expressed high levels of ICAM-1 (Figs. 6-9). There was a decrease (10-30%) in the number of cells
i
lOO
[]
8o 60
rtl
~
20 ~, 0
I
0
I
|
I
I
I
10 20 30 40 50 60 70
i
o 1
I
10
80 Concenlration of TNF-a
"nine CO) Fig. 5. H B M E C expression of ICAM-1 after incubation with 10 U / m l (D), 1 U / m l (zx), or 0.1 U / m l (©) IL-1/3. Values represent mean +_SE of triplicate wells. * P < 0.05 compared to expression at 0 h.
100
(U/ml)
]Fig.6. Combined effects of 500 U / m l ([]), 100 U / m l ( A ), or 0 U / m l (©) IFN-N and i00 U/ml, I0 U/ml, or I U / m l TNF-a on the ICAM-I expression on H B M E C after 24 h. Values represent mean + SE. * P < 0.05 compared to expression at 0 h.
17
100
--//
//
100 8O
60
60
40
40
cD
20
8
r
2O
8 o --//
. 0.OOl
.
.
.
0.01
0.1
~. 1
0
//
0
10
Concentration of LPS (tag/ml)
1
i
1
10
100
Concentration of TNF-a CO/ml)
Fig. 7. Reversibility of ICAM-1 expression on HBMEC induced by incubation with LPS for 2 days then 3 days without LPS (o), with LPS for 2 days (D), with LPS for 5 days (zx), or without LPS (9). Values represent mean_+SE of duplicate wells.
Fig. 9. Reversibility of ICAM-1 expression on HBMEC induced by incubation with TNF-a for 2 days then 3 days without TNF-a (o), with TNF-ot for 2 days ([]), with TNF-a for 5 days (z~), or without TNF-a (~v). Values represent mean + SE of triplicate wells.
expressing ICAM-1 at all concentrations of LPS (Fig. 7). Reversibility of expression was minimal when 5 /zg/ml LPS was used. Reversibility of upregulation was not observed with IFN-7 at 100 U/ml, while 500 U / m l was associated with a 19% increase instead (Fig. 8). Stimulation followed by withdrawal of TNF-a resulted in decrease in ICAM-1 expression ranging from 6% to
23%. TNF-a concentration of 100 U / m l was associated with the smallest decrease while 1 U / m l induced the greatest (Fig. 9). Paradoxically, incubation of endothelial cells in IL-lfl-free media for 3 days following stimulation resulted in increase in the number of labelled cells ranging
//
100
!
4
8O
IO0
--//
"
6o
o
o o
40
40
8 0
-//
o
-// 100
1000
Concentration of H~N-7 (U/ml) Fig. 8. Reversibility of ICAM-1 expression on HBMEC induced by incubation with IFN-7 for 2 days then 3 days without IFN-7 (o), with IFN-y for 2 days (D), with IFN-y for 5 days ( zx), or without IFN-7 (*). Values represent mean + SE of triplicate wells.
0
i
0.1
10
Concen~afion of IL-113 (U/ml) Fig. 10. Reversibility of ICAM-1 expression on HBMEC induced by incubation with IL-1/3 for 2 days then 3 days without IL-1/~ (o), with IL-1/3 for 2 days (t~), with IL-1/3 for 5 days ( 4 ) , or without IL-lfl (~). Values represent mean_+SE of triplicate wells.
18 from a low of 1% with 10 U IL-1/3/ml to 17% with 0.1 U / m l (Fig. 10).
Discussion
The present study describes the induction of ICAM-1 expression on primary cultures of HBMEC by LPS and cytokines, and the reversibility of this expression. Intact, unstimulated endothelial cells constitutively express low levels of ICAM-1. Incubation with LPS, TNF-a, IL-1/3 and to a lesser extent IFN-T, results in upregulation of ICAM-1 expression which is time- and concentration-dependent and persists in the continuous presence of mediators in the culture. ICAM-1 is an inducible glycoprotein of the immunoglobulin superfamily, present at low levels on a variety of cell types including endothelial cells. It is important in both adhesion and transendothelial migration of lymphocytes, monocytes, neutrophils, basophils and eosinophils in vivo and in vitro (Albelda and Buck, 1990; A1belda, 1991; Bochner et al., 1991; Pober and Cotran, 1991; Shimizu et al., 1991). Antibodies against ICAM-1 and LFA-1 (the leukocyte receptor for ICAM-1) block the extravasation of eosinophils in vivo (Albelda, 1991) and prevent the adhesion of leukocytes to endothelial cells in vitro (Bochner et al., 1991). The pattern of ICAM-1 expression in inflammatory disorders of the nervous system was recently investigated in guinea pigs and mice during the acute and chronic phases of EAE. O'Neill et al. (1991) observed the upregulation of MALA-2 (the murine homologue of ICAM-1) on endothelial cells and mononuclear infiltrates in spinal cord lesions during the active episodes of chronic relapsing EAE in the Biozzi A B / H mouse. Recent studies on adoptively transferred E A E in the S J L / J mouse have shown that attachment and extravasation of mostly LFA-1 positive lymphocytes correlated with the appearance of MALA-2 on CNS vessels and the onset of clinical signs (Raine et al., 1990). In fact, the expression of MALA-2 fluctuated according to the clinical phase of EAE, with upregulation occurring with each relapse and down-regulation during remissions (Cannella et al., 1991). In both studies, endothelial and glial
cells of normal mice showed low levels of ICAM-1. Similarly, the normal spinal cord vasculature of guinea pigs has only a few ICAM-l-positive endothelial cells, while ICAM-1 was present on the endothelium of both lesion- and non-lesion-associated blood vessels of animals during the acute phase of E A E (Wilcox et al., 1990). In the same study, cultured endothelial cells isolated from guinea pig brains did not express ICAM-1 unless incubated in a lymphocyte-conditioned medium. The above animal studies suggest that upregulation of ICAM-1 may be important in the initial influx of lymphocytes into the brain in autoimmune demyelinating CNS disorders. The role of ICAM-1 in orchestrating leukocyte traffic across the BBB in the course of inflammatory diseases of the human CNS has not yet been fully investigated. In a recent neuropathological report, based on the examination of post-mortem material from normal and pathological brain tissue, only a small number of microvessels were ICAM-1 positive within the normal CNS. In contrast, numerous ICAM-l-positive vessels were observed in MS plaque edges, viral encephalitis lesions and infarcts (Sobel et al., 1990). Recently, high IFN-7 levels have been observed in EAE (Cannella et al., 1991), high TNF-a levels have been reported in some MS plaques (Brosnan et al., 1988; Hofman et al., 1989), and IL-1 has been found in MS lesions (Hofman et al., 1986). The upregulation of ICAM-1 induced by LPS and TNF-a on H B M E C is very drastic even by 4 h, and is maximal by 12 h, while IL-1/3 did not induce a maximal level until 24 h. IFN-7 led to a slight increase in expression which was not very significant statistically. In human umbilical vein ECs (HUVEC) and human dermal microvessel ECs (HDMEC), LPS caused a rise in ICAM-1 that did not peak until 24 h (Wellicome et al., 1990; Swerlick et al., 1991). Swerlick et al. (1991) and Pober et al. (1986) observed a similar rise of ICAM-1 in H U V E C and H D M E C following treatment with TNF-a and IL-1/3, in contrast to Wellicome et al. (1990) who showed that TNF and IL-1 induced a peak in ICAM-1 expression on H U V E C by 10 h. The slight upregulation of ICAM-1 observed after treatment of H B M E C with IFN-T is in keeping with previous reports indicating that IFN-T is a weak inducer of
19 ICAM-1 upregulation on extracerebral endothelium. Thus, IFN-y did not induce a noticeable change in ICAM-1 levels in HDMEC until 72 h (Swerlick et al., 1991), and caused a 'slower' increase than IL-1 and TNF in HUVEC (Pober et al., 1986). It is evident that the highest levels of ICAM-1 induced by LPS, TNF-a and IFN-y appear earlier in HBMEC than in HDMEC and some HUVEC cultures. The time course of ICAM-1 expression generated by IL-1/3 is similar in all three culture systems. The optimal concentration of cytokines and LPS required for the upregulation of ICAM-1 is different in cerebral and extracerebral endothelial cultures. Thus, LPS at a concentration of 0.1 p.g/ml induces maximal ICAM-1 levels in HUVEC (Wellicome et al., 1990) while a slightly higher concentration (1 ~ g / m l ) was required for HBMEC in our study. In contrast, HUVEC and HDMEC require 100 U T N F / m l (Pober et al., 1987; Detmer et al., 1990) and 100 U IFN-y/ml (Ruszczak et al., 1990) for maximal effect, while HBMEC showed no noticeable change in ICAM-1 level with 1-100 U TNF-a/ml, or 10-500 U IFN-y/ml. Human retinal capillary ECs, on the other hand, showed maximal induction of ICAM-1 upregulation with 5 U IFN-y/ml (Liversidge et al., 1990). IL-1 was required in a concentration greater than 10 U / m l for maximal effect on HUVEC (Pober et al., 1986; Wellicome et al., 1990), while 10 U / m l induced highest levels on HBMEC. HBMEC are thus more sensitive to cytokines and less sensitive to LPS than HUVEC and HDMEC with respect to ICAM-1 upregulation. The staining intensity of HBMEC varied even within monolayers derived from a single isolation. As a rule, the labelling of larger, older endothelial cells was stronger than that of actively proliferating cells. Such heterogeneity in ICAM-1 expression has also been reported in cultured human saphenous vein endothelial cells and HUVEC (Dustin and Springer, 1988) and in cultured fibroblasts (Dustin et al., 1986). The basal expression of ICAM-1 in HBMEC (22-40% of cells) is higher than the 18% in h u m a n retinal capillary ECs (Liversidge et al., 1990) and the 11-19% found in HDMEC (Ruszczak et al., 1990), which in turn are higher
than the expression in HUVEC (Swerlick et al., 1991). The maximal level of ICAM-1 expression in HBMEC is greatest with LPS (95% of cells labelled), intermediate with TNF-a and IL-1/3 (85%), and lowest with IFN- 7 (58%). This is in great contrast to HDMEC which expressed ICAM-1 to almost a 100% level with 100 U TNF/ml (Detmer et al., 1990), and showed a 78% labelling after incubation with 100 U IFN7 / m l and 93% with 1000 U IFN-y/ml (Ruszczak et al., 1990). HBMEC express much lower levels of ICAM-1 than HDMEC when activated with LPS or cytokines. Following incubation of HBMEC with a combination of TNF-a and IFN-7, 99 to 100% of the cells became labelled for ICAM-1. A synergistic effect of these two cytokines has also been observed in HDMEC (Detmer et al., 1990) and HUVEC (Doukas and Pober, 1990). In the latter study, it was also found that IFN- 7 does not significantly enhance the expression of ICAM-1 induced by IL-1/3. Other cell types within the brain respond much differently than ECs to cytokine stimulation. ICAM-1 expression on oligodendrocytes and astrocytes was markedly increased by IFN-y, while TNF-a, IL-la and LPS were less effective (Satoh et al., 1991). We found that removal of cytokines and LPS from the culture media was associated with variable expression of ICAM-1 after 3 days. Thus, there seemed to be a slight decrease in ICAM-1 with LPS (10-30%) and TNF-a (6-23%), and a slight increase with IFN- 7 (0-19%) and IL-1/3 (1-17%). Dustin and Springer (1988) observed that IL-1 and TNF caused ICAM-1 expression in HUVEC to become elevated, but returned to basal levels when the mediators were washed out. It would seem that HBMEC maintain their ICAM-1 expression much longer than HUVEC. The mechanism(s) of ICAM-1 induction by cytokines is presently poorly understood and rather controversial. Thus, recent studies indicate that activation of protein kinase C (PKC) may play a role in the induction of ICAM-1 expression on large vessel endothelium by IL-I, TNF-a, LPS and INF-y (Lane et al., 1990; Renkonen et al., 1990a). However, according to other investigators, IL-1 may operate via the cAMP (Renkonen et al., 1990b) and activation of PKC may not be the necessary pathway for ICAM-1 induction by
20 T N F - a ( R i t c h i e et al., 1991). W h e t h e r s i m i l a r m e c h a n i s m s a r e r e s p o n s i b l e for t h e i n d u c t i o n o f I C A M - 1 e x p r e s s i o n by c y t o k i n e s o n m i c r o v e s s e l e n d o t h e l i u m a n d m o r e specifically o n c e r e b r a l e n d o t h e l i u m r e m a i n s to b e i n v e s t i g a t e d . Our results indicate that HBMEC, compared to o t h e r E C s , t e n d to e x p r e s s a h i g h e r b a s a l l e v e l o f I C A M - 1 , a r e m o r e s e n s i t i v e to c y t o k i n e s , a n d less s e n s i t i v e to L P S , e x p r e s s I C A M - 1 e a r l i e r b u t at a l o w e r level w h e n i n d u c e d , a n d fail to r e v e r s e its e x p r e s s i o n w i t h i n a s h o r t t i m e u p o n t h e removal of these agents. These findings indicate t h a t u p r e g u l a t i o n o f I C A M - 1 by c y t o k i n e s o r L P S in v i v o m a y f a c i l i t a t e i n t e r a c t i o n s b e t w e e n L F A 1-positive l e u k o c y t e s a n d b r a i n m i c r o v e s s e l e n d o t h e l i a l cells a n d p l a y a n i m p o r t a n t r o l e in leukocyte trafficking across the blood-brain barr i e r in C N S i n f l a m m a t i o n .
Acknowledgements T h e a u t h o r s t h a n k D r s . P. R a o a n d T. S p r i n g e r for p r o v i d i n g t h e a n t i - I C A M - 1 a n t i b o d i e s , M r s . R. P r a m e y a for aid in t h e i s o l a t i o n a n d c u l t u r e o f H B M E C . T h i s s t u d y was s u p p o r t e d by g r a n t s from the National Institutes of Health (RO1NS23746), the Medical Research Council of Canada (MA-9823), and the British Columbia Health Care Research Foundation.
References Albelda, S.M. (1991) Endothelial and epithelial cell adhesion molecules. Am. J. Respir. Cell Mol. Biol. 4, 195-203. Albelda, S.M. and Buck, C.A. (1990) Integrins and other cell adhesion molecules. FASEB J, 4, 2868-2880. Bochner, B.S., Luscinskas, F.W., Gimbrone, M.A., Jr., Newman, W., Sterbinsky, S.A., Derse-Anthony, C.P., Klunk, D. and Schleimer, R.P. (1991) Adhesion of human basophils, eosinophils and neutrophils to interleukin 1-activated human vascular endothelial cell: contributions of endothelial cell adhesion molecules. J. Exp. Med. 173, 1553-1556. Brosnan, C.F., Selmej, K. and Raine, C.S. (1988) Hypothesis: a role of tumor necrosis factor in immune-mediated demyelination and its relevance to multiple sclerosis. J. Neuroimmunol. 18, 87-94. Cannella, B., Cross, A.H. and Raine, C.S. (1991) Adhesion-related molecules in the central nervous system. Up-regu-
lation correlates with inflammatory cell influx during relapsing experimental autoimmune encephalomyelitis. Lab. Invest. 65, 23-31. Detmer, M., Imcke, E., Ruszczak, Z. and Orfanos, C.E. (1990) Effects of recombinant tumor necrosis factor-alpha on cultured microvascular endothelial cells derived from human dermis. J. Invest. Dermatol. 95, 219S-222S. Dorovini-Zis, K., Prameya, R. and Bowman, P. (1991) Culture and characterization of microvessel endothelial cells derived from human brain. Lab. Invest. 64, 425-436. Doukas, J. and Pober, J.S. (1990) IFN-3, enhances endothelial activation induced by tumor necrosis factor but not IL-1. J. Immunol. 145, 1727-1733. Dustin, M.L. and Springer, T.A. (1988) Lymphocyte functionassociated antigen-1 (LFA-1) interaction with intercellular adhesion molecule-1 (ICAM-I) is one of at least three mechanisms for lymphocyte adhesion to cultured endothelial cells. J. Cell. Biol. 107, 321-331. Dustin, M.L., Rothlein, R., Bhan, A.K., Dinarello, C.A. and Springer, T.A. (1986) Induction by IL-1 and interferon, tissue distribution, biochemistry and function of a natural adherence molecule (ICAM-1). J. Immunol. 137, 245-254. Hofman, F.M., von Hanwehr, R.I., Dinarello, C.A., Mizel, S.B., Hinton, D. and Merrill, J.E. (1986) Immunoregulatory molecules and IL-2 receptors identified in multiple sclerosis brain. J. Immunol. 136, 3239-3245. Hofman, F.M., Hinton, D.R., Johnson, K. and Merrill, J.E. (1989) Tumor necrosis factor identified in multiple sclerosis brain. J. Exp. Med. 170, 607-612. Lane, T.A., Lamkin, G.E. and Wancewicz, E.V. (1990) Protein kinase C inhibitors block the enhanced expression of intercellular adhesion molecule-1 on endothelial cells activated by interleukin-1, lipopolysaccharide and tumor necrosis factor. Biochem. Biophys. Res. Commun. 172, 1273-1281. Liversidge, J., Sewell, H.F. and Forrester, J.V. (1990) Interactions between lymphocytes and cells of the blood-retina barrier: mechanisms of T lymphocyte adhesion to human retinal capillary endothelial cells and retinal pigment epithelial cells in vitro. Immunology 71,390-396. Munro, J.M., Pober, J.S. and Cotran, R.S. (1991) Recruitment of neutrophils in the local endotoxin response: association with de not,o endothelial expression of endothelial leukocyte adhesion molecule-1. Lab. Invest. 64, 295-299. O'Neill, J.K., Butter, C., Baker, D., Gschmeissner, S.E., Kraal, G., Butcher, E.C. and Turk, J.L. (199I) Expression of vascular addressins and ICAM-1 by endothelial cells in the spinal cord during chronic relapsing experimental allergic encephalomyelitis in the Biozzi A B / H mouse. Immunology 72, 520-525. Pober, J.S. and Cotran, R.S. (1990) The role of endothelial cells in inflammation. Transplantation 50, 537-544. Pober, J.S. and Cotran, R.S. (1991) What can be learned from the expression of endothelial adhesion molecules in tissues? Lab. Invest. 64, 301-305. Pober, J.S., Gimbrone, M.A. Jr., Lapierre, L.A., Mandrick, D.L., Fiers, W., Rothlein, R. and Springer, T.A. (1986) Overlapping patterns of activation of human endothelial
21 cells by interleukin 1, tumor necrosis factor, and immune interferon. J. Immunol. 137, 1893-1896. Pober, J.S., Lapierre, L.A., Stolpen, A.H., Brook, T.A., Springer, T.A., Fiers, W., Bevilacqua, M.P., Mendrick, D.L. and Gimbrone, M.A. Jr. (1987) Activation of cultured human endothelial cells by recombinant lymphotoxin: comparison with tumor necrosis factor and interleukin 1 species. J. Immunol. 136, 3319-3324. Raine, C.S., Cannella, B., Duijvestijn, A.M. and Cross. A.H. (1990) Homing to central nervous system vasculature by antigen-specific lymphocytes. II. Lymphocyte/endothelial cell adhesion during the initial stages of autoimmune demyelination. Lab. Invest. 63, 476-489. Renkonen, R., Mennander, A., Ustinov, J. and Mattila, P. (1990a) Activation of protein kinase C is crucial in the regulation of ICAM-1 expression on endothelial cells by interferon--/. Int. Immunol. 2, 719-724. Renkonen, R., Mennander, A., Mattila, P. and Ustinov, J. (1990b) Signal transduction during in vitro lymphocyte homing. Human Immunol. 28, 134-140. Ritchie, A.J., Johnson, D.R., Ewenstein, B.M. and Pober, J.S. (1991) Tumor necrosis factor induction of endothelial cell surface antigen is independent of protein kinase C activation or inactivation. Studies with phorbol myristate acetate and staurosporine. J. Immuno|. 146, 3056-3062. Ruszczak, Z., Detmer, M., Imcke, E. and Orfanos, C.E. (1990) Effects of rlFN alpha, beta and gamma on the morphology, proliferation and cell surface antigen expression of human dermal microvascular endothelial cells in vitro. J. Invest. Dermatol. 95, 693-699. Satoh, J.-l., Kastrukoff, L.F. and Kim, S.U. (1991) Cytokine-
induced expression of intercellular adhesion molecule-1 (ICAM-1) in cultured human oligodendrocytes and astrocytes. J. Neuropathol. Exp. Neurol. 3, 215-226. Shimizu, Y., Newman, W., Gopal, T.V., Horgan, K.J., Graber, N., Beall, L.D., van Seventer, G.A. and Shaw, S. (1991) Four molecular pathways of T cell adhesion to endothelial cells: roles of LFA-1, VCAM-I and ELAM-1 and changes in pathway hierarchy under different activation conditions. J. Cell. Biol. 113, 1203-1212. Sobel, R.A., Mitchell, M.E. and Fondien, G. (1990) Intercellular adhesion molecule-1 (ICAM-1) in cellular immune reactions in the human central nervous system. Am. J. Pathol. 136, 1309-1316. Springer, T.A. (1990) Adhesion receptor of the immune system. Nature 346, 425-434. Swerlick, R.A., Garcia-Gonzalez, E., Kubota, Y., Xu, Y. and Lanky, T.J. (1991) Studies of the modulation of MHC antigen and cell adhesion molecule expression on human dermal microvascular endothelial cells. J. Invest. Dermatol. 97, 190-196. Wellicome, S.M., Thornhill, M.H., Pitzalis, C., Thomas, D.S., Lanchbury, J.S.S., Panayi, G.S. and Haskard, D.O. (1990) A monoclonal antibody that detects a novel antigen on endothelial cells that is induced by tumor necrosis factor, IL-I, or lipopolysaccharide. J. Immunol. 144, 2558-2565. Wilcox, C.E., Ward, A.M., Evans, A., Baker, D., Rothlein, R. and Turk, J.L. (1990) Endothelial cell expression of the intercellular adhesion molecule (ICAM-1) in the central nervous system of guinea pigs during acute and chronic relapsing experimental allergic encephalomyelitis. J. Neuroimmunol. 30, 43-51.
11
JNI 02193
Upregulation of intercellular adhesion molecule-1 (ICAM-1) expression in primary cultures of human brain microvessel endothelial cells by cytokines and lipopolysaccharide D ona l d W ong and Katerina Dorovini-Zis Department of Pathology, Section of Neuropathology, the University of British Columbia and VancouL,er General Hospital, Vancouver, B.C., Canada (Received 11 November 1991) (Revised received 3 January and 10 February 1992) (Accepted 10 February 1992)
Key words: Cerebral endothelium; Intercellular adhesion molecule-I; Cytokine; Lipopolysaccharide
Summary The expression of intercellular adhesion molecule-1 (ICAM-1) by human cerebral endothelium was studied in primary cultures of human brain microvessel endothelial cells following treatment with bacterial lipopolysaccharide (LPS), tumor necrosis factor-a (TNF-a), interleukin-1/3 (IL-1/3) and interferon- 7 (IFN-y). Surface expression of ICAM-1 was examined with the immunogold silver staining technique. Intact cerebral endothelial cells constitutively express low levels of ICAM-1. Stimulation with LPS and cytokines induces upregulation of ICAM-1 which is minimal with IFN-7 and maximal with LPS or a combination of IFN-7 and TNF-a. Upregulation of ICAM-1 expression is concentration- and time-dependent, is observed as early as 4 h following incubation and persists for up to 72 h in the continuous presence of LPS or cytokines. The ICAM-I expression is not reversed by 3 days after removal of the LPS or cytokines. These findings may be relevant to the interactions between leukocytes and brain microvessel endothelial cells in inflammatory and demyelinating diseases of the CNS.
Introduction Several disorders of the central nervous system (CNS) including infectious, inflammatory and autoimmune demyelinating diseases are characterized by the migration of acute and chronic inflammatory cells from blood into brain with inva-
Correspondence to: K. Dorovini-Zis, Department of Pathology, Section of Neuropathology, Vancouver General Hospital, 855 West 12th Avenue, Vancouver, B.C., Canada V5Z 1M9.
sion of the extravascular tissue. The factors responsible for the regulation of leukocyte traffic across the cerebral endothelial barrier that normally excludes white blood cells from entering the brain, remain largely unknown. Recent studies on in vitro interactions between leukocytes and extracerebral endothelium suggest that de novo expression or upregulation of certain surface glycoproteins - adhesion molecules - - by endothelial cells mediates leukocyte adhesion to endothelium. Expression of intercellular adhesion molecules has been induced in vitro on extracere-
12 bral small and large vessel endothelial cells by cytokines and bacterial lipopolysaccharide (Albelda and Buck, 1990; Pober and Cotran, 1990, 1991) and in vivo following intradermal injection of endotoxin (Munro et al., 1991). A number of endothelial adhesion molecules have been described and characterized, including intercellular adhesion molecule-1 (ICAM-1), endothelial leukocyte adhesion molecule-1 (ELAM-1) and vascular cell adhesion molecule-1 (VCAM-1). ICAM-1 is a member of the immunoglobulin gene superfamily. It binds to leukocyte functionassociated molecule 1 (LFA-1), an integrin on the surface of neutrophils, lymphocytes and monocytes, and has been shown to play a role in the adhesion of polymorphonuclear leukocytes and lymphocytes to cultured endothelium (Albelda, 1991; Pober and Cotran, 1990, 1991). ICAM-1 is expressed at low levels basally by human umbilical vein, dermal and retinal microvessel endothelial cells in culture (Pober et al., 1986, 1987; Detmer et al., 1990; Liversidge et al., 1990; Ruszczak et al., 1990; Wellicome et al., 1990; Swerlick et al., 1991). Following activation by interleukin-1 (IL-1), tumor necrosis factor-a (TNF-a), lipopolysaccharide (LPS) or interferon-7 (IFN-7), ICAM-1 expression reaches maximal levels after 24 h (Albelda and Buck, 1990; Pober and Cotran, 1990; Springer, 1990). Differences in its expression have been observed among endothelial cells from different vascular beds. In the nervous system, the role of ICAM-1 in the initiation of the inflammatory response has not yet been fully investigated. Expression of ICAM-1 and the addressin MECA-325 by CNS endothelium has been implicated in the adhesion of lymphocytes to endothelial cells during the acute and chronic relapsing phases of experimental allergic encephalomyelitis, the animal model of multiple sclerosis (Raine et al., 1990; Wilcox et al., 1990; Cannella et al., 1991). In addition, immunolocalization of ICAM-1 in human cerebrovascular endothelium has been reported in necrotic and inflammatory CNS lesions (Sobel et al., 1990). The results of these studies suggest that the cerebral endothelium, through its ability to express certain adhesion proteins that interact with corresponding leukocyte membrane molecules, may play an important role in modu-
lating leukocyte traffic across the blood-brain barrier. In the present study, we investigated the expression of ICAM-1 by human brain microvessel endothelial cells (HBMEC) in primary culture following stimulation with LPS, IFN-7, TNF-a and IL-1/3. Previous studies have shown that HBMEC in culture form confluent monolayers that retain important endothelial and blood-brain barrier characteristics including paucity of cytoplasmic vesicles and presence of tight junctions that restrict the intercellular movement of macromolecules (Dorovini-Zis et al, 1991). Intact HBMEC showed minimal basal expression of ICAM1. Incubation with LPS and cytokines led to upregulation of ICAM-1 expression, which was minimal following IFN-7 stimulation and maximal after treatment with LPS or a combination of TNF-a and IFN-y. This expression was sustained in the continuous presence of LPS or cytokines in culture. Our results indicate that human cerebral endothelial cells are capable, upon activation, of expressing intercellular adhesion molecules for circulating leukocytes and may thus play an important role in the initiation of CNS inflammation.
Materials and methods
Isolation and culture of human brain microL'essel endothelial cells (HBMEC) Primary cultures of microvessel endothelial cells were established from human brains at autopsy and from temporal lobectomy specimens by methods previously described (Dorovini-Zis et al., 1991). The endothelial origin of the cells was determined by their strongly positive, granular, perinuclear immunofluorescence for Factor VIII antigen and binding of Ulex europaeus agglutinin. The isolated clumps of endothelial cells were seeded onto fibronectin-coated 24-well plates (Corning Plastics, Corning, NY) and maintained in culture in minimum essential medium alpha (aMEM, Gibco, Burlington, ON) supplemented with 10% horse plasma-derived serum (PDS, Hyclone Laboratories, Logan, UT), 25 mM Hepes, 10 mM sodium bicarbonate, 100 /zg/ml heparin, 20/xg/ml endothelial cell growth supplement (all
13 from Sigma, St. Louis, MO), 100 Izg/ml penicillin, 100 /zg/ml streptomycin, and 2.5 /xg/ml amphotericin B (Gibeo) at 37°C in a humidified 2.5% CO2/97.5% air atmosphere. Culture media were changed every other day. Endothelial cells reached confluency 7-9 days after plating and were used for studies on or after the 10th day.
ICAM-1 induction by cytokines and LPS Bacterial lipopolysaccharide (LPS from Escherichia coli O55:B5, Sigma), human recombinant interferon-y (IFN-7, Collaborative Research Inc., Bedford, MA), human recombinant tumor necrosis factor-a (TNF-a, Sigma) or human recombinant interleukin-1/3 (IL-1/3, Boehringer Mannheim, Laval, PQ) were dissolved in complete media at a final concentration of 0.001, 0.01, 0.1, 1, 5 /zg LPS/ml; 10, 50, 100, 500 U IFN-y/ml; 1, 10, 100 U TNF-a/ml; or 0.1, 1, 10 U IL-1/3/ml. 10-11-day-old cultures were incubated with different concentrations of LPS and cytokines for 4, 12, 2 4 , 4 8 and 72 h. In addition, separate wells were incubated with various concentrations of TNF-a and IFN-y (100 U IFN-y + 100 U TNF-ce/ml; 500 U IFN-7 + 1 U TNF-a/ml; 500 U IFN-7 + 10 U TNF-a/ml; or 500 U IFN-y + 100 U TNF-a/ml) for 24 h to examine the combined effects of these two cytokines. Cytokines or LPS were replenished after media change every other day. Treatments were given to duplicate or triplicate wells. Several primary cultures derived from different autopsy brains and surgical specimens were utilized in these experiments. Reversibility of ICAM-1 expression The reversibility of ICAM-1 expression was studied by incubating confluent, 8-10-day-old monolayers with LPS (0.001, 0.01, 0.1, 1, 5 p~g/ml); IFN-y (100, 500 U/ml); TNF-a (1, 10, 100 U / m l ) or IL-1/3 (0.1, 1, 10 U/ml) for 48 h. Following incubation, the cultures were washed three times with medium 199 (M199, Gibco) to remove the cytokine and LPS, placed in complete media and then returned to the incubator for 3 days prior to ICAM-1 detection. Control monolayers of the same age, originating from the same culture, were treated with identical concentrations of cytokines or LPS for 48 h. The unstimu-
lated expression of ICAM-1 was also studied in identical cultures grown in separate wells.
Antibodies Mouse anti-human ICAM-1 antibody OKT27, a gift from Dr. P. Rao (Ortho Diagnostics, Raritan, NJ), and RR1/1, a gift from Dr. T. Springer (Harvard Medical School, Boston, MA), were used as the primary antibodies. Goat anti-mouse IgG coupled to 5-nm gold particles (Auroprobe T M LM GAM IgG, Janssen/Cedarlane Labs Ltd., Hornby, ON) was used as the secondary antibody. Immunocytochemistry Following incubation with cytokines or LPS, cultures were washed briefly three times with buffer containing phosphate-buffered saline (PBS) with 1% bovine albumin and 1% normal goat serum (PBS/BSA/NGS) (both from Sigma), then incubated for 40 min at room temperature with anti-ICAM-i antibody at a final concentration of 2 / z g / m l in carrier buffer containing PBS with 5% BSA and 4% NGS. At the end of the incubation period, the monolayers were washed for 3 rain twice with PBS/BSA/NGS, and then incubated with the secondary antibody (Auroprobe T M LM GAM IgG) at a 1:40 dilution in carrier buffer for 1 h at room temperature. Subsequently, cultures were washed with PBS/BSA/NGS, fixed in 9.25% formaldehyde and 45% acetone in PBS for 30 s, washed with distilled deionized water and incubated in silver enhancing solution (IntenSE MTM, Janssen/Cedarlane) for 22-26 min. Cultures were washed again with distilled deionized water, counterstained with Giemsa and coverslipped using JB-4 plus ( P o l y s c i e n c e s / A n a l y c h e m , Markham, Ont.) as a mountant. Controls included cultures incubated in growth media in the absence of cytokines and LPS and monolayers incubated with (1) normal mouse IgG (Cedarlane) at the same concentration as the primary antibody, (2) irrelevant antibody (antihuman pituitary follicle stimulating hormone IgG, Biogenex Lab, CA), and (3) carrier buffer instead of the primary antibody. Stained cells were examined under a Nikon Labophot light microscope. Quantitation of ICAM-1 expression was performed by counting
14
O A
Fig. 1. A. HBMEC expression of ICAM-1 without LPS or cytokine, detected by immunogold silver staining, expression of ICAM-1 after incubation with 0.1 /xg LPS/ml for 24 h, detected by immunogold silver staining, expression of ICAM-1 after incubation with 500 U IFN-y/ml for 12 h, detected by immunogold silver staining, expression of ICAM-1 after incubation with 10 U TNF-a/ml for 24 h, detected by immunogold silver staining, expression of ICAM-1 after incubation with 10 U IL-1/3/ml for 24 h, detected by immunogold silver staining. expression of ICAM-1 after incubation with 100 U TNF-a/ml and 500 U IFN-y/ml for 24 h, detected by staining. × 560. (Publisher's magnification is 0.97.)
x560. B. HBMEC x 560. C. HBMEC x 560. D. I-tBMEC x560. E. HBMEC ×560. F. HBMEC immunogold silver
15
cells as labelled or unlabelled in four peripheral and one central randomly selected fields with an ocular grid under 20 × magnification. Prior to counting, the identity of all wells was masked, so that all measurements were performed blindly. Student's t-test was applied to the data to determine the significance of the differences.
Results
so
1
I"
2O
ICAM-1 expression In control cultures incubated in growth media in the absence of cytokines and LPS, 20-40% of endothelial cells expressed ICAM-1 (Figs. 1-5). The degree of basal expression varied among primary cultures derived from different individuals, but remained constant in a given culture. Treatment of HBMEC with LPS, IFN-7, TNF-a, and IL-1/3 led to significant upregulation of 1CAM-1 expression. Positively stained cells displayed diffuse surface staining in the form of finely granular, dark-brown to black deposits (Fig. 1). There was some variability in the intensity of labelling among individual endothelial cells, the plumper, older ceils being decorated with the most dense reaction product. The overall intensity of staining was higher in experimental than in control cultures (Fig. la). LPS was the most potent inducer of ICAM-1 upregulation (Figs. 1A, B, 2). 5 ~g LPS/ml resulted in expression of ICAM-1 by 95% of the cells within 12 h (P < 0.05), and this level remained constant up to 72 h in the continuous presence of LPS. The high level of ICAM-1 induced by 1 /~g and 0.1 /.~g LPS/ml seemed to decline slightly after 24 h. 0.01 /~g LPS/ml produced an intermediate rise in ICAM1 which dropped slowly after 12 h. A statistically significant (P < 0.05) rise in ICAM-1 (40%) was not observed with 0.001/~g LPS/ml until after 48 h. In contrast to the large increase in ICAM-1 expression observed with LPS, treatment with IFN-y induced only a slight increase (Fig. 1C, 3), with maximal levels of expression reaching only 58%. By 12 h, a statistically significant level of upregulation to 47% was observed with all concentrations (P < 0.05), but at longer incubation times, the relatively larger variance made any
• ,
o 0
10 20 30 40 50 60 70
80
Time Fig. 2. HBMEC expression of ICAM-1 after incubation with 5 txg/ml (©), 1/xg/ml (~), 0.1 ,~g/ml (D), 0.01 txg/ml (zx), or 0.001 ~ g / m l (©) LPS. Values represent mean+SE of duplicate wells. * P < 0.05 compared to expression at 0 h.
increases in expression not to be statistically significant. The extent of ICAM-1 upregulation by both TNF-a (Figs. 1D, 4) and IL-1/3 (Figs. 1E, 5) was intermediate between that of LPS and IFN-y. By
100
N
o 0
10 20 30 4O 5O 60 70
80
Fig. 3. HBMEC expression of 1CAM-1 after incubation with 500 U / m l (a), 100 U / m l (•), 50 U / m l (zx), or 10 U / m l (©) IFN-y. Values represent mean + SE of triplicate wells. * P < 0.05 compared to expression at 0 h.
16
100
20
~,
0
,
0
,
,
,
,
,
,
10 20 30 40 50 60 70
80
T i m e CO)
nificant. The increase caused by 1 U TNF-a/ml was not statistically significant until 12 h, and after incubation for 72 h, the increase declined to non-significant levels. The upregulation of ICAM-1 elicited by IL-1/3 reached maximal levels at 24 h (84% for 10 U, 76% for 1 U, and 67% for 0.1 U, all P < 0.05), and then declined. Incubation of endothelial cells with TNF-a and IFN-y combined resulted in marked increase in ICAM-1 expression that was even greater than that induced by LPS (Fig. 1F, 6). The number of immunostained cells ranged from 99 to 100% with 500 U IFN-T/ml plus 1 U to 100 U TNFa / m l and reached 97% with 100 U IFN-y/ml plus 100 U TNF-a/ml.
Reversibility
Fig. 4. H B M E C expression of ICAM-1 after incubation with 100 U / m l (El), 10 U / m l (z~), or 1 U / m l ((3) TNF-a. Values represent mean + SE of triplicate wells. * P < 0.05 compared to expression at 0 h.
24 h, 85% of cells showed positive staining (P < 0.05), after which there was a general trend of slight decline. The increased staining induced by 10 U and 100 U TNF-a/ml was statistically sig-
11111
Stimulation of HBMEC with LPS or cytokines for 2 days followed by withdrawal and culture of cells in regular growth media for 3 days was associated with variable degrees of reversal of ICAM-1 upregulation following LPS and TNF-a incubation, and lack of reversal following IL-1/3 and IFN-y treatment. These differences were due to cytokine and LPS removal since cultures incubated with cytokine or LPS for 5 days expressed high levels of ICAM-1 (Figs. 6-9). There was a decrease (10-30%) in the number of cells
i
lOO
[]
8o 60
rtl
~
20 ~, 0
I
0
I
|
I
I
I
10 20 30 40 50 60 70
i
o 1
I
10
80 Concenlration of TNF-a
"nine CO) Fig. 5. H B M E C expression of ICAM-1 after incubation with 10 U / m l (D), 1 U / m l (zx), or 0.1 U / m l (©) IL-1/3. Values represent mean +_SE of triplicate wells. * P < 0.05 compared to expression at 0 h.
100
(U/ml)
]Fig.6. Combined effects of 500 U / m l ([]), 100 U / m l ( A ), or 0 U / m l (©) IFN-N and i00 U/ml, I0 U/ml, or I U / m l TNF-a on the ICAM-I expression on H B M E C after 24 h. Values represent mean + SE. * P < 0.05 compared to expression at 0 h.
17
100
--//
//
100 8O
60
60
40
40
cD
20
8
r
2O
8 o --//
. 0.OOl
.
.
.
0.01
0.1
~. 1
0
//
0
10
Concentration of LPS (tag/ml)
1
i
1
10
100
Concentration of TNF-a CO/ml)
Fig. 7. Reversibility of ICAM-1 expression on HBMEC induced by incubation with LPS for 2 days then 3 days without LPS (o), with LPS for 2 days (D), with LPS for 5 days (zx), or without LPS (9). Values represent mean_+SE of duplicate wells.
Fig. 9. Reversibility of ICAM-1 expression on HBMEC induced by incubation with TNF-a for 2 days then 3 days without TNF-a (o), with TNF-ot for 2 days ([]), with TNF-a for 5 days (z~), or without TNF-a (~v). Values represent mean + SE of triplicate wells.
expressing ICAM-1 at all concentrations of LPS (Fig. 7). Reversibility of expression was minimal when 5 /zg/ml LPS was used. Reversibility of upregulation was not observed with IFN-7 at 100 U/ml, while 500 U / m l was associated with a 19% increase instead (Fig. 8). Stimulation followed by withdrawal of TNF-a resulted in decrease in ICAM-1 expression ranging from 6% to
23%. TNF-a concentration of 100 U / m l was associated with the smallest decrease while 1 U / m l induced the greatest (Fig. 9). Paradoxically, incubation of endothelial cells in IL-lfl-free media for 3 days following stimulation resulted in increase in the number of labelled cells ranging
//
100
!
4
8O
IO0
--//
"
6o
o
o o
40
40
8 0
-//
o
-// 100
1000
Concentration of H~N-7 (U/ml) Fig. 8. Reversibility of ICAM-1 expression on HBMEC induced by incubation with IFN-7 for 2 days then 3 days without IFN-7 (o), with IFN-y for 2 days (D), with IFN-y for 5 days ( zx), or without IFN-7 (*). Values represent mean + SE of triplicate wells.
0
i
0.1
10
Concen~afion of IL-113 (U/ml) Fig. 10. Reversibility of ICAM-1 expression on HBMEC induced by incubation with IL-1/3 for 2 days then 3 days without IL-1/~ (o), with IL-1/3 for 2 days (t~), with IL-1/3 for 5 days ( 4 ) , or without IL-lfl (~). Values represent mean_+SE of triplicate wells.
18 from a low of 1% with 10 U IL-1/3/ml to 17% with 0.1 U / m l (Fig. 10).
Discussion
The present study describes the induction of ICAM-1 expression on primary cultures of HBMEC by LPS and cytokines, and the reversibility of this expression. Intact, unstimulated endothelial cells constitutively express low levels of ICAM-1. Incubation with LPS, TNF-a, IL-1/3 and to a lesser extent IFN-T, results in upregulation of ICAM-1 expression which is time- and concentration-dependent and persists in the continuous presence of mediators in the culture. ICAM-1 is an inducible glycoprotein of the immunoglobulin superfamily, present at low levels on a variety of cell types including endothelial cells. It is important in both adhesion and transendothelial migration of lymphocytes, monocytes, neutrophils, basophils and eosinophils in vivo and in vitro (Albelda and Buck, 1990; A1belda, 1991; Bochner et al., 1991; Pober and Cotran, 1991; Shimizu et al., 1991). Antibodies against ICAM-1 and LFA-1 (the leukocyte receptor for ICAM-1) block the extravasation of eosinophils in vivo (Albelda, 1991) and prevent the adhesion of leukocytes to endothelial cells in vitro (Bochner et al., 1991). The pattern of ICAM-1 expression in inflammatory disorders of the nervous system was recently investigated in guinea pigs and mice during the acute and chronic phases of EAE. O'Neill et al. (1991) observed the upregulation of MALA-2 (the murine homologue of ICAM-1) on endothelial cells and mononuclear infiltrates in spinal cord lesions during the active episodes of chronic relapsing EAE in the Biozzi A B / H mouse. Recent studies on adoptively transferred E A E in the S J L / J mouse have shown that attachment and extravasation of mostly LFA-1 positive lymphocytes correlated with the appearance of MALA-2 on CNS vessels and the onset of clinical signs (Raine et al., 1990). In fact, the expression of MALA-2 fluctuated according to the clinical phase of EAE, with upregulation occurring with each relapse and down-regulation during remissions (Cannella et al., 1991). In both studies, endothelial and glial
cells of normal mice showed low levels of ICAM-1. Similarly, the normal spinal cord vasculature of guinea pigs has only a few ICAM-l-positive endothelial cells, while ICAM-1 was present on the endothelium of both lesion- and non-lesion-associated blood vessels of animals during the acute phase of E A E (Wilcox et al., 1990). In the same study, cultured endothelial cells isolated from guinea pig brains did not express ICAM-1 unless incubated in a lymphocyte-conditioned medium. The above animal studies suggest that upregulation of ICAM-1 may be important in the initial influx of lymphocytes into the brain in autoimmune demyelinating CNS disorders. The role of ICAM-1 in orchestrating leukocyte traffic across the BBB in the course of inflammatory diseases of the human CNS has not yet been fully investigated. In a recent neuropathological report, based on the examination of post-mortem material from normal and pathological brain tissue, only a small number of microvessels were ICAM-1 positive within the normal CNS. In contrast, numerous ICAM-l-positive vessels were observed in MS plaque edges, viral encephalitis lesions and infarcts (Sobel et al., 1990). Recently, high IFN-7 levels have been observed in EAE (Cannella et al., 1991), high TNF-a levels have been reported in some MS plaques (Brosnan et al., 1988; Hofman et al., 1989), and IL-1 has been found in MS lesions (Hofman et al., 1986). The upregulation of ICAM-1 induced by LPS and TNF-a on H B M E C is very drastic even by 4 h, and is maximal by 12 h, while IL-1/3 did not induce a maximal level until 24 h. IFN-7 led to a slight increase in expression which was not very significant statistically. In human umbilical vein ECs (HUVEC) and human dermal microvessel ECs (HDMEC), LPS caused a rise in ICAM-1 that did not peak until 24 h (Wellicome et al., 1990; Swerlick et al., 1991). Swerlick et al. (1991) and Pober et al. (1986) observed a similar rise of ICAM-1 in H U V E C and H D M E C following treatment with TNF-a and IL-1/3, in contrast to Wellicome et al. (1990) who showed that TNF and IL-1 induced a peak in ICAM-1 expression on H U V E C by 10 h. The slight upregulation of ICAM-1 observed after treatment of H B M E C with IFN-T is in keeping with previous reports indicating that IFN-T is a weak inducer of
19 ICAM-1 upregulation on extracerebral endothelium. Thus, IFN-y did not induce a noticeable change in ICAM-1 levels in HDMEC until 72 h (Swerlick et al., 1991), and caused a 'slower' increase than IL-1 and TNF in HUVEC (Pober et al., 1986). It is evident that the highest levels of ICAM-1 induced by LPS, TNF-a and IFN-y appear earlier in HBMEC than in HDMEC and some HUVEC cultures. The time course of ICAM-1 expression generated by IL-1/3 is similar in all three culture systems. The optimal concentration of cytokines and LPS required for the upregulation of ICAM-1 is different in cerebral and extracerebral endothelial cultures. Thus, LPS at a concentration of 0.1 p.g/ml induces maximal ICAM-1 levels in HUVEC (Wellicome et al., 1990) while a slightly higher concentration (1 ~ g / m l ) was required for HBMEC in our study. In contrast, HUVEC and HDMEC require 100 U T N F / m l (Pober et al., 1987; Detmer et al., 1990) and 100 U IFN-y/ml (Ruszczak et al., 1990) for maximal effect, while HBMEC showed no noticeable change in ICAM-1 level with 1-100 U TNF-a/ml, or 10-500 U IFN-y/ml. Human retinal capillary ECs, on the other hand, showed maximal induction of ICAM-1 upregulation with 5 U IFN-y/ml (Liversidge et al., 1990). IL-1 was required in a concentration greater than 10 U / m l for maximal effect on HUVEC (Pober et al., 1986; Wellicome et al., 1990), while 10 U / m l induced highest levels on HBMEC. HBMEC are thus more sensitive to cytokines and less sensitive to LPS than HUVEC and HDMEC with respect to ICAM-1 upregulation. The staining intensity of HBMEC varied even within monolayers derived from a single isolation. As a rule, the labelling of larger, older endothelial cells was stronger than that of actively proliferating cells. Such heterogeneity in ICAM-1 expression has also been reported in cultured human saphenous vein endothelial cells and HUVEC (Dustin and Springer, 1988) and in cultured fibroblasts (Dustin et al., 1986). The basal expression of ICAM-1 in HBMEC (22-40% of cells) is higher than the 18% in h u m a n retinal capillary ECs (Liversidge et al., 1990) and the 11-19% found in HDMEC (Ruszczak et al., 1990), which in turn are higher
than the expression in HUVEC (Swerlick et al., 1991). The maximal level of ICAM-1 expression in HBMEC is greatest with LPS (95% of cells labelled), intermediate with TNF-a and IL-1/3 (85%), and lowest with IFN- 7 (58%). This is in great contrast to HDMEC which expressed ICAM-1 to almost a 100% level with 100 U TNF/ml (Detmer et al., 1990), and showed a 78% labelling after incubation with 100 U IFN7 / m l and 93% with 1000 U IFN-y/ml (Ruszczak et al., 1990). HBMEC express much lower levels of ICAM-1 than HDMEC when activated with LPS or cytokines. Following incubation of HBMEC with a combination of TNF-a and IFN-7, 99 to 100% of the cells became labelled for ICAM-1. A synergistic effect of these two cytokines has also been observed in HDMEC (Detmer et al., 1990) and HUVEC (Doukas and Pober, 1990). In the latter study, it was also found that IFN- 7 does not significantly enhance the expression of ICAM-1 induced by IL-1/3. Other cell types within the brain respond much differently than ECs to cytokine stimulation. ICAM-1 expression on oligodendrocytes and astrocytes was markedly increased by IFN-y, while TNF-a, IL-la and LPS were less effective (Satoh et al., 1991). We found that removal of cytokines and LPS from the culture media was associated with variable expression of ICAM-1 after 3 days. Thus, there seemed to be a slight decrease in ICAM-1 with LPS (10-30%) and TNF-a (6-23%), and a slight increase with IFN- 7 (0-19%) and IL-1/3 (1-17%). Dustin and Springer (1988) observed that IL-1 and TNF caused ICAM-1 expression in HUVEC to become elevated, but returned to basal levels when the mediators were washed out. It would seem that HBMEC maintain their ICAM-1 expression much longer than HUVEC. The mechanism(s) of ICAM-1 induction by cytokines is presently poorly understood and rather controversial. Thus, recent studies indicate that activation of protein kinase C (PKC) may play a role in the induction of ICAM-1 expression on large vessel endothelium by IL-I, TNF-a, LPS and INF-y (Lane et al., 1990; Renkonen et al., 1990a). However, according to other investigators, IL-1 may operate via the cAMP (Renkonen et al., 1990b) and activation of PKC may not be the necessary pathway for ICAM-1 induction by
20 T N F - a ( R i t c h i e et al., 1991). W h e t h e r s i m i l a r m e c h a n i s m s a r e r e s p o n s i b l e for t h e i n d u c t i o n o f I C A M - 1 e x p r e s s i o n by c y t o k i n e s o n m i c r o v e s s e l e n d o t h e l i u m a n d m o r e specifically o n c e r e b r a l e n d o t h e l i u m r e m a i n s to b e i n v e s t i g a t e d . Our results indicate that HBMEC, compared to o t h e r E C s , t e n d to e x p r e s s a h i g h e r b a s a l l e v e l o f I C A M - 1 , a r e m o r e s e n s i t i v e to c y t o k i n e s , a n d less s e n s i t i v e to L P S , e x p r e s s I C A M - 1 e a r l i e r b u t at a l o w e r level w h e n i n d u c e d , a n d fail to r e v e r s e its e x p r e s s i o n w i t h i n a s h o r t t i m e u p o n t h e removal of these agents. These findings indicate t h a t u p r e g u l a t i o n o f I C A M - 1 by c y t o k i n e s o r L P S in v i v o m a y f a c i l i t a t e i n t e r a c t i o n s b e t w e e n L F A 1-positive l e u k o c y t e s a n d b r a i n m i c r o v e s s e l e n d o t h e l i a l cells a n d p l a y a n i m p o r t a n t r o l e in leukocyte trafficking across the blood-brain barr i e r in C N S i n f l a m m a t i o n .
Acknowledgements T h e a u t h o r s t h a n k D r s . P. R a o a n d T. S p r i n g e r for p r o v i d i n g t h e a n t i - I C A M - 1 a n t i b o d i e s , M r s . R. P r a m e y a for aid in t h e i s o l a t i o n a n d c u l t u r e o f H B M E C . T h i s s t u d y was s u p p o r t e d by g r a n t s from the National Institutes of Health (RO1NS23746), the Medical Research Council of Canada (MA-9823), and the British Columbia Health Care Research Foundation.
References Albelda, S.M. (1991) Endothelial and epithelial cell adhesion molecules. Am. J. Respir. Cell Mol. Biol. 4, 195-203. Albelda, S.M. and Buck, C.A. (1990) Integrins and other cell adhesion molecules. FASEB J, 4, 2868-2880. Bochner, B.S., Luscinskas, F.W., Gimbrone, M.A., Jr., Newman, W., Sterbinsky, S.A., Derse-Anthony, C.P., Klunk, D. and Schleimer, R.P. (1991) Adhesion of human basophils, eosinophils and neutrophils to interleukin 1-activated human vascular endothelial cell: contributions of endothelial cell adhesion molecules. J. Exp. Med. 173, 1553-1556. Brosnan, C.F., Selmej, K. and Raine, C.S. (1988) Hypothesis: a role of tumor necrosis factor in immune-mediated demyelination and its relevance to multiple sclerosis. J. Neuroimmunol. 18, 87-94. Cannella, B., Cross, A.H. and Raine, C.S. (1991) Adhesion-related molecules in the central nervous system. Up-regu-
lation correlates with inflammatory cell influx during relapsing experimental autoimmune encephalomyelitis. Lab. Invest. 65, 23-31. Detmer, M., Imcke, E., Ruszczak, Z. and Orfanos, C.E. (1990) Effects of recombinant tumor necrosis factor-alpha on cultured microvascular endothelial cells derived from human dermis. J. Invest. Dermatol. 95, 219S-222S. Dorovini-Zis, K., Prameya, R. and Bowman, P. (1991) Culture and characterization of microvessel endothelial cells derived from human brain. Lab. Invest. 64, 425-436. Doukas, J. and Pober, J.S. (1990) IFN-3, enhances endothelial activation induced by tumor necrosis factor but not IL-1. J. Immunol. 145, 1727-1733. Dustin, M.L. and Springer, T.A. (1988) Lymphocyte functionassociated antigen-1 (LFA-1) interaction with intercellular adhesion molecule-1 (ICAM-I) is one of at least three mechanisms for lymphocyte adhesion to cultured endothelial cells. J. Cell. Biol. 107, 321-331. Dustin, M.L., Rothlein, R., Bhan, A.K., Dinarello, C.A. and Springer, T.A. (1986) Induction by IL-1 and interferon, tissue distribution, biochemistry and function of a natural adherence molecule (ICAM-1). J. Immunol. 137, 245-254. Hofman, F.M., von Hanwehr, R.I., Dinarello, C.A., Mizel, S.B., Hinton, D. and Merrill, J.E. (1986) Immunoregulatory molecules and IL-2 receptors identified in multiple sclerosis brain. J. Immunol. 136, 3239-3245. Hofman, F.M., Hinton, D.R., Johnson, K. and Merrill, J.E. (1989) Tumor necrosis factor identified in multiple sclerosis brain. J. Exp. Med. 170, 607-612. Lane, T.A., Lamkin, G.E. and Wancewicz, E.V. (1990) Protein kinase C inhibitors block the enhanced expression of intercellular adhesion molecule-1 on endothelial cells activated by interleukin-1, lipopolysaccharide and tumor necrosis factor. Biochem. Biophys. Res. Commun. 172, 1273-1281. Liversidge, J., Sewell, H.F. and Forrester, J.V. (1990) Interactions between lymphocytes and cells of the blood-retina barrier: mechanisms of T lymphocyte adhesion to human retinal capillary endothelial cells and retinal pigment epithelial cells in vitro. Immunology 71,390-396. Munro, J.M., Pober, J.S. and Cotran, R.S. (1991) Recruitment of neutrophils in the local endotoxin response: association with de not,o endothelial expression of endothelial leukocyte adhesion molecule-1. Lab. Invest. 64, 295-299. O'Neill, J.K., Butter, C., Baker, D., Gschmeissner, S.E., Kraal, G., Butcher, E.C. and Turk, J.L. (199I) Expression of vascular addressins and ICAM-1 by endothelial cells in the spinal cord during chronic relapsing experimental allergic encephalomyelitis in the Biozzi A B / H mouse. Immunology 72, 520-525. Pober, J.S. and Cotran, R.S. (1990) The role of endothelial cells in inflammation. Transplantation 50, 537-544. Pober, J.S. and Cotran, R.S. (1991) What can be learned from the expression of endothelial adhesion molecules in tissues? Lab. Invest. 64, 301-305. Pober, J.S., Gimbrone, M.A. Jr., Lapierre, L.A., Mandrick, D.L., Fiers, W., Rothlein, R. and Springer, T.A. (1986) Overlapping patterns of activation of human endothelial
21 cells by interleukin 1, tumor necrosis factor, and immune interferon. J. Immunol. 137, 1893-1896. Pober, J.S., Lapierre, L.A., Stolpen, A.H., Brook, T.A., Springer, T.A., Fiers, W., Bevilacqua, M.P., Mendrick, D.L. and Gimbrone, M.A. Jr. (1987) Activation of cultured human endothelial cells by recombinant lymphotoxin: comparison with tumor necrosis factor and interleukin 1 species. J. Immunol. 136, 3319-3324. Raine, C.S., Cannella, B., Duijvestijn, A.M. and Cross. A.H. (1990) Homing to central nervous system vasculature by antigen-specific lymphocytes. II. Lymphocyte/endothelial cell adhesion during the initial stages of autoimmune demyelination. Lab. Invest. 63, 476-489. Renkonen, R., Mennander, A., Ustinov, J. and Mattila, P. (1990a) Activation of protein kinase C is crucial in the regulation of ICAM-1 expression on endothelial cells by interferon--/. Int. Immunol. 2, 719-724. Renkonen, R., Mennander, A., Mattila, P. and Ustinov, J. (1990b) Signal transduction during in vitro lymphocyte homing. Human Immunol. 28, 134-140. Ritchie, A.J., Johnson, D.R., Ewenstein, B.M. and Pober, J.S. (1991) Tumor necrosis factor induction of endothelial cell surface antigen is independent of protein kinase C activation or inactivation. Studies with phorbol myristate acetate and staurosporine. J. Immuno|. 146, 3056-3062. Ruszczak, Z., Detmer, M., Imcke, E. and Orfanos, C.E. (1990) Effects of rlFN alpha, beta and gamma on the morphology, proliferation and cell surface antigen expression of human dermal microvascular endothelial cells in vitro. J. Invest. Dermatol. 95, 693-699. Satoh, J.-l., Kastrukoff, L.F. and Kim, S.U. (1991) Cytokine-
induced expression of intercellular adhesion molecule-1 (ICAM-1) in cultured human oligodendrocytes and astrocytes. J. Neuropathol. Exp. Neurol. 3, 215-226. Shimizu, Y., Newman, W., Gopal, T.V., Horgan, K.J., Graber, N., Beall, L.D., van Seventer, G.A. and Shaw, S. (1991) Four molecular pathways of T cell adhesion to endothelial cells: roles of LFA-1, VCAM-I and ELAM-1 and changes in pathway hierarchy under different activation conditions. J. Cell. Biol. 113, 1203-1212. Sobel, R.A., Mitchell, M.E. and Fondien, G. (1990) Intercellular adhesion molecule-1 (ICAM-1) in cellular immune reactions in the human central nervous system. Am. J. Pathol. 136, 1309-1316. Springer, T.A. (1990) Adhesion receptor of the immune system. Nature 346, 425-434. Swerlick, R.A., Garcia-Gonzalez, E., Kubota, Y., Xu, Y. and Lanky, T.J. (1991) Studies of the modulation of MHC antigen and cell adhesion molecule expression on human dermal microvascular endothelial cells. J. Invest. Dermatol. 97, 190-196. Wellicome, S.M., Thornhill, M.H., Pitzalis, C., Thomas, D.S., Lanchbury, J.S.S., Panayi, G.S. and Haskard, D.O. (1990) A monoclonal antibody that detects a novel antigen on endothelial cells that is induced by tumor necrosis factor, IL-I, or lipopolysaccharide. J. Immunol. 144, 2558-2565. Wilcox, C.E., Ward, A.M., Evans, A., Baker, D., Rothlein, R. and Turk, J.L. (1990) Endothelial cell expression of the intercellular adhesion molecule (ICAM-1) in the central nervous system of guinea pigs during acute and chronic relapsing experimental allergic encephalomyelitis. J. Neuroimmunol. 30, 43-51.