Expression and functional significance of adhesion molecules on cultured endothelial cells in response to ionizing radiation

OBJECTIVE

Upregulation of adhesion molecules on endothelial cells following irradiation has been shown, but the functional significance of this upregulation in various endothelial cell lines is not clear. We have developed an in vitro flow model to study the functional consequences of the radiation-induced upregulation of E-selectin and intercellular adhesion molecule (ICAM-1).

METHODS

Human dermal microvascular endothelial cells (HDMEC), human umbilical vein endothelial cells (HUVEC), or transformed human microvascular endothelial cells (HMEC-1) were grown in 35-mm dishes and irradiated with a single dose of 10 Gy. HL-60 (human promyelocytic leukemia) cells were perfused over the irradiated endothelial cells in a parallel plate flow chamber at shear stress ranging from 0.5 to 2.0 dynes/cm2. Flow cytometry was used to quantify the expression of E-selectin and ICAM-1 on the various endothelial cells.

RESULTS

Flow cytomeric analysis revealed an upregulation of ICAM-1 expression on all three cell types postirradiation (post-IR), and an upregulation of E-selectin expression only on HDMEC post-IR. E-selectin expression was detected on control HDMEC, but at a lower level than that detected on post-IR HDMEC. Flow assays revealed a significant increase in the number of rolling and firmly adherent HL-60 cells on post-IR HDMEC at shear stress < or =1.5 dynes/cm2; pretreatment of control and irradiated HDMEC with antibodies to E-selectin and ICAM-1 significantly diminished the number of rolling and firmly adherent HL-60 cells, respectively. No rolling or firm adhesion of HL-60 cells was observed on HUVEC or HMEC-1 monolayers post-IR.

CONCLUSION

These findings suggest that ICAM-1 is upregulated on irradiated HDMEC, HUVEC, and HMEC-1. E-selectin is upregulated to a functional level only on irradiated HDMEC, and not on irradiated HUVEC or HMEC-1.

Limited adhesion of biodegradable microspheres to E- and P-selectin under flow

In a variety of disease settings the expression of the endothelial selectins E- and P-selectin appears to be increased. This feature makes these molecules attractive targets around which to design directed drug-delivery schemes. One possible approach for achieving such delivery is to use polymeric biodegradable microspheres bearing a humanized monoclonal antibody (MAb) for E- and P-selectin, MAb HuEP5C7.g2. Perhaps the simplest technique for "coupling" HuEP5C7.g2 to the microspheres is via nonspecific adsorption. Previous studies suggest, however, that the adsorption of proteins onto microspheres fabricated in the presence of a stabilizer such as poly(vinyl alcohol) (PVA) is limited. It is unclear to what extent this limited level of adsorbed HuEP5C7.g2 would be able to support adhesion to E- and P-selectin under flow conditions. To explore this issue, we prepared microspheres from the biodegradable polymer, poly(epsilon-caprolactone) (PCL), using a single emulsion process and PVA as a stabilizer. We then incubated the PCL microspheres with HuEP5C7.g2 and studied the adhesion of the resulting HuEP5C7.g2 microspheres to E- and P-selectin under in vitro flow conditions. We found that the HuEP5C7.g2 PCL microspheres exhibit specific adhesion to Chinese hamster ovary cells stably expressing P-selectin (CHO-P) and 4-h IL-1beta-activated human umbilical vein endothelial cells (HUVEC). In contrast, HuEP5C7.g2 PCL microspheres exhibit little adhesion to parental CHO cells or unactivated HUVEC. The attachment efficiency to the selectin substrates was quite low, with appreciable attachment occurring only at low shear (0.3 dyn/cm(2)). Other supporting data strongly suggest that the limited attachment efficiency is due to a low level of HuEP5C7.g2 adsorbed to the PCL microspheres. Although the attachment was limited, a significant percentage of the HuEP5C7.g2 PCL microspheres were able to remain adherent at relatively high shear (8 dyn/cm(2)). Combined, our data suggest that HuEP5C7.g2 PCL microspheres exhibit selective limited adhesion to cellular substrate expressing E- and P-selectin.

Ligand coated nanosphere adhesion to E- and P-selectin under static and flow conditions

The heterogeneous distribution of endothelial cell adhesion molecules (ECAMs) on the lumenal surface of vascular endothelium provides an opportunity to deliver drugs to select tissues. The targeting could be achieved by using carriers whose outer surface has a ligand for a selectively expressed ECAM. The carriers would interact with the endothelium in a fluid dynamic environment and in many of these schemes nanoparticles would be used. It is unclear what role various parameters (e.g., ligand-ECAM chemistry, fluid shear) will have on the adhesion of the nanoparticles to the endothelium. To facilitate studies in this area, we have developed a prototypical in vitro model that allows investigation of nanoparticle adhesion. We coated polystyrene nanospheres with a humanized mAb (HuEP5C7.g2) that recognizes the ECAMs E- and P-selectin. Adhesion assays revealed that HuEP5C7.g2 nanospheres exhibit augmented, specific adhesion to selectin presenting cellular monolayers and that the adhesion can be affected by the fluid shear. These results; (i) strongly suggest that HuEP5C7.g2 could be used to target nanoparticles to selectin presenting endothelium; (ii) demonstrate that fluid shear can affect nanoparticle adhesion; and (iii) define a system which can be used to study the effects of various system parameters on nanoparticle adhesion.

Particle diameter influences adhesion under flow

The diameter of circulating cells that may adhere to the vascular endothelium spans an order of magnitude from approximately 2 microm (e.g., platelets) to approximately 20 microm (e.g., a metastatic cell). Although mathematical models indicate that the adhesion exhibited by a cell will be a function of cell diameter, there have been few experimental investigations into the role of cell diameter in adhesion. Thus, in this study, we coated 5-, 10-, 15-, and 20-microm-diameter microspheres with the recombinant P-selectin glycoprotein ligand-1 construct 19.ek.Fc. We compared the adhesion of the 19.ek.Fc microspheres to P-selectin under in vitro flow conditions. We found that 1) at relatively high shear, the rate of attachment of the 19.ek.Fc microspheres decreased with increasing microsphere diameter whereas, at a lower shear, the rate of attachment was not affected by the microsphere diameter; 2) the shear stress required to set in motion a firmly adherent 19.ek.Fc microsphere decreased with increasing microsphere diameter; and 3) the rolling velocity of the 19.ek.Fc microspheres increased with increasing microsphere diameter. These results suggest that attachment, rolling, and firm adhesion are functions of particle diameter and provide experimental proof for theoretical models that indicate a role for cell diameter in adhesion.

Alpha4beta1-integrin activation is necessary for high-efficiency T-cell subset interactions with VCAM-1 under flow

OBJECTIVE

The purpose of this study was to examine the relationship between alpha4beta1-integrin state of activation on CD4+ T-cell subsets and their adhesive interaction to VCAM-1 under flow.

METHODS

Human CD4+ memory and naive T-cells were freshly isolated and effector-helper T-cell subsets. Th1 and Th2 cells, were differentiated in vitro from CD4+ naive T-cells. The expression of activation/ligand induced epitopes on beta1-integrins of each T-cell subset was assessed using mAb HUTS21 and mAb 15/7. T-cell subsets attachment and rolling on VCAM-1 was determined under defined flow conditions and the rates of attachment (ka), accumulation, and instantaneous rolling velocities were correlated to their beta1-integrin activation epitope expression.

RESULTS

A subset of memory T-cells constitutively express activation/ligand induced epitopes on beta1-integrins recognized by mAb HUTS21 and 15/7, whereas expression levels on naive T-cells is low or not detectable. Consistent with an activated phenotype, memory T-cells exhibit significantly higher rates of attachment and accumulation on VCAM-1 under flow as compared to naive T-cells. Interestingly, the expression of activation/ligand induced epitopes on beta1-integrins on Th2 cells and the ability of these cells to interact with VCAM-1 are comparable to memory T-cells. In contrast, Th1 cells did not interact as efficiently with VCAM-1, which correlated with lower expression of activation/ligand induced epitopes on these cells. VCAM-1 interactions are inhibited completely by pretreatment of the T-cells with blocking mAb to alpha4-integrins or beta1-integrins, indicating that alpha4beta1 is the predominant T-cell integrin involved.

CONCLUSIONS

Memory T-cells express constitutively active alpha4beta1-integrins, as compared to naive T-cells, which mediate high rates of initial attachment and sustained high-affinity adhesive interactions with VCAM-1 under flow conditions in vitro. Similarly, in vitro differentiated Th2 cells but not Th1 cells, which also express elevated levels of activated alpha4beta1-integrins, are capable of sustaining high-affinity adhesive interactions with VCAM-1. The differences observed in beta1-integrin activation on T-cell subsets may underlie selective recruitment patterns of T-cell subsets in vivo.

CD11b/CD18-coated microspheres attach to E-selectin under flow

Neutrophils can attach to E-selectin under flow. Proposed ligands for E-selectin carry SLe(x)-type glycans. The leukocyte beta2 integrins are glycosylated with SLe(x). Thus, we speculated that beta2 integrins could support attachment to E-selectin. To test this hypothesis, we coated 10-microm-diameter microspheres with purified CD11b/CD18 (alphaMbeta2) and investigated the adhesion of the resulting alphaMbeta2 microspheres to E-selectin. Under in vitro flow conditions, the alphaMbeta2 microspheres attached to Chinese hamster ovary cells expressing E-selectin (CHO-E) and 4-h interleukin-1beta-activated human umbilical vein endothelial cells (HUVEC). At a shear stress of 1.8 dynes/cm2, the attachment events were eliminated by pretreatment of the cellular monolayers with a mAb to E-selectin. alphaMbeta2 microspheres did not attach to untransfected CHO cells or unactivated HUVEC at 1.8 dynes/cm2. Taken together, the results strongly suggest that the CD11b/CD18-E-selectin bond has sufficient biophysical properties to mediate attachment of neutrophil-sized particles to E-selectin under flow.

Isolated P-selectin glycoprotein ligand-1 dynamic adhesion to P- and E-selectin

Leukocyte adhesion to vascular endothelium under flow involves an adhesion cascade consisting of multiple receptor pairs that may function in an overlapping fashion. P-selectin glycoprotein ligand-1 (PSGL-1) and L-selectin have been implicated in neutrophil adhesion to P- and E-selectin under flow conditions. To study, in isolation, the interaction of PSGL-1 with P- and E-selectin under flow, we developed an in vitro model in which various recombinant regions of extracellular PSGL-1 were coupled to 10-microm-diameter microspheres. In a parallel plate chamber with well defined flow conditions, live time video microscopy analyses revealed that microspheres coated with PSGL-1 attached and rolled on 4-h tumor necrosis factor-alpha-activated endothelial cell monolayers, which express high levels of E-selectin, and CHO monolayers stably expressing E- or P-selectin. Further studies using CHO-E and -P monolayers demonstrate that the first 19 amino acids of PSGL-1 are sufficient for attachment and rolling on both E- and P-selectin and suggest that a sialyl Lewis x-containing glycan at Threonine-16 is critical for this sequence of amino acids to mediate attachment to E- and P-selectin. The data also demonstrate that a sulfated, anionic polypeptide segment within the amino terminus of PSGL-1 is necessary for PSGL-1-mediated attachment to P- but not to E-selectin. In addition, the results suggest that PSGL-1 has more than one binding site for E-selectin: one site located within the first 19 amino acids of PSGL-1 and one or more sites located between amino acids 19 through 148.

Sialyl Lewis(x)/E-selectin-mediated rolling in a cell-free system

Selections mediate transient adhesion of neutrophils to stimulated endothelial cells at sites of inflammation by binding counter-receptors that present carbohydrates such as sialyl Lewis(x). We have developed a cell-free adhesion assay using sialyl Lewis(x)-coated microspheres and E-selection-IgG chimera-coated substrates to investigate the premise that rolling primarily results from functional properties of selection-carbohydrate bonds, whereas cellular morphology and signaling act as secondary effects. Sialyl Lewis(x)-coated microspheres attach to and roll over E-selectin-IgG chimera-coated substrates between the physiological wall shear stresses of 0.7 and 2 dynes/cm2. Rolling velocities vary with time and depend on E-selectin-IgG chimera site density and wall shear stress. Our results show that sialyl Lewis(x) is a minimal functional recognition element required for rolling on E-selectin under flow.

A human colon carcinoma cell line exhibits adhesive interactions with P-selectin under fluid flow via a PSGL-1-independent mechanism

It has been postulated that endothelial cell adhesion molecules involved in leukocyte recruitment play a role in metastasis. Using an in vitro flow model, we studied the adhesion of the human colon carcinoma cell line KM12-L4 to P-selectin, an inducible endothelial-expressed adhesion molecule involved in leukocyte recruitment. Recombinant forms of P-selectin and Chinese hamster ovary cells stably expressing P-selectin supported attachment and rolling of KM12-L4 cells at 1 to 2 dynes/cm2. The adhesive interactions to P-selectin were abolished by pretreatment of the KM12-L4 cells with neuraminidase but were unaltered by pretreatment of the KM12-L4 cells with O-sialoglycoprotein endopeptidase, an enzyme that cleaves mucin type glycoproteins such as P-selectin glycoprotein ligand-1 (PSGL-1). PSGL-1 is the only counter-receptor for P-selectin known to mediate myeloid cell adhesion to P-selectin under flow. Flow cytometric and Northern blot analyses revealed that KM12-L4 cells did not express PSGL-1 and monoclonal antibody PL1, a function-blocking monoclonal antibody to PSGL-1, had no inhibitory effect on KM12-L4 adhesion to P-selectin under flow. Compared with HL-60 cells, which express PSGL-1, the KM12-L4 cells exhibited a slightly lower rate of attachment to P-selectin and rolled at a significantly higher velocity. In summary, KM12-L4 human colon carcinoma cells interact with P-selectin, under flow, through a PSGL-1-independent adhesion pathway.

Lu-ECAM-1-mediated adhesion of melanoma cells to endothelium under conditions of flow

Lu-ECAM-1 is a lung-derived, venular endothelial cell adhesion molecule. It promotes the selective adhesion of lung-metastatic B16-F10 melanoma cells to endothelium under static conditions and mediates colonization of the lungs by the same tumor cells. To test whether Lu-ECAM-1 by itself is sufficient to cause vascular arrest of B16-F10 cells, we measured here under conditions of flow tumor cell adhesion to endothelia that express different amounts of Lu-ECAM-1 on their surfaces. At physiological shear stresses, adhesion of B16-F10 melanoma cells to endothelia correlates positively with the amount of Lu-ECAM-1 expression on the endothelial cell surface and inversely with the level of the applied shear stress. Tumor cell trajectories are biphasic; i.e., B16-F10 melanoma cells initially move along the endothelial surface with a velocity similar to the theoretical velocity, then arrest within a fraction of a second. Arrest is permanent for most B16-F10 melanoma cells at all shear stresses tested. Tumor cells never engaged in a rolling motion prior to arrest. Masking of the Lu-ECAM-1 ligand on the surface of B16-F10 melanoma cells with soluble Lu-ECAM-1 impedes arrest of tumor cells on the surface of the test endothelium. Purified Lu-ECAM-1 also mediates B16-F10 arrest, but arrest is mostly transient at shear stresses of 0.59 dynes/cm2 and higher, implying adhesion by single receptor/ligand bonds. Our data suggest that Lu-ECAM-1 plays a critical role in the recognition and initial arrest of murine melanoma cells in lung venules.

An E-selectin-IgG chimera supports sialylated moiety dependent adhesion of colon carcinoma cells under fluid flow

To further characterize the molecular mechanisms that govern carcinoma cell adhesion to stimulated endothelium, we studied the adhesion of a human colon carcinoma cell line, KM12-L4, to an E-selectin-IgG1 chimera and interleukin (IL)-1 beta-stimulated human umbilical vein endothelial cells (HUVEC) under in vitro fluid flow conditions. Between 0.6 and 1.8 dynes/cm2, KM12-L4 cells attach to and roll on IL-1 beta-stimulated HUVEC. The adhesion is E-selectin dependent and diminished upon pretreatment of the KM12-L4 cells with neuraminidase (neuraminidase sensitive). Between 0.7 and 1.8 dynes/cm2, surfaces coated with an E-selectin-IgG1 chimera support attachment and rolling of KM12-L4 cells. The adhesion to the E-selectin-IgG1 chimera is blocked by an antibody to the lectin domain of E-selectin and is neuraminidase sensitive. Rolling KM12-L4 cells exhibit variable velocity motion over both IL-1 beta-stimulated HUVEC and E-selectin-IgG1 chimera-coated surfaces. Our results provide the first direct evidence that sialylated moieties are involved in the adhesion of carcinoma cells to IL-1 beta-stimulated endothelium under flow conditions; E-selectin-IgG1 chimeras can support cell attachment and rolling under defined flow conditions; the topology of the endothelium is not the sole cause of variable velocity motion observed in cell rolling systems.

Dynamics of neutrophil rolling over stimulated endothelium in vitro

Prior to extravasation at sites of acute inflammation, neutrophils roll over activated endothelium. Neutrophil rolling is often characterized by the average rolling velocity. An additional dynamic feature of rolling that has been identified but not extensively studied is the fluctuation in the rolling velocity about the average. To analyze this characteristic further, we have measured the instantaneous velocity of bovine neutrophils interacting with lipopolysaccharide-stimulated bovine aortic endothelium at shear stresses of 1, 2, 3, and 4 dynes/cm2. The average velocities are quantitatively similar to those reported for human neutrophils rolling over reconstituted P-selectin at a surface density of 400 sites/microns 2. At all shear stresses tested, the population average variance in the instantaneous velocity is at least 2 orders of magnitude higher than the theoretical variance generated from experimental error, indicating that the neutrophils translate with a nonconstant velocity. Possible sources of the variance are discussed. These include "macroscopic" sources such as topological heterogeneity in the endothelium and microscopic sources, such as inherent stochastic formation and breakage of the receptor-ligand bonds that mediate the rolling. Regardless of the ultimate source of the variance, these results justify the use of mathematical models that incorporate stochastic processes to describe bond formation and breakage between the neutrophil and the endothelium and hence are able to generate variable velocity trajectories.