Mice lacking the signaling molecule CalDAG-GEFI represent a model for leukocyte adhesion deficiency type III

Single gene mutations in β integrins can account for functional defects of individual cells of the hematopoietic system. In humans, mutations in β₂ integrin lead to leukocyte adhesion deficiency (LAD) syndrome and mutations in β₃ integrin cause the bleeding disorder Glanzmann thrombasthenia. However, multiple defects in blood cells involving various β integrins (β₁, β₂, and β₃) occur simultaneously in patients with the recently described LAD type III (LAD-III). Here we show that the product of a single gene, Ca²⁺ and diacylglycerol-regulated guanine nucleotide exchange factor I (CalDAG-GEFI), controlled the activation of all 3 integrins in the hematopoietic system. Neutrophils from CalDAG-GEFI^–/– mice exhibited strong defects in Rap1 and β₁ and β₂ integrin activation while maintaining normal calcium flux, degranulation, and ROS generation. Neutrophils from CalDAG-GEFI–deficient mice failed to adhere firmly to stimulated venules and to migrate into sites of inflammation. Furthermore, CalDAG-GEFI regulated the activation of β₁ and β₃ integrins in platelets, and CalDAG-GEFI deficiency caused complete inhibition of arterial thrombus formation in mice. Thus, mice engineered to lack CalDAG-GEFI have a combination of defects in leukocyte and platelet functions similar to that of LAD-III patients.

Integrins are heterodimeric cell-surface receptors that mediate adhesion to the extracellular matrix and cell-cell interactions. On circulating blood cells, most integrins are expressed in a resting, low-affinity state. Cellular stimulation induces a high-affinity state in the integrins, which enables them to bind to their ligands. The molecular basis of this inside-out activation of integrins is only partially understood (1). In blood cells, integrins are critical for the formation of the immunological synapse (2), in the extravasation of circulating immune cells from the bloodstream (3), and in the formation of platelet plugs at sites of vascular damage (4). The functional importance of integrins expressed on blood cells is best documented in patients with germline mutations in the gene encoding β₂ integrins (leukocyte adhesion deficiency type I; LAD-I; ref. 5), who experience recurrent infections, and in patients with mutations in the genes encoding α_IIb or β₃ integrins (Glanzmann thrombasthenia; ref. 6), who have bleeding diathesis. Similar phenotypes have been observed in mice with specific deletions in the genes coding the β₂ (7) or β₃ (8) integrins.

Recently, patients with normal expression but defective activation of β₁, β₂, and β₃ integrins have been identified (9–13). Affected patients exhibit clinical symptoms such as severe recurrent infections, a heightened tendency to bleed, and marked leukocytosis. It has been proposed that this group of integrin activation disorders be designated LAD-III (9) based on the nomenclature for LAD-I and LAD-II, which describes patients with impaired expression of β₂ integrins (5) and defective fucosylation of selectin ligands (14), respectively. Because the expression of β₁, β₂, and β₃ integrins appears to be normal in LAD-III patients, it seems likely that a genetic defect in 1 or more intracellular signaling molecules involved specifically in the activation of leukocyte and platelet integrins is the basis of the LAD-III syndrome.

We have recently identified Ca²⁺ and diacylglycerol-regulated guanine nucleotide exchange factor I (CalDAG-GEFI; also referred to as RasGRP2) as crucial for β₃ integrin activation in platelets (15). CalDAG-GEFI is a member of the CalDAG-GEF/RasGRP family of intracellular signaling molecules, containing binding sites for calcium and diacylglycerol as well as a guanine nucleotide exchange factor (GEF) domain that catalyzes the exchange of GTP for GDP bound to Rap1 or Rap2 (16, 17). Rap1 is the major isozyme in both platelets and neutrophils (18, 19). Interestingly, in contrast to Rap1, which is ubiquitously expressed in hematopoietic and nonhematopoietic cells, CalDAG-GEFI appears to be specifically expressed in platelets and megakaryocytes as well as neutrophils within the hematopoietic system as well as in neurons, especially in the striatum of the basal ganglia (15, 17). CalDAG-GEFI^–/– mice are characterized by severely impaired hemostasis, caused by defective activation of Rap1 and integrin α_IIbβ₃ in platelets, and by mild neutrophilia (15). This neutrophilia indicated to us that neutrophils from CalDAG-GEFI^–/– mice might have a defect in extravasation similar to that observed in selectin-deficient mice (20), causing neutrophil accumulation in the bloodstream. Therefore, we evaluated the role of CalDAG-GEFI in the activation of leukocyte integrins, specifically members of the β₂ integrin family. Our results pointed to significant defects in β₁ integrin– and β₂ integrin–mediated adhesion of CalDAG-GEFI–deficient neutrophils in vitro and in vivo, which caused a markedly impaired response to acute inflammation. Furthermore, we examined the function of CalDAG-GEFI in platelets and found that CalDAG-GEFI was essential for the activation of β₁ integrins on platelets and that arterial thrombus formation was completely abolished in CalDAG-GEFI^–/– mice. Thus CalDAG-GEFI regulates β₁, β₂, and β₃ integrins, which suggests this gene may be defective in patients with LAD-III.

Normal expression of agonist receptors, calcium flux, and ROS formation in CalDAG-GEFI–deficient neutrophils. Integrin activation plays a key role in the firm adhesion and extravasation of PBLs (3). We have previously shown that CalDAG-GEFI is expressed in platelets and neutrophils and that it is critical for β₃ integrin activation in platelets (15). To determine whether CalDAG-GEFI plays a role in integrin activation in neutrophils, we studied neutrophil responses to activation both in vitro and in vivo in CalDAG-GEFI^–/– mice and their littermate WT controls.

We first compared the surface expression of key adhesion receptors in neutrophils from CalDAG-GEFI^–/– and WT mice. As shown in Table 1, CalDAG-GEFI–deficient and WT neutrophils expressed comparable levels of P-selectin glycoprotein ligand–1 (PSGL-1) and β₂ integrins on their surface. No difference in agonist-induced upregulation of α_Mβ₂ (Mac-1) on the cell surface was observed, demonstrating that knockout neutrophils were fully capable of recruiting intracellular pools of integrins by granule release (Figure 1A). Compared with WT neutrophils, CalDAG-GEFI–deficient neutrophils expressed less L-selectin and β₁ integrin on their cell surface.

Figure 1

Degranulation, calcium flux, and ROS production in stimulated neutrophils. (A) Mac-1 expression. PBLs from WT and CalDAG-GEFI^–/– mice were kept resting or were activated for 10 minutes with LTB₄ (300 nM), C5a (50 ng/ml), or PMA (200 nM), stained with antibodies against Mac-1, and immediately analyzed by flow cytometry. n = 6. (B) Calcium flux. PBLs from WT and CalDAG-GEFI^–/– mice were loaded with the calcium-sensitive dye Fluo-3, activated with the indicated doses of LTB₄ or C5a, and immediately analyzed by flow cytometry. Results are representative of 6 individual experiments. (C) ROS formation. PBLs from WT and CalDAG-GEFI^–/– mice were loaded with the ROS-sensitive agent DCFDA, activated with PMA (2 μM) or fMLP (5 μM) for 30 minutes at 37°C, and immediately analyzed by flow cytometry. n = 5. ROS production is expressed as fold increase of mean fluorescence intensity over unstimulated cells. No significant differences in Mac-1 expression, calcium flux, or ROS production were observed between WT and CalDAG-GEFI–deficient neutrophils.

Table 1

Expression of adhesion receptors on peripheral blood neutrophils

To study signaling responses triggered by surface-expressed agonist receptors, we tested several agonists for their ability to induce calcium flux in neutrophils from WT and CalDAG-GEFI^–/– mice. No differences were observed in calcium flux in response to various doses of C5a or leukotriene B₄ (LTB₄; Figure 1B) or formyl-methionylleucylphenylalanine (fMLP; data not shown) between WT and mutant cells, nor did we detect significant differences in the formation of ROS between WT and CalDAG-GEFI–deficient neutrophils stimulated with fMLP or PMA (Figure 1C). This finding is in line with previous studies showing that the small GTPase Rac (not a target of CalDAG-GEFI) is required for the activation of the respiratory burst in neutrophils (21, 22). These data establish that the expression and function of agonist receptors and intracellular signaling molecules upstream of calcium flux are normal in CalDAG-GEFI–deficient neutrophils and that cellular functions such as degranulation and ROS formation are also intact.

CalDAG-GEFI mediates Rap1 activation in stimulated neutrophils. After establishing that agonist-triggered events upstream of calcium flux were normal in CalDAG-GEFI–deficient neutrophils, we next studied Rap1 activation in these cells. Within seconds, stimulation of leukocytes triggers the activation of Rap1, i.e., the exchange of GDP for GTP bound to the small GTPase (23, 24). We detected trace amounts of activated Rap1 in lysates of resting WT and CalDAG-GEFI–deficient neutrophils. Upon stimulation with LTB₄, C5a, or a low dose of platelet activating factor (PAF; 3 nM), a strong increase in Rap1-GTP was observed in lysates from WT but not CalDAG-GEFI–deficient neutrophils (Figure 2). The defect in PAF-induced activation of Rap1 observed in CalDAG-GEFI–deficient neutrophils was partially overcome by increasing the concentration of the agonist to 30 nM. Furthermore, we precipitated Rap1-GTP from lysates preincubated with GTP-γS, a form of GTP that cannot be hydrolyzed to GDP by the intrinsic GTPase activity of Rap1. We detected equivalent levels of GTP-γS–loaded Rap1 in mutant and WT samples, demonstrating that Rap1 from the neutrophils of CalDAG-GEFI^–/– mice did not have an intrinsic activation defect. The levels of total Rap1 were comparable in whole cell lysates from WT and CalDAG-GEFI–deficient neutrophils (Figure 2). Together, these data demonstrate that CalDAG-GEFI plays an important role in Rap1 activation in neutrophils stimulated by various agonists and that the defect in Rap1 activation observed in CalDAG-GEFI–deficient neutrophils may be overcome by increasing the concentration of the agonist.

Figure 2

Impaired Rap1 activation in CalDAG-GEFI–deficient neutrophils. Western blots of affinity-precipitated Rap1-GTP showing strongly decreased Rap1 activation in CalDAG-GEFI–deficient neutrophils (–/–) stimulated with LTB₄ (300 nM for 30 seconds), C5a (75 ng/ml for 30 seconds), or PAF (3 or 30 nM for 30 seconds) relative to that of WT neutrophils (+/+). Equivalent loading of GTP onto Rap1 in neutrophils from WT and CalDAG-GEFI^–/– mice was shown by preincubation of lysates with GTP-γS. Total Rap1 levels were determined in whole cell lysates of WT and CalDAG-GEFI–deficient neutrophils. Results are representative of 3 individual experiments.

Impaired activation of β₁ and β₂ integrins in CalDAG-GEFI–deficient neutrophils in vitro. To determine whether CalDAG-GEFI deficiency also affected integrin activation in neutrophils, we first studied β₁ integrin–mediated adhesion of isolated CalDAG-GEFI–deficient neutrophils to fibronectin. Neutrophils isolated from the bone marrow of WT and CalDAG-GEFI^–/– mice were activated with 300 nM LTB₄, and adhesion to fibronectin-coated plates was monitored. As shown in Figure 3A, adhesion of CalDAG-GEFI–deficient neutrophils was significantly impaired compared with that of WT neutrophils. We next examined β₂ integrin–mediated adhesion to fibrinogen, as it was shown previously that leukocyte binding to this ligand is mediated by the β₂ integrin receptors α_Mβ₂ (Mac-1) and α_Xβ₂ (25). Compared with WT neutrophils, CalDAG-GEFI–deficient neutrophils activated with LTB₄ or PAF showed impaired adhesion to fibrinogen (Figure 3B).

Figure 3

CalDAG-GEFI deficiency causes impaired β₁ integrin– and β₂ integrin–mediated adhesion of neutrophils in vitro. (A and B) Neutrophil adhesion to fibronectin (A) or fibrinogen (B) in vitro. WT or CalDAG-GEFI–deficient neutrophils isolated from bone marrow were added to fibronectin- or fibrinogen-coated plates and incubated for 30 minutes in the presence or absence (i.e., resting) of 300 nM LTB₄ or 3 or 30 nM PAF. Plates were washed and adherent neutrophils were counted. n = 4. *P < 0.05, **P < 0.01, ***P < 0.001.

CalDAG-GEFI is critical for β₂ integrin–mediated firm adhesion of neutrophils in vivo. We next examined the adhesion of leukocytes to activated mesenteric venules in the CalDAG-GEFI^–/– and WT mice. Venules were superfused with 300 nM LTB₄ in PBS, and rolling as well as firmly adherent rhodamine 6G–labeled leukocytes were counted over a period of 20 minutes. Significantly more rolling leukocytes were observed in CalDAG-GEFI^–/– than in WT mice (P < 0.04; Figure 4A). However, this may not be a result of increased adhesiveness of the mutant cells, but rather the approximately 2-fold greater number of peripheral neutrophils circulating in these animals (15). It is not surprising that CalDAG-GEFI–deficient leukocytes rolled normally along stimulated venules, as this process depends largely on the binding of leukocyte PSGL-1 to endothelial selectins (26), and PSGL-1 is a constitutively active receptor that we found to be expressed normally in CalDAG-GEFI^–/– mice (Table 1). In contrast, firm leukocyte adhesion induced by LTB₄ requires the rapid activation of integrins (27, 28) via intracellular signaling pathways. We found that WT leukocytes adhered firmly to mesenteric venules within minutes of LTB₄ superfusion, while firm adhesion of CalDAG-GEFI–deficient leukocytes was reduced by more than 90% (Figure 4B). Infusion of WT mice with a blocking antibody against β₂ integrins reduced the firm adhesion of leukocytes by more than 80% (Figure 4C), confirming the important role of β₂ integrins in leukocyte adhesion previously shown in venules of CD18^–/– (29) and Mac1^–/– mice (28). Our results indicate that CalDAG-GEFI is needed for the activation of β₂ integrins both in vitro and in vivo.

Figure 4

Impaired firm adhesion of leukocytes to mesenteric venules in CalDAG-GEFI^–/– mice. (A) Leukocyte rolling in vivo. The number of rolling leukocytes in WT or CalDAG-GEFI^–/– mice was determined by intravital microscopy. Mice were infused with rhodamine 6G to label circulating leukocytes. Leukocyte rolling was quantified 1–5 minutes after superfusion with 300 nM LTB₄. (B) Firm adhesion. WT or CalDAG-GEFI–deficient leukocytes were considered firmly adherent when they remained stationary for more than 30 seconds. n = 6. (C) Blocking antibodies to β₂ integrin markedly reduce leukocyte adhesion to mesenteric venules. Leukocyte adhesion within the first 5 minutes after LTB₄ superfusion was studied in WT mice infused with PBS or 40 μg anti-β₂ antibodies (β₂ Ab). n = 3. *P < 0.05, ***P < 0.001. No significant difference in leukocyte adhesion was observed between untreated and control IgG–treated WT mice (not shown).

Impaired inflammatory response in CalDAG-GEFI^–/– mice. To test the ability of CalDAG-GEFI^–/– mice to respond to acute inflammation, we challenged WT and mutant mice by intraperitoneal injection of thioglycollate (TG). Using this model, several groups have shown a key role for β₂ integrins in the rapid recruitment of neutrophils to the peritoneal cavity (30–32). Others did not find a marked reduction in the total number of neutrophils recruited to the peritoneum of CD18^–/– mice (7). However, CD18^–/– mice are characterized by neutrophil counts elevated by greater than 10-fold compared with controls, suggesting that the efficiency of TG-induced neutrophil recruitment was also impaired in these studies. We found few neutrophils in the unchallenged peritoneal cavities of WT and CalDAG-GEFI^–/– mice (Figure 5A). In WT mice, we found robust intraperitoneal infiltration of neutrophils 4 hours after challenge. In contrast, CalDAG-GEFI^–/– mice exhibited approximately 80% reduction in neutrophil infiltration compared with WT mice after challenge (P < 0.001; Figure 5A), demonstrating a pivotal function for CalDAG-GEFI in neutrophil recruitment to inflamed peritoneum.

Figure 5

CalDAG-GEFI regulates neutrophil extravasation. (A) TG-induced peritonitis. Infiltrating leukocytes were isolated from the peritoneum of WT and CalDAG-GEFI^–/– mice 4 hours after injection of PBS or 3% TG. Neutrophil counts were determined by morphological analyses of Diff-Quik–stained cytocentrifuge preparations by an observer blinded to the genotype. (B) Croton oil–induced dermatitis. Ears of WT and CalDAG-GEFI^–/– mice were painted with croton oil, and infiltrating neutrophils were counted 6 hours later in H&E-stained ear sections. Few neutrophils were found in vehicle-treated ears. n = 3–5. ***P < 0.001.

To test CalDAG-GEFI^–/– mice in a specifically β₂ integrin–dependent inflammation model (7), we studied neutrophil infiltration into croton oil–irritated ears. As shown in Figure 5B, the number of neutrophils counted in the croton oil–painted ears of CalDAG-GEFI^–/– mice was about 60% that of WT controls (P < 0.001). These findings strengthen the case for CalDAG-GEFI as a critical intracellular signaling molecule upstream of β₂ integrin activation.

Impaired β₁ integrin activation in CalDAG-GEFI–deficient platelets. To determine whether CalDAG-GEFI also regulates β₁ integrin function in platelets, we examined platelet adhesion to established substrates of β₁ integrins. Platelets express various members of the β₁ integrin subfamily, including α₂β₁ (collagen as the main ligand), α₅β₁ (fibronectin as the main ligand), and α₆β₁ (laminin as the main ligand) (33, 34). The expression levels of β₁ integrins were comparable in platelets from CalDAG-GEFI^–/– and WT mice (mean fluorescence intensity, 240.5 ± 8.0 and 242.7 ± 14.8, respectively). We studied the adhesion of activated WT and CalDAG-GEFI–deficient platelets to a laminin-coated surface by stimulating platelets with PAR4-activating peptide (PAR4p; GYPGKF) or a combination of ADP and the thromboxane A₂ mimetic U46619. In order to avoid adhesion/aggregation of platelets mediated by α_IIbβ₃ integrin, the experiments were performed in the presence of a blocking antibody against this receptor (35). In contrast to activated WT platelets, which showed robust adhesion to laminin, the adhesion of CalDAG-GEFI–deficient platelets stimulated with either agonist was significantly inhibited (Figure 6A). The dependence of the adhesion process on α₆β₁ integrin was demonstrated by blocking this receptor on WT platelets (Figure 6A). With the same experimental setup, we also tested the adhesion of CalDAG-GEFI–deficient platelets to fibronectin, a process mediated by integrins α₅β₁ and α_IIbβ₃ (34, 36). Because we are not aware of any specific reagents that inhibit α_IIbβ₃ integrin–mediated adhesion of mouse platelets to fibronectin, we used EDTA in control experiments to inhibit both β₁ integrin– and β₃ integrin–mediated adhesion. The specificity of the adhesion process for β₃ and β₁ integrins was verified in studies with activated platelets lacking β₃ integrins, which adhered to fibronectin in an α₅β₁ integrin–dependent manner (data not shown). Again, U46619/ADP-stimulated CalDAG-GEFI–deficient platelets showed significantly impaired adhesion to fibronectin (Figure 6B). No significant difference in the adhesion of WT and CalDAG-GEFI–deficient platelets stimulated with PAR4p was observed, indicating that 1 or more alternative signaling pathways mediated β₁ integrin activation in platelets stimulated by PAR4. These results provide strong evidence that CalDAG-GEFI is an important signaling molecule mediating activation of α₆β₁ and α₅β₁ integrins in platelets activated by some, but not all, platelet agonists.

Figure 6

Impaired activation of β₁ integrins in CalDAG-GEFI–deficient platelets. (A and B) Biotinylated WT and CalDAG-GEFI–deficient platelets were stimulated with PAR4p (2 mM) or U46619/ADP (5 or 10 μM) in the presence of a blocking antibody to α_IIbβ₃ and allowed to adhere for 30 minutes under static conditions to laminin (A) or fibronectin (B) in microtiter plates. A separate group of WT platelets was pretreated with a blocking antibody to α₆ (adhesion to laminin) or with EDTA (adhesion to fibronectin) to demonstrate the integrin dependency of the adhesion process. Adherent platelets were quantified colorimetrically for peroxidase activity. Data are mean ± SEM of 3 individual experiments in triplicate wells. **P < 0.01, ***P < 0.001. Similar results were observed with nonbiotinylated platelets when the number of adhesive platelets was determined by light microscopy (not shown).

CalDAG-GEFI^–/– mice do not form thrombi in an experimental arterial thrombosis model. To determine the effect of defective β₁ and β₃ integrin activation on thrombus formation in vivo, we studied WT and CalDAG-GEFI^–/– mice in a model of FeCl₃-induced arterial thrombosis (37, 38). In this model, FeCl₃ causes endothelial denudation leading to the exposure of extracellular matrix, which in turn promotes the adhesion and activation of circulating platelets. We did not observe a significant difference between WT and CalDAG-GEFI^–/– mice in the frequency of platelet tethering (Figure 7, A and B). This was not surprising, as the initial tethering of platelets depends on the constitutively active GPIb-V-IX receptor complex (33), which was expressed in high copy numbers on both WT and mutant platelets (mean fluorescence intensity, CalDAG-GEFI^–/–, 907 ± 37.8; WT, 886.7 ± 35.9). Firm adhesion and/or aggregation of platelets at sites of vascular injury, however, requires agonist-induced inside-out activation of platelet integrins. The first thrombi formed in WT mice approximately 10 minutes after application of FeCl₃ (data not shown), and vessel occlusion was observed in all WT mice within 20 minutes (mean occlusion time, 14 ± 1.3 minutes), whereas thrombus formation was completely absent in CalDAG-GEFI^–/– mice for the duration of the 40-minute observation period (Figure 7, A and C). Video of real-time platelet adhesion/thrombus formation in injured arterioles is provided as supplemental data (Supplemental Video 1; available online with this article; doi:10.1172/JCI30575DS1). Thus, CalDAG-GEFI is a major regulator of β₁ and β₃ integrin activation, and its absence has a profound impact on platelet function in vivo.

Figure 7

Platelet adhesion and thrombus formation in FeCl₃-injured arterioles. (A) WT and CalDAG-GEFI^–/– mice were injected with calcein-green-labeled platelets of the respective genotype, and platelet adhesion was monitored in mesenteric arterioles (diameter, WT, 90.7 ± 3.5 μm; CalDAG-GEFI^–/–, 96.4 ± 4.0 μm) upon application of FeCl₃. Images show platelet adhesion and thrombus formation in arterioles at the indicated times after application of FeCl₃. The direction of blood flow is from top to bottom. (B and C) Comparison of the number of tethering platelets (B) and the time of occlusion (C) in FeCl₃-treated mesenteric arterioles of WT and CalDAG-GEFI^–/– mice. No thrombi formed in the mutant mice during the 40-minute observation period. ***P < 0.001.

We found that CalDAG-GEFI was a major regulator of the activation of β₁, β₂, and β₃ integrins on platelets and neutrophils both in vitro and in vivo. As a consequence, mice lacking CalDAG-GEFI had defective inflammatory responses and markedly impaired ability to form thrombi in response to vascular injury. This phenotype is strikingly similar to that described for LAD-III patients, which combines a mild LAD with a Glanzmann-like bleeding disorder (9–13). In all cases reported to date, the clinical symptoms of LAD-III seem to be caused by a defect in the activation, but not the expression or structure, of β₁, β₂, and β₃ integrins on leukocytes and platelets (9). The successful treatment of LAD-III patients by bone marrow transplantation strongly suggests that the genetic defect underlying the disease resides in hematopoietic cells (9). Thus, it has been speculated that the absence or malfunction of a key integrin regulator expressed in hematopoietic cells is responsible for the phenotype observed in LAD-III patients (9). We propose here that CalDAG-GEFI is such a regulator and that genetic deficiency of CalDAG-GEFI may be present in patients with LAD-III.

Like leukocytes from LAD-III patients (10–12), neutrophils from CalDAG-GEFI^–/– mice behaved normally in many functional assays including ROS formation, intracellular calcium flux, and granule release (Figure 1), demonstrating that the cells were fully capable of activating intracellular signaling pathways. CalDAG-GEFI–deficient neutrophils expressed normal levels of β₂ integrins and PSGL-1 on the cell surface, while the surface expression of β₁ integrins and L-selectin was significantly reduced (Table 1). The decrease in L-selectin expression was similar to that previously observed in P/E-selectin–deficient mice, which also show a defect in neutrophil extravasation (20). Thus, it may reflect shedding of the receptor from activated blood leukocytes that are unable to extravasate into tissue. Interestingly, reduced L-selectin expression was also reported for leukocytes from 1 LAD-III patient (13). As a consequence of their defective integrin activation, CalDAG-GEFI–deficient neutrophils showed an impaired response to acute inflammation. Compared with controls, significantly fewer CalDAG-GEFI–deficient neutrophils migrated into the inflamed tissues of mice subjected to experimental peritonitis or dermatitis (Figure 5). It is interesting to note, however, that neutrophil recruitment in the dermatitis model was only partially reduced in CalDAG-GEFI^–/– mice (approximately 40% of control), while previous studies showed a complete inhibition in neutrophil recruitment in this model in CD18^–/– mice (7). These data suggest that CalDAG-GEFI plays an important and specific role in the activation of β₂ integrins in neutrophils and that, at least in some inflammatory situations, other signaling molecules such as PKC family members (39, 40) or other Rap-GEFs (41) can serve as alternative pathways leading to β₂ integrin activation in the absence of CalDAG-GEFI. Similarly, we have previously shown robust aggregation of CalDAG-GEFI–deficient platelets stimulated with thrombin or collagen (15). In further studies, we identified signaling by PKC as an independent pathway allowing for α_IIbβ₃ activation in the absence of CalDAG-GEFI (Bergmeier et al., unpublished observations).

We now provide evidence that CalDAG-GEFI was also critical for the activation of platelet β₁ integrins (Figure 6). Activation of β₁ integrins was almost completely inhibited in CalDAG-GEFI–deficient platelets stimulated with ADP and the thromboxane A₂ analog U46619, while it was only partially inhibited in CalDAG-GEFI–deficient platelets activated by PAR4 receptors (Figure 6). Our results confirm observations by Gruner et al., who showed that β₁ integrins are expressed on resting mouse platelets in a low-affinity state and that cellular activation is required for these integrins to shift to a high-affinity state (34). Previous work with human platelets suggested that integrins α₅β₁ and α₆β₁ may be expressed constitutively in a high-affinity state, as these cells spontaneously adhere to fibronectin and laminin, respectively (36, 42). However, these studies only analyzed the adhesion of unstimulated platelets and did not investigate whether platelet activation would further increase the number of adherent cells. Based on our findings that strong agonists induced significant activation of β₁ and β₃ integrins in knockout platelets in vitro, we were surprised that CalDAG-GEFI deficiency led to complete inhibition of arterial thrombus formation (Figure 7). Probably the simplest explanation for this finding is that the collagen and thrombin concentrations that a platelet encounters at sites of vascular damage are much lower than those used in our in vitro studies. Alternatively, the strong defect in platelet adhesion observed in CalDAG-GEFI^–/– mice could reflect the critical role that CalDAG-GEFI plays in outside-in signaling by ligand-occupied integrins in platelets (43) or that CalDAG-GEFI–deficient platelets are not capable of activating their integrins on the subsecond scale required for successful adhesion under conditions of flow. Such a defect has been suggested for leukocytes from LAD-III patients, which showed a prominent adhesion deficiency under flow (13). We similarly found the most striking defects in the adhesion of CalDAG-GEFI–deficient neutrophils in vivo rather than in vitro under static conditions (Figures 3 and 4).

Although we observed strong defects in platelet and neutrophil adhesion in CalDAG-GEFI^–/– mice, we did not observe spontaneous bleeding or apparent infections in these mice. These findings may appear in contrast to the phenotype described for patients with LAD-III, who have mild to severe bleeding complications as well as recurrent infections (9). However, in patients, bleeding or infection is most likely caused by accidental tissue damage or exposure to pathogens, challenges that mice held in a restricted, pathogen-free living space hardly encounter. Spontaneous bleeding complications are also rare in mouse models of Bernard-Soulier syndrome or severe thrombocytopenia, while strongly impaired hemostasis is observed in tail bleed assays in these mice (44, 45).

The small GTPase Rap1 has recently been identified as a critical molecule regulating integrin activation in many cell types, including platelets, megakaryocytes, and neutrophils (23, 24). However, the widespread expression of Rap1 and its reported importance in nonhematopoietic cells (23, 24) make it unlikely that Rap1 itself is mutated in patients with LAD-III (9). In addition, a recent study by Kinashi et al. demonstrated normal expression but impaired activation of Rap1 in transformed PBLs derived from a patient with LAD-III (46). The authors concluded that a Rap1-GEF activity essential for Rap1 and integrin activation was defective in these cells. We have shown that CalDAG-GEFI is a major regulator of Rap1 and β₁ and β₃ integrin activation in platelets (Figure 6 and ref. 15) and that it is also critical for activation of Rap1 (Figure 2) and β₁ and β₂ integrin (Figures 3 and 4) in neutrophils. In addition, the expression of CalDAG-GEFI is much more tissue specific than that of Rap1 (15, 17), which may explain why mice lacking CalDAG-GEFI do not show obvious defects during embryonic development, whereas mice lacking Rap1b display approximately 85% embryonic lethality (47). Thus, CalDAG-GEFI could be the defective Rap-GEF in platelets and leukocytes of LAD-III patients.

Other proteins involved in the inside-out activation of more than 1 integrin family include cytoskeletal proteins such as talin or filamin (48, 49), integrin-linked kinase (ILK) (50), or members of the PKC family (39, 40). These proteins, however, are not restricted to cells of the hematopoietic lineage; thus, mutations in the genes encoding them are not likely responsible for LAD-III. Furthermore, normal expression of various cytoskeletal adapters, PKC isoforms, and ILK were reported in leukocytes from 1 patient with LAD-III (12).

Different phenotypes have been described in individual patients with LAD-III (10–13), which suggests that there are variations in disease penetrance or that more than 1 gene may be implicated. For example, Harris et al. (11) found that a patient’s integrins had an intrinsic defect in their avidity to bind ligands that could not be overcome by activating antibodies or exogenous cations and that neutrophils from this patient showed an impaired chemotactic response. Such defects have not to our knowledge been reported by other groups. Alon et al. (13) found that leukocytes from another patient with LAD-III exhibited impaired adhesion under physiological flow conditions, but that their adhesion under static conditions was similar to that of controls. Adhesion under static conditions has been found to be severely impaired in all other reported cases of LAD-III (10–12). In CalDAG-GEFI^–/– mice, the defect in neutrophil adhesion was most prominent under conditions of flow, but it was also observed under static conditions.

Our finding that CalDAG-GEFI was critical to the activation of β₁, β₂, and β₃ integrins in mouse platelets and neutrophils suggests this gene may be defective in patients with LAD-III. With their defects in thrombus formation and leukocyte recruitment to sites of inflammation, mice deficient in CalDAG-GEFI represent an animal model for this disease.

View Supplemental data

We are grateful to Tanya Mayadas for advice and helpful discussions and Richard Hynes for critical reading of the manuscript and advice. We thank Lesley Cowan for help with preparing the manuscript. This work was funded by a Scientist Development grant from the American Heart Association (to W. Bergmeier); by stipend GO1360 from Deutsche Forschungsgemeinschaft (to T. Goerge); by National Heart, Lung, and Blood Institute, NIH, grant R37-HL41002 (to D.D. Wagner); by National Institute of Child Health and Development, NIH, grant R01-HD28341 (to A.M. Graybiel); by the Simons Foundation (to A.M. Graybiel); and by National Institutes of Mental Health, NIH, grant F32-MH065815 (to J.R. Crittenden).

Address correspondence to: Denisa D. Wagner or Wolfgang Bergmeier, CBR Institute for Biomedical Research, 800 Huntington Avenue, Boston, Massachusetts 02115, USA. Phone: (617) 278-3344; Fax: (617) 278-3368; E-mail: [email protected] (D. Wagner). Phone: (617) 278-6617; Fax: (617) 278-3368; E-mail: [email protected] (W. Bergmeier).

Nonstandard abbreviations used: CalDAG-GEFI, Ca²⁺ and diacylglycerol-regulated guanine nucleotide exchange factor I; fMLP, formyl-methionylleucylphenylalanine; GEF, guanine nucleotide exchange factor; LAD, leukocyte adhesion deficiency; LTB₄, leukotriene B₄; PAF, platelet activating factor; PAR4p, PAR4-activating peptide; PSGL-1, P-selectin glycoprotein ligand–1; TG, thioglycollate.

Conflict of interest: The authors have declared that no conflict of interest exists.

Reference information: J. Clin. Invest.117:1699–1707 (2007). doi:10.1172/JCI30575

1. Hynes, R.O. 2002. Integrins: bidirectional, allosteric signaling machines. Cell. 110:673-687.
   View this article via: CrossRef PubMed Google Scholar
2. Sims, T.N., Dustin, M.L. 2002. The immunological synapse: integrins take the stage. Immunol. Rev. 186:100-117.
   View this article via: CrossRef PubMed Google Scholar
3. Simon, S.I., Green, C.E. 2005. Molecular mechanics and dynamics of leukocyte recruitment during inflammation. Annu. Rev. Biomed. Eng. 7:151-185.
   View this article via: CrossRef PubMed Google Scholar
4. Savage, B., Cattaneo, M., Ruggeri, Z.M. 2001. Mechanisms of platelet aggregation. Curr. Opin. Hematol. 8:270-276.
   View this article via: CrossRef PubMed Google Scholar
5. Anderson, D.C., Springer, T.A. 1987. Leukocyte adhesion deficiency: an inherited defect in the Mac-1, LFA-1, and p150,95 glycoproteins. Annu. Rev. Med. 38:175-194.
   View this article via: CrossRef PubMed Google Scholar
6. Tomiyama, Y. 2000. Glanzmann thrombasthenia: integrin alpha IIb beta 3 deficiency. Int. J. Hematol. 72:448-454.
   View this article via: PubMed Google Scholar
7. Mizgerd, J.P., et al. 1997. Neutrophil emigration in the skin, lungs, and peritoneum: different requirements for CD11/CD18 revealed by CD18-deficient mice. J. Exp. Med. 186:1357-1364.
   View this article via: CrossRef PubMed Google Scholar
8. Hodivala-Dilke, K.M., et al. 1999. β3-integrin-deficient mice are a model for Glanzmann thrombasthenia showing placental defects and reduced survival. J. Clin. Invest. 103:229-238.
   View this article via: JCI PubMed Google Scholar
9. Alon, R., Etzioni, A. 2003. LAD-III, a novel group of leukocyte integrin activation deficiencies. Trends Immunol. 24:561-566.
   View this article via: CrossRef PubMed Google Scholar
10. Kuijpers, T.W., et al. 1997. Leukocyte adhesion deficiency type 1 (LAD-1)/variant. A novel immunodeficiency syndrome characterized by dysfunctional β₂ integrins. J. Clin. Invest. 100:1725-1733.
    View this article via: JCI PubMed Google Scholar
11. Harris, E.S., et al. 2001. A novel syndrome of variant leukocyte adhesion deficiency involving defects in adhesion mediated by beta1 and beta2 integrins. Blood. 97:767-776.
    View this article via: CrossRef PubMed Google Scholar
12. McDowall, A., et al. 2003. A novel form of integrin dysfunction involving β1, β2, and β3 integrins. J. Clin. Invest. 111:51-60.
    View this article via: JCI PubMed Google Scholar
13. Alon, R., et al. 2003. A novel genetic leukocyte adhesion deficiency in subsecond triggering of integrin avidity by endothelial chemokines results in impaired leukocyte arrest on vascular endothelium under shear flow. Blood. 101:4437-4445.
    View this article via: CrossRef PubMed Google Scholar
14. Karsan, A., et al. 1998. Leukocyte adhesion deficiency type II is a generalized defect of de novo GDP-fucose biosynthesis. Endothelial cell fucosylation is not required for neutrophil rolling on human nonlymphoid endothelium. J. Clin. Invest. 101:2438-2445.
    View this article via: JCI PubMed Google Scholar
15. Crittenden, J.R., et al. 2004. CalDAG-GEFI integrates signaling for platelet aggregation and thrombus formation. Nat. Med. 10:982-986.
    View this article via: CrossRef PubMed Google Scholar
16. Springett, G.M., Kawasaki, H., Spriggs, D.R. 2004. Non-kinase second-messenger signaling: new pathways with new promise. Bioessays. 26:730-738.
    View this article via: CrossRef PubMed Google Scholar
17. Kawasaki, H., et al. 1998. A Rap guanine nucleotide exchange factor enriched highly in the basal ganglia. Proc. Natl. Acad. Sci. U. S. A. 95:13278-13283.
    View this article via: CrossRef PubMed Google Scholar
18. Torti, M., Lapetina, E.G. 1994. Structure and function of rap proteins in human platelets. Thromb. Haemost. 71:533-543.
    View this article via: PubMed Google Scholar
19. Quinn, M.T., et al. 1989. Association of a Ras-related protein with cytochrome b of human neutrophils. Nature. 342:198-200.
    View this article via: CrossRef PubMed Google Scholar
20. Frenette, P.S., Mayadas, T.N., Rayburn, H., Hynes, R.O., Wagner, D.D. 1996. Susceptibility to infection and altered hematopoiesis in mice deficient in both P- and E-selectins. Cell. 84:563-574.
    View this article via: CrossRef PubMed Google Scholar
21. Knaus, U.G., Heyworth, P.G., Kinsella, B.T., Curnutte, J.T., Bokoch, G.M. 1992. Purification and characterization of Rac 2. A cytosolic GTP-binding protein that regulates human neutrophil NADPH oxidase. J. Biol. Chem. 267:23575-23582.
    View this article via: PubMed Google Scholar
22. Abo, A., et al. 1991. Activation of the NADPH oxidase involves the small GTP-binding protein p21rac1. Nature. 353:668-670.
    View this article via: CrossRef PubMed Google Scholar
23. Bos, J.L., de Rooij, J., Reedquist, K.A. 2001. Rap1 signalling: adhering to new models. Nat. Rev. Mol. Cell Biol. 2:369-377.
    View this article via: CrossRef PubMed Google Scholar
24. Caron, E. 2003. Cellular functions of the Rap1 GTP-binding protein: a pattern emerges. J. Cell Sci. 116:435-440.
    View this article via: CrossRef PubMed Google Scholar
25. Ugarova, T.P., Yakubenko, V.P. 2001. Recognition of fibrinogen by leukocyte integrins. Ann. N. Y. Acad. Sci. 936:368-385.
    View this article via: PubMed Google Scholar
26. Butcher, E.C. 1991. Leukocyte-endothelial cell recognition: three (or more) steps to specificity and diversity. Cell. 67:1033-1036.
    View this article via: CrossRef PubMed Google Scholar
27. Tager, A.M., et al. 2000. BLTR mediates leukotriene B (4)-induced chemotaxis and adhesion and plays a dominant role in eosinophil accumulation in a murine model of peritonitis. J. Exp. Med. 192:439-446.
    View this article via: CrossRef PubMed Google Scholar
28. Coxon, A., et al. 1996. A novel role for the beta 2 integrin CD11b/CD18 in neutrophil apoptosis: a homeostatic mechanism in inflammation. Immunity. 5:653-666.
    View this article via: CrossRef PubMed Google Scholar
29. Forlow, S.B., Foley, P.L., Ley, K. 2002. Severely reduced neutrophil adhesion and impaired host defense against fecal and commensal bacteria in CD18^–/–P-selectin^–/– double null mice. FASEB J. 16:1488-1496.
    View this article via: CrossRef PubMed Google Scholar
30. Walzog, B., Scharffetter-Kochanek, K., Gaehtgens, P. 1999. Impairment of neutrophil emigration in CD18-null mice. Am. J. Physiol. 276:G1125-G1130.
    View this article via: PubMed Google Scholar
31. Graf, J.M., Smith, C.W., Mariscalco, M.M. 1996. Contribution of LFA-1 and Mac-1 to CD18-dependent neutrophil emigration in a neonatal rabbit model. J. Appl. Physiol. 80:1984-1992.
    View this article via: PubMed Google Scholar
32. Henderson, R.B., et al. 2001. The use of lymphocyte function-associated antigen (LFA)-1-deficient mice to determine the role of LFA-1, Mac-1, and alpha4 integrin in the inflammatory response of neutrophils. J. Exp. Med. 194:219-226.
    View this article via: CrossRef PubMed Google Scholar
33. Ruggeri, Z.M. 2002. Platelets in atherothrombosis. Nat. Med. 8:1227-1234.
    View this article via: CrossRef PubMed Google Scholar
34. Gruner, S., et al. 2003. Multiple integrin-ligand interactions synergize in shear-resistant platelet adhesion at sites of arterial injury in vivo. Blood. 102:4021-4027.
    View this article via: CrossRef PubMed Google Scholar
35. Bergmeier, W., et al. 2002. Flow cytometric detection of activated mouse integrin alphaIIbbeta3 with a novel monoclonal antibody. Cytometry. 48:80-86.
    View this article via: CrossRef PubMed Google Scholar
36. Beumer, S., IJsseldijk, M.J., de Groot, P.G., Sixma, J.J. 1994. Platelet adhesion to fibronectin in flow: dependence on surface concentration and shear rate, role of platelet membrane glycoproteins GP IIb/IIIa and VLA-5, and inhibition by heparin. Blood. 84:3724-3733.
    View this article via: PubMed Google Scholar
37. Denis, C., et al. 1998. A mouse model of severe von Willebrand disease: defects in hemostasis and thrombosis. Proc. Natl. Acad. Sci. U. S. A. 95:9524-9529.
    View this article via: CrossRef PubMed Google Scholar
38. Ni, H., et al. 2000. Persistence of platelet thrombus formation in arterioles of mice lacking both von Willebrand factor and fibrinogen. J. Clin. Invest. 106:385-392.
    View this article via: JCI PubMed Google Scholar
39. Kolanus, W., Zeitlmann, L. 1998. Regulation of integrin function by inside-out signaling mechanisms. Curr. Top. Microbiol. Immunol. 231:33-49.
    View this article via: PubMed Google Scholar
40. Quinton, T.M., et al. 2002. Protein kinase C- and calcium-regulated pathways independently synergize with Gi pathways in agonist-induced fibrinogen receptor activation. Biochem. J. 368:535-543.
    View this article via: CrossRef PubMed Google Scholar
41. Bos, J.L., et al. 2003. The role of Rap1 in integrin-mediated cell adhesion. Biochem. Soc. Trans. 31:83-86.
    View this article via: PubMed Google Scholar
42. Geberhiwot, T., et al. 1999. Blood platelets contain and secrete laminin-8 (alpha4beta1gamma1) and adhere to laminin-8 via alpha6beta1 integrin. Exp. Cell Res. 253:723-732.
    View this article via: CrossRef PubMed Google Scholar
43. Bernardi, B., et al. 2006. The small GTPase Rap1b regulates the cross talk between platelet integrin alpha2beta1 and integrin alphaIIbbeta3. Blood. 107:2728-2735.
    View this article via: CrossRef PubMed Google Scholar
44. Nieswandt, B., Bergmeier, W., Rackebrandt, K., Gessner, J.E., Zirngibl, H. 2000. Identification of critical antigen-specific mechanisms in the development of immune thrombocytopenic purpura in mice. Blood. 96:2520-2527.
    View this article via: PubMed Google Scholar
45. Ware, J., Russell, S., Ruggeri, Z.M. 2000. Generation and rescue of a murine model of platelet dysfunction: the Bernard-Soulier syndrome. Proc. Natl. Acad. Sci. U. S. A. 97:2803-2808.
    View this article via: CrossRef PubMed Google Scholar
46. Kinashi, T., et al. 2004. LAD-III, a leukocyte adhesion deficiency syndrome associated with defective Rap1 activation and impaired stabilization of integrin bonds. Blood. 103:1033-1036.
    View this article via: CrossRef PubMed Google Scholar
47. Chrzanowska-Wodnicka, M., Smyth, S.S., Schoenwaelder, S.M., Fischer, T.H., White (2nd.), G.C. 2005. Rap1b is required for normal platelet function and hemostasis in mice [erratum 2005, 115:2296]. J. Clin. Invest. 115:680-687.
    View this article via: JCI PubMed Google Scholar
48. Liu, S., Calderwood, D.A., Ginsberg, M.H. 2000. Integrin cytoplasmic domain-binding proteins. J. Cell Sci. 113:3563-3571.
    View this article via: PubMed Google Scholar
49. Ratnikov, B.I., Partridge, A.W., Ginsberg, M.H. 2005. Integrin activation by talin. J. Thromb. Haemost. 3:1783-1790.
    View this article via: CrossRef PubMed Google Scholar
50. Wu, C., Dedhar, S. 2001. Integrin-linked kinase (ILK) and its interactors: a new paradigm for the coupling of extracellular matrix to actin cytoskeleton and signaling complexes. J. Cell Biol. 155:505-510.
    View this article via: CrossRef PubMed Google Scholar
51. Bradley, P.P., Priebat, D.A., Christensen, R.D., Rothstein, G. 1982. Measurement of cutaneous inflammation: estimation of neutrophil content with an enzyme marker. J. Invest. Dermatol. 78:206-209.
    View this article via: CrossRef PubMed Google Scholar
52. Mayadas, T.N., Johnson, R.C., Rayburn, H., Hynes, R.O., Wagner, D.D. 1993. Leukocyte rolling and extravasation are severely compromised in P selectin-deficient mice. Cell. 74:541-554.
    View this article via: PubMed CrossRef Google Scholar