AUTHORS
Amrita Sarkar, PhD, Diana V Hnin, N/A2, Isabelle Chen, N/A3, Preetha Rani MR, PhD4, Guohua Zhao, PhD5, Jenna Oberg6, Arin Cox7, Charles-Antoine Assenmacher, PhD8, Jason S. Knight, MD, PhD9, Thomas L. Ortel, MD, PhD10, Keith McCrae, MD11, Lubica Rauova, MD, PhD12, Mortimer Poncz, MD13
AFFILIATIONS
1 CHOP/UPENN, Philadelphia, Pennsylvania, United States
2 CHOP, Philadelphia, Pennsylvania, United States
3 CHOP, Philadelphia, Pennsylvania, United States
4 CHOP, Philadelphia, Pennsylvania, United States
5 CHOP, Philadelphia, Pennsylvania, United States
6 CHOP, Philadelphia, Pennsylvania, United States
7 University of Pennsylvenia, Philadelphia, United States
8 University of Pennsylvenia, Philadelphia, United States
9 University of Michigan, Ann Arbor, United States
10 Duke University, Durham, United States
11 Cleveland Clinic Lerner College of Medicine, Cleveland, United States
12 CHOP/UPENN, Philadelphia, Pennsylvania, United States
13 CHOP/UPENN, Philadelphia, Pennsylvania, United States
ABSTRACT
Background: Antiphospholipid syndrome (APS) is an autoimmune disorder broadly associated with two life-threatening complications: (i) arterial/venous thrombosis and (ii) maternal/fetal injury. We recently reported that β2-glycoprotein I (β2GPI)-specific APS antibodies selectively target PF4:β2GPI:neutrophil extracellular trap (NET) triplexes, such that PF4 is required for APS-associated thrombosis. While thrombosis is a component of OAPS, accumulating evidence suggests that placental injury is also driven by β2GPI and NET-mediated targeting of trophoblastic cells, resulting in inflammation, complement activation, placental dysfunction and adverse pregnancy outcomes.
Aim: We now wish to extend our findings that PF4 is essential for thrombotic complications in APS to the role of PF4 in OAPS using in vitro and in vivo models and to evaluate anti-PF4 blockade as a potential therapeutic strategy in OAPS.
Methods: We used an in vitro BioFlux model with trophoblast cells (HTR-8) to examine trophoblastic dysfunction during APS. The contribution of PF4:β2GPI:NET triplexes to trophoblast injury following addition of IgG from healthy controls or triple-positive APS patients was evaluated for cellular damage by lactate dehydrogenase (LDH) release, apoptosis by AnnexinV-PI, and downstream inflammatory changes and for PF4 and complement deposition. Protection against these changes by G4KKO, an Fc-modified anti-PF4 antibody, which we have shown prevents APS IgG binding to PF4:b2GPI:NET triplexes, was also assessed.
In vivo studies were performed using a passive immunization model of OAPS. Pregnant mice received either IgG isolated from three triple-positive APS patients (500 μg/mouse) or saline intravenously on gestational day 9 (GD9). Studies were conducted in littermate strains consisting of: (i) murine PF4 knockout (mPF4 KO) mice; (ii) mPF4 KO mice transgenically expressing human PF4 (hPF4); and (iii) hPF4 mice additionally expressing FcγRIIA (hPFm). Outcomes, included maternal weight gain, litter size, fetal growth, and survival to weaning. Placental pathology was evaluated in a blinded manner. Additional studies determined whether G4KKO could prevent OAPS-adverse outcomes.
Results: Only in the presence of both PF4 and β2GPI did APS IgG increase LDH release, indicating trophoblast injury with minimal We occurance of apoptosis, suggesting cellular dysfunction without overt trophoblastic cell death. Additionally, substantial PF4 and complement deposition on HTR-8 cells supported development of a proinflammatory microenvironment. G4KKO significantly attenuated these effects.
APS IgG induced a robust OAPS phenotype in vivo in hPF4 and hPFm mice that was not seen in the mPF4 KO mice or all three mice phenotypes injected with saline. The hPF4 and hPFm mice post-APS IgG exhibited no maternal weight gain during pregnancy with reduced litter sizes. Fetuses exhibited poor growth at GD17 and, in a separate cohort allowed to undergo natural birth, impaired survival during the first 7 days of life. Disease severity was greater in hPFm mice than in hPF4 mice, suggesting that FcγRIIA-mediated cellular activation contributes to OAPS pathogenesis. Histopathologic examination of placentas from APS-treated hPF4 and hPFm mice demonstrated extensive hemorrhage, fibrin deposition, and tissue necrosis. These findings were absent in APS-treated mPF4 KO mice, whose placentas were indistinguishable from those of saline-treated controls. Therapeutic intervention with G4KKO significantly reduced placental pathology and reversed maternal and fetal outcomes, maintaining normal physical parameters in both dams and offspring.
Conclusion: We previously showed that PF4 forms an antigenic PF4:β2GPI:NET triplex that is needed to observe APS thrombosis. We now demonstrate, for the first time, a critical role for PF4 in OAPS pathogenesis. In a microfluidic system, PF4 is required as well as b2GPI to lead to trophoblastic cell injury. Genetic absence of PF4 protected against APS-induced maternal and fetal injury, whereas expression of human PF4 restored disease susceptibility in both hPF4 and hPFm mice. Blockade of PF4 with the monoclonal antibody G4KKO prevented trophoblastic dysfunction, reduced placental injury, and improved pregnancy outcomes. These findings provide new mechanistic insight into OAPS pathogenesis and identify PF4 as a promising therapeutic target that may complement or potentially replace current non-specific anticoagulant approaches to prevent OAPS.
DISCLOSURES
Amrita Sarkar, PhD: No relevant conflicts of interest to declare.