- Email: [email protected]
Adhesion properties of Mono Mac 6, a monocytic cell line with characteristics of mature human monocytes
ATHEROSCLEROSIS
ELSEVIER
Atherosclerosis 113 (1995) 99-107
Adhesion properties of Mono Mac 6, a monocytic cell line with characteristics of mature human monocytes Wolfgang Erl*, Christian Weber, Celina Wardemann, Peter C. Weber Institut
Jlir Prophylaxe
und Epidemiologic
der Kreislaufkrankheiten, Germany
Universitiit
Miinchen,
Pettenkoferstr.
9, 80336 M&hen,
Received 9 May 1994; revision received 16 August 1994; accepted 2 September 1994
-Abstract
Progressin the understandingof blood cell - endothelialcell interactionshasbeenachievedby the development of in-vitro modelsystems.We describeadhesionpropertiesof the recentlyestablished humanmonocyticcellline Mono Mac 6. Thesecellsshowedincreasedadherenceto unstimulatedand tumour necrosisfactor (TNF)-(r (50 U/ml) stimulated humanumbilicalvein endothelialcells(HUVEC) (9.4% f 0.4%and 56.5%f 3.3%),ascomparedto U937cells (2.6%f 0.8%and 40.0%f 8.4%).The valuesweresimilarto freshlyisolatedhumanblood monocytes(18.8%& 7.5% and 55.7%f 9.3%,respectively).Maximal bindingwas6.2 f 0.6 Mono Mac 6 cellsper HUVEC, which was34%less than U937cells(8.9 f 0.3). The lower numberof adherentMono Mac 6 cellsper HUVEC couldbe dueto their larger size,asassessed by flow cytometry. Blockingexperimentswith monoclonalantibody (mAb) directedagainstE-selectin, VCAM-1 andICAM- on HUVEC and CD1lb or CD14on Mono Mac 6 cellsdemonstratedthe contribution of these moleculesto Mono Mac 6 adherence.Reducedbinding after 24 h parallelsthe declineof E-selectinexpressionin HUVEC. Linearity of cell binding wasconfirmed from 0.2 x lo6 to 1.0 x lo6 Mono Mac 6 cells. Expressionof CD11b and CD14 in Mono Mac 6 cellsand in isolatedhumanmonocytesbut not in U937cellsleadingto interaction with ICAM- on HUVEC appearsto beresponsiblefor the increasedadhesionof Mono Mac 6, ascomparedto U937 cells.In conclusion,Mono Mac 6 cellsexhibit adhesioncharacteristicssimilarto humanmonocytesand thereforeseem to be a cell line well suitedfor studying monocyte-endothelialcell interactions. Keywords:Humanumbilical vein endothelialcells; Tumor necrosisfactor; Adhesionmolecules
1. Introdwtioo * Corresponding author.‘Tel.:49 89 5160 4350; Fax: 49 89 5160 4352. Abbreviations: BCECF/AM, 2’,7’-bis-(carboxyethyl)-5(6’)carboxyfluorescein pentacetoxy methylester; ECs,, halfmaximal efkctive concentration; EC, endothelial cell(s); FACS, fluorescence activatedcell sorter; HUV, human umbilical vein; mAb, monoclonal antibody(s); sMF1, specific mean fluorescence intensity; TEM, transmission electron microscope; TNF, tumor necrosis factor a
The molecular interactions determining monocyte adhesion to vascular endothelium are becoming increasingly evident [I]. However, in vitro studies using isolated human blood monocytes are accompanied by major problems, such as a high and varying basal adherence to endothelial cells. Several human tumor cell lines with monocyte-
0021-9150/95/.W9.50 0 19!)5 Elsevier Science Ireland Ltd. All rights reserved SSDI 002 I-!)1 50(94)05434-K
loo
W. Ed et al. /Atherosclerosis
related functions have been established, such as U937 [2], HL-60 [3] and THP-1 [4]. U937 cells are well defined with respect to adhesive properties [5] and commonly used in assessing monocytic cell adhesion to human umbilical vein endothelial cell (HUVEC) monolayers. The prolonged culture of this cell line over the past 20 years has caused selection of different clones following mutations under varying conditions. The occurrence of less mature U937 clones which lack certain monocytic characteristics has prompted an intensive search for other monocytic cell lines [6,7]. Recently, Mono Mac 6, a more mature monocytic cell line, has been described [8]. Untreated Mono Mac 6 cells exhibit many aspects closely related to human monocytes, such as expression of NaF-sensitive non-specific esterase, production of reactive oxygen, staining for My4, as well as many other monocyte-specific mAb, secretion of cytokines and phagocytosis in 80% of cells. This led us to investigate the as yet undefined adhesion properties of the Mono Mac 6 cell line in a model of cell-cell interaction using unstimulated and tumor necrosis factor OL(TNF-(r) stimulated HUVEC. The results show that Mono Mac 6 cells exhibit increased adhesion to stimulated and unstimulated HUVEC as compared to U937 cells, indicating properties of a more mature monocyte. The specific adhesion properties can be explained by the distinct expression of adhesion molecules on Mono Mac 6 cells. 2. Materials and methods 2.1. Cell culture
HUVEC were obtained from human umbilical cord veins by digestion with a-chymotrypsin, cultured in low-serum EGM-medium (Promo Cell, Heidelberg, FRG) using T-25 and T-150 flasks precoated with collagen in a 5% CO2 atmosphere at 37°C [9]. Cell purity was assessed by morphology (‘cobblestone’) and factor VIII staining. Confluent HUVEC passage 2 were detached by 0.05% trypsin/0.02% EDTA and grown in T-25 flasks or 24 well plates for treatment with human recombinant TNF (8.7 x lo6 U/ml in EGM, provided by Knoll-BASF, Ludwigshafen, FRG) at indicated concentrations and periods. For Bow cytometry, cells were harvested by careful treatment with
I13 (1995)
99-107
trypsin/EDTA, antagonized by immediate addition of foetal calf serum (FCS), and counted in triplicate. U937 cells were grown in RPM1 1640 medium with 2 mM L-glutamine and 10% FCS in suspension [lo]. Mono Mac 6 cells (kindly provided by Dr. H.W.L. Ziegler-Heitbrock) were cultured in RPM1 1640 with 10% FCS, 1 mM oxalacetate, 1 mM Na-pyruvate, 1 x nonessential amino acid solution, 2 x penicillin/streptomycin solution and 2 mM L-glutamine, using 24 well plates (Falcon, Becton Dickinson, Heidelberg, FRG) in a humidified 5% CO2 atmosphere. Before addition of FCS, the medium was ultrafiltrated through a Gambro 200 column (Hechingen, FRG) to eliminate lipopolysaccharide (LPS) [ll]. The LPS content of the medium was determined to be <2 pg/ml by Limulus amaebocyte assay. Human monocytes were freshly isolated from healthy donors by Ficoll density gradient centrifugation and subsequent magnetic cell separation (Miltenyi Biotec, Bergisch Gladbach, FRG), yielding a purity of >90%. Under all conditions, cell viability was >95% as judged by ethidium bromide/a&din orange fluorescence. 2.2. Immunofuorescence
Cells (2 x 105) were treated for 30 min with saturating amounts of mouse anti-VCAM-1, antiICAM- 84HlO (Dianova, Hamburg, FRG), antiCD62E (E-selectin) mAb (HUVEC), anti-CD1 1b (Mac-l), anti-CD18 (&), antXDw49d (VLA-4), anti-CD29 mAb (0,) or IgGi, isotype control (all Camon, Wiesbaden, FRG), FITC-anti-CD14 mAb (Mo2) or FITC-IgM isotype control (Coulter, Hamburg, FRG) (Mono Mac 6, U937) in phosphate buffered serum (PBS) with 0.5% bovine serum albumin (BSA) on ice. For staining, cells were reacted with goat anti-mouse FITCIgG,, (Camon) on ice after washing twice with PBS. To avoid unspecific binding, cells were preincubated with 5% human serum in PBS for 15 min on ice. Samples were washed twice with FACS buffer (Be&on Dickinson), fixed in 2% paraformaldehyde to prevent aggregation and analysed with 10 000 cells/sample in a FACScan, as described [9]. After correction for unspecific binding (isotype control), specific mean fluorescence intensity (sMF1) was expressed in channels.
W. Ed et al. /Atherosclerosis
2.3. Monocytic
ceil adhesion assay
HWEC were seeded in 24 well plates 48 h before the experiment. Only confluent monolayers were used, as confirmed by microscopic inspection. Fresh medium was provided before stimulation with TNF. Monocytic cells (3 x 107) were incubated in RPM1 1640 medium (6 ml) containing 2% FCS and 10 &ml of the fluorescent dye BCECF/A.M (Boehringer, Mannheim, FRG) at 37°C for 30 min [9]. Dye loading was stopped by adding 44 ml cold RPM1 1640 (2% FCS). Fluorescence labeled cells were pelleted and resuspended (0.1-3.0 >: 106/ml) in medium 199 (Gibco-BRL) with 10 mM HEPES buffer (M199H). HUVEC were washed twice with M199H before addition of loaded cells and incubated at 37°C 5% CO2 at 90% humidity. After 30 min, cell suspensions were withdrawn, HWEC were gently washed with M199H and inverted plates were centrifuged (50 x g, 55min, 4°C) [12]. Cells were lysed with 1 ml of 0.1% Triton X-100 in 0.1 M tris buffer/well (pH 8.0) and fluorescence was measured by a PTI deltascan spectrofluorimeter (excitation 485 nm, emission 535 nm). The number of adherent cells/well was calculated by comparing the amount of fluorescence to a standard curve of BCECF activity/cell (and expressed as % adhesion of added cells per well. Some HUVEC were pretreated with saturating amounts of anti-E-selectin (Camon), anti-.VCAM-1 1Gll or anti-ICAMmAb 84HlO (Dianova) 10 min before addition of U937 or Mono Mac 6 cells. Some monocytic cells were pretreated with anti-CD1 lb (Mac-l) or anti-CD14 mAb (UCHM-1) 10 min before addition to HUVEC. All other reagents were from Sigma, Deisenhofen, FRG.
II3
(1995)
101
99-107
TNF to obtain more adhesive HUVEC. As observed for U937 cells [5], binding of Mono Mac 6 cells to TNF (50 U/ml)-stimulated HUVEC rapidly increased within the first 20 min of coincubation at 37°C peaked at 30 min and remained almost unchanged for up to 2 h (Fig. 1). Binding experiments at 4°C revealed a more delayed onset of attachment (not shown). The time required for maximal binding at 4°C was not determined, because with longer incubation periods at this temperature, HWEC tended to detach from the inverted well plates during centrifugation. of monocytic cell binding To determine maximal binding and the optimal number of Mono Mac 6 cells to be employed in our assay, we incubated TNF (50 U/ml)-stimulated HWEC with increasing cell numbers (Fig. 2). The average number of HUVEC per well was 6.6 f 0.9 x lo4 (mean f SD, n = 8). The number of Mono Mac 6 cells bound at an excess of cells employed (3.0 x 106) was 4.1 f 0.4 x lo5 (n = 5), which is equivalent to 6.2 f 0.6 Mono Mac 6 cells per EC. The higher maximal binding of U937 cells (5.8 f 0.2 x lo’, n = 10; i.e. 8.9 f 0.3 U937 cells per HUVEC) suggests different spatial requirements for U937 and Mono
3.2. Saturability
2.4. Statistics
Data were analyzed by Student’s t-test.
,r 0
3. Results 3.1. Time course of binding of Mono Mac 6
To evaluate optimal and reproducible conditions for adhesion, we first characterized the time course of Mono Mac 6 cell binding to HWEC. As expected, Mono Mac 6 cell adherence in unstimulated EC was comparably low. We therefore used
50lJ/ml,6h
TNF-a, 0 10
20
30
time
40
50
60
3 120
(min)
Fig. 1. Time course of Mono Mac 6 cell binding. HUVEC in 24 well plates were stimulated with TNF (50 U/ml). Each well was treated 6 h before addition of 7.5 x 10’ fluorescence labelled Mono Mac 6 cells and incubated for designated periods. Adhesion was determined as in Materials and methods. Values represent the mean of 3 separate experiments run in triplicate. Percentage of adherent cells is expressed as mean f S.D.
W. Ed et al. /Atherosclerosis 113 (1995) 99-107
102
Mac 6 cells. This is most probably due to different cell size, as indicated by FACS forward scatter analysis. As a relative measure of cell size, the mean value in channels was 315 f 6 (n = 3) for Mono Mac 6 and 254 f 4 for U937 cells, revealing that Mono Mac 6 cells are approximately 25% bigger than U937 cells. This is in accordance with an average size of 16.0 and 12.5 pm found by TEM for Mono Mac 6 [8] and U937 cells [2], respectively. This difference may explain the 34% lower maximal binding of Mono Mac 6 as compared to U937 cells. We found a wide linear correlation between added and adherent Mono Mac 6 cells in a range of 0.2 x lo6 to 1.0 x lo6 (Fig. 2, insert). This “0
6 A
0.5 Mono
1.0 1.5 2.0 Mac 6 cells, free (xl@)
2.5
range was smaller in U937 cells (0.2 X lo6 to 0.7 x 106). In our experiments, not more than 0.8 x lo6 monocytic cells/ml were employed. 3.3. Dose and time dependenceof TNF-stimulated adhesion
TNF induced a dose- and time-dependent enhancement of Mono Mac 6 and U937 cell adhesion. As demonstrated in Fig. 3, basal adhesion of Mono Mac 6 cells was 9.4% f 0.4% (all values in % mean i SD, n = 3), stimulation of HUVEC with TNF (1 U/ml) for 6 h showed no effect (10.3% f l.O%), a rapid increase in adhesion above 1 U/ml, and a plateau at 10 U/ml (55.1% f 3.6%). TNF did not further increase adhesion at 50 or 100 U/ml (56.5% * 3.3% and 54.7% * 2.8%, respectively). In U937 cells, TNF increased adhesion at 1 U/ml compared to basal
3.0
r 60
.o=
t
40-
3 e;
1
_ 4
20-
k 0
0.5
1.0
1.5
2.0
2.5
24 h
3.0
U937 cells, free (x1$)
Fig. 2. Saturability and linearity of monocytic cell binding. HUVEC were pretreated with 50 U/ml TNF for 6 h. Increasing numbers (0.1 x IO6 to 3.0 x 106) of fluorescence labelled monocytic cells (A: Mono Mac 6, B: U937) were added for 30 min co-incubation at 37OC and adhesion was measured as in Materials and methods. Linear binding was determined by a correlation coefficient (r2) > 0.995, as shown in insert of Figs. 2A and B. Values represent the mean of 3 separate experiments run in triplicate. Data are expressed as mean + SD.
0
5 10
25
50
TNF U/ml
Fig. 3. Time and dose. dependence of Mono Mac 6 (-6) and U937 (-0-) cell binding. HUVEC were incubated with increasing concentrations of TNF for 6 and 24 h. Fluorescence labelled monocytic cells were added and % adhesion determined as in Materials and methods. Values represent 4 independent experiments performed in triplicate and are expressed as mean f SD.
W. Ed et al. /Atherosclerosis
values (12..0% f 4.0% vs. 2.6% f 0.8%). Adherence was further enhanced at 10 U/ml and maximal at 50 U/ml (40.0% * 8.4%) or at 100 U/ml (38.0% f 7.7%). As shown in Fig. 3, treatment of HWEC with TNF for 24 h resulted in similar binding characteristics, but lower adhesion for both cell lines as compared to 6 h. Again, TNF caused maximal attachment at 50 U/ml, and did not further increase adhesion at 100 U/ml (data not shown). At 6 h, the EC,, for TNF was approximately 2 U/ml for both cell lines. Stimulation with TNF for 24 h exhibited an ECSo of < 3 U/ml for Mono Mac 6 and >6 U/ml for U937 cells, In comparison, freshly isolated human monocytes showed (an adhesion of 18.8% f 7.5% to unstimulated and of 55.7% f 9.3% to TNF (50 U/ml, 6 h&stimulated HWEC. 3.4. Flow cytometric analysis and blocking of adhesion moleculeson stimulated HUVEC
To elucidate these differences in adhesion, we investigated the time dependent induction of adhesion molecules. Endothelial leukocyte adhesion E-selectin
I13 (1995)
103
99-107
molecule 1 (ELAM- 1, E-selectin) and vascular cell adhesion molecule 1 (VCAM-1) are absent in unstimulated HWEC, but contribute signiticantly to monocyte adhesion on EC [ 131. We used TNF (50 U/ml) to induce these molecules and analyzed surface expression at 6 and 24 h by flow cytometry (Fig. 4). E-selectin peaked at 6 h and declined to basal values at 24 h. VCAM-1 was induced after 6 h and remained increased at 24 h. Intercellular adhesion molecule 1 (ICAM-1) was constitutively expressed and upregulated by TNF at 6 and 24 h. These distinct kinetics of adhesion molecule expression on HWEC can explain the decreased cell adherence seen after 24 h. The contribution of E-selectin, VCAM-1 and ICAMto Mono Mac 6 cell adhesion was further investigated by incubation of TNF (50 U/ml)stimulated HWEC with blocking mAb. As shown in Fig. 5, employment of mAb to a single molecule showed only slight inhibition of adhesion. Combination of 2 mAb exhibited a significant decrease of adherent Mono Mac 6 cells, which was not further pronounced using a set of 3 mAb. VCAM-1
ICAM-
Fig. 4. Surface molecule expression of HUVEC stimulated by TNF. HUVEC were stimulated for 0, 6, and 24 h with 50 U/ml TNF, stained for E-selectin (left), VCAM-1 (middle), ICAM-I (right) or isotype control (reflecting unspecific binding) and analyzed by flow cytometry. Shown are original histograms representative of 3 similar experiments. Relative cell numbers are vertical and fluorescence intensity on a log 10 scale is horizontal.
104
W. Ed et al. /Atherosclerosis 113 (1995) 99-107
60
-I
TNF50Wml anti-E-se&tin mAb anti-VCAM-1 mAb anti-ICAM- mAb
T
-
+++ -
+ -
+ -
+++++ + + + -
+ +
+ +
+ + +
Fig. 5. Effect of anti-E-se&tin, anti-VCAM-1 and anti-ICAMmAb on Mono Mac 6 adhesion to TNF-stimulated EC. HUVEC were incubated with 50 U/ml TNF for 6 h. Medium was changed and mAb were reacted for 10 min at room temperature. Labelled Mono Mac 6 cells (7.5 x 10’ per well) were added and co-incubated for 30 min at 37°C. Nonadherent cells were removed and % adhesion determined as described. Values are means of 3 separate experiments run in duplicate and are expressed as mean f S.D.(*P c 0.05, l *P c 0.01 vs. 50 U/ml TNF).
3.5. Flow cytometric analysis of adhesionmolecules on Mono Mac 6 cells
A monocytic counterreceptor for E-selectin is sialyl-Lewis X, for VCAM-1 very late activation antigen 4 (VLA-4) and for ICAMthe CD1 1b/ CD18 integrin. Sialyl-Lewis X is expressed on untreated U937 cells [14], as is the BZintergrin Table 1 Adhesion molecules on Mono Mac 6 cells and U937 cells
CD1 lb (Mac-l) CD18
632)
CDw49d (VLA-4) CD29 (8,) CD14 (Mo2)
Mono Mac 6
u937
139 f 11 105 f 18 82 l 12 175 l 15 97 + 4
5zt6 114 * II 174 f 13 246zt 16 4*3
Mono Mac 6 and U937 cells were cultured for 72 h, stained with mAb directed against indicated molecules or respective isotype control and analyzed by flow cytometry. Data are expressed as sMF1 corrected for unspecific binding in channels and represent mean f S.D. of four separate experiments.
CD18, whereas the corresponding monomer CD1 lb is virtually absent on U937 cells. In contrast, CD1 1b is markedly expressed on Mono Mac 6 cells (Table 1) and on freshly isolated monocytes (sMFI 259 f 41 channels). Since CD1 lb is found in U937 cells only after differentiation [ 151, this indicates a more mature phenotype of Mono Mac 6 cells. Consistent with results showing downregulation of VLA4 during monocyte differentiation [ 161, a less marked expression of VLA4 is found in Mono Mac 6 as compared to U937 cells. CD14 on mature monocytes has recently been shown to contribute to monocyte adhesion to cytokine stimulated HUVEC [ 171. We therefore measured CD14 expression in both cell lines. CD14 was not detectable in U937 cells, where it can be induced by differentiation [ 151, but constitutively expressed on Mono Mac 6 cells and mature monocytes [8]. Consistently, preincubation of Mono Mac 6 cells and human monocytes but not U937 cells with anti-CD14 or anti CD1 lb mAb significantly reduced adhesion (Fig. 6). In contrast, Polymyxin B (100 U/ml) had no effect, excluding cell activa-
W. Ed
et al. /Atherosclerosis
Mono
antXD14mAb anti-CD11 b mAb
-
Mac 6
+ -
-’ +
113 (1995)
99-107
105
blc+od monocytes
-
+ -
+
_
+ -
+
Fig. 6. Effect of anti-CD14 and anti-CD1 lb mAb on monocytic cell adhesion to TNF-stimulated EC. HUVEC were incubated with 50 U/ml TNF for 6 h. Labelled monocytic cells were preincubated with mAb for 10 min prior to addition and co-incubation with HUVEC for 30 min at 37OC. Nonadherent cells were removed and % adhesion was determined as described. Values are means of 3 separate experiments run in duplicate and are expressed as mean f S.D.(*P < 0.05 vs. 50 U/ml TNF).
tion by LPS. Taken together, the data suggest that differences in adhesion properties of U937 and Mono Mac 6 cells to HWEC are primarily due to a different contribution of CD1 1b/ICAM- 1 dependent interaction. 4. Discmion As a consequence of the immature nature of U937 cells which includes several subclones, a cell line with characteristics of mature monocytes would be useful to further investigate the interactions of white blood cells with vascular endothelial cells. In our study, we describe the adhesion properties of the novel human monocytic cell line Mono Mac 6, in comparison to the widely used U937 cell line. The time course of Mono Mac 6 cell binding to TNF-stim.ulated HWEC was comparable to that of U937 cells, suggesting that similar mechanisms of attachment are involved. This can also be derived from cross-competition experiments in which both cell types were able to block adhesion of each other when equal cell numbers were simultaneously incubated with HWEC (unpublished results). All experiments were performed at 30 min, when maximal adhesion occurred for both Mono Mac 6 and U937 cells. To investigate firm adhesion, we
performed our experiments at 37”C, since at 4°C HWEC showed a tendency to detach from wells during inverse centrifugation. After 30 min at 37”C, spreading of both cell types could rarely be detected by microscopic inspection. Mono Mac 6 cells more readily adhere to unstimulated H WEC than U937 cells. The higher basal adhesion may be due to enhanced interaction with constitutively expressed HWEC adhesion molecules, such as ICAM- 1. A high basal adhesion reported for U937 cells [18] may be attributed to the use of proliferating EC, which are more adhesive than confluent cells [19], or culture medium with high serum content, which may promote expression of binding sites and EC proliferation beyond contact inhibition (unpublished results). Hence, Mono Mac 6 adhesion properties may be reflective of a more mature monocytic cell, as similar adhesion values for isolated blood monocytes were demonstrated in our experiments and have recently been published [20]. The high and varying basal adhesion of human blood monocytes [21] may result from activation during preparation procedures, which was minimized by the use of magnetic cell separation in our study. However, Mono Mac 6 cells show markedly less variance in binding as compared to human monocytes [20,21 J, possibly due to a more uniform population during
106
W. Ed et al. /Atherosclerosis
cell culture. Mono Mac 6 cells also exhibit a higher maximal adhesion to TNF-stimulated HUVEC as compared to U937 cells. The assessment of the molecular moieties on both cell lines provides an explanation for these observations, since the ICAMcounterreceptor CD1 lb is markedly expressed on Mono Mac 6 cells and monocytes, but not detectable on U937 cells. The CD1 l-associated /32 integrin CD18 is also found on U937 cells, indicating the presence of at least one CD1 1 chain. Since only an activated subpopulation of CD1 lb has been described to mediate adherence to ICAM[22], the increased adhesion of Mono Mac 6 cells may be due to the occurrence of more CD1 1 molecules in an adhesive conformation. The marked expression of CD14 in Mono Mac 6 cells and inhibition of adhesion by anti-CD14 mAb further support a contribution of CD14 to adhesion of more mature monocytic cells to cytokinestimulated HUVEC [17]. In this context, it has been suggested that binding of CD14 to surface structures of stimulated EC can generate crosslinking of CD14, resulting in activation of CDllI CD18 and ICAMdependent monocyte adherence [23]. Inhibition experiments with blocking mAb demonstrated that several ligand pairs are involved in maximal adherence, since blocking of one EC adhesion molecule only slightly affected adhesion, while combinations of 2 Ab proved to be very effective. Consistently, marked inhibition of monocyte adhesion required combination of anti-Eselectin and anti-VCAM-1 mAb, indicating the dependence on both molecules [ 121. An interpretation of this finding would be that other systems can compensate, as long as only one type of binding site is restricted. This is in accordance with observations that VLA4 and CD1 l/CD18 act interdependently in monocyte transmigration [24]. The incomplete reduction of adhesion may be due to the blocking properties, shedding or incorporation of mAb, or the result of yet unidentified adhesion molecules on stimulated HUVEC. Furthermore, preincubation of monocytic cells with either anti-CD1 1b or anti-CD14 mAb attenuated adhesion of Mono Mac 6, but not of U937 cells. In conclusion, we have described the adhesion characteristics of the mature human monocytic
113 (1995)
99-107
cell line Mono Mac 6 to HUVEC monolayers, which may be well suited for studying biochemical and functional interactions of monocytes and EC. Acknowledgements This study was partially supported by grants from ‘Bundesministerium fur Forschung und Technologie’ BMFT 07ERGO3 and by ‘AugustLenz-Stiftung’. This work partially fulfills requirements for the doctoral theses of W. Erl and C. Wardemann. We thank E. Bretzke for secretarial assistance. References 111 Beekhuizen, H. and van Furth, R., Monocyte adherence
to human vascular endothelium, J. Leukoc. Biol., 54 (1993) 363. PI Sundstrom, C. and Nilsson, K., Establishment and characterization of a human histiocytic lymphoma cell line (U-937), Int. J. Cancer, 17 (1976) 565. [31 Collins, S.J., Ciallo, R.C. and Gallagher, R.E., Continuous growth and differentiation of human myeloid leukaemic cells in suspension culture, Nature., 270 (1977) 347. 141 Tsuchiya, S., Yamabe, M., Yamaguchi, Y., Kobayashi, Y., Konno, T. and Tada, K., Establishment and characterization of a human acute leukemia cell line (THP-I), Int. J. Cancer, 26 (1980) 171. [51 DiCorletto, P.E. and de la Motte, C.A., Characterization of the adhesion of the human monocytic cell line U937 to cultured endothelial cells, J. Clin. Invest., 75 (1985) 1153. 161 Ohno, H., Fukuhara, S., Doi, S. et al., Establishment of a novel acute monoblastic leukemia cell line (YK-M2) having a near triploid karyotype, Cancer Res., 46 (1986) 6400. [71 Ohta, M., Furukawa, Y., Ide, C., Akiyama, N., Utakoji, T., Miura, Y. and Saito, M., Establishment and characterization of four human monocytoid cell lines (JOSK-I, -S, -M, and -K) with capabilities of monocytemacrophage lineage differentiation and constitutive production of interleukin 1, Cancer Res., 46 (1986) 3067. Is1 Ziegler-Heitbrock, H.W.L., Thiel, E., Fiitterer, A., Herzog, V., Wirtz, A. and Riethmtiller, G., Establishment of a human cell line (Mono Mac 6) with characteristics of mature monocytes, Int. J. Cancer, 41 (1988) 456. 191 Weber, C., Erl, W., Pietsch, A., Striibel, M., ZieglerHeitbrock, H.W.L. and Weber, P.C., Antioxidants inhibit monocyte adhesion by suppressing nuclear factor-KB mobilization and induction of vascular cell adhesion molecule 1 in endothelial cells stimulated to generate radicals, Arteriosclerosis Thromb., 14 (1994) 1665. 1101Weber, C., Aepfelbacher, M., Lux, I., Zimmer B. and
W. Erl et al. /Atherosclerosis
[I I]
[12]
[13]
[14]
[15] [16]
Weber, P.C., Docosahexaenoic acid inhibits PAF and LTD4-stimulated [Ca++]i-increase in differentiated monocytic U937 cells, Biochim. Biophys. Acta, 1133 (1991) 38. Weber, C., Aepfelbacher, M., Haag, H., Ziegler-Heitbrock, H.W.L. and Weber, PC., Tumor necrosis factor induces enhanced responses to platelet-activating factor and differentiation in human monocytic Mono Mac 6 cells, Eur. J. Immunol., 23 (1993) 852. Charo, I.F., Yuen, C. and Goldstein, I.M., Adherence of human polymorphonuclear leukocytes to endothelial monolayers: effect of temperature, divalent cations, and chemotactic factors on the strength of adherence measured with a new centrifugation assay, Blood, 65 (1985) 473. Carlos, T., Kovach, N., Schwartz, B. et al., Human monocytes bind to two cytokine-induced adhesive ligands on cultured human endothelial cells: endothelialleukocyte adhesion molecule-l and vascular cell adhesion molecule-l, Blood, 77 (1991) 2266. Fukushima, K., Hirota, M., Terasaki, P.I., Wakisaka, A., Togashi, H. and Chia, D., Characterization of sialosylated Lewis x as a new tumor-associated antigen, Cancer Res., 44 (1984) 5279. &erg, F., Botling, J. and Niisson, K., Macrophages and the cyt.okine network, Transplant. Proc., 25 (1993) 2044. Bauvois, B., Rouillard, D., Sanceau, J. and Wietzerbin, J., IFN--r and transforming growth factor-b differentially regulate tibronectin and laminin receptors of human differentiating monocytic cells, J. Immunol., 148 (1992) 3912.
I13 (1995)
99- 107
107
[17] Beekhuizen, H., Blokland, I., Corsel-van Tilburg, A., Koning, F. and van Furth, R., CD14 contributes to the adherence of human monocytes to cytokine-stimulated endothelial cells, J. Immunol., 147 (1991) 3761. [18] DiCorletto, P.E. and de la Motte, CA., Thrombin causes increased monocytic-cell adhesion to endothelial cells through a protein kinase C-dependent pathway, B&hem. J., 264 (1989) 71. [19] &Bono, D.P. and Green, C., The adhesion of different cell types to cultured vascular endothelium: effects of culture density and age, Br. J. Exp. Pathol., 65 (1984) 145. [20] Couffinhal, T., Duplaa, C., Labat, L., Moreau, C., Bietz, I. and Bonnet, J., Effect of low density lipoprotein on monocyte adhesiveness to endothelial cells in vitro, Atherosclerosis, 99 (1993) 35. [21] McGregor, P.E., Agrawal, D.K. and Edwards, J.D., Attenuation of human leukocyte adherence to endothelial cell monolayers by tyrosine kinase inhibitors, B&hem. Biophys. Res. Commun., 198 (1994) 359. [22] Diamond, MS. and Springer, T.A., A subpopulation of Mac-l (CD1 I b/CDlS) molecules mediates neutrophil adhesion to ICAM- and fibrinogen, J. Cell Biol., 120 (1993) 545. (231 Beekhuizen, H., Blokland, I. and van Furth, R., Crosslinking of CD14 molecules on monocytes results in a CD1 l/CD18- and ICAM-l-dependent adherence to cytokine-stimulated human endothelial cells, J. Immunol., 150 (1993) 950. [24] Winn, R.K. and Harlan, J.M., CD18-dependent neutrophil and mononuclear leukocyte emigration into the peritoneum of rabbits. J. Clin. Invest., 92 (1993) 1168.
ELSEVIER
Atherosclerosis 113 (1995) 99-107
Adhesion properties of Mono Mac 6, a monocytic cell line with characteristics of mature human monocytes Wolfgang Erl*, Christian Weber, Celina Wardemann, Peter C. Weber Institut
Jlir Prophylaxe
und Epidemiologic
der Kreislaufkrankheiten, Germany
Universitiit
Miinchen,
Pettenkoferstr.
9, 80336 M&hen,
Received 9 May 1994; revision received 16 August 1994; accepted 2 September 1994
-Abstract
Progressin the understandingof blood cell - endothelialcell interactionshasbeenachievedby the development of in-vitro modelsystems.We describeadhesionpropertiesof the recentlyestablished humanmonocyticcellline Mono Mac 6. Thesecellsshowedincreasedadherenceto unstimulatedand tumour necrosisfactor (TNF)-(r (50 U/ml) stimulated humanumbilicalvein endothelialcells(HUVEC) (9.4% f 0.4%and 56.5%f 3.3%),ascomparedto U937cells (2.6%f 0.8%and 40.0%f 8.4%).The valuesweresimilarto freshlyisolatedhumanblood monocytes(18.8%& 7.5% and 55.7%f 9.3%,respectively).Maximal bindingwas6.2 f 0.6 Mono Mac 6 cellsper HUVEC, which was34%less than U937cells(8.9 f 0.3). The lower numberof adherentMono Mac 6 cellsper HUVEC couldbe dueto their larger size,asassessed by flow cytometry. Blockingexperimentswith monoclonalantibody (mAb) directedagainstE-selectin, VCAM-1 andICAM- on HUVEC and CD1lb or CD14on Mono Mac 6 cellsdemonstratedthe contribution of these moleculesto Mono Mac 6 adherence.Reducedbinding after 24 h parallelsthe declineof E-selectinexpressionin HUVEC. Linearity of cell binding wasconfirmed from 0.2 x lo6 to 1.0 x lo6 Mono Mac 6 cells. Expressionof CD11b and CD14 in Mono Mac 6 cellsand in isolatedhumanmonocytesbut not in U937cellsleadingto interaction with ICAM- on HUVEC appearsto beresponsiblefor the increasedadhesionof Mono Mac 6, ascomparedto U937 cells.In conclusion,Mono Mac 6 cellsexhibit adhesioncharacteristicssimilarto humanmonocytesand thereforeseem to be a cell line well suitedfor studying monocyte-endothelialcell interactions. Keywords:Humanumbilical vein endothelialcells; Tumor necrosisfactor; Adhesionmolecules
1. Introdwtioo * Corresponding author.‘Tel.:49 89 5160 4350; Fax: 49 89 5160 4352. Abbreviations: BCECF/AM, 2’,7’-bis-(carboxyethyl)-5(6’)carboxyfluorescein pentacetoxy methylester; ECs,, halfmaximal efkctive concentration; EC, endothelial cell(s); FACS, fluorescence activatedcell sorter; HUV, human umbilical vein; mAb, monoclonal antibody(s); sMF1, specific mean fluorescence intensity; TEM, transmission electron microscope; TNF, tumor necrosis factor a
The molecular interactions determining monocyte adhesion to vascular endothelium are becoming increasingly evident [I]. However, in vitro studies using isolated human blood monocytes are accompanied by major problems, such as a high and varying basal adherence to endothelial cells. Several human tumor cell lines with monocyte-
0021-9150/95/.W9.50 0 19!)5 Elsevier Science Ireland Ltd. All rights reserved SSDI 002 I-!)1 50(94)05434-K
loo
W. Ed et al. /Atherosclerosis
related functions have been established, such as U937 [2], HL-60 [3] and THP-1 [4]. U937 cells are well defined with respect to adhesive properties [5] and commonly used in assessing monocytic cell adhesion to human umbilical vein endothelial cell (HUVEC) monolayers. The prolonged culture of this cell line over the past 20 years has caused selection of different clones following mutations under varying conditions. The occurrence of less mature U937 clones which lack certain monocytic characteristics has prompted an intensive search for other monocytic cell lines [6,7]. Recently, Mono Mac 6, a more mature monocytic cell line, has been described [8]. Untreated Mono Mac 6 cells exhibit many aspects closely related to human monocytes, such as expression of NaF-sensitive non-specific esterase, production of reactive oxygen, staining for My4, as well as many other monocyte-specific mAb, secretion of cytokines and phagocytosis in 80% of cells. This led us to investigate the as yet undefined adhesion properties of the Mono Mac 6 cell line in a model of cell-cell interaction using unstimulated and tumor necrosis factor OL(TNF-(r) stimulated HUVEC. The results show that Mono Mac 6 cells exhibit increased adhesion to stimulated and unstimulated HUVEC as compared to U937 cells, indicating properties of a more mature monocyte. The specific adhesion properties can be explained by the distinct expression of adhesion molecules on Mono Mac 6 cells. 2. Materials and methods 2.1. Cell culture
HUVEC were obtained from human umbilical cord veins by digestion with a-chymotrypsin, cultured in low-serum EGM-medium (Promo Cell, Heidelberg, FRG) using T-25 and T-150 flasks precoated with collagen in a 5% CO2 atmosphere at 37°C [9]. Cell purity was assessed by morphology (‘cobblestone’) and factor VIII staining. Confluent HUVEC passage 2 were detached by 0.05% trypsin/0.02% EDTA and grown in T-25 flasks or 24 well plates for treatment with human recombinant TNF (8.7 x lo6 U/ml in EGM, provided by Knoll-BASF, Ludwigshafen, FRG) at indicated concentrations and periods. For Bow cytometry, cells were harvested by careful treatment with
I13 (1995)
99-107
trypsin/EDTA, antagonized by immediate addition of foetal calf serum (FCS), and counted in triplicate. U937 cells were grown in RPM1 1640 medium with 2 mM L-glutamine and 10% FCS in suspension [lo]. Mono Mac 6 cells (kindly provided by Dr. H.W.L. Ziegler-Heitbrock) were cultured in RPM1 1640 with 10% FCS, 1 mM oxalacetate, 1 mM Na-pyruvate, 1 x nonessential amino acid solution, 2 x penicillin/streptomycin solution and 2 mM L-glutamine, using 24 well plates (Falcon, Becton Dickinson, Heidelberg, FRG) in a humidified 5% CO2 atmosphere. Before addition of FCS, the medium was ultrafiltrated through a Gambro 200 column (Hechingen, FRG) to eliminate lipopolysaccharide (LPS) [ll]. The LPS content of the medium was determined to be <2 pg/ml by Limulus amaebocyte assay. Human monocytes were freshly isolated from healthy donors by Ficoll density gradient centrifugation and subsequent magnetic cell separation (Miltenyi Biotec, Bergisch Gladbach, FRG), yielding a purity of >90%. Under all conditions, cell viability was >95% as judged by ethidium bromide/a&din orange fluorescence. 2.2. Immunofuorescence
Cells (2 x 105) were treated for 30 min with saturating amounts of mouse anti-VCAM-1, antiICAM- 84HlO (Dianova, Hamburg, FRG), antiCD62E (E-selectin) mAb (HUVEC), anti-CD1 1b (Mac-l), anti-CD18 (&), antXDw49d (VLA-4), anti-CD29 mAb (0,) or IgGi, isotype control (all Camon, Wiesbaden, FRG), FITC-anti-CD14 mAb (Mo2) or FITC-IgM isotype control (Coulter, Hamburg, FRG) (Mono Mac 6, U937) in phosphate buffered serum (PBS) with 0.5% bovine serum albumin (BSA) on ice. For staining, cells were reacted with goat anti-mouse FITCIgG,, (Camon) on ice after washing twice with PBS. To avoid unspecific binding, cells were preincubated with 5% human serum in PBS for 15 min on ice. Samples were washed twice with FACS buffer (Be&on Dickinson), fixed in 2% paraformaldehyde to prevent aggregation and analysed with 10 000 cells/sample in a FACScan, as described [9]. After correction for unspecific binding (isotype control), specific mean fluorescence intensity (sMF1) was expressed in channels.
W. Ed et al. /Atherosclerosis
2.3. Monocytic
ceil adhesion assay
HWEC were seeded in 24 well plates 48 h before the experiment. Only confluent monolayers were used, as confirmed by microscopic inspection. Fresh medium was provided before stimulation with TNF. Monocytic cells (3 x 107) were incubated in RPM1 1640 medium (6 ml) containing 2% FCS and 10 &ml of the fluorescent dye BCECF/A.M (Boehringer, Mannheim, FRG) at 37°C for 30 min [9]. Dye loading was stopped by adding 44 ml cold RPM1 1640 (2% FCS). Fluorescence labeled cells were pelleted and resuspended (0.1-3.0 >: 106/ml) in medium 199 (Gibco-BRL) with 10 mM HEPES buffer (M199H). HUVEC were washed twice with M199H before addition of loaded cells and incubated at 37°C 5% CO2 at 90% humidity. After 30 min, cell suspensions were withdrawn, HWEC were gently washed with M199H and inverted plates were centrifuged (50 x g, 55min, 4°C) [12]. Cells were lysed with 1 ml of 0.1% Triton X-100 in 0.1 M tris buffer/well (pH 8.0) and fluorescence was measured by a PTI deltascan spectrofluorimeter (excitation 485 nm, emission 535 nm). The number of adherent cells/well was calculated by comparing the amount of fluorescence to a standard curve of BCECF activity/cell (and expressed as % adhesion of added cells per well. Some HUVEC were pretreated with saturating amounts of anti-E-selectin (Camon), anti-.VCAM-1 1Gll or anti-ICAMmAb 84HlO (Dianova) 10 min before addition of U937 or Mono Mac 6 cells. Some monocytic cells were pretreated with anti-CD1 lb (Mac-l) or anti-CD14 mAb (UCHM-1) 10 min before addition to HUVEC. All other reagents were from Sigma, Deisenhofen, FRG.
II3
(1995)
101
99-107
TNF to obtain more adhesive HUVEC. As observed for U937 cells [5], binding of Mono Mac 6 cells to TNF (50 U/ml)-stimulated HUVEC rapidly increased within the first 20 min of coincubation at 37°C peaked at 30 min and remained almost unchanged for up to 2 h (Fig. 1). Binding experiments at 4°C revealed a more delayed onset of attachment (not shown). The time required for maximal binding at 4°C was not determined, because with longer incubation periods at this temperature, HWEC tended to detach from the inverted well plates during centrifugation. of monocytic cell binding To determine maximal binding and the optimal number of Mono Mac 6 cells to be employed in our assay, we incubated TNF (50 U/ml)-stimulated HWEC with increasing cell numbers (Fig. 2). The average number of HUVEC per well was 6.6 f 0.9 x lo4 (mean f SD, n = 8). The number of Mono Mac 6 cells bound at an excess of cells employed (3.0 x 106) was 4.1 f 0.4 x lo5 (n = 5), which is equivalent to 6.2 f 0.6 Mono Mac 6 cells per EC. The higher maximal binding of U937 cells (5.8 f 0.2 x lo’, n = 10; i.e. 8.9 f 0.3 U937 cells per HUVEC) suggests different spatial requirements for U937 and Mono
3.2. Saturability
2.4. Statistics
Data were analyzed by Student’s t-test.
,r 0
3. Results 3.1. Time course of binding of Mono Mac 6
To evaluate optimal and reproducible conditions for adhesion, we first characterized the time course of Mono Mac 6 cell binding to HWEC. As expected, Mono Mac 6 cell adherence in unstimulated EC was comparably low. We therefore used
50lJ/ml,6h
TNF-a, 0 10
20
30
time
40
50
60
3 120
(min)
Fig. 1. Time course of Mono Mac 6 cell binding. HUVEC in 24 well plates were stimulated with TNF (50 U/ml). Each well was treated 6 h before addition of 7.5 x 10’ fluorescence labelled Mono Mac 6 cells and incubated for designated periods. Adhesion was determined as in Materials and methods. Values represent the mean of 3 separate experiments run in triplicate. Percentage of adherent cells is expressed as mean f S.D.
W. Ed et al. /Atherosclerosis 113 (1995) 99-107
102
Mac 6 cells. This is most probably due to different cell size, as indicated by FACS forward scatter analysis. As a relative measure of cell size, the mean value in channels was 315 f 6 (n = 3) for Mono Mac 6 and 254 f 4 for U937 cells, revealing that Mono Mac 6 cells are approximately 25% bigger than U937 cells. This is in accordance with an average size of 16.0 and 12.5 pm found by TEM for Mono Mac 6 [8] and U937 cells [2], respectively. This difference may explain the 34% lower maximal binding of Mono Mac 6 as compared to U937 cells. We found a wide linear correlation between added and adherent Mono Mac 6 cells in a range of 0.2 x lo6 to 1.0 x lo6 (Fig. 2, insert). This “0
6 A
0.5 Mono
1.0 1.5 2.0 Mac 6 cells, free (xl@)
2.5
range was smaller in U937 cells (0.2 X lo6 to 0.7 x 106). In our experiments, not more than 0.8 x lo6 monocytic cells/ml were employed. 3.3. Dose and time dependenceof TNF-stimulated adhesion
TNF induced a dose- and time-dependent enhancement of Mono Mac 6 and U937 cell adhesion. As demonstrated in Fig. 3, basal adhesion of Mono Mac 6 cells was 9.4% f 0.4% (all values in % mean i SD, n = 3), stimulation of HUVEC with TNF (1 U/ml) for 6 h showed no effect (10.3% f l.O%), a rapid increase in adhesion above 1 U/ml, and a plateau at 10 U/ml (55.1% f 3.6%). TNF did not further increase adhesion at 50 or 100 U/ml (56.5% * 3.3% and 54.7% * 2.8%, respectively). In U937 cells, TNF increased adhesion at 1 U/ml compared to basal
3.0
r 60
.o=
t
40-
3 e;
1
_ 4
20-
k 0
0.5
1.0
1.5
2.0
2.5
24 h
3.0
U937 cells, free (x1$)
Fig. 2. Saturability and linearity of monocytic cell binding. HUVEC were pretreated with 50 U/ml TNF for 6 h. Increasing numbers (0.1 x IO6 to 3.0 x 106) of fluorescence labelled monocytic cells (A: Mono Mac 6, B: U937) were added for 30 min co-incubation at 37OC and adhesion was measured as in Materials and methods. Linear binding was determined by a correlation coefficient (r2) > 0.995, as shown in insert of Figs. 2A and B. Values represent the mean of 3 separate experiments run in triplicate. Data are expressed as mean + SD.
0
5 10
25
50
TNF U/ml
Fig. 3. Time and dose. dependence of Mono Mac 6 (-6) and U937 (-0-) cell binding. HUVEC were incubated with increasing concentrations of TNF for 6 and 24 h. Fluorescence labelled monocytic cells were added and % adhesion determined as in Materials and methods. Values represent 4 independent experiments performed in triplicate and are expressed as mean f SD.
W. Ed et al. /Atherosclerosis
values (12..0% f 4.0% vs. 2.6% f 0.8%). Adherence was further enhanced at 10 U/ml and maximal at 50 U/ml (40.0% * 8.4%) or at 100 U/ml (38.0% f 7.7%). As shown in Fig. 3, treatment of HWEC with TNF for 24 h resulted in similar binding characteristics, but lower adhesion for both cell lines as compared to 6 h. Again, TNF caused maximal attachment at 50 U/ml, and did not further increase adhesion at 100 U/ml (data not shown). At 6 h, the EC,, for TNF was approximately 2 U/ml for both cell lines. Stimulation with TNF for 24 h exhibited an ECSo of < 3 U/ml for Mono Mac 6 and >6 U/ml for U937 cells, In comparison, freshly isolated human monocytes showed (an adhesion of 18.8% f 7.5% to unstimulated and of 55.7% f 9.3% to TNF (50 U/ml, 6 h&stimulated HWEC. 3.4. Flow cytometric analysis and blocking of adhesion moleculeson stimulated HUVEC
To elucidate these differences in adhesion, we investigated the time dependent induction of adhesion molecules. Endothelial leukocyte adhesion E-selectin
I13 (1995)
103
99-107
molecule 1 (ELAM- 1, E-selectin) and vascular cell adhesion molecule 1 (VCAM-1) are absent in unstimulated HWEC, but contribute signiticantly to monocyte adhesion on EC [ 131. We used TNF (50 U/ml) to induce these molecules and analyzed surface expression at 6 and 24 h by flow cytometry (Fig. 4). E-selectin peaked at 6 h and declined to basal values at 24 h. VCAM-1 was induced after 6 h and remained increased at 24 h. Intercellular adhesion molecule 1 (ICAM-1) was constitutively expressed and upregulated by TNF at 6 and 24 h. These distinct kinetics of adhesion molecule expression on HWEC can explain the decreased cell adherence seen after 24 h. The contribution of E-selectin, VCAM-1 and ICAMto Mono Mac 6 cell adhesion was further investigated by incubation of TNF (50 U/ml)stimulated HWEC with blocking mAb. As shown in Fig. 5, employment of mAb to a single molecule showed only slight inhibition of adhesion. Combination of 2 mAb exhibited a significant decrease of adherent Mono Mac 6 cells, which was not further pronounced using a set of 3 mAb. VCAM-1
ICAM-
Fig. 4. Surface molecule expression of HUVEC stimulated by TNF. HUVEC were stimulated for 0, 6, and 24 h with 50 U/ml TNF, stained for E-selectin (left), VCAM-1 (middle), ICAM-I (right) or isotype control (reflecting unspecific binding) and analyzed by flow cytometry. Shown are original histograms representative of 3 similar experiments. Relative cell numbers are vertical and fluorescence intensity on a log 10 scale is horizontal.
104
W. Ed et al. /Atherosclerosis 113 (1995) 99-107
60
-I
TNF50Wml anti-E-se&tin mAb anti-VCAM-1 mAb anti-ICAM- mAb
T
-
+++ -
+ -
+ -
+++++ + + + -
+ +
+ +
+ + +
Fig. 5. Effect of anti-E-se&tin, anti-VCAM-1 and anti-ICAMmAb on Mono Mac 6 adhesion to TNF-stimulated EC. HUVEC were incubated with 50 U/ml TNF for 6 h. Medium was changed and mAb were reacted for 10 min at room temperature. Labelled Mono Mac 6 cells (7.5 x 10’ per well) were added and co-incubated for 30 min at 37°C. Nonadherent cells were removed and % adhesion determined as described. Values are means of 3 separate experiments run in duplicate and are expressed as mean f S.D.(*P c 0.05, l *P c 0.01 vs. 50 U/ml TNF).
3.5. Flow cytometric analysis of adhesionmolecules on Mono Mac 6 cells
A monocytic counterreceptor for E-selectin is sialyl-Lewis X, for VCAM-1 very late activation antigen 4 (VLA-4) and for ICAMthe CD1 1b/ CD18 integrin. Sialyl-Lewis X is expressed on untreated U937 cells [14], as is the BZintergrin Table 1 Adhesion molecules on Mono Mac 6 cells and U937 cells
CD1 lb (Mac-l) CD18
632)
CDw49d (VLA-4) CD29 (8,) CD14 (Mo2)
Mono Mac 6
u937
139 f 11 105 f 18 82 l 12 175 l 15 97 + 4
5zt6 114 * II 174 f 13 246zt 16 4*3
Mono Mac 6 and U937 cells were cultured for 72 h, stained with mAb directed against indicated molecules or respective isotype control and analyzed by flow cytometry. Data are expressed as sMF1 corrected for unspecific binding in channels and represent mean f S.D. of four separate experiments.
CD18, whereas the corresponding monomer CD1 lb is virtually absent on U937 cells. In contrast, CD1 1b is markedly expressed on Mono Mac 6 cells (Table 1) and on freshly isolated monocytes (sMFI 259 f 41 channels). Since CD1 lb is found in U937 cells only after differentiation [ 151, this indicates a more mature phenotype of Mono Mac 6 cells. Consistent with results showing downregulation of VLA4 during monocyte differentiation [ 161, a less marked expression of VLA4 is found in Mono Mac 6 as compared to U937 cells. CD14 on mature monocytes has recently been shown to contribute to monocyte adhesion to cytokine stimulated HUVEC [ 171. We therefore measured CD14 expression in both cell lines. CD14 was not detectable in U937 cells, where it can be induced by differentiation [ 151, but constitutively expressed on Mono Mac 6 cells and mature monocytes [8]. Consistently, preincubation of Mono Mac 6 cells and human monocytes but not U937 cells with anti-CD14 or anti CD1 lb mAb significantly reduced adhesion (Fig. 6). In contrast, Polymyxin B (100 U/ml) had no effect, excluding cell activa-
W. Ed
et al. /Atherosclerosis
Mono
antXD14mAb anti-CD11 b mAb
-
Mac 6
+ -
-’ +
113 (1995)
99-107
105
blc+od monocytes
-
+ -
+
_
+ -
+
Fig. 6. Effect of anti-CD14 and anti-CD1 lb mAb on monocytic cell adhesion to TNF-stimulated EC. HUVEC were incubated with 50 U/ml TNF for 6 h. Labelled monocytic cells were preincubated with mAb for 10 min prior to addition and co-incubation with HUVEC for 30 min at 37OC. Nonadherent cells were removed and % adhesion was determined as described. Values are means of 3 separate experiments run in duplicate and are expressed as mean f S.D.(*P < 0.05 vs. 50 U/ml TNF).
tion by LPS. Taken together, the data suggest that differences in adhesion properties of U937 and Mono Mac 6 cells to HWEC are primarily due to a different contribution of CD1 1b/ICAM- 1 dependent interaction. 4. Discmion As a consequence of the immature nature of U937 cells which includes several subclones, a cell line with characteristics of mature monocytes would be useful to further investigate the interactions of white blood cells with vascular endothelial cells. In our study, we describe the adhesion properties of the novel human monocytic cell line Mono Mac 6, in comparison to the widely used U937 cell line. The time course of Mono Mac 6 cell binding to TNF-stim.ulated HWEC was comparable to that of U937 cells, suggesting that similar mechanisms of attachment are involved. This can also be derived from cross-competition experiments in which both cell types were able to block adhesion of each other when equal cell numbers were simultaneously incubated with HWEC (unpublished results). All experiments were performed at 30 min, when maximal adhesion occurred for both Mono Mac 6 and U937 cells. To investigate firm adhesion, we
performed our experiments at 37”C, since at 4°C HWEC showed a tendency to detach from wells during inverse centrifugation. After 30 min at 37”C, spreading of both cell types could rarely be detected by microscopic inspection. Mono Mac 6 cells more readily adhere to unstimulated H WEC than U937 cells. The higher basal adhesion may be due to enhanced interaction with constitutively expressed HWEC adhesion molecules, such as ICAM- 1. A high basal adhesion reported for U937 cells [18] may be attributed to the use of proliferating EC, which are more adhesive than confluent cells [19], or culture medium with high serum content, which may promote expression of binding sites and EC proliferation beyond contact inhibition (unpublished results). Hence, Mono Mac 6 adhesion properties may be reflective of a more mature monocytic cell, as similar adhesion values for isolated blood monocytes were demonstrated in our experiments and have recently been published [20]. The high and varying basal adhesion of human blood monocytes [21] may result from activation during preparation procedures, which was minimized by the use of magnetic cell separation in our study. However, Mono Mac 6 cells show markedly less variance in binding as compared to human monocytes [20,21 J, possibly due to a more uniform population during
106
W. Ed et al. /Atherosclerosis
cell culture. Mono Mac 6 cells also exhibit a higher maximal adhesion to TNF-stimulated HUVEC as compared to U937 cells. The assessment of the molecular moieties on both cell lines provides an explanation for these observations, since the ICAMcounterreceptor CD1 lb is markedly expressed on Mono Mac 6 cells and monocytes, but not detectable on U937 cells. The CD1 l-associated /32 integrin CD18 is also found on U937 cells, indicating the presence of at least one CD1 1 chain. Since only an activated subpopulation of CD1 lb has been described to mediate adherence to ICAM[22], the increased adhesion of Mono Mac 6 cells may be due to the occurrence of more CD1 1 molecules in an adhesive conformation. The marked expression of CD14 in Mono Mac 6 cells and inhibition of adhesion by anti-CD14 mAb further support a contribution of CD14 to adhesion of more mature monocytic cells to cytokinestimulated HUVEC [17]. In this context, it has been suggested that binding of CD14 to surface structures of stimulated EC can generate crosslinking of CD14, resulting in activation of CDllI CD18 and ICAMdependent monocyte adherence [23]. Inhibition experiments with blocking mAb demonstrated that several ligand pairs are involved in maximal adherence, since blocking of one EC adhesion molecule only slightly affected adhesion, while combinations of 2 Ab proved to be very effective. Consistently, marked inhibition of monocyte adhesion required combination of anti-Eselectin and anti-VCAM-1 mAb, indicating the dependence on both molecules [ 121. An interpretation of this finding would be that other systems can compensate, as long as only one type of binding site is restricted. This is in accordance with observations that VLA4 and CD1 l/CD18 act interdependently in monocyte transmigration [24]. The incomplete reduction of adhesion may be due to the blocking properties, shedding or incorporation of mAb, or the result of yet unidentified adhesion molecules on stimulated HUVEC. Furthermore, preincubation of monocytic cells with either anti-CD1 1b or anti-CD14 mAb attenuated adhesion of Mono Mac 6, but not of U937 cells. In conclusion, we have described the adhesion characteristics of the mature human monocytic
113 (1995)
99-107
cell line Mono Mac 6 to HUVEC monolayers, which may be well suited for studying biochemical and functional interactions of monocytes and EC. Acknowledgements This study was partially supported by grants from ‘Bundesministerium fur Forschung und Technologie’ BMFT 07ERGO3 and by ‘AugustLenz-Stiftung’. This work partially fulfills requirements for the doctoral theses of W. Erl and C. Wardemann. We thank E. Bretzke for secretarial assistance. References 111 Beekhuizen, H. and van Furth, R., Monocyte adherence
to human vascular endothelium, J. Leukoc. Biol., 54 (1993) 363. PI Sundstrom, C. and Nilsson, K., Establishment and characterization of a human histiocytic lymphoma cell line (U-937), Int. J. Cancer, 17 (1976) 565. [31 Collins, S.J., Ciallo, R.C. and Gallagher, R.E., Continuous growth and differentiation of human myeloid leukaemic cells in suspension culture, Nature., 270 (1977) 347. 141 Tsuchiya, S., Yamabe, M., Yamaguchi, Y., Kobayashi, Y., Konno, T. and Tada, K., Establishment and characterization of a human acute leukemia cell line (THP-I), Int. J. Cancer, 26 (1980) 171. [51 DiCorletto, P.E. and de la Motte, C.A., Characterization of the adhesion of the human monocytic cell line U937 to cultured endothelial cells, J. Clin. Invest., 75 (1985) 1153. 161 Ohno, H., Fukuhara, S., Doi, S. et al., Establishment of a novel acute monoblastic leukemia cell line (YK-M2) having a near triploid karyotype, Cancer Res., 46 (1986) 6400. [71 Ohta, M., Furukawa, Y., Ide, C., Akiyama, N., Utakoji, T., Miura, Y. and Saito, M., Establishment and characterization of four human monocytoid cell lines (JOSK-I, -S, -M, and -K) with capabilities of monocytemacrophage lineage differentiation and constitutive production of interleukin 1, Cancer Res., 46 (1986) 3067. Is1 Ziegler-Heitbrock, H.W.L., Thiel, E., Fiitterer, A., Herzog, V., Wirtz, A. and Riethmtiller, G., Establishment of a human cell line (Mono Mac 6) with characteristics of mature monocytes, Int. J. Cancer, 41 (1988) 456. 191 Weber, C., Erl, W., Pietsch, A., Striibel, M., ZieglerHeitbrock, H.W.L. and Weber, P.C., Antioxidants inhibit monocyte adhesion by suppressing nuclear factor-KB mobilization and induction of vascular cell adhesion molecule 1 in endothelial cells stimulated to generate radicals, Arteriosclerosis Thromb., 14 (1994) 1665. 1101Weber, C., Aepfelbacher, M., Lux, I., Zimmer B. and
W. Erl et al. /Atherosclerosis
[I I]
[12]
[13]
[14]
[15] [16]
Weber, P.C., Docosahexaenoic acid inhibits PAF and LTD4-stimulated [Ca++]i-increase in differentiated monocytic U937 cells, Biochim. Biophys. Acta, 1133 (1991) 38. Weber, C., Aepfelbacher, M., Haag, H., Ziegler-Heitbrock, H.W.L. and Weber, PC., Tumor necrosis factor induces enhanced responses to platelet-activating factor and differentiation in human monocytic Mono Mac 6 cells, Eur. J. Immunol., 23 (1993) 852. Charo, I.F., Yuen, C. and Goldstein, I.M., Adherence of human polymorphonuclear leukocytes to endothelial monolayers: effect of temperature, divalent cations, and chemotactic factors on the strength of adherence measured with a new centrifugation assay, Blood, 65 (1985) 473. Carlos, T., Kovach, N., Schwartz, B. et al., Human monocytes bind to two cytokine-induced adhesive ligands on cultured human endothelial cells: endothelialleukocyte adhesion molecule-l and vascular cell adhesion molecule-l, Blood, 77 (1991) 2266. Fukushima, K., Hirota, M., Terasaki, P.I., Wakisaka, A., Togashi, H. and Chia, D., Characterization of sialosylated Lewis x as a new tumor-associated antigen, Cancer Res., 44 (1984) 5279. &erg, F., Botling, J. and Niisson, K., Macrophages and the cyt.okine network, Transplant. Proc., 25 (1993) 2044. Bauvois, B., Rouillard, D., Sanceau, J. and Wietzerbin, J., IFN--r and transforming growth factor-b differentially regulate tibronectin and laminin receptors of human differentiating monocytic cells, J. Immunol., 148 (1992) 3912.
I13 (1995)
99- 107
107
[17] Beekhuizen, H., Blokland, I., Corsel-van Tilburg, A., Koning, F. and van Furth, R., CD14 contributes to the adherence of human monocytes to cytokine-stimulated endothelial cells, J. Immunol., 147 (1991) 3761. [18] DiCorletto, P.E. and de la Motte, CA., Thrombin causes increased monocytic-cell adhesion to endothelial cells through a protein kinase C-dependent pathway, B&hem. J., 264 (1989) 71. [19] &Bono, D.P. and Green, C., The adhesion of different cell types to cultured vascular endothelium: effects of culture density and age, Br. J. Exp. Pathol., 65 (1984) 145. [20] Couffinhal, T., Duplaa, C., Labat, L., Moreau, C., Bietz, I. and Bonnet, J., Effect of low density lipoprotein on monocyte adhesiveness to endothelial cells in vitro, Atherosclerosis, 99 (1993) 35. [21] McGregor, P.E., Agrawal, D.K. and Edwards, J.D., Attenuation of human leukocyte adherence to endothelial cell monolayers by tyrosine kinase inhibitors, B&hem. Biophys. Res. Commun., 198 (1994) 359. [22] Diamond, MS. and Springer, T.A., A subpopulation of Mac-l (CD1 I b/CDlS) molecules mediates neutrophil adhesion to ICAM- and fibrinogen, J. Cell Biol., 120 (1993) 545. (231 Beekhuizen, H., Blokland, I. and van Furth, R., Crosslinking of CD14 molecules on monocytes results in a CD1 l/CD18- and ICAM-l-dependent adherence to cytokine-stimulated human endothelial cells, J. Immunol., 150 (1993) 950. [24] Winn, R.K. and Harlan, J.M., CD18-dependent neutrophil and mononuclear leukocyte emigration into the peritoneum of rabbits. J. Clin. Invest., 92 (1993) 1168.