Monocytes complexed to platelets differentiate into functionally deficient dendritic cells

Abstract In addition to their role in hemostasis, platelets store numerous immunoregulatory molecules such as CD40L, TGFβ, β2‐microglobulin, and IL‐1β and release them upon activation. Previous studies indicate that activated platelets form transient complexes with monocytes, especially in HIV infected individuals and induce a proinflammatory monocyte phenotype. Because monocytes can act as precursors of dendritic cells (DCs) during infection/inflammation as well as for generation of DC‐based vaccine therapies, we evaluated the impact of activated platelets on monocyte differentiation into DCs. We observed that in vitro cultured DCs derived from platelet‐monocyte complexes (PMCs) exhibit reduced levels of molecules critical to DC function (CD206, dendritic cell‐specific intercellular adhesion molecule‐3‐grabbing nonintegrin, CD80, CD86, CCR7) and reduced antigen uptake capacity. DCs derived from PMCs also showed reduced ability to activate naïve CD4+ and CD8+ T cells, and secrete IL‐12p70 in response to CD40L stimulation, resulting in decreased ability to promote type‐1 immune responses to HIV antigens. Our results indicate that formation of complexes with activated platelets can suppress the development of functional DCs from such monocytes. Disruption of PMCs in vivo via antiplatelet drugs such as Clopidogrel/Prasugrel or the application of platelet‐free monocytes for DCs generation in vitro, may be used to enhance immunization and augment the immune control of HIV.

Platelets are small anucleate cells, which store or produce multiple immune mediators including CD40L, TGF , 2 -microglobulin ( 2 M), platelet factor 4, and IL-1 , among others. Platelets contain the largest levels of TGF in the body and patients with immune thrombocytopenia have low levels of circulating TGF . [30][31][32] In addition, almost 99% of circulating CD40L and most of the 2 M are of platelet origin. [33][34][35] These molecules are stored in platelet -granules in their bioactive form, and are secreted through the open canalicular system immediately upon cellular activation. Platelets also store highly stable RNAs for molecules such as IL-1 and contain the complete machinery for splicing and translation. [36][37][38] Factors such as CD40L, TGF , and IL-1 , are known to modulate DC function. Being in close proximity to platelets in the form of PMCs might significantly increase the exposure of monocytes and their DC progeny to these biomolecules, thus making them prone to platelet-mediated functional alterations.
In order to test the impact of platelets on DC development, we established an in vitro coculture system in which DCs were derived from platelet complexed vs. noncomplexed monocytes. The resulting immature DCs were matured using two different cytokine cocktails, one which leads to generation of standard DCs 39 and one which induces their type 1 polarization. [40][41][42] Our results indicate that in vitro cultured DCs derived from PMCs exhibit reduced levels of several molecules critical to DC function (e.g., CD206, DC-SIGN, CD80, CD86, CCR7) as well as reduced antigen uptake capacity, deficiency in inducing the proliferation of naïve CD4 and CD8 T cells, and decreased IL-12 secretion resulting in reduced ability to induce responses to HIV antigens.

Ethics statement and demographics
All studies involving human samples were reviewed and approved by the University of Rochester Medical Center Research Subjects Review Board. Blood samples were acquired from adults after written informed consent carried out in accordance of the Declaration of Helsinki. Unless indicated, the donors were healthy individuals, not infected with HIV (n = 33). The mean age of donors was 40 (range . A total of 52.17% participants were males whereas 43.4% were females. In addition, key experiments (data points shown in red color in Figs. 4C, D, and 5) were also performed using blood obtained from HIV infected individuals (n = 4, 1 male and 3 females, age range [38][39][40][41][42][43][44][45][46][47][48][49][50][51][52]. Approximately, 40-60 ml of whole blood was collected in acid citrate dextrose (ACD) vacutainers and processed within 2 h of collection. The blood was incubated at room temperature with slow shaking until then.

Cell culture 2.3.1 Isolation of platelets
A total of 10 ml of whole blood was centrifuged at 250 ×g for 15 min, and platelet-rich plasma was collected. Following addition of PGI2 (1 g/ml of platelet-rich plasma) to maintain platelet quiescence, plateletrich plasma was centrifuged at 1000 ×g for 10 min to pellet the platelets. The platelet pellet was then washed and resuspended using Tyrode's salt solution (Sigma-Aldrich) supplemented with ACD anticoagulant and PGI 2 . Subsequently, washed platelets were centrifuged once again at 1000 ×g for 10 min, and the remaining purified platelet pellet was resuspended in Tyrode's salt solution without supplements.
The purity of isolated platelets was determined using a Sysmex KX-21N hematology analyzer (Sysmex Inc, Hyogo, Japan) and was found to be 99% pure. The platelets were resuspended at a concentration of 100 million/ml and activated using 1 g/ml of recombinant human soluble CD40L (henceforth termed as CD40L) at 37 • C for 15-20 min. 14 When indicated, platelets were also activated with thrombin (0.1 U/ml) for 20 min at 37 • C. After 20 min thrombin activity was inhibited by adding 0.1 U/ml of hirudin.

Isolation of monocytes
Monocytes were isolated from 30 ml of whole blood. PBMCs were isolated using Ficoll-Histopaque density gradient centrifugation. Monocytes were isolated using CD14 microbeads (Miltenyi Biotech) as per the manufacturer's instructions.

Isolation of T lymphocytes
Monocyte depleted PBMCs were used to isolate either naïve CD4 T cells or naïve CD8 T cells (using naïve CD4 and CD8 T isolation kits from Miltenyi Biotec). The cells were cryopreserved in RPMI containing 20% FBS and 10% DMSO using Mr. Frosty for overnight at −80 • C and were then stored in vapor phase liquid nitrogen freezer until required.

Platelet-monocyte coculture and differentiation to DCs
Freshly isolated monocytes were cocultured with activated platelets at the ratio of 1:10 and at 1 million monocytes/ml concentration (day 0).
Monocytes not mixed with platelets, monocytes treated with CD40L and monocytes treated with Thr/Hir were used as controls. Human recombinant IL-4 and GM-CSF (50 ng/ml) were added every alternate day for 5 d to obtain immature DCs. Immature DCs (day 5) were used for flow cytometric analysis of cell surface markers, antigen uptake assay using IgG coated Fluoresbrite microspheres, electron microscopy, and for generating mature DCs. Cells were matured using two different cytokine cocktails for 2 d. Standard cocktail, comprised IL-1 (25 ng/ml), IL-6 (2000IU/ml), TNF (100 ng/ml), and PGE 2 (2 g/ml), and is commonly used to generate standard DCs. 39 Type 1 cocktail comprised IL-1 (25 ng/ml), TNF (100 ng/ml), IFN (50 ng/ml), IFN (3000 U/ml), and Poly I:C (20 g/ml), which is known to drive DCs toward a type 1 phenotype. [40][41][42] Immature and mature DCs were washed three time with RPMI containing 10% FBS prior to performing all the assays, in order to remove the growth factors, unbound platelets and HIV.

Flow cytometry
Immature and mature DCs were processed for flow cytometric analysis on days 5 and 7, respectively. The cells were washed once with staining buffer (PBS with 0.1% BSA) and stained with titrated amounts of antibodies against CD61, CD14, CD16, CD206, HLADR, CD80, CD40, DC-SIGN, MHCI, CCR5, CCR7, CD86, LILRB2, CCR7, CLEC9A, and CD163. The cells were acquired using BD Accuri C6 flow cytometer BD Biosciences, San Jose, CA, USA. A total of 10,000 gates events were acquired on forward and side scatter plot at slow rate. The results were analyzed using Flow Jo software (version 10). For analysis, cells were first gated on forward and side scatter, followed by expression of CD61, a platelet marker (graphs from one representative experiment are shown in Supporting Information Fig. S1A). Data is shown as fold change of percentage of cells expressing respective markers and/or mean fluorescence intensity (MFI) as compared to nontreated (NT) cells.

Antigen uptake assay
Immature DCs (day 5) were incubated with IgG-coated fluorescent beads (1 m Fluoresbrite YG carboxylate microspheres) for 45 min at 37 • C to allow for phagocytosis. 43 The phagocytosis was stopped by adding ice cold PBS. Cells were placed on ice and stained with antibodies against CD61. Internalized beads were distinguished from externally bound beads by using PE-conjugated anti-human IgG. Cells were then washed and analyzed by flow cytometry as described earlier.

Electron microscopy 2.6.1 Scanning electron microscopy (SEM)
Immature DCs complexed to platelets (day 5) were allowed to adhere onto poly-L-lysine-coated coverslips for 2 h and were subsequently placed into 0.1 M sodium cacodylate-buffered 2.5% glutaraldehyde at 4 • C for overnight fixation. The cells on the cover glasses were postfixed using the same buffer in 1.0% osmium tetroxide and then transitioned through a graded series of ethanol to 100% (×2), then through a graded series of 100% ethanol/hexamethyldisilazane (HMDS) and finally into three changes of pure HMDS. The last change was allowed to evaporate off of the cover glasses overnight in a fume hood. The cover glasses were then mounted onto aluminum stubs and sputter coated with gold. Imaging was performed using a Zeiss Auriga field emission SEM (Carl Zeiss AG, Oberkochen, Germany).

Transmission electron microscopy (TEM)
Immature DCs complexed to platelets (day 5) were allowed to adhere to a two-chamber slide for 2 h, following which media was removed and immediately replaced with room temperature fixative composed

3D flow chamber migration assay
The assay was performed as described previously. 44  according to the manufacturer's protocol. Labelled noncomplexed and platelet complexed DCs for each individual's cocktail were combined together (2.5 × 10 5 cells/group) in a total volume of 20 ml. Cells were run through the 3D flow chamber device for 60 min using a flow rate of 0.9 mL/min (2.9 dyne/cm 2 ). The 3D flow chamber was disassembled.
Transmigrated cells were harvested from the lower compartment of the wells, fixed in 4% PFA, and analyzed by volumetric flow cytometry using Accuri C6 flow cytometer (BD Biosciences).

IL-12 and IL-10 ELISA
DCs were matured using the standard and type 1 cocktail as described

Generation of HIV-1 virus stock
HIV-1 NL4-3 ΔEnv reporter vector was obtained through the NIH AIDS Reagent Program (# 11100). 45 Plasmid expressing vesicular stomatitis G protein (VSV-G) was obtained from Dr. Dykes. 46 The virus stocks were generated as described previously with some modifications. 47,48 Briefly, human embryonic kidney cells, 293T, were cotransfected with 40 g each of NL4-3 vector and VSV-G vector using polyethyleneimine. Virus supernatant was harvested after 72 h and concentrated using PEG6000 as described and resuspended in PBS.
HIV-1 p24 protein quantitation was performed by ELISA according to manufacturer's instructions.

T cell stimulation
Autologous naïve CD4 or CD8 T cells were revived a day before setting up the assay and were incubated at 37 • C overnight in RPMI con- labelled T cells, stimulated with CD3/CD28 beads were used as con-trols. Supernatants from day 3 and day 7 post-DC-CD4 T cell cocultures were used to measure p24 levels by ELISA.

Statistical analysis
GraphPad Prism v7 was used for statistical analysis. The data are shown as scatter data plots with the error bar representing mean and SD. Normality analysis was done by Shapiro-Wilk normality test. All the data passed the normality test. It was further analyzed by repeated measures 1-way ANOVA or 1-way ANOVA followed by Tukey's multiple comparisons test. Paired t-test was used analyze migration assay.

Immature DCs derived from PMCs exhibit reduced antigen uptake ability
DCs employ multiple receptors to capture antigens 43  DCs that were derived from monocytes cocultured with platelets and expressed CD61, a platelet marker, are termed "Plt." DCs that were derived from monocytes cocultured with platelets, did not express CD61, but were still exposed to platelet-derived biomolecules are called "Plt Releasates." DCs derived from noncomplexed monocytes (not treated; NT), and CD40L treated monocytes (CD40L) were used as controls. Cells from each group were stained with respective monoclonal antibodies and analyzed using Accuri C6 flow cytometer and FlowJo software v10. The data is shown as raw data or fold change as compared to NT (HIV uninfected individuals, n = 6-9, 4 males and 5 females). Results were analyzed by repeated measures 1-way ANOVA followed by Tukey's multiple comparisons test using GraphPad prism v7. * indicates P < 0.05, ** indicates P < 0.01, and *** indicates P < 0.00. Percentages and mean fluorescence intensity (MFI) for expression of (

Platelet complexed DCs mature into phenotypically inferior DCs as compared to noncomplexed DCs
Antigenic stimuli trigger the process of DC maturation, which is associated with increased expression of MHC molecules and costimulatory molecules followed by migration to lymph nodes and terminal differentiation to initiate a T cell immune response. Various combinations of cytokines have been used in vitro to mimic this complex process in monocyte derived DCs (MoDCs). The most commonly used cocktail, comprises IL-1 , IL-6, TNF , and PGE 2 (standard cocktail). 39 In this experiment, the immature DCs were matured for an additional 48 h (day 7) using this cocktail and followed by phenotypic characterization (HIV uninfected individuals, n = 6-9 donors, 6 males and 3 females).
Results indicate that similar to immature DCs, about 40% of the mature DCs were found complexed to platelets on day 7 (Supporting Information Fig. S2A), whereas NT and CD40L treated DCs showed less than 2% complexes. DCs complexed to platelets exhibited reduced percentage of CD80 + ( Fig Recently, there have been speculations about the functional quality of DCs generated using the standard cocktail. For example, these DCs fail to secrete IL-12p70 upon stimulation with CD40L, which is a very important attribute that defines their ability to induce CTL immune responses. 40 We have previously shown that a cytokine cocktail con- Please also note that for all experiments henceforth, the "Plt" cells are a mixed population of platelet complexed DCs (approximately 40%) and cells exposed to platelet origin biomolecules (because they were cocultured in the same environment).

DCs derived from monocytes complexed to activated platelets exhibit decreased ability to secrete IL-12
In the next set of experiments we measured different readouts of DC function such as cytokine secretion and ability to stimulate

F I G U R E 2 Phenotypic characterization of dendritic cells (DCs) matured using standard cocktail.
(2A-2I) Immature DCs were obtained as described in methods section. Cells were matured using standard cocktail comprising IL-1 (25 ng/ml), IL-6 (2000 IU/ml), TNF (100 ng/ml), and PGE 2 (2 g/ml) for two more days followed by flow cytometric analysis for expression of various cell surface proteins (HIV uninfected individuals, n = 6-9, 6 males and 3 females). Results were analyzed by repeated measures 1-way ANOVA followed by Tukey's multiple comparisons test using GraphPad prism v7. * indicates P < 0.05, ** indicates P < 0.01, *** indicates P < 0.001, and **** indicates P < 0.0001. Percentages and mean fluorescence intensity (MFI) for expression of ( better able to stimulate CTLs. 49 We also measured IL-10 produced  Fig. S3F, G). These results indicate that upon contact with activated platelets, there is a reduction in the ability of DCs to secrete IL-12, both in HIV infected and uninfected individuals.  Figure S3H and I. DCs obtained using monocytes that were cocultured with not activated platelets also showed defects in T cell stimulation to some extent but it did not reach statistical significance. Cells obtained from HIV infected donors also showed a similar loss in T cell stimulation by DCs derived from PMCs.

PMC-DCs exhibit reduced capacity to induce HIV-specific responses
We also measured p24 levels in day 3 and day 7 post-DC-CD4 T cell coculture supernatants, in order to make sure that the VSV pseudotyped virus that was used as a source of antigen for the DCs, did not infect CD4 T cells resulting in reduced number of proliferating T cells.
The levels of p24 detected were very negligible (Supporting Information Fig. S3J) and we did not see any increase in p24 levels from day 3 to day 7.
These results indicate that platelets, particularly, activated platelets impair the differentiation and maturation of monocytes (that they are complexed with) into functional DCs, enhance their migration across the endothelium and that these cells are less efficient at stimulating the proliferation of naïve CD4 and CD8 T cells in response to HIV antigen. This phenomenon can be observed using cells isolated from both HIV infected and uninfected donors.

DISCUSSION
More than 35 yr after the identification of HIV and despite massive research efforts, RV144 is the only vaccine so far that has showed 31% reduction in new infections. 50  Studies have suggested that DCs from controllers are better able to capture antigens as well as cross-present these to Th1 CD4 and CD8 T cells. [63][64][65] In addition to the genetic predisposition of study participants (e.g., HLA alleles and genetic polymorphisms [66][67][68][69][70] ), the quality of DCs injected is an important factor that determines the outcomes of the clinical trials. This has been demonstrated by the close correlations between levels of IL-12 produced by vaccine candidates and better disease outcomes during the clinical trials in cancer patients by our group 41 and others. 71 To date there have been at least 16 early-phase clinical trials of DCbased HIV vaccines. [51][52][53][54][55][56][57][58][59][60] DCs were uniformly well tolerated, but the immune response was insufficient to control the virus and achieve a "functional cure," which indicates that although promising, there are factors, as of yet unknown, which affect the outcomes of these trials.
Although there is a high variability in terms of the immunization routes, immunogens, participant profiles, and number of DCs injected, all protocols used monocytes as the source of DCs. 72 In that respect, we and others have previously shown that HIV infection is associated with aberrant platelet activation [73][74][75][76][77] and results in increased levels of circulating PMCs. [12][13][14][15][16] Interaction of monocytes with activated platelets drove the monocytes toward a proinflammatory, promigratory phenotype that accentuated neuroinflammation. 14 Based on these observations, we hypothesized that platelet-induced functional defects of monocytes, dampen the immunogenicity of these cells even after their subsequent differentiation into DCs.
Indeed, we show that DCs which were complexed to platelets or were exposed to platelet-derived biomolecules expressed significantly expression, which is a marker of DCs proficient at antigen crosspresentation. This functional characteristic is especially critical for F I G U R E 5 T cell activation by mature dendritic cells (DCs) derived from platelet complexed and noncomplexed monocytes. Immature DCs obtained from monocytes that were either complexed with not-activated, CD40L activated or thrombin activated platelets or reagent controls were pulsed with vesicular stomatitis virus (VSV) pseudo-typed, single cycle replication recombinant HIV and exposed to the two maturation cocktails for 2 d. Naïve, autologous CD4 or CD8 T cells were labelled with CFSE and mixed at the ratio of 10:1 with DCs. The mixed culture was incubated for seven more days followed by measurement of CFSE lo , that is, proliferated T cells using volumetric flow analysis (HIV infected and uninfected individuals, n = 3-8, 4 males and 4 females). Data obtained from HIV infected subjects is denoted in red. Results were analyzed by 1-way ANOVA followed by Tukey's multiple comparisons test using GraphPad prism v7. * indicates P < 0.05, ** indicates P < 0.01, *** indicates P < 0.001, and **** indicates P < 0.0001. (5A) Flow plots from one representative experiment for CD8 T cells. Cumulatively, these findings demonstrate that interaction with activated platelets in the form of PMCs, results in increased migration of immunogenically inferior DCs. Thus, it is possible that these cells might induce T cell anergy, instead of generating viable antigen-specific immune responses against HIV.
During the generation of DCs for immunotherapy, specifically when monocytes are used as their source, contamination of the preparation with residual platelets has always been a concern 104,105 ; however,

DISCLOSURES
The authors declare no conflicts of interest.