Antiplatelet effects of the CEACAM1-derived peptide QDTT

Abstract

Carcinoembryonic antigen-related cell adhesion molecule 1 (CEACAM1) restricts platelet activation via platelet collagen receptor GPVI/FcRγ-chain. In this study, screening against collagen-induced platelet aggregation was performed to identify functional CEACAM1 extracellular domain fragments. CEACAM1 fragments, including Ala-substituted peptides, were synthesized. Platelet assays were conducted on healthy donor samples for aggregation, cytotoxicity, adhesion, spreading, and secretion. Mice were used for tail bleeding and FeCl₃-induced thrombosis experiments. Clot retraction was assessed using platelet-rich plasma. Extracellular segments of CEACAM1 and A1 domain-derived peptide QDTT were identified, while N, A2, and B domains showed no involvement. QDTT inhibited platelet aggregation. Ala substitution for essential amino acids (Asp139, Thr141, Tyr142, Trp144, and Trp145) in the QDTT sequence abrogated collagen-induced aggregation inhibition. QDTT also suppressed platelet secretion and “inside-out” GP IIb/IIIa activation by convulxin, along with inhibiting PI3K/Akt pathways. QDTT curtailed FeCl₃-induced mesenteric thrombosis without significantly prolonging bleeding time, implying the potential of CEACAM1 A1 domain against platelet activation without raising bleeding risk, thus paving the way for novel antiplatelet drugs.

Plain Language Summary

What is the context?

• The study focuses on Carcinoembryonic antigen-related cell adhesion molecule 1 (CEACAM1) and its role in platelet activation, particularly through the GPVI/FcRγ-chain pathway.The research aims to identify specific fragments of CEACAM1’s extracellular domain that could restrict platelet activation, without increasing bleeding risk.

What is new?

• The researchers identified a peptide called QDTT derived from the A1 domain of CEACAM1’s extracellular segment. This peptide demonstrated the ability to inhibit platelet aggregation, secretion, and GP IIb/IIIa activation.

• The study also revealed that specific amino acids within the QDTT sequence were essential for its inhibitory effects on collagen-induced aggregation.

What is the impact?

• The findings suggest that the A1 domain-derived peptide QDTT from CEACAM1 could serve as a potential basis for developing novel antiplatelet drugs. This peptide effectively limits platelet activation and aggregation without significantly prolonging bleeding time, indicating a promising approach to managing thrombosis and related disorders while minimizing bleeding risks.

Keywords:

• Antithrombotic effects
• carcinoembryonic antigen-related cell adhesion molecule 1
• convulxin
• peptides
• platelet activation

Introduction

Cardiovascular diseases are an important cause of morbidity and death worldwide.¹ Thrombosis is considered a major complication of cardiovascular disease, which leads to acute ischemic stroke, myocardial infarction, and venous thromboembolism.^2,3 Platelets, which have a central role in hemostasis and thrombosis, have long been a primary target for therapeutic intervention for preventing occlusive thrombotic events.^1,2 Over the years, various antiplatelet drugs, such as clopidogrel, aspirin, and tirofiban, have been developed. Yet, their side effects and drug resistance demand the introduction of novel, effective treatments.⁴

Carcinoembryonic antigen-related cell adhesion molecule 1 (CEACAM1) is a transmembrane glycoprotein expressed in platelets, endothelial cells, and immune cells.⁵ Most epithelial, endothelial, and lymphocyte cells express CEACAM1-4 L and CEACAM1-4S among the 11 isoforms that make up the CEACAM1 family, all sharing the same extracellular domains,⁶ comprising the N terminus of the immunoglobulin V (IgV)-like domain (N domain) and three immunoglobulin constant-region-type-2 (IgC2)-like domains (A1, B, and A2 domains). The differences in isoforms lie in the intracellular cytoplasmic tails of CEACAM1-4 L and CEACAM1-4S. The CEACAM1-4 L tail includes two immunoreceptor tyrosine-based inhibitory motifs (ITIMs), whereas the CEACAM1-4S tail does not but has sequences for calmodulin and tropomyosin and may interact with the cytoskeleton.^7,8 Previous studies using CEACAM1-deficient mice showed that CEACAM1 inhibits collagen-platelet interactions.⁹ Furthermore, CEACAM1 is essential for platelet function through GP IIb/IIIa.^9,10 The intracellular and extracellular domains of CEACAM1 has substantial role in regulating various cellular functions. It has been also demonstrated that the IgV-like N terminal domain of human CEACAM1 directly contributes to the implantation of cancer cells through its homotypic and heterotypic binding capabilities.^11–15 In addition, a study showed that the IgC2-like domain of CEACAM1 is crucial for regulating the cell-binding abilities of the protein.¹⁶ Nevertheless, the regulatory mechanisms and critical domains involved in platelet activation have not been comprehensively explored.

Glycoprotein VI (GPVI) is a membrane protein expressed only in bone marrow-derived megakaryocytes and platelets. It is the primary platelet receptor for collagen.¹⁷ A damaged vessel wall exposes matrix collagen, which binds to GPVI to stimulate platelet activation.¹⁷ GPVI then mediates the activation of immunoreceptor tyrosine-based activation motif (ITAM) signaling pathways and promotes intracellular Ca²⁺ mobilization and platelet shape change.¹⁷ GPVI induces platelet secretion of adenosine diphosphate (ADP), thromboxane, and soluble platelet agonists.^18,19 These soluble factors can further activate and recruit circulation platelets adhering to the injured vessel wall surface.²⁰ Activating signaling molecules in platelet cytoplasm, known as “inside-out” signal transduction, ultimately makes integrin αIIbβ3 (GP IIb/IIIa) to adopt its high-affinity conformation.²¹ The high-affinity conformations of GP IIb/IIIa bind to fibrinogen, facilitating the “outside-in” activation of platelets to form stable thrombi.²¹ Antagonizing GPVI receptor function can effectively prevent thrombosis and slightly influence coagulation function.¹⁷ In addition, antagonizing the function of the GPVI receptor is also conducive to reversing platelet participation in the inflammatory response.^17,22–24 Therefore, GPVI has become a hot target for new antithrombotic therapies.

A study showed that matrix metallopeptidase-12 (MMP-12) could cleave recombinant human CEACAM1 (rhCEACAM1, aa A34-G252) into several peptide fragments in vitro.¹⁴ One of these peptides, QLSNGNRTLT (QLSN), derived from an IgC2 domain (A1 domain), inhibits collagen-induced platelet activation.²⁵ Still, the rhCEACAM1 peptide fragments were only tested against platelet activation by a high dose of collagen (5 μg/mL), not by lower doses. Therefore, it is essential to determine whether other extracellular CEACAM1 domains can influence platelet activation under physiological conditions. Hence, the present study examined the effects of rhCEACAM1 protein and peptides on agonist-induced platelet aggregation. In addition, the effects of one specific fragment, QDTT, were examined on GPVI-mediated platelet activation and thrombus formation. Our findings provide novel insights into the mechanisms of thrombosis and identify potential new therapeutic targets.

Materials and methods

Antibodies and reagents

rhCEACAM1 was reconstituted in sterile phosphate-buffered saline (PBS, ST448) purchased from the Beyotime Institute of Biotechnology (Shanghai, China). GeneTex Inc. (Irvine, California, USA) provided the human FcγRIIa monoclonal antibody (IV.3, GTX14572). Convulxin (ALX-350-100-C050) was supplied by Enzo Life Sciences (Farmingdale, New York, USA). Chrono-Log Corp. (Havertown, PN, USA) provided collagen (P/N 385), ADP (P/N 384), thrombin (P/N 386), arachidonic acid (AA, P/N 390), and luciferin/luciferase (P/N 395). Prostaglandin E1 (PGE1, 900100P), fibrinogen (F3879), apyrase (A6410), and calcium ionophore (C9275) were supplied by Sigma-Aldrich (St. Louis, MO, USA). Fluo-3 AM (F880) and rhodamine (R7921) were purchased from Solarbio Life Science (Beijing, China). Thermo Fisher Scientific Inc. (Waltham, MA, US) provided Goat anti-Mouse (GAM) IgG F(ab’)2 (31166). Triton X-100 (A110694) was supplied by Sangon Biotech (Shanghai, China) Cell Signaling Technology Inc. (Danvers, Massachusetts, US) provided the antibodies for Src (2110), phospho-Src (Tyr 416, 6943), Akt (4691), phospho-Akt (Thr 308, 2965), Syk (12358), phospho-Syk (Tyr 525/526, 2710), and PLCγ2 (3872). Antibodies against phospho-PLCγ2 (Tyr 759, GTX133463) were from GeneTex Inc (Irvine, California, USA).

Synthesis, purification, and storage of the peptides

In order to explore the other active peptides, synthetic peptide sequences were designed according to the cleavage fragments of CEACAM1 (Table S2). As previously reported,¹³ we also successfully synthesized 29 peptides of 14 amino acids in the length of extracellular CEACAM1 (Table S1). The peptides were produced using a peptide synthesizer, and their sequences were determined according to a standard method.

A Rainin Symphony Peptide Synthesizer (Mettler-Toledo, Columbus, OH, USA) was used to create the CEACAM1-derived peptides, Ala-substituted peptides, and scrambled peptides, as previously described.²⁵ Crude peptides were purified using a high-performance liquid chromatography (HPLC) column (Shimadzu LC-2010A) using a gradient elution of acetonitrile and triethylamine phosphate aqueous solution (flow rate of 1 ml/min). The peak samples were collected at A220 nm. The samples were tested for molecular weight and identification of peptide composition using MS (Shimadzu LCMS-2020). The peptides (purity ≥ 95%) were stored at −20°C. The peptides were dissolved in dimethyl sulfoxide (DMSO) and were diluted with 0.9% normal saline.

Platelet aggregation assay

All human-based experiments were performed in accordance with the Declaration of Helsinki. The study protocol was approved by the Institutional Review Board of the First Affiliated Hospital of Kunming Medical University (Kunming, China; approval #2020-L-17). All volunteers signed an informed consent form before donating blood samples.

For the determination of rhCEACAM1 concentration, previous studies determined that the serum CEACAM1 levels in healthy humans were about 7 ng/ml, while the serum CEACAM1 levels in patients with myocardial infarction were >11 ng/ml.^26,27 On the other hand, Ergün et al.²⁸ used supraphysiological concentrations of rhCEACAM1 in the range of 100–600 ng/ml to treat human dermal microvascular endothelial cells (HDMECs) for 72 h before observing the promoter growth of HDMECs. In the present study, we used rhCEACAM concentrations ranging from 80 pM-50 nM (i.e., 3.6–2230 ng/ml) and observed that the incubation of rhCEACAM1 at 2 nM (i.e., 89 ng/ml, within the range of the physiological and pathological concentrations) for 10 min markedly inhibited collagen-induced platelet aggregation. For the determination of peptide concentration, the concentration of peptide that could inhibit collagen-induced platelet aggregation was screened using the peptides from Table S2, and the concentration was gradually increased until the concentration with the maximum collagen inhibition was reached.

Human apheresis platelets were obtained at the Yunnan Kunming Blood Center (Kunming, China) from healthy volunteers who had not taken any medication within 2 weeks.^14,25 The methods are shown in the Supplementary Materials. ADP-induced platelet aggregation was performed using diluted human apheresis platelets, as previously described.²⁵ For FcγRIIa-mediated platelet aggregation, the washed platelets were first mixed with IV.3 (1.25 μg/mL), stirred at 1,200 rpm for 1 min, and then 20 μg/mL of GAM IgG F(ab’)2 antibody was added.²⁹

Platelet cytotoxicity assay

The measurement of lactate dehydrogenase (LDH) leakage was performed as previously reported.³⁰ The methods are shown in the Supplementary Materials.

Platelet adhesion and spreading assay

The methods are shown in the Supplementary Materials and performed as previously described.^31,32

Platelet secretion assay

The release of ATP was detected during platelet aggregation using the Chrono-log Whole Blood/Optical Lumi-Aggregation System (Chrono-Log Corp.).²⁵ The methods are shown in the Supplementary Materials. The platelets were detected by flow cytometry, as described previously.²⁹

Western blotting

The suspended platelets were incubated with QDTT or DMSO for 10 min at 37°C. Eptifibatide (HY-B0686, MedChemExpress, Monmouth Junction, NJ, USA) prevents platelet aggregation.³³ The methods are shown in the Supplementary Materials.

Tail bleeding

Forty male C57BL/6 mice (4–6 weeks, 18–20 g) were purchased from SJA Laboratory Animal Company (Hunan, China). All animal procedures were carried out in accordance with institutional guidelines. The study was approved by the Experimental Animal Ethics Committee of Kunming Medical University (approval #kmmu2021737). All the animals were housed in an environment with a temperature of 18–26°C, a relative humidity of 40%–70%, and a light/dark cycle of 12/12 hr, and had free access to water and food.

The measurement of the bleeding time in mice was performed according to a previously reported method.³⁴ Male mice were randomly divided into three groups. The mice were anesthetized using an intraperitoneal injection of pentobarbital (45 mg/kg). After administration of QDTT (10 mg/kg), aspirin (100 mg/kg), or DMSO via the tail vein for 30 min, the tail was cut off 5 mm from the distal end of the mouse tail and soaked in saline at 37°C. Blood coagulation was assumed to occur when there was no blood flow for 60 s.

Mouse model of FeCl₃-injured mesenteric arteriole thrombus

Injured mesenteric arterioles were observed and recorded, as previously described.³⁵ C57BL/6 mice aged 6–8 weeks were anesthetized by an intraperitoneal injection of pentobarbital (45 mg/kg). Injection of QDTT (10 mg/kg), aspirin (100 mg/kg), or DMSO and rhodamine (2.5 mg/kg) into the caudal vein was performed 30 min before thrombosis induction (10 mice/group). A filter paper (5 × 2 mm) saturated with 7.5% FeCl₃ solution was placed on the mesenteric arterioles with a 60–200 μm diameter for 2 min to induce vascular injury. Within 2 min of removing the filter paper, a blood clot appeared in the vessel at the site of contact with the filter paper, indicating that the modeling was successful. During thrombosis, images were captured every 5 min using an MSHOT MD50 camera (Micro-Shot Technology Ltd.) and an Olympus i×73microscope (Olympus Corp., Tokyo, Japan).

Clot retraction

Clot retraction tests were carried out using a previously described technique with slight changes.³⁶ Platelet-rich plasma (PRP) was prepared at 500 × 10⁹/L and incubated with QDTT at various doses (50, 250, and 500 μM) or DMSO for 10 min. Then, 200 μL of PRP, 1 mM CaCl₂, 0.1 U/ml thrombin, and a siliconized clear glass tube were added, thoroughly mixed at 37°C, and images were taken every 10 min. Image J was used to determine the clot size.

Statistical analysis

GraphPad Prism 9 software (GraphPad Software Inc., San Diego, CA, USA) was used to analyze the experimental data. The results were expressed as means ± standard errors of the mean (SEM). The differences between the two experimental groups were analyzed using the t-test. One-way analysis of variance (ANOVA) and Tukey’s post hoc test were used to compare more than two groups. P-values < .05 were considered statistically significant.

Results

Screening of the extracellular domain of rhCEACAM1 and effects of the synthetic peptides on platelet aggregation

We attempted to determine whether rhCEACAM1 (aa Gln35-Gly428) contains the whole extracellular domain of amino acids of CEACAM1 and regulates platelet aggregation. We compared the aggregation of platelets in the presence or absence of 0 (vehicle: PBS), 0.08, 2, or 50 nM rhCEACAM1 over a range of acid-soluble type I collagen concentrations (0.625 or 1 μg/mL). Similar experiments were performed using GPCR agonists, including ADP (5 or 50 μM), thrombin (0.05 or 0.1 U/mL), and arachidonic acid (AA; 0.167 or 0.5 mM). As shown in Figure 1a, platelets displayed similar aggregation profiles in response to thrombin and AA in the presence of 0.08, 2, or 50 nM rhCEACAM1. Yet, 50 nM rhCEACAM1 treatment decreased platelet aggregation responses to collagen 0.625 µg/mL and ADP 5 µM but not to collagen 1 µg/mL and ADP 50 µM (Figure 1a).

Figure 1. The effects of rhCEACAM1 and A1 domain of CEACAM1-derived peptides on collagen-induced platelet aggregation. (a) Platelet aggregation induced by various agonists. The results are expressed as mean ± SEM (n = 3). *p < .05, **p < .01, vs. The vehicle group. (b, c) Platelet aggregation after treatment with various peptides and collagen stimulation. The rate of aggregation was recorded at the end of 10 min. The results are expressed as mean ± SEM (n = 4–6). *p < .05, **p < .01, ****p < .0001, vs. The vehicle group.

We then screened the identified peptide fragments (Table S2) in low doses of collagen (1 μg/mL). Five out of the 10 peptides cleaved by MMP-12 were identified, and the remaining peptides were synthesized. The QNPV and CD66a-10 peptides were insoluble, and thus, they were not further investigated. As illustrated in Figure 1b, 50 μM of each peptide in Table S2 was screened for its effect on 1 μg/mL collagen-induced platelet aggregation. QDTTYLWW (QDTT) and QLSNGNRTLT (QLSN) significantly impaired collagen-induced platelet aggregation. Still, platelet aggregation in response to collagen stimulation was not substantially influenced by other peptides from the A1 domain of CEACAM1 (Figure 1b).

We also screened the data from Skubitz et al.¹³ who reported 29 predicted peptide fragments (Table S1) of N, A1, B, and A2 ectodomains of CEACAM1 on collagen-induced platelet aggregation. Except for CD66a-12, which contains the QLSN amino acid sequence, the remaining 27 CEACAM1-derived peptides had no significant effect on collagen-induced platelet aggregation, even at the concentration of 200 μM (Figure S1).

Mapping the critical amino acids of QDTT

We attempted to identify the critical residues within QDTT by replacing each residue with alanine, one at a time. Our results showed that the replacement of the asparagine (D), the second threonine (T), tyrosine (Y) in the fifth position, and one or two tryptophan (W) residues in the N-terminus position resulted in the loss of the inhibitory effect of QDTT (Figure 1c). In addition, the scrambled peptide TLYWDWQT did not inhibit platelet aggregation (Figure 1c). The results show that QDTT exhibits a sequence-specific activity.

QDTT inhibits various agonist-induced platelet aggregation

rhCEACAM1 (Figure 1a) and QDTT (Figure 1b) significantly inhibited collagen-induced platelet aggregation. PGE1, which significantly hinders platelet activation,^37,38 was used here as a positive control. Next, we examined the ability of QDTT to inhibit different doses of collagen-induced platelet responses. Figure 2a–c shows that 500 μM QDTT strongly inhibited platelet aggregation compared with the vehicle. Nevertheless, the efficacy of QDTT significantly decreased with the increase in collagen concentration (Figure 2d). QDTT also appeared to inhibit convulxin-induced platelet aggregation (Figure 2e). As displayed in Figure 1f–j, compared with the vehicle, 250–500 μM of QDTT reduced thrombin-, ADP-, and arachidonic acid-induced platelet aggregation while slightly influencing calcium ionophore-induced platelet aggregation and having no effect on immune-activated receptor FcγRIIa-induced platelet aggregation. Moreover, QDTT inhibited the static adhesion of platelets to a collagen surface (Figure 2l), and the half-maximal inhibitory concentration (IC₅₀) was calculated as approximately 100 μM. Finally, the LDH release assay showed that 50–500 μM QDTT did not have toxicity effects on platelets under the conditions used (Figure 2k). The results of the present study indicate that QDTT substantially inhibits the aggregation and adhesion of platelets induced by collagen.

Figure 2. The effects of QDTT on various agonists-induced platelet aggregations and its cytotoxicity. (a-j) Platelet aggregation after treatment with QDTT and activation with various agonists. The results are presented as mean ± SEM (n = 4–6). *p < .05, **p < .01, ***p < .001, ****p < .0001, vs. The vehicle group. (k) LDH assay after QDTT treatment. The results were presented as mean ± SEM (n = 3). ns: not significant versus each group. (i) Platelet adhesion after QDTT treatment. The results are expressed as mean ± SEM (n = 4). **p < .01, ****p < .0001 vs. The vehicle group.

QDTT inhibits GPVI-mediated platelet secretion

GPVI reacts with collagen and activates platelets.³⁹ Since QDTT inhibits platelet aggregation induced by various agonists, we investigated how QDTT could affect platelet secretion induced by various agonists. As shown in Figure 3, platelet P-selectin expression (representing α-granule release) and ATP release induced by convulxin were inhibited by QDTT. Next, we examined how QDTT could inhibit convulxin and other agonists-induced GP IIb/IIIa-mediated “inside-out” activation. It was revealed that both PAC-1 expression and Ca²⁺ release were increased in response to convulxin. PGE1 could decrease those effects, and QDTT significantly inhibited convulxin-induced platelet PAC-1 expression and Ca²⁺ release (Figure 4a,b,d,e). QDTT had little effect on ADP-, AA-, and thrombin-induced platelet PAC-1 expression (Figure 4c), and QDTT did not impair AA- and thrombin-induced platelet Ca²⁺ release (Figure 4f). The findings show that the inhibitory effects of QDTT on platelet activation are associated with GPVI-activated GP IIb/IIIa.

Figure 3. Effects of QDTT on convulxin-induced platelet secretion. Histogram (a) and a bar graph (b) present the effects of QDTT on platelet membrane P-selectin expression after convulxin stimulation, as assessed by flow cytometry using phycoerythrin (PE)-conjugated anti-CD62P monoclonal antibody. (c) Bar graph presents the effects of QDTT on convulxin-induced ATP release. The results were expressed as mean ± SEM (n = 3). *p < .05, ***p < .001, and ****p < .0001 vs. The vehicle group.

Figure 4. Effects of QDTT on GP IIb/IIIa-mediated human platelet “inside-out” activation. Histogram (a) and a bar graph (b, c) present the effects of QDTT on platelet membrane PAC-1 expression after agonist stimulation (c). (d) Line graph representing the effect of QDTT on convulxin-induced calcium ion release in real-time. (e and f) the bar graph represents the percentage of peak RFU at different concentrations of QDTT compared with the vehicle group. The results are presented as mean ± SEM (n = 3–4). *p < .05, **p < .01, ***p < .001, and ****p < .0001 vs. The vehicle group.

QDTT inhibits convulxin-induced platelet ITAM signaling

This study evaluated the hypothesis that QDTT might block convulxin-induced platelet ITAM signaling in human platelets, as QDTT was shown to reduce convulxin-induced platelet aggregation and secretion. The tyrosine phosphorylation of platelet ITAM signaling was examined under convulxin stimulation to examine this hypothesis. As shown in Figure 5a,b, the influence of QDTT on platelet ITAM signaling stimulated by convulxin was assessed. Convulxin stimulation activated Src, Syk, Akt, and PLCγ2 in platelets, and the addition of QDTT restrained the effect. Secondary mediators (such as ADP and thromboxane) are released during GPVI activation.⁴⁰ Therefore, to indicate whether QDTT could inhibit GPVI signaling, the signaling was investigated with inhibitors of secondary mediators by adding apyrase and indomethacin. Integrilin was added to both groups to prevent platelet aggregation during the experiment. As shown in Figure 5c,d, QDTT inhibited the activation of platelet ITAM signaling molecules induced by GPVI in the presence of apyrase and indomethacin. The evidence suggests that QDTT represses GPVI-mediated platelet activation via the Src-Syk-Akt-PLCγ2 signaling pathway.

Figure 5. Effects of QDTT on GPVI-mediated platelet signal transduction. (a, b) After incubation with QDTT, 2 ng/mL convulxin was added to stimulate washed human platelets while stirred at 1,200 rpm. In order to block feedback from ADP and thromboxane during convulxin-medicated GPVI activation, we added apyrase (1 U/mL), indomethacin (10 μM), and integrin (20 μg/mL) before stimulation of platelets by convulxin at 37°C (c, d). The results are presented as mean ± SEM (n = 3). *p < .05 and **p < .01 vs. The vehicle group. (e, f) After incubation with QDTT, PI3K inhibitor wortmannin (Wort.), or DMSO, collagen was added to washed human platelets. The corresponding aggregation rate (e) was recorded, and Akt phosphorylation (f) was measured. ImageJ software was used to calculate the density of bands. The results are presented as mean ± SEM (n = 3). *p < .05, **p < .01, ***p < .001, and ****p < .0001 vs. DMSO (vehicle)-treated platelets. ^####p < .0001 vs. The Wort. (50 nM) group.

Effects of QDTT on regulating phosphatidylinositol 3‑kinase (PI3K)/protein kinase B (Akt) signaling pathway

The PI3K pathway is crucial for GPVI-mediated platelet activation.^41,42 The effects of QDTT on the PI3K signaling pathway were investigated using the PI3K-specific broad-spectrum antagonist Wort. and Akt activation because Akt is the main molecule downstream of the PI3K signaling pathway. Figure 5e shows that Wort. and QDTT reduced collagen-induced platelet aggregation and the addition of Wort. increased the inhibitory effects of QDTT. The phosphorylation levels of Akt were significantly different between the vehicle and Wort. groups. The Wort.+QDTT group had no difference in Akt activation compared with the Wort. group (Figure 5f). Hence, the PI3K/Akt signaling pathway could be associated with the effects of QDTT on collagen stimulation.

QDTT inhibits GP IIb/IIIa-mediated “outside-in” platelet activation

The “inside-out” activation of platelets induced by different activators causes conformational changes in GP IIb/IIIa and binding to its ligands, such as fibrinogen.⁴³ The ligand-bound integrins, in turn, cause clot retraction and spreading and “outside-in” mediated platelet activation. Therefore, we explored the effect of QDTT on platelet clot retraction and spreading. As shown in Figure 6, QDTT significantly inhibited platelet spreading on fibrinogen-coated slides (Figure 6c–h) and also significantly inhibited platelet clot retraction (Figure 6a,b). The above findings suggest that QDTT inhibits the function of platelet membrane protein GP IIb/IIIa.

Figure 6. QDTT inhibited platelet spreading on fibrinogen-coated surfaces and clot retraction. (a,b) Clot retraction reaction. Each experimental group was repeated three times, and the pictures represent three experiments. **p < .01 vs. The control group (vehicle). (c-h) platelet adherence and spreading after QDTT treatment (magnification × 40, twelve 0.03 mm² fields of view were taken for each experimental group for counting). (c-f) pictures of QDTT (50, 250, or 500 μM), DMSO-incubated platelets spreading on fibrinogen-coated substrates. (g-h) histogram of platelet spreading inhibition by QDTT. Each experimental group was repeated three times, and the pictures represent three experiments. Spread platelets/filed and total area expressed as mean ± SEM (n = 3). *p < .05, ***p < .001 and ****p < .0001 vs. The DMSO group (vehicle).

QDTT inhibits FeCl₃-induced mouse mesenteric arteriole thrombosis formation

Platelet activation suggested that QDTT could participate in platelet activation. Thus, the effects of QDTT on thrombosis in mice were explored using a FeCl₃-induced mesenteric artery thrombosis model. As shown in Figure 7a,b, QDTT 10 mg/kg significantly prolonged FeCl₃-induced stable mesenteric artery thrombosis. However, it was also found that QDTT was not as good as aspirin 100 mg/kg in inhibiting vessel occlusion; yet, different doses were applied.

Figure 7. Effects of QDTT on tail bleeding and FeCl₃-induced mesenteric arteriolar thrombosis in vivo. (a) QDTT inhibits FeCl₃-induced mesenteric arteriolar thrombosis (magnification × 40). (b) summary of the time of arterial occlusion. n = 10 for each group. The results were presented as mean ± SEM. **p < .01, and ****p < .0001 vs. The vehicle group. (c) QDTT, aspirin, or DMSO (control) was injected through the caudal vein 30 min before the experiment. n = 10 for each group. The results are presented as mean ± SEM. *p < .05 vs. The vehicle group. The yellow line represents the artery, and the red arrow represents the small artery thrombosis.

QDTT did not prolong tail bleeding time in mice

The experimental results suggested that QDTT has an inhibitory effect on thrombosis. The present study explored the effects of QDTT on tail bleeding time in mice. As shown in Figure 7c, aspirin significantly prolonged the tail bleeding time in mice compared with DMSO, but QDTT did not significantly prolong tail bleeding time in mice compared with DMSO or aspirin. Also, no adverse events occurred or led to protocol modifications.

Discussion

The present study investigated the effects of the CEACAM1 extracellular domains on platelet function using rhCEACAM1 and synthetic peptides. By combining with earlier observations,²⁵ the results showed that the extracellular A1 domain of CEACAM1 contains at least two sites involved in collagen-induced platelet activation. A1 domain-derived peptide QDTT reduced platelet aggregation, and the release was mediated by convulxin in vitro and effectively inhibited thrombosis in vivo. This effect was attributed to the ITAM-dependent and PI3K/AKT signaling pathways. These results confirm that CEACAM1’s A1 domain is involved in the interaction between collagen and platelets, indicating that QDTT might be a valuable marker for developing antiplatelet drugs.

Studies using rhCEACAM1 and synthetic peptides were conducted to investigate the role of the CEACAM1 extracellular domain in collagen-induced platelet aggregation.^9,25 Previous studies have shown that serum expression of soluble CEACAM1 is increased in various diseases closely related to thromboses, such as acute myocardial infarction and preeclampsia.^27,44 Moreover, Skubitz et al.^12,13 predicted and synthesized the extracellular peptide of CEACAM1 and screened seven peptides that could promote the activation and adhesion of neutrophils; these active peptides were all from the D1 domain, which mediates intercellular adhesion.^{11–13,15,45–47} The present study revealed that rhCEACAM1 and A1 domain-derived polypeptides (QDTT and QLSN) were similar, inhibiting collagen-induced platelet aggregation. The findings confirmed that Asp139, Thr141, Tyr142, Trp144, and Trp145 are critical for the biological activities of QDTT on platelet aggregation. Also, the present study showed that a range of peptides generated from the N-terminal domain had no significant effect on collagen-induced platelet aggregation, and the same A2 and B domain-derived peptides did not have the functions mentioned above. This suggests that A1 domain-derived peptides (QLSN and QDTT) have a more profound role in the collagen-induced platelet response. Yet, the precise mechanism through which the A1 domain regulates platelet activity remains elusive. Multiple IgC2 domains and signals emerging from the IgC2 domain have an increased affinity for homophilic binding,¹⁶ suggesting that the receptor’s capacity for intercellular binding is regulated by the number of existing IgC2 domains.⁴⁸

Next, we examined the regulatory mechanisms of QDTT involved in platelet activation. Our data suggested that the PI3K/Akt signaling and ITAM-dependent signaling pathways are associated with the fundamental molecular processes of the inhibitory effect of QDTT on platelet function. PI3K/Akt signaling pathway is pivotal in collagen-GPVI-induced platelet activation.^21,39,42 Human platelets express three ITAM-containing receptors: C-type lectin-like receptor 2 (CLEC-2), GPVI, and FcγRIIa,^49,50 all having important roles in platelet function and thrombosis. The activation of these receptors involves Src kinases that promote the phosphorylation of two tyrosine residues within ITAM, and the phosphorylated receptors provide docking sites for signaling proteins containing the SH2 structural domain, such as the tyrosine kinase Syk, to initiate a signaling cascade based on intracellular tyrosine kinases, ultimately leading to the phosphorylation and activation of PLCγ2 and a cellular response.⁵¹ Canobbio et al.⁵¹ showed that the stimulation of platelets by agonists acting on the G protein-coupled seven transmembrane structural domain receptor (GPCR) induces the tyrosine phosphorylation of FcγRIIa, and a common feature of agonists acting on the GPCR is their ability to cause the activation of tyrosine kinases (particularly the members of the Src family) and promote tyrosine phosphorylation of a wide range of intracellular substrates. In this study, we found that QDTT represses GPVI-mediated platelet activation via the Src-Syk-Akt-PLCγ2 signaling pathway. Thus, we concluded that QDTT may indirectly inhibit platelet aggregation via the GPCR. Nevertheless, QDTT does not necessarily inhibit FcγRIIa activation during platelet activation, although the GPVI and FcγRIIa receptors share an ITAM signaling pathway.

The inside-out signaling of αIIbβ3 on platelets can be initiated by a variety of soluble agonists, e.g., ADP, thrombin, etc., which bind to the GPCR, where ligand binding induces integrin aggregation and subsequently promotes “outside-in” signaling, initiating and amplifying signaling cascade responses to drive essential platelet functions, e.g., spreading, aggregation, clot retraction, and thrombus coagulation, and integrin αIIbβ3 is the final common pathway for platelet activation.^21,52 In this study, QDTT inhibited collagen- and convulxin-induced platelet aggregation at low concentrations, whereas it inhibited or did not inhibit platelet aggregation induced by other platelet agonists only at high concentrations. Therefore, it can be speculated that high concentrations of QDTT may inhibit platelet aggregation induced by other platelet agonists through integrin αIIbβ3, the final pathway of platelet activation. Hence, QDTT may inhibit platelet aggregation primarily through GPVI receptors and act by inhibiting integrin αIIbβ3. Still, more studies are needed to determine the exact signaling pathways involved in the effects of QDTT on platelets.

This study has a few limitations. First, the available experimental data cannot explain the polymerization state of QDTT. In future research, we plan to investigate whether QDTT may self-aggregate and clarify the effect of the aggregation state on platelet function. Second, various doses should be explored in vivo, and toxicity in vivo should be further investigated. Third, the exact mechanisms for the effects of QDTT should be explored and confirmed using knockdown, siRNAs, and overexpression of the identified genes. Multi-omics experiments could also be conducted.

In conclusion, the present study revealed that QDTT noticeably reduces convulxin-induced platelet activation. The PI3K/Akt and ITAM-dependent signaling pathways are associated with the fundamental molecular processes of QDTT’s inhibitory effects on platelet function. These results suggest that QDTT may be crucial against disorders caused by abnormal platelet activation and thrombosis.

Author contributions

Zhaohui Meng, Yujia Ye, Shengjie Chai, and Min Leng carried out the studies, participated in collecting data, and drafted the manuscript. Lihong Yang and Longcheng Ren performed the statistical analysis and participated in its design. Wen Wan, Huawei Wang, Longjun Li, and Chaozhong Li participated in the acquisition, analysis, or interpretation of data and drafted the manuscript. All authors read and approved the final manuscript.

Abbreviations

AA:	=	arachidonic acid
BSA:	=	bovine serum albumin
CEACAM1:	=	carcinoembryonic antigen-related cell adhesion molecule 1
DMSO:	=	dimethyl sulfoxide
GPVI:	=	Glycoprotein VI
GPIIb/IIIa:	=	Glycoprotein IIb/IIIa
IgC:	=	immunoglobulin C
IgG:	=	immunoglobulin G
ITAM:	=	immunoreceptor tyrosine-based activation motif
ITIMs:	=	immunoreceptor tyrosine-based inhibitory motifs
LDH:	=	lactate dehydrogenase
PBS:	=	phosphate-buffered saline
QDTT:	=	QDTTYLWW
QLSN:	=	QLSNGNRTLT
rhCEACAM1:	=	recombinant human carcinoembryonic antigen-related cell adhesion molecule 1

Ethical statement

The study was approved by the Experimental Animal Ethics Committee of Kunming Medical University (approval #kmmu2021737) and by the Institutional Review Board of the First Affiliated Hospital of Kunming Medical University (Kunming, China; approval #2020-L-17). All methods were carried out in accordance with relevant guidelines and regulations.

Data declaration

All data generated or analyzed during this study are included in this published article [and its supplementary information files].

Acknowledgments

We thank Dr. Peng Sang and Ph.D. Yirui Yin (College of Agriculture and Biological Science, Dali University, Dali, P. R. China.) for their insightful comments and suggestions in preparing this manuscript.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Supplementary material

Supplemental data for this article can be accessed online at https://doi.org/10.1080/09537104.2024.2308635.

Additional information

Funding

This study was supported by the National Natural Science Foundation of China [grant #82360074 and grant #81860074], the Yunnan Provincial Clinical Medical Center of Cardio-cerebral Vascular Diseases under Grant [grant #ZX2019-03-01], the Yunnan Fundamental Research Projects [grant # 202101AU070105], the Education Department of Yunnan Province Natural Science Fund [grant #2021J0245], the Priority Union Foundation of Yunnan Provincial Science, and the Technology Department and Kunming Medical University [grant #202101AY070001-118 and grant #202001AY070001-208].