Friday, August 19, 2011

Research of Eric F. Grabowski, MD, DSci

Eric F. Grabowski, MD, DSci

Eric F. Grabowski, MD, DSci

Dr. Grabowski is a practicing pediatric hematologist/oncologist  at MGHfC, who treats children with major deep venous thrombosis, bleeding disorders, and stroke.  He also is Chairman of the New England Retinoblastoma Group, and evaluates and,   when necessary, provides chemotherapy to children with retinoblastoma, the most common eye tumor of childhood.  As Director of the Cardiovascular Thrombosis Laboratory, however, he is studying the critical role of the tissue factor pathway of coagulation in the pathophysiology of the childhood hemolytic uremic syndrome, and the use of flowing blood to develop an assay by which von Willebrand disease might be more easily diagnosed.  Towards these ends his laboratory, based on Jackson 13, has special expertise in vascular endothelial cell culture and function, and in platelet-endothelial cell interactions in flowing blood.

Dr. Grabowski is immediate past Chair of the Scientific Committee on Thrombosis and Vascular Biology of the American Society of Hematology, and Co-Chair of the Pediatric Committee of the Transfusion Medicine-Hemostasis network.  He is a member of the Steering Committee of the Thrombosis and Hemostasis Summit of North America to be held in May 2012, and a member of the Board of Directors of the Hemophilia and Thrombosis Research Society of North America. 

Childhood hemolytic uremic syndrome (HUS). One project, supported by NHLBI (R01HL089332, “Tissue Factor, Flow, and Platelet Adhesion/Aggregation on Activated Endothelium”), seeks to establish a pathophysiology for the little understood childhood hemolytic uremic syndrome (HUS).  Background:  E. coli-related HUS is a thrombotic microangiopathy driven by Shiga toxin (Stx).  We hypothesize that increased expression of tissue factor (TF) on endothelium plays a major role in the pathophysiology of this disorder.  Methods:  We are using primary cultures (passage 3 or 4) of monolayers of human umbilical vein endothelial cells (HUVECs) to examine gene expression for TF and TF pathway inhibitor (TFPI) under several experimental conditions--1) control, 2) activation with 20 ng/ml  tumor necrosis factor (TNFa) alone x 22 hours, 3) activation with 10 pM Stx-1 alone for hours 18-22, 4) activation with the same concentrations of TNFa x 22 hours and Stx-1 for hours 18-22, and 5) activation with the same concentration of TNFa x 22 hours and 1.0 µg/ml lipopolysaccharide (LPS) with Stx-1 for hours 18-22.  Fold changes in mRNAs have been calculated using the comparative ct method, with a standard deviation in these fold changes derived from values for N different sets of pooled HUVECs.  In parallel with these studies and as a function of the above experimental conditions, we have performed a chromogenic assay for functional TF, measured as fmol of factor Xa generated from factor X per cm2 of monolayer per min, measured total cellular antigenic TF and TFPI, measured secretion rates of TFPI into supernatant, and assessed endothelial cell surface TF and TFPI antigens via immunofluorescence.  Results:  We find that TNFa alone leads to a fold increase in TF mRNA of 2.57 ± 0.84 (N=10), while Stx-1 alone induces no significant change (1.25 ± 0.32, 9).  However, the combination of TNFa and Stx-1 causes a fold change of 5.82 ± 2.64, 14, while TNFa, Stx-1, and LPS in combination induces a fold change of 15.50 ± 4.78 (14).  Stx-containing treatments also lead to modest yet significant fold reductions in TFPI mRNA (half-life of 48-72 hrs, vs 3 hrs for TF mRNA).  Factor Xa production (mean ± SD, N) for these same treatments is 0 ± 0 (56), 0.609 ± 0.720 (53), 1.064 ± 0.882 (46), 2.717 ± 1.824 (39) fmol/cm2/min.  Figure 1 shows these fold-changes, as well as those in control experiments with LPS, a mutant Stx-1 (Sm), and heat-inactivated Stx-1 (Sh).  In three experiments, total cellular TF antigen increases from 107 ± 42 (control) to 273 ± 70 with TNFa, to 638 ± 121 pg/ml with TNFa plus Stx-1.  In six experiments, total TFPI antigen for these same conditions is 313 ± 96, 313 ± 31, and 257 ± 41 pg/ml.  Secretion of full-length TFPI into supernatant is diminished in the presence of Stx-1.  Finally, immunofluorescence studies show changes in TF and TFPI antigen expression which parallel those for total celluar TF and TFPI antigen.  Conclusions:  Stx-1 augments functional TF on endothelium by a combination of two processes.  First, this toxin increases TF gene expression, yielding enhanced TF protein synthesis.  Second, Stx-1 impairs synthesis, secretion and cell-surface association of TFPI.  Blockade of the tissue factor pathway of coagulation may therefore afford a therapeutic approach to HUS, for which at present there is only supportive therapy.

Figure 1. Fold changes in TF mRNA; Figure 2. 3-D reconstruction of platelet adhesion/aggregation in flowing blood

Improved laboratory diagnosis of von Willebrand Disease (vWD).  A second project (supported by Behring Foundation) is the use of real-time epifluorescent digital video microscopy in flowing blood as a means to improve diagnostic accuracy in mild von Willebrand disease.  Background:  vWD is one of the most common bleeding disorders worldwide, yet a precise approach to the diagnosis of this condition remains elusive. One important reason is that vWD is a blood flow related disorder: a vW Factor-platelet GPIb binding defect in this condition exists under the high shear rate (>1000 sec-1 in whole blood; >3000 sec-1 in PRP) conditions of physiologic blood flow which exist in the arterioles of mucous membranes, from which most bleeding in vWD occurs.  Methods:  We studied 28 patients (mean 18.9 yrs ) with vWD, diagnosed according to the 2007 NHLBI clinical guidelines, and 26 healthy controls (mean 17.5 yrs).  Blood was collected into a plastic tube containing 4 U/ml FC dalteparin, 1.75 µg/ml of the Tab (anti-CD41) monoclonal antibody directed against platelet GPIIb, 1.0 µg/ml of an ALEXA 555-conjugated rabbit anti-mouse second antibody, and PBS (9 parts blood:1 part PBS). Within 30-90 min, the blood was then withdrawn at 670 and 1330 sec-1 through a special flow chamber designed by Dr. Grabowski which allows for blood exposure to 150 µm glass cover clips precoated with microfibrillar collagen.  Using epifluorescence digital videomicroscopy, we imaged platelets interacting in real time with the collagen substrate, and quantified with MetaMorph software (Universal Imaging) the percent area(PA) covered by adherent platelet aggregates, and the number of platelets/mm2 (total volume, TV) of these aggregates within a 435 µm x 580 µm field of view.  Resolution was better than 1 µm. Clinical bleeding score (ICTH) and ristocetin cofactor activity were also recorded.  Results:  At 670 sec-1 PA and TV were similar for patients and controls, but at 1330 sec-1 after 1 min PA was 9.32 ± 4.21 (mean ± SD) for patients, a value significantly (p = 0.00072) lower than the 12.8 ± 3.39 for controls.  TV was (1.43 ± 0.91)x104 for patients, a value also significantly (p = 0.00056) lower than the (2.22 ± 0.77)x104 for controls.  Eight patients (29%) had a PA below the lower 95th percentile, and 11 (39%) had a TV below the lowest control value.  Figure 2 shows platelet adhesion/aggregation in flowing whole blood for a normal control at 670 sec-1.  Flow direction is from top-to-bottom, with the varying local aggregate thickness represented by pseudo-color. Conclusions:  With considerations of bleeding history set aside, the present approach detected more patients than did the same-day ristocetin cofactor activities, of which only four (14%) were <30%, and allowed direct visualization of forming platelet thrombi.  The present method, moreover, allows real-time imaging of platelet aggregate structure in flowing blood.


1.  Grabowski EF, Carter CC, Tsukurov O, Conroy N, Abbott WM and Orkin RW:  A comparison of the tissue factor pathway in human umbilical vein and adult saphenous vein endothelial cells:  Implications for newborn hemostasis and for laboratory models of endothelial cell function.  Ped Res 1999;46:742-747.

2.  Grabowski EF:  The hemolytic-uremic syndrome—toxin, thrombin, and thrombosis. [Editorial]  NEJM 2002; 346(1):58-61.

3.  E.F. Grabowski, M. Cheung, E.E. Biliouris, R.I. Kushak, and E. Van Cott.  Pattern and reduced volume of platelet aggregation in a high shear rate blood flow chamber allows for the sensitive and specific diagnosis of vWD.  Presented at the 51st Annual Meeting of the American Society of Hematology, New Orleans, 2009.

Back to Top