About

Douglas B. Cines, M.D., is a faculty member in the Department of Pathology and Laboratory Medicine at the Perelman School of Medicine, University of Pennsylvania. He holds a background in history and pre-med from New York University (A.B., 1968) and earned his M.D. from New York University School of Medicine in 1972. His research expertise focuses on thrombosis, fibrinolysis, vascular biology, and immunohematology, with clinical expertise in coagulation and immune hematological disorders. Dr. Cines has contributed to the understanding of immune-mediated thrombotic conditions, including vaccine-induced immune thrombotic thrombocytopenia (VITT), antiphospholipid syndrome (APS), and heparin-induced thrombocytopenia (HIT). His work involves exploring the mechanisms of immunothrombosis and developing therapeutic strategies for related disorders.

Research topics

• Medicine
• Chemistry
• Immunology
• Internal medicine
• Molecular biology

Selected publications

• Structural and functional changes underlying activation of monocytes in heparin-induced thrombocytopenia

  Journal of Thrombosis and Haemostasis · 2025-02-09 · 1 citations

  articleOpen access
  PublisherOA PDFDOI

• Both platelet factor 4 (PF4) and b2-glycoprotein I (b2GPI) bind to von Willebrand factor (VWF) and inhibit ADAMTS13 cleavage: Mechanistic and clinical implications in thrombotic thrombocytopenic purpura (TTP)

  Blood · 2025-11-03

  articleOpen access

  Abstract Background: PF4 is a platelet-specific chemokine released in large amounts from activated platelets that binds to the A2 domain of VWF. β2-glycoprotein I (β2GPI) is an abundant circulating protein known to bind to the A1 domain of VWF. PF4 has also been shown to complex with b2GPI, an observation we confirmed in studies using dynamic light scattering (DLS). Whether PF4:VWF or b2GPI:VWF or PF4:b2GPI:VWF alter proteolytic cleavage of VWF by ADAMTS13 has not been reported. Aim: To test whether PF4 and b2GPI form complexes with VWF and whether these complexes inhibit ADAMTS13 proteolytic activity, potentially exacerbating the prothrombotic risk of TTP. Methods: Particle sizes were analyzed by DLS on a fixed scattering angle Zetasizer Nano-ZS system (Malvern Instruments Ltd). The Z-average size distribution (i.e., hydrodynamic diameters) of particles based on volume, were measured at RT with light backscattering of 173°. At least 25 repetitive measurements were made at the studied timepoints. Data analysis was performed using the Zetasizer software, version 7.03 (Malvern Inst Ltd). Recombinant human D'D3-A1-A2 VWF and PF4, and plasma-derived b2GPI were studied. ADAMTS13 activity was tested using plasma-derived, full-length VWF (Hematologic Technologies) denatured using 1.5 M urea in cleavage buffer to expose the A2 domain. PF4 (0-100 µg/mL) and β2GPI (0-200 µg/mL) were added to the VWF (Haematologic Technologies) alone or in combination, prior to ADAMTS13 (5nM final conc, rh-ADAMTS13, R&D) exposure. Cleavage products were analyzed by gel electrophoresis and western blot with a polyclonal anti-VWF-HRP antibody (Dako). In some studies, 100 µg/mL of the heparin-induced thrombocytopenia (HIT)-like anti-PF4 monoclonal antibody KKO or an isotype control TRA to further support role for PF4complexed to VWF on ADAMTS13 protease activity. A previously described hematoporphyrin photochemically injured human umbilical vein endothelial cell (HUVEC)-lined microfluidic channels system (PMC7146020) was perfused with washed human platelets with and without added ADAMTS13 (0.5 or 0.7µg/mL) with PF4, b2GPI and the above moAbs. VWF strand persistence on the HUVECs and platelet adhesion were visualized and quantified using confocal and epifluorescence microscopy. Results: DLS studies showed that both PF4 and b2GPI bind to the D'D3-A1-A2 fragment individually and as a triplex. PF4 inhibited ADAMTS13 activity in a dose-dependent manner, reducing cleavage by 38% at 10 µg/mL and 70% at 25 µg/mL. β2GPI alone reduced cleavage by 42% at 20 µg/mL and 91% at 50 µg/mL. When 20 µg/mL of b2GPI was added in addition to 10 µg/mL of PF4, cleavage was reduced by nearly 90%. The addition of KKO (100µg/mL), but not control TRA, fully prevented ADAMTS13-induced proteolysis whenever PF4 was present, independent of the presence of b2GPI. In the HUVEC-based microfluidic model, PF4 (25µg/mL) promoted VWF strand retention on injured endothelium, with further enhancement when combined with β2GPI (200µg/mL). VWF retention correlated with increased platelet adhesion. The addition of ADAMTS13 (0.7µg/mL) reduced both VWF retention and platelet binding, unless PF4 and/or β2GPI were present, especially, when both were present. The addition of KKO further enhanced PF4:VWF retention in the microfluidic channel as well as platelet binding. Conclusion: Our data suggest that PF4 and β2GPI each inhibit ADAMTS13 cleavage of VWF. PF4 reportedly binds directly to the A2 domain, whereas b2GPI reportedly binds to the A1 domain but still inhibits PF4 and ADAMTS13 cleavage within the A2 domain. Inhibition of ADAMTS13 is enhanced when both PF4 and b2GP1 are present, perhaps by forming generating tripartite complexes with VWF. At least one anti-PF4 monoclonal antibody KKO blocks proteolysis of PF4:VWF by ADAMTS13 as well. Our observations suggest that platelet activation with release of PF4 may contribute to the acute onset and unexplained clinical exacerbations in patients with TTP at varied and sometimes relatively stable levels of ADAMTS13 activity. Additionally, our data suggests that clinical assays measuring ADAMTS13 activity relying on short VWF A2 fragments in the absence of PF4 and b2GP1 may not fully characterize the risk of thrombosis. Inhibition of ADAMTS13 proteolysis of PF4:VWF by the HIT-like monoclonal antibody KKO also suggests that a similar process may contribute to the prothrombotic nature of HIT associated with increased levels of high molecular weight VWF multimers.

  PublisherOA PDFDOI

• HTRS2025.O6B.1 Abstract Travel Award Antiphospholipid syndrome (APS) is another platelet factor 4 (PF4) :neutrophil extracellular trap (NET)-dependent immunothrombotic disorder

  Research and Practice in Thrombosis and Haemostasis · 2025-11-01

  articleOpen access

  thrombin-mediated platelet activation similarly resulted in decreased intraluminal platelet accumulation during the post-hemostatic period (p=0.0009,WT vs Par4-/-) that was associated with hemostatic plug failure and rebleeding episodes.Finally, we found that increasing platelet sensitivity to thrombin via deletion of the PAR regulator GRK6 ameliorated excess bleeding in the context of FVIII-deficiency (p=0.0256,WT vs PF4Cre;Grk6fl/fl, both treated with FVIII inhibitor).Conclusions: Taken together, these results indicate that thrombin generated downstream of the intrinsic tenase is essential for sustained platelet responses over time after the initial cessation of bleeding.A critical role for the intrinsic tenase is to localize thrombin generation within the intraluminal component of hemostatic plugs to facilitate PARmediated platelet activation that is required to maintain hemostasis.These findings provide new insights into distinct roles of the extrinsic and intrinsic tenases in regulating the spatiotemporal distribution of thrombin generation during hemostasis.They further show that enhancing platelet sensitivity to thrombin may be a novel rebalancing approach to ameliorate bleeding in coagulation disorders.

  PublisherOA PDFDOI

• Antiphospholipid syndrome (APS) is a platelet factor 4 (PF4)-centric immunothrombotic disorder

  bioRxiv (Cold Spring Harbor Laboratory) · 2025-11-09

  preprintOpen access

  Abstract Antiphospholipid syndrome (APS) is an immunothrombotic disorder, frequently attributed to autoantibodies that bind β2-glycoprotein I (β2GPI). A study showed that the platelet-specific chemokine, platelet factor 4 (PF4), binds to β2GPI, enhancing recognition of β2GPI by APS antibodies. APS antibodies induce the release of neutrophil extracellular traps (NETs), webs of decondensed chromatin that bind both PF4 and β2GPI. We propose that PF4 bridges β2GPI to NETs (and other PF4-targeted polyanions), leading to the formation of prothrombotic PF4:β2GPI:NET immunotargets in APS. Dynamic light-scattering studies of isolated IgGs from four patients with triple-positive APS show formation of PF4:β2GPI:NET complexes that bind APS antibodies. NETs released in a microfluidic system bound β2GPI, but only in the presence of PF4, forming a multimolecular APS antigenic target. Whole blood infused through a photochemically-injured, endothelium-lined microfluidic channel formed platelet-, fibrin-, and complement-rich thrombi that bound APS antibody only in the presence of PF4. Thrombi were reduced in size if either ADAMTS13 or DNase1 was infused. In a murine APS model, wildtype and transgenic mice expressing platelet human PF4 ± FcγRIIA developed more intense neutrophil rolling along veins, and more extensive thrombus formation following laser injury to cremaster arterioles and venules, whereas mice lacking PF4 did not. Three antigenically distinct anti-hPF4 monoclonal antibodies blocked thrombosis in vitro , and neutrophil rolling and thrombosis in vivo . Our studies provide new insights into the basis of APS that has mechanistic parallels to other known PF4 immunothrombotic disorders and offer potential diagnostic and non-anticoagulant therapeutic strategies for clinical management. Key points PF4 enhances β2GPI binding to NETs and these complexes are central to APS immunothrombosis. Anti-PF4 monoclonal antibodies block APS immunothrombosis in microfluidic and murine studies.

  PublisherOA PDFDOI

• Neonatal Fc Receptor — Biology and Therapeutics

  New England Journal of Medicine · 2025-04-23 · 26 citations

  review
  PublisherDOI

• Destabilization of PF4-antigenic complexes in heparin-induced thrombocytopenia

  Blood · 2025-03-26 · 9 citations

  articleSenior author

  ABSTRACT: Heparin-induced thrombocytopenia (HIT) is initiated by antibodies that recognize large antigenic complexes composed of multiple molecules of cationic platelet factor 4 (PF4) and polyanions such as unfractionated heparin (UFH) that bind to each other primarily through electrostatic interactions. We asked whether the formation and stability of these HIT antigenic or ultralarge immune complexes (ULICs) would be inhibited by biocompatible synthetic polycationic molecules shown previously to dissociate UFH from antithrombin III and to inhibit polyphosphates. Members of this family of molecules, designated universal heparin reversal agents (UHRAs), inhibited formation and dissociated preformed ultralarge PF4-UFH (antigenic) complexes (ULCs), dissociated ULICs composed of the HIT-like monoclonal antibody KKO and ULCs, blocked binding of human HIT immunoglobulin G antibodies to PF4/heparin, binding of KKO to platelets, KKO-induced adhesion of platelets to activated human endothelium under flow, and microvascular thrombosis induced by KKO in a mouse model of HIT. These data suggest that UHRAs might provide a rationale intervention that acts at an early step in the pathogenesis of HIT to enhance the benefits and lessen the risks of nonheparin anticoagulants. Destabilization of immune complexes using polycationic inhibitors might also find a role in management of other polyanion PF4-antibody-mediated conditions, including vaccine-induced thrombocytopenia/thrombosis, postviral, and autoimmune HIT.

  PublisherDOI

• Platelet-activating antibodies to the chemokine, platelet factor 4, in cerebral malaria differ from those in heparin-induced thrombocytopenia and vaccine-induced thrombotic thrombocytopenia.

  Blood · 2025-11-03

  articleOpen access

  Abstract Background: Cerebral malaria (CM) is a severe complication of Plasmodium falciparum infection, primarily affecting children, with a mortality rate near 20% and no targeted treatment. Recent studies have identified platelet-activating autoantibodies in patients with CM, detected using commercial heparin-induced thrombocytopenia (HIT) assay (Immunocor), suggesting a potential autoimmune component in CM-associated thromboinflammation (PMID:38652559). We hypothesized, that in CM, platelet factor 4 (PF4) released from activated platelets serves as an immunogenic antigen that triggers the production of anti-PF4 antibodies contributing to thromboinflammation and disease severity. Aim: To characterize the epitope specificity of anti-PF4 antibodies in CM, and determine whether their properties mirror those associated with HIT or vaccine-induced immune thrombotic thrombocytopenia (VITT). Methods & Results: We studied 66 plasma samples from Malawi that were previously used in our study (IV, KS & KK; along with five CM samples and five control samples from asymptomatic community children in Uganda. The Malawi samples included 22 CM with positive (OD>0.4) and 22 CM samples negative (OD≤0.4) antibodies to human (h) PF4/polyanion complexes (hPF4/P) and 22 uncomplicated malaria (UM) samples also negative for hPF4/P antibodies, as determined using the Immucor PF4 ELISA. All five Uganda CM samples—and none of the control samples—tested positive for hPF4 antibodies. We found that anti-PF4 antibodies in CM patients react with PF4 alone independent of heparin. To assess the pathogenic potential, we tested whether CM plasma positive for hPF4 antibodies could activate murine platelets derived from PF4-deficient mice, with or without expression of human FcγRIIA. Platelet activation was dependent on the presence of both FcγRIIA and exogeneous hPF4 (p < 0.0001). To determine whether anti-hPF4 antibodies in CM patient samples recognize the previously defined HIT or VITT PF4 epitope, we performed competitive binding assays using increasing amounts of a HIT-like monoclonal antibody KKO (0-5 µg/ml), and a VITT-like monoclonal antibody 1E12 (0-5 µg/ml), respectively. Neither KKO nor 1E12 inhibited the binding of CM-derived anti-hPF4 antibodies. Increasing concentrations of 1E12 actually enhanced CM anti-PF4 IgG binding to hPF4 up to a threefold increase observed at ≥1 µg/ml of 1E12 compared to binding in the absence of 1E12 (p<0.01). Finally, we assessed cross-reactivity to related chemokines stored in platelet α-granules—namely mouse PF4 (mPF4) and both human and mouse neutrophil-activating peptide 2 (NAP2/CXCL7). Using ELISA, we found that CM samples positive for hPF4 antibodies exhibited significantly higher binding to mPF4 (p < 0.05), hNAP2 (p < 0.001), and mNAP2 (p < 0.05), compared to healthy controls. Conclusions: These findings identify a subset of patients with CM that had high-titer, FcγRIIA-dependent, platelet-activating anti-PF4 antibodies that do not map to the canonical HIT or VITT epitopes. Their cross-reactivity with conserved regions of related chemokines supports the hypothesis that the shared antigenic target lies within the conserved C-terminal heparin-binding domain of these chemokines.and suggests a distinct autoimmune response potentially driven by malaria-induced platelet activation. This highlights a novel mechanism of thromboinflammation in CM and points to PF4-targeted autoimmunity as a contributor to disease severity. Future studies will aim to map this antigenic site and evaluate the thrombogenic potential of CM antibodies in vivo as well as the ability of anti-PF4 blocking antibodies to ameliorate the severity of the CM.

  PublisherOA PDFDOI

• PAI-1 deficiency drives pulmonary vascular smooth muscle remodeling and pulmonary hypertension

  American Journal of Physiology-Lung Cellular and Molecular Physiology · 2024-06-11 · 8 citations

  articleOpen access

  This study identifies a novel role for the deficiency of plasminogen activator inhibitor (PAI)-1 and resultant unrestricted uPA activity in PASMC remodeling and PH in vitro and in vivo, provides novel mechanistic link from PAI-1 loss through uPA-induced Akt/mTOR and TGFβ-Smad3 upregulation to pulmonary vascular remodeling in PH, and suggests that inhibition of uPA to rebalance the uPA-PAI-1 tandem might provide a novel approach to complement current therapies used to mitigate this pulmonary vascular disease.

  PublisherOA PDFDOI
• RETRACTED

  Retraction Note: Release of IL-6 After Stroke Contributes to Impaired Cerebral Autoregulation and Hippocampal Neuronal Necrosis Through NMDA Receptor Activation and Upregulation of ET-1 and JNK

  Translational Stroke Research · 2024-06-05 · 1 citations

  articleSenior author
  PublisherOA PDFDOI

• Complement Activation As a Biomarker for Platelet Activating Antibodies in Heparin-Induced Thrombocytopenia (HIT)

  Blood · 2024-11-05 · 1 citations

  articleOpen access

  Background: IgG antibodies (Abs) to platelet factor 4 and heparin (PF4/H) commonly occur after heparin exposure, but cause life-threatening complications of HIT in only a subset of patients. To date, only platelet activation assays can reliably distinguish anti-PF4/heparin Abs that cause disease (HIT Abs) from those more likely to be asymptomatic (AAbs). In recent studies, we showed that complement activation by HIT Abs is essential for downstream FcgRIIA-mediated cellular activation. Study Objectives: As pathogenic HIT Abs also bind and cross-link platelet FcgRIIA, we asked if there was a correlation between complement and platelet activating properties of anti-PF4/heparin Abs. To address this question, we studied a clinically annotated cohort of patients with (HIT; n=8) or without HIT (AAb+, n=14) to compare clinical and laboratory features with serologic properties of Ab titers, platelet and complement activation. Methods: Patients testing positive for polyclonal anti-PF4/heparin Abs in the Duke Coagulation Laboratory were consented and included in the study based on sample availability. Plasma or whole blood from consented healthy donors was used as controls or as a source of complement in complement activation assays. Demographics, platelet counts, 4Ts and HEP score, % decline in platelet count, serologic data (anti-PF4/heparin polyclonal and IgG assay, and serotonin release assay or SRA results) were recorded. Endpoint titers were calculated using an in-house ELISA. Complement activation was measured, as previously described (Khandelwal, Blood 2021), using patient plasma (10% v/v) added to undiluted healthy donor plasma with buffer or PF4 (25 µg/ml) and heparin to generate in situ immune complexes (ICs). Complement-fixed ICs were measured using an immunocapture assay, using KKO, a monoclonal anti-PF4/heparin Ab, as the capture Ab and an anti-C3c (Quidel, San Diego) Ab for detection of captured immune complexes. MMP9 and IL-8 were measured after incubation of patient plasma (10% v/v) added to undiluted whole blood for 30 minutes (MMP9) or 6 hours (IL-8), with respective proteins detected using commercial immunoassays (R&D Systems, Minneapolis, MN). Results: As compared to AAb+ patients, HIT patients had significantly lower mean + standard deviation (SD) platelet counts (AAb+ vs. HIT: 67 +31 v39 +16; p<0.02), greater % drop in platelet counts (AAb+ vs. HIT: 65% + 14% v 79% +12%; p<0.03), higher 4Ts (AAb+ vs. HIT: 3.4 + 0.8 v 6.5 +1.1; p<0.0001), HEP scores (AAb+ vs. HIT: 4.4 +1.7 v 11.1 +2.5; p<0.0002), anti-PF4 polyclonal (AAb+ vs. HIT: 3.9 + 4.0 v 8.5 + 5.8; p<0.03) and IgG Ab levels (AAb+ vs. HIT: 0.7 +0.7 v 2.2 +0.9; p<0.008). Serologic properties also significantly differed between the two cohorts, with HIT patients showing median higher Ab titers (AAb+ vs. HIT: 900 v 7400; p<0.009), higher mean + SD% serotonin release or SRA+ (AAb+ vs. HIT: 5+7 v 91 +8; p<0.0001) and greater mean + SD complement activation, as determined by a C3c immunoassay (AAb+ vs. HIT: 0.8 +0.6 v 3.4 +0.6; p<0.0002). All HIT patients (8/8) showed strong complement activation, while 2/14 AAb+ patients showed increased complement activation over background levels of healthy donors. The extent of complement activation closely correlated with % serotonin release by anti-PF4/heparin Abs (r=0.754; p<0.001 by Spearman), as well as other clinical and laboratory parameters including, clinical 4Ts score (r = 0.64, p = 0.001), % drop in platelet count (r = 0.256, p = .023) and anti-PF4/heparin IgG levels (r = 0.444, p = .038). Complement activation also showed strong correlation with properties of cellular activation as gauged by MMP9 secretion (r = 0.680, p = 0.001) and release of IL-8 (r = 0.799, p = 0.0001) in whole blood. Conclusions: Our findings support that complement activation strongly correlates with other cellular activation endpoints, including platelet and monocyte/neutrophil activation and importantly distinguish anti-PF4/heparin Abs with and without disease potential. If these findings are confirmed in a larger cohort, complement activation can likely serve as “functional” biomarker for detecting pathogenic HIT Abs, perhaps obviating the need for a functional platelet activation assay.

  PublisherOA PDFDOI

Recent grants

• NIH Grant R01HL116916

  NIH · $1.2M · 2018

• NIH Grant R01HL050790

  NIH · $882k · 1999

• NIH Grant R21HL081864

  NIH · $426k · 2010

• NIH Grant R03TW001468

  NIH · $121k · 2005

• NIH Grant R01HL034044

  NIH · $392k · 1988

Frequent coauthors

• Mortimer Poncz

  Children's Hospital of Philadelphia

  342 shared

• Khalil Bdeir

  University of Pennsylvania

  198 shared

• Vladimir R. Muzykantov

  Translational Therapeutics (United States)

  195 shared

• Gowthami M. Arepally

  Duke Medical Center

  183 shared

• Abd Al‐Roof Higazi

  University of Pennsylvania

  182 shared

• Alice A. Kuo

  University of California, Los Angeles

  159 shared

• Juan-Carlos Murciano

  Instituto Ramón y Cajal de Investigación Sanitaria

  153 shared

• Lubica Rauova

  Children's Hospital of Philadelphia

  151 shared