Procoagulant in vitro effects of clinical cellular therapeutics in a severely injured trauma population

Abstract Clinical trials in trauma populations are exploring the use of clinical cellular therapeutics (CCTs) like human mesenchymal stromal cells (MSC) and mononuclear cells (MNC). Recent studies demonstrate a procoagulant effect of these CCTs related to their expression of tissue factor (TF). We sought to examine this relationship in blood from severely injured trauma patients and identify methods to reverse this procoagulant effect. Human MSCs from bone marrow, adipose, and amniotic tissues and freshly isolated bone marrow MNC samples were tested. TF expression and phenotype were quantified using flow cytometry. CCTs were mixed individually with trauma patients' whole blood, assayed with thromboelastography (TEG), and compared with healthy subjects mixed with the same cell sources. Heparin was added to samples at increasing concentrations until TEG parameters normalized. Clotting time or R time in TEG decreased relative to the TF expression of the CCT treatment in a logarithmic fashion for trauma patients and healthy subjects. Nonlinear regression curves were significantly different with healthy subjects demonstrating greater relative decreases in TEG clotting time. In vitro coadministration of heparin normalized the procoagulant effect and required dose escalation based on TF expression. TF expression in human MSC and MNC has a procoagulant effect in blood from trauma patients and healthy subjects. The procoagulant effect is lower in trauma patients possibly because their clotting time is already accelerated. The procoagulant effect due to MSC/MNC TF expression could be useful in the bleeding trauma patient; however, it may emerge as a safety release criterion due to thrombotic risk. The TF procoagulant effect is reversible with heparin.


| INTRODUCTION
Traumatic injury is the leading cause of death in people ages 1 to 44 years old in the United States. 1 Advances in resuscitation and hemorrhage control strategies have improved survival after injury. 2 However, after initial stabilization, there are few treatments for the morbidity associated with major trauma. For example, therapy for sequela of traumatic brain injury (TBI) or hemorrhagic shock remains largely supportive. 3,4 Clinical cellular therapeutics (CCTs) like human mesenchymal stem cells (MSCs) and mononuclear cells (MNCs) may treat these deadly and morbid conditions. [5][6][7] However, CCTs express tissue factor (TF) which is a potent activator of the coagulation cascade.
TF expression by CCTs has been linked with procoagulant effects in non-trauma patients. 8 In a previous broad survey of multiple tissue sources of CCTs from multiple donors, we demonstrated a causal relationship of CCT-associated TF with accelerated clot formation using thromboelastography (TEG) and accelerated thrombin production using a calibrated thrombogram. 9 Previous to that, Christy et al demonstrated similar findings in human adipose and bone marrow MSCs. 10 In addition, Moll et al examined placental decidual and bone marrow stem cells, showing a relationship between TF expression and increased blood clotting. 11,12 However, the effects of TF expressed by MSCs or MNCs in a population of severely injured trauma patients are unknown.
The purpose of this study is to investigate the procoagulant activity of CCTs in a severely injured trauma population and identify methods of its reversal. We hypothesize that CCTs will have a procoagulant effect in trauma patients similar to healthy uninjured subjects and that this effect is reversible with standard anticoagulation drugs. We use TEG to measure the procoagulant effect of CCTs in trauma patients and add heparin to reverse this effect.

| Human subjects
This study was conducted at the Memorial Hermann Hospital within the Texas Medical Center, Houston, Texas. Prior to the study, approval was obtained from the Institutional Review Board (IRB) for blood collection from trauma subjects (HSC-GEN-12-0059) and healthy controls (HSC-MS-10-0190). Patients meeting the highest level of trauma team activation were included in the study from August 2017 to March 2018. Patients were excluded from the study if they were younger than 16 years, pregnant, prisoners, enrolled in other studies, found to be pharmacologically anticoagulated by antiplatelet agents such as aspirin or clopidogrel, or declined to give informed consent. The patients from whom we could not obtain an admission blood draw were also excluded from the study. Informed consent was obtained from the patient or a legally authorized representative within 72 hours of admission. A waiver of the informed consent was obtained for those patients who were discharged or died within 24 hours. In the remaining cases in which the informed consent could not be obtained, the patients were excluded from the study and their blood samples were destroyed.

| Blood sample collection
Blood samples from trauma subjects were drawn with the initial sample within 5 minutes after arrival to the hospital. Two milliliters of blood were drawn through venipuncture into a Vacutainer tube containing 3.2% citrate and inverted to assure proper anticoagulation.
Samples obtained from healthy subjects were collected in a similar manner. Vital signs, injury data, Injury Severity Score (ISS), laboratory data, and demographic variables were also obtained from trauma subjects.

| CCT preparation
CCTs were sourced from four different human tissues including amniotic fluid derived MSCs, adipose-derived MSCs (ADP MSCs), bone marrow-derived MSCs, and bone marrow MNCs. All tissues were acquired either from commercial sources or under IRB-approved protocols.
Processing of the amniotic fluid-derived MSCs (AF MSCs) was carried out in an ISO Class 7 human cell production facility in compliance with current Good Manufacturing Practice (GMP) guidelines of the FDA. Amniotic fluid samples were collected through the approved IRB protocol HSC-MS-11-0593. All reagents used were of GMP grade, and risk analysis of the manufacturing process was performed as previously described. 13 In brief, amniotic fluid was centrifuged at 400g for 15 minutes and the pellet was resuspended in sterile-filtered complete TheraPEAK XenoFree chemically defined mesenchymal stromal cell growth medium (Lonza, Walkersville, Maryland) supplemented with 20% allogeneic pooled human AB serum (Valley Biomedical, Winchester, Pennsylvania) and 5 ng/mL basic fibroblast growth factor (CellGenix, Freiburg, Germany). Cells were plated on Corning (Corning, New York) CellBIND surface and incubated at 37 C in a 5% CO 2 and 95% relative humidity environment. Nonadherent

Significance statement
Stem cells are currently under investigation as a treatment for sequela of trauma like brain or lung injury. However, stem cells express tissue factor (TF) that causes rapid blood clotting. It is demonstrated that stem cells make blood from trauma patients, which clot faster. A potential antidote to this effect is heparin, a common and inexpensive blood thinner. It is believed that stem cells used in trauma studies should be risk-stratified based on their TF expression. cells were removed after 48 hours, and growth medium was changed (Fraction, Gibco) for 55 minutes at 37 C/5% CO 2 . For every 3 g of the tissue, 10 mL of digestion buffer was used. After incubation, the tubes were centrifuged at 400g for 15 minutes at room temperature.
The cell pellet was plated at a density of 9 g tissue/225 cm 2 Flasks (Thermo). Cells were expanded in 5% Platelet Lysate (Gulf coast blood bank) in alpha-MEM, 1000 U/mL heparin, and 10 μg/mL gentamicin.
Passage 0 was maintained at 37 C/5% CO 2 , fed every third day until confluence reached 70%. Upon reaching the desired confluence, the medium was discarded, the cultures were washed with PBS, and the adherent cells harvested with 0.25% trypsin/1 mM EDTA for Cell preparation for testing occurred on weekday mornings at 8 AM when at least one trauma patient had arrived overnight after 4 AM For experiments described below, cells were thawed, counted using a hemocytometer, and resuspended in PBS to a working concentration of 10 6 cells/mL. Cells were not used beyond 12 hours after preparation. There was a single donor for each type of cell, with 1 × 10 6 cells aliquoted to individual vials to allow testing on multiple days. CaCl 2 was added to the TEG cup first followed by recombinant TF dilutions or CCTs and blood was added last. The sample was mixed twice using a pipette before starting the assay. The final molarity of CaCl 2 was 0.01 M. The final concentration of CCTs was 10 5 cells/mL, which is similar to clinical trials administering CCTs which use a concentration of 1 to 10 × 10 6 cells/kg. 7,10 Control assays had an identical volume of PBS vehicle added to each sample. Each sample was run in duplicate, and values for TEG parameters were averaged. Each trauma patient and control sample were assayed with the four different CCTs and compared to their controls. TEG was performed within 6 hours of collection for trauma patients and within 2 hours of collection for healthy controls. These TEG data are separate from the rapid TEG performed in the hospital, presented in Table 2. 7AAD was added to exclude dead cells, and the solution was diluted to a total volume of 1 mL. Flow analysis was performed on a Gallios (Beckman Coulter) and analyzed using the Kaluza v. 1.5a analysis software (Beckman Coulter). Results presented are percent of expression relative to controls of unstained cells, isotype controls, and fluorescence minus one controls (Table 1). Flow cytometry was performed on CCT samples at 9 AM on weekday mornings after their preparation.

| Calculation of TF load
TF load in CCT samples was measured in a manner described previously. 9 Briefly, the product of the mean fluorescent intensity (MFI) and

| Data Analysis
Correlation between TF load and TEG for both trauma and healthy subjects was determined with Pearson's product moment correlations. Statistical significance was set to P < .05. Nonlinear regression curves were fitted to TF load and TEG data for trauma and control subjects. Based on variance of the source data, 95% confidence intervals were calculated for each curve at TF steps of 250. Analysis was performed using

| CCTs accelerate clot formation in trauma patients
Blood samples from the 36 trauma patients were assayed in standard TEG with the four different CCTs. TEG R time decreases as TF load increases with a Pearson's product-moment correlation r value of .84 and P < .0001 ( Figure 1A)

| CCT procoagulant effects in trauma patients vs healthy controls
Blood samples from 10 healthy subjects were also assayed in TEG with the four different CCTs. A Pearson's product-moment correlation had an r value of .72 and P < .0001 ( Figure 1B) decreases in TEG R time when treated with the same CCTs ( Figure 2).
Error bars represent 95% confidence intervals, and the cohorts demonstrate no overlap from TF load of 250 to 1700 relative units. Raw TEG data for trauma patients and healthy controls are presented in Table 3 as an average R time in minutes with and without each CCT treatment.

| Heparin reverses the procoagulant effect of CCTs
The procoagulant effect of CCTs is potentially reversible with stan-  Doses of heparin required to restore R time increased as TF load of the CCT increased. This study is clinically relevant because CCTs are being introduced as novel therapy for TBI 7 and acute respiratory distress syndrome. 14 We show that CCTs accelerate clot formation in vitro in a cohort of trauma patients. In addition, we demonstrate that heparin can reverse this procoagulant effect. The procoagulant effects of CCTs in trauma patients must be considered as a safety release criterion for their clinical use. Consideration should be given to choosing cell sources expressing less TF and cell expansion methods that minimize TF expression should also be explored. 8,15 The present study in trauma patients is distinct from our previous study in a healthy subject. In our previous study, blood samples from a single healthy donor were treated in vitro with 33 different CCT donors from six different tissue sources. 9 The focus of that study was to establish the causal relationship of CCT-associated TF and acceleration of blood clotting. In the current study, blood samples from patients with hypercoaguable TEG on admission have higher rates of DVT than those that do not (15.6% vs 8%, P = .039). 16 Selby et al studied the coagulation markers thrombin, prothrombin, and soluble fibrin at different time points in trauma subjects. They found that soluble fibrin was increased in those who developed DVT compared with those who did not. 17 This finding relates to TEG in that R time reflects initial formation of the fibrin clot. A limitation of this study was that CCTs are rarely given clinically in a single treatment, rather they are given in multiple doses over time.
In this study we only modeled a single dose of CCTs.

| CONCLUSIONS
Previous studies demonstrate an in vitro procoagulant response to CCTs related to their TF expression. These studies have been limited to blood donated by healthy subjects. Our study expands on previous work by confirming this effect in trauma patients. We demonstrate acceleration of clot formation in trauma patients as TF load of a CCT treatment increases. In addition, we demonstrate heparin as a potential reversal agent to this procoagulant effect.

DATA AVAILABILITY STATEMENT
The data that support the findings of this study are available from the corresponding author upon reasonable request.