Polymorphonuclear phenotypical expression of CD18,at baseline and after in vitro activation,in several clinical disorders: Revision of our case series

1 Introduction

Since 2000 and following for some years one of the interests of our research group has been the determination of beta 2 integrins on polymorphonuclear leukocyte, at a time when experimental research, but also clinical research in various clinical disorders, had aimed to use specific monoclonal antibodies directed towards one of the main beta 2 integrins (CD11a/CD18, CD11b/CD18, CD11c/CD18, CD11c/CD18) or to a subunit with results that at the beginning seemed encouraging and promising (with particular regard to acute ischemic stroke or deep venous thrombosis) but that subsequently were almost completely abandoned. Taking into account this past and not negligible interest we wanted to revisit our previous data, supplemented with some more recent information such as that relating to venous ulcers of the lower limbs, in the light of the most recent literature but limiting now the attention only on the integrin subunit CD18 present in all beta 2-integrins. Integrins are heterodimeric type I transmembrane proteins consisting of alpha and beta subunits and they are expressed in all nucleated cells with a key role in adhesion, cell communication, and migration. Integrins have large extracellular domains which contain the ligand-binding sites, and the ability to bind the ligands is regulated through three conformational changes: inactive (bent-closed), intermediate (extended-closed), and active state (extended-open), with the active conformation (extended-open) has a 4,000-fold increase in ligand affinity compared to the other two states[1, 2].

The leukocyte-restricted β2 integrins comprise four members: αLβ2 (CD11aCD18, LFA-1), αMβ2 (CD11bCD18, Mac-1, Mo-1), αXβ2 (p150,95, CD11cCD18) and βis expressed on all leucocytes, but expression levels vary with cellular activation and differentiation state. It binds ICAM-1, ICAM- 2, ICAM-3, ICAM-4, ICAM-5 and JAM-1 (junctional adhesion molecule-1). It has a pivotal role in leukocyte trans-endothelial migration and in immune synapse formation [3]. Integrin αMβ2, also known as CR3, is expressed primarily on myeloid cells. Unlike integrin αLβ2, αMβ2 binds an array of ligands that do not share canonical binding sequences, including ICAM-1, fibrinogen, JAM-3, denatured proteins, microbial lipopolysaccharide and zymosan. Integrin αMβ2 is a well-established receptor on phagocytes, is involved in leukocyte extravasation across the endothelium, in the degradation and remodelling of ECM and in immune tolerance. Integrin αXβ2 is expressed in myeloid cells, dendritic cells, NK cells and populations of activated T- and B-cells. The expression of αXβ2 serves as a marker to distinguish mature from immature dendritic cells. It mediates phagocytosis of iC3b-opsonized particles and is involved in the adhesion of monocytes to endothelial cells. The last member of the β2 integrins, αDβ2, is expressed at moderate level on myelomonocytic cells and high levels on foamy macrophages in aortic fatty streaks and on eosinophils [4]. The beta2-integrins are involved in numerous pathophysiological processes, [5], including the leukocyte trafficking and leukocyte extravasation into tissues, especially in acute inflammation and sepsis [6–10].

In relation to their fundamentally important role in leukocyte trafficking, beta2-integrins also mediate other cell-cell contacts that are essential for immunological processes. Beta2-integrins (e.g., CD11a/CD18-integrin; LFA-1) are central components of the immunological synapse which forms between an antigen presenting cell (APC) and a T cell, between a B cell and a T cell and between an NK cell and its target cell. In brief, the cell-cell interactions mediated by CD11a/CD18 on the T cell enables T cell activation, by binding to ICAM-1 on the APC.

In myeloid cells such as macrophages, beta2-integrins can initiate intracellular signaling pathways leading to cytokine secretion, either by themselves or together with Toll like receptors (TLRs). In addition, many neutrophil functions such as cytokine release and oxidative burst are dependent on beta2-integrins. Beta2-integrins have also been associated with many immunosuppressive functions. In fact, they can inhibit TLR signaling in macrophages through negative feedback loops, either directly or indirectly, through the anti-inflammatory cytokine IL-10.

While most of the findings concerning the immunoregulatory role of beta2 integrins have been made in myeloid cells, also some lymphocyte subgroups express CD11b/CD18. Indeed, in B cells, CD11b/CD18 has been shown to negatively regulate B cell receptor signaling to maintain autoreactive B cell tolerance. Therefore, manipulating integrin activation pharmacologically could be an efficient therapeutic approach in treating certain inflammatory or autoimmune diseases.

Finally, recent data suggest that beta2- integrins modulate neutrophil apoptosis, an important process for the resolution of inflammation.

In relation to the different and important roles of the beta2 integrins, not only in several immunological functions but, in general, in the pathophysiology of leukocytes, our aim, in this paper, is to re-examine some our data relating to the phenotipycal expression of β2 integrins on the surface of polymorphonucleated, in the context of different clinical conditions, in the light of the most recent findings on their function.

Specifically, we have focused on the expression of CD18, which represents the beta subunit of all members of the beta2 integrins above described. On the cell surface of polymorphonuclear cells, using a cytofluorimetric method, we evaluated the expression of CD18 at baseline and after activation from 15 min at 37° with 4-phorbol 12-myristate 13-acetate (PMA) in 9 different groups of medical subjects.

2 Matherials and methods

3 Results

4 Discussion

The re-examination of previously published data underlines their actuality, since all recent data regarding the analysis of the polymorphonuclear (PMN) beta-2 integrins, addressed on the one hand to their structure, their functional regulation and their signalling properties [3] and the other hand to their counter receptors, such as intercellular cell adhesion molecules (ICAM-1 and ICAM-2), junctional adhesion molecules (JAM-A, JAM-C) and receptors for advanced glycation end products (RAGE), have been almost only examined within particular topics such as immunity, neoplasms, autoimmune diseases and leukemia [16].

In comparison with the earlier published in several clinical disorders, which had taken into account at baseline and after activation, in vitro, with PMA and fMLP, prolonged for 5 and 15 minutes, respectively CD11a, CD11b, CD11c and CD18, in this new revision of our case studies we have considered the integrin subunit CD18, that plays a key role making part of the beta2 integrins LFA-1, Mac-1, p 150,95 and Alpha D beta2 (CD11d/CD18), examined at baseline but also after in vitro activation with PMA, prolonged for 15 min.

At the beginning of these conclusive considerations, we must point out that in all control group subjects compared from time to time with the nine groups showing different clinical disorders, after activation, in vitro, with PMA prolonged for 15 min has always been observed, even if with different percentages, a significant increase in the phenotypical expression of CD18.

For the first group, that includes type 1 diabetes mellitus, the literature data about this aspect are quite small. Previously, we had considered them apart [17] but before then other authors had examined almost always the PMN integrin profile in diabetics not divided for type [18, 19] and unfortunately the same approach was followed by other authors [20] who have examined in basal conditions the behaviour of CD18 in relation to the 5 stages of diabetic retinopathy. Other authors, instead, have demonstrated an slight increase of the expression of CD18 in a group of type 1 diabetic patients [21]. Are well known the characteristics that typify the PMNs of type 1 diabetes about chemotaxis, adhesion, oxidative burst activity, migration, phagocytosis and neutrophil-related adhesion molecules [22] but in our group the phenotypical expression of CD18 was, at baseline, significantly lower in comparison with that of control and did not undergo any change after in vitro activation, prolonged for 15 min, with PMA but also with fMLP (data not shown). It is possible that non-response of CD18 to activation in vitro with PMA (not receptor-mediated) and fMLP (receptor-mediated) which is similar to what happens to CD11a and CD11b (data not shown) with the same activators, can be attributed to an alteration of the transduction mechanism. As it is known, the mechanotransduction and/or the signal transduction with particular regard to integrins play a key role in leukocyte recruitment [23, 24]. Prior to the study of integrins, we noticed in the same clinical condition a functional dysfunction of the PMN ascertained by examining their membrane fluidity and their cytosolic calcium concentration both in basal condition and after in vitro activation [25, 26]. In addition to what has been said so far about the role played by integrins, not only in intercellular connections, which are essential in defence mechanisms, but in particular in leukocyte recruitment some authors [27] have reported that a gene-CD18 targeted deficiency protects against multiple low-dose streptozotocin that in mice induces autoimmune diabetes. In the same study the authors report that only in 10% of these mice genetically missing of the CD18 develop diabetes against 95% of the control group and that in these mice there is no evidence of insulits suggesting the pivotal role of this integrin subunit in the pancreatic leukocyte infiltration.

Till now, there have been several studies on the trend of PMN integrins on clinical conditions equivalent to vascular atherosclerotic disease (VAD) [28–31] whereas in the group (second group) enrolled by us [32] there were subjects with peripheral artery disease (PAD), subjects with chronic cerebrovascular disease (CVD) and subjects with PAD associated with CVD; in some of these subjects, also an atherosclerotic coronary disease with chronic course has been observed. Before the publication of our data on the topic, other authors [33–35] had examined respectively the PMN integrin profile in diffuse atherosclerosis, in ischemic heart disease and in coronary artery disease with results not always convergent also considering the employed techniques. In VAD group, at baseline, the phenotypical expression of the CD18 is increased compared to control group, even if such phenotypical expression reduces after in vitro activation with PMA. It is not easy to explain the behaviour of CD18 after activation, even if it is known, that some integrin subunits during activation move from the secretory granules to the cell surface and others, including CD18, can internalize [36]. The interest of the evaluation of beta2 integrins in atherosclerosis is supported by some data of particular attention published in the last decade and especially considering that atherosclerosis turns out to be an immune-inflammatory disease. In animal models is evident that in those knockout for the low-density-lipoprotein receptor the beta2 integrins are able to modulate both the initiation and progression of atherosclerotic lesions and therefore to perform an effective proaterogenic action [37]. In the same animal models (mice) other authors [38] have demonstrated, using an interesting and sophisticated model, that the overregulation of CD11d/CD18 on inflammatory macrophages is responsible for their retention in vascular injury and therefore in the atherosclerosis development.

For convenience, we will treat the trend of CD18 in type 2 diabetes mellitus without and with macrovascular complications (MVC). Previously to our paper, about type 2 diabetes some authors [39] had described the behaviour of CD11b and of CD69, at baseline and after activation with PMA, especially in relation to hypertension and microalbuminuria. After the publication of our data on the topic [40–42], many other authors have examined PMN integrin subunits, in particular the CD18, in type 2 diabetes mellitus without and with MVC almost always in basal condition and obtaining results not always overlapping [43–50]. Of all these works mentioned, those of Horvath et al. [47] and de Vries et al. [48] take on particular interest and in fact these authors have ascertained how the glycemic levels modulate the phenotypical expression of PMN CD11b, while a successful study of Horvath et al. [49] has showed how intravenous glucose infusion, alone, had little effect on the CD11b expression while the infusion of lipids associated or not with glucose increases significantly its cell expression. Our data regarding the phenotypical expression of CD18 found in type 2 diabetes mellitus without (third group) and with (fourth group) MVC, show in both groups such as CD18, that at baseline is significantly increase compared to control, but does not undergo further variation after activation with PMA. Without wishing to draw any particular conclusions, these data seem almost to support what has recently been described by some authors [51] who underscore the deregulation of leukocyte trafficking in type 2 diabetes.

Since the role of PMNs in renal failure is well known [52–55], most of the work has examined the behaviour of integrins on PMNs, on monocytes or basophiles in dialyzed subjects and especially in relation to the pre-and post-dialytic trend and also according to the type of used dialytic membranes [56–64]. There are instead little information about the PMN integrin profile in chronic renal failure (CRF) under conservative treatment. In the paper of Dou et al. [58] in seventeen non dialyzed subjects with CRF there is evident an increased phenotypical expression of CD11b which suggests a state of activation not observed, however, in dialyzed subjects in pre-dialytic time. In our study (fifty group) we have observed [65] that, at baseline, the phenotypical expression of CD18 was increased compared to control and that showed a further increase after activation with PMA. Although this datum may be, at baseline, charged to the spontaneous activation of PMN we must, however, remember that the uremic toxins besides to interfere with leukocyte function such as chemotaxis [66] can influence, increasing them, the cellular expression of integrins [67]. Furthermore, we must underline that in CRF the high-density lipoproteins can modify the expression of some integrin subunits such as CD11b [68].

Beyond their number the leukocytes, and especially PMNs, play a pivotal part in arterial hypertension; their activation, their role as regards the oxidative stress, their altered rheology, the modified expression not only of soluble adhesion molecules but also of beta2 integrins is the common denominator of this clinical disorder. The data regarding the phenotypical expression of CD18 at baseline and after activation with PMA and fMLP described by our research group (sixth group) many years ago [69, 70] and reconsidered a few years later [71] highlights the particular response of PMN beta2 integrin subunits to these two activators in arterial hypertension. As reports in the section of results, the percentage increase in PMN expression of CD18 after activation with PMA exceed 140% in comparison with baseline value and this difference is highest of those observed in our case records. In experimental models of arterial hypertension, such as that obtained in genetically normotensive animals, is present an up-regulation of phenotypical expression of CD11a and CD18 [72]. In obese hypertensives an increased expression of CD11b on circulating monocytes and of CD68 on macrophages in the adipose tissue has been found [73]. CD11b is considered a marker of leukocyte activation and play an important role in the adhesion of circulating cells to the endothelium. Other authors instead have found the in arterial hypertension the leukocyte integrin profile is related to the tumour necrosis factor-alpha (TNF-alpha) [74] and to the adrenergic modulation [75]. It should be noted that in the work of Rea and co-workers the phenotypical expression of CD11b on PMNs and monocytes is significantly related to the systolic blood pressure values as well as is interesting to observe that of PMNs, of monocytes and lymphocytes the expression of the CD11b increases in a dose-dependent manner after stimulation with TNF-alpha. In the paper of Scanzano and co-workers, it is evident, instead, that the PMN expression of CD11b and CD18 modulated by adrenergic agents is antagonized by yohimbine and propanolol and increased by prazosin. The treatment of hypertensives with telmisartan, angiotensin II receptor type 1 antagonist, reduces the lymphocyte expression of CD11b/CD18 [76], affects the monocyte CD11b and reduces their adhesion [77], while in endothelial cells of human umbilical vein, telmisartan reduces the synthesis of TNF-alpha and the monocyte adhesion [78]. Also the hydralazine (antihypertensive drug) in animal models of arterial hypertension acts on CD18, reduces leukocyte migration as well as the ICAM-1 and the P-selectin expression [79]. Similar data had previously been obtained by other authors [80], who in the same animal model (spontaneously hypertensive rats), have ascertained that hydralazine, besides to reduce the adhesion of monocytes to thoracic aorta endothelium, affects the relative gene expression of ICAM-1 and VCAM but not that of E-selectin. Also the therapy with calcium channel blockers modifies the leukocyte adhesion molecules. As regards some authors [81] have demonstrated that lacidipine reduces ICAM-1 level as well as neutrophil and monocyte infiltration. Other authors [82] have instead ascertained that the mibefradil, non-selective calcium channel blocker, reduce the surface expression of beta2 integrins and L-selectin and prevents the excessive adhesion of leukocytes. In the clinical practice instead the amlodipine (5–10 mg/day for 8 weeks) increases CD11a/LFA-1 in all three cell types (lymphocyte, PMN, monocyte), besides to reduce the blood pressure values [83]; in this last study, the authors have found a negative correlation between monocyte phenotypical expression of CD11a and systolic blood pressure values.

In the genesis of deep venous thrombosis (DVT) are involved both the inflammatory response and the coagulative cascade and relatively to the inflammatory one non recent studies give particular importance to the leukocyte activation [84, 85]. In experimental models of DVT, PMNs accumulate early in the venous wall site of thrombosis, and in the same site cytokines such as TNF-alpha and IL-6 tend to increase. In the same models, leukocyte infiltration may be attenuated by anti-TNFalpha antibodies, anti-ICAM-1 antibodies, anti CD18 antibodies, anti-P-selectin antibodies, and by IL-10 [86]. In experimental models of venous thrombosis, knock-out mice for E- and P- selectins were found to develop smaller thrombi associated with reduced leukocyte infiltration of the venous wall. In the same experimental models of venous thrombosis a cross talk between monocyte, neutrophils, and platelets results responsible for the initiation and amplification of TVT and for influencing its unequalled clinical characteristics [87]. Using animal models, and in particular mice with Mac-1 deficiency, some authors [88] have demonstrate how the leukocyte integrin Mac-1 (CD11b/CD18) regulates the thrombotic process via interaction with platelet GPIb-alpha. In human experimental venous thrombosis, induced by thrombin, some authors [89] have observed a significant increase in soluble E-selectin, soluble L-selectin and soluble ICAM-1 in the refluent venous blood. An interesting reflection has come to our observation from the data of Song and co-workers [90] who have demonstrated in 120 subjects with clinical proven acute venous thromboembolism (VTE), of those 72 (60%) with DVT, in comparison with 120 non-VTE subjects, a significant increase in lymphocyte and polymorphonuclear integrins. In our group (seventh group) of DVT subjects [91, 92], clinical disorder in which even before taking into consideration other DVT subjects who had examined [93, 94] the leukocyte rheology by evaluating filtration parameters, membrane fluidity and cytosolic calcium, no variation of the phenotypical expression of CD18, at baseline, was observed in comparison with control group, while after in vitro activation with PMA, prolonged for 15 min, the expression of integrin subunit CD18 significantly increased. The data concerning the phenotypical expression of CD18, at baseline, in this group of DVT subjects can perhaps be explained by the fact that the determination was performed, not immediately, but within three days of what was diagnosed as the acute event, while the activation in vitro with PMA seems to demonstrate a trend that has the profile of regularity all the more since even with the activator fMLP (data non shown), there is an increase in its expression.

In the past years, a relationship has been observed between leukocyte counts and the risk of acute ischemic stroke (AIS), suggesting their role in the onset of this disorder. An increase in leukocyte count and aggregation is an unfavourable prognostic marker for the clinical evolution of AIS [95–98]. In experimental animals, PMNs begin to accumulate in the cerebral ischemic outbreak 6 hours after the onset of ischemia, the infiltration reaches its peak after 48 hours and in many studies it seems that the accumulation follows and does not precede the ischemic damage. In experimental animals [99] the depletion of PMNs by immunological method induces a 40–66% reduction and some studies have also observed a functional improvement. In the same experimental animal models studies with antibodies anti-ICAM-1 and antibodies anti CD11/CD18 showed a reduction in infarct volume [100]. Also in other animal models (pigs) the use of a monoclonal antibody to the leukocyte CD11/CD18 severely reduces the leukocyte adhesion and vascular permeability in pig cerebral circulation [101]. In humans, instead, it has been found that the variation in neurological deficit is associated to the PMN accumulation pattern evaluated by single-photon emission computed tomography [102]. Previously, we have examined [103, 104] PMN beta2 integrins profile in a group (eighth group) of AIS subjects in which, at baseline, the phenotypical expression of CD18 is significantly increased in comparison with control group and that increased after activation with PMA. The our data related to the overexpression of CD18, at baseline, confirm the results obtained by other authors [105–107]. The CD11b/CD 18 data observed on PMN and monocytes on different days following the acute event in a group of 65 AIS subjects have been illuminating on the doctrinal level [108]; in the same study, however, information on the integrin profile was found to be incomplete when the entire group was divided according to the early neurological deterioration and according to the 3-month outcome. Other authors [109] starting from the assumption that the activation of circulating PMNs is integral part of the systemic inflammatory response present in ischemic stroke, among other things, they have analyzed the behaviour of the integrins, the correlation which exists between the peripheral and cerebral activation of PMNs, the possibility that the extent of their activation may give useful information on the recovery and outcome of these patients; and especially that the inhibition of this peripheral activation of PMNs may constitute a potential ischemic stroke treatment. However, as regards the use of molecules capable of carrying out this inhibition in humans, the authors agree that the studies carried out so far do not seem encouraging. To this conclusion had reached some years before also del Zoppo [110] that had analyzed, revisiting several trials, the possibility of using molecules capable of inhibiting the interrelations between PMN leukocytes and endothelial cells during ischemic stroke. Recently other authors [111] in experimental animals have clarified a pathophysiological issue that same time had been suspended and that is that the adhesion of the leukocyte to the cerebral venules precedes the neuronal death and seems sufficient to begin the tissue damage that occurs after cerebral ischemia.

Already in the Saharay’s study [112] the variation of the phenotypical expression of PMN CD11b before, during and after provoked leg venous hypertension had been observed as well as in the same years [113] the expression of some adhesion molecules present in the skin have been related to the stages of chronic venous insufficiency. In the chronic venous insufficiency, examined in 18 subjects, some authors [114] have found on PMNs an overexpression of CD11b and CD18 associated at none variation of the CD11a. In 8 subjects with venous ulcer of the leg and in 11 normal subjects, both subjected to a condition of experimental venous hypertension by an orthostatic stress test, the authors [115] both charged with the venous blood of the foot and the cubital venous blood, have not found any significant difference regarding the phenotypical expression of CD18,CD11a,CD11b and CD11c of neutrophils; different is the behaviour of L-selectin on lymphocytes observed in the group of normal controls. Other authors [116] treating subjects with healed venous ulcers with flavonoids have described an overexpression of CD11b on PMNs. Although the activity of PMNs in venous ulcers of the lower limbs has been repeatedly emphasized [117–119] however, little information are available about the PMN integrin profile in this clinical disorder that in United States seems to involve approximately 600.000 people ever year. As it is known, the pathogenetic factors of chronic venous ulcers are to be found in increased capillary hydrostatic pressure, increased vascular permeability, microcirculation alterations, chronic hypoxia, persistent inflammation, impaired activity of metalloproteases and their inhibitors and leukocyte activation. In our group of 20 subjects with venous leg ulcers (ninth group) we have evaluated [120] the phenotypical expression of CD18, which, at baseline, was not different from that found in the control group, and that did not change with the activation in vitro with PMA; the same trend has been observed with fMLP activation (data not shown). It is not to exclude that the same altered transduction mechanism that we have called for other groups of subjects including in our survey may be involved in this chronic clinical disorder.

In conclusion, the review of our data shows the following information. In type 1 diabetes mellitus, at baseline, CD18 is underexpressed and not vary after PMA activation; in VAD subjects the CD18 that, at baseline, is overexpressed does not undergo any variation after activation; in type 2 diabetes mellitus, without and with MVC,CD18 which at baseline is up-regulated does not change after activation; in CRF subjects the CD18 overexpressed at baseline increases further after in vitro activation; in EH subjects CD18 already up-regulated at baseline shows the highest rate of expression after activation; in DET subjects the phenotypical expression of CD18 not different from control group at baseline, increases after in vitro activation; in AIS subjects the expression of CD18, increased significantly at baseline, spikes after activation; in VLU the expression of CD18 not different, at baseline, from that of the control group, does not vary after activation. All these data regarding the phenotypical expression of polymorphonuclear CD18, at baseline and after activation, in several groups of subjects with different clinical disorders, at first glance, seem dispersive and disjointed but nevertheless seem orient on some specificities typifying the role of PMN CD18 in vascular and metabolic disorders. Its trend moreover is similar in chronic renal failure and in essential hypertension, while results contrasting in the acute disorders such as deep venous thrombosis and acute ischemic stroke; the behaviour of the phenotypical expression of CD18 in venous leg ulcers seems instead referable to the non responsivity, in vitro, of PMNs to these chemotactic agents employed in our laboratory.