If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Corresponding author at: Medical University of Vienna, Institute of Vascular Biology and Thrombosis Research, Schwarzspanierstrasse 17, 1090 Vienna, Austria.
Platelets are primarily recognized for their role in hemostasis, but also regulate immune responses by interacting with leukocytes. Their highly sensitive nature enables platelets to rapidly respond to micro-environmental changes, which is crucial under physiological condition but can jeopardize in vitro analyses. Thus, we tested how platelet count and changes in pH and temperatures, which are commonly experienced during inflammation and infection but also affected by ex vivo analyses, influence platelet-leukocyte interaction and immunomodulation.
Reducing platelet count by up to 90 % slightly decreased platelet activation and platelet-leukocyte aggregate formation, but did not affect CD11b activation nor CD62L shedding of monocytes or neutrophils. Acidosis (pH 6.9) slightly elevated platelet degranulation and binding to innate leukocytes, though pH changes did not modulate leukocyte activation. While platelet responsiveness was higher at room temperature than at 37 °C, incubation temperature did not affect platelet-leukocyte aggregate formation. In contrast, platelet-mediated CD11b activation and CD62L expression increased with temperature.
Our data thus demonstrate the importance of standardized protocols for sample preparation and assay procedure to obtain comparable data. Further, unspecific physiologic responses such as thrombocytopenia, acidosis or temperature changes may contribute to platelet dysfunction and altered platelet-mediated immunomodulation in inflammatory and infectious disease.
Platelets govern both hemostatic/thrombotic and inflammatory processes. Due to their sheer number, wide receptor repertoire and their versatile and highly sensitive nature, elevated platelet activation is indicative of various pathologies of vascular, inflammatory and infectious origin [
]. Activated platelets readily bind leukocytes, in particular monocytes/macrophages and neutrophils, and these platelet-leukocyte aggregates (PLA) contribute to host responses against invading pathogens, including cytokine release, oxidative burst, neutrophil extracellular trap formation and phagocytosis [
Platelet activation and dysfunction in infectious diseases may be caused by pathogen-specific molecular mechanisms, including recognition by specific receptors [
]. However, platelet function may also be influenced by general physiologic responses during infection such as thrombocytopenia, fever and acidosis.
Several studies have demonstrated a correlation of platelet count and aggregation, though threshold levels for count-dependent aggregation deficits depend on the specific method, ranging from <150,000/μL (impedance-based) to <10,000/μL (flow cytometry-based) [
Platelet function may also be influenced by other inflammatory host responses such as fever or acidosis, however in vitro findings are partly contradictory. Overall, elevated temperature renders platelets hypo-responsive with impaired degranulation, adhesion and aggregation relative to 37 °C [
]. In contrast, hypothermic conditions ranging from 34 °C to room temperature are associated with augmented GPIIb/IIIa activation, shape change and adhesion in vitro as well as accelerated thrombosis in vivo [
]. Of note, platelet analyses are frequently performed at room temperature or 37 °C depending on the method and individual laboratory routine, which could impact results.
Although pH levels are strictly regulated in blood, certain diseases including infections can result in local or systemic pH disturbances. Alkalosis is associated with elevated capacitative calcium entry and enhanced platelet aggregation [
]. Contrarily, acidosis impairs calcium flux, spreading, adhesion and aggregation of washed platelets in vitro, though its impact on platelet degranulation remains unclear [
]. Further, anticoagulation with acid citrate-dextrose (ACD) instead of citrate, which yields lower pH, better preserves platelet function and microstructure, thus facilitating platelet aggregation [
Although several controlled in vitro studies report on the modulation of pro-thrombotic platelet functions by cell count, temperature and pH, little is known about the impact of these variables on platelet-mediated immunomodulation. Acidosis is associated with augmented platelet binding to isolated neutrophils [
]. Intriguingly, PLA formation was also found to be upregulated upon both hyper- and hypothermia, irrespective of their differential effects on pro-thrombotic platelet functions [
Understanding the impact of cell count, temperature and pH on platelets and platelet-leukocyte interactions is pivotal to I) discriminate between disease-specific and unspecific modulation of platelet function in inflammation and infection and to II) optimize laboratory protocols for evaluating platelet function and platelet-leukocyte interplay in patients and in vitro experiments. Protocols employed by different groups and institutions often vary regarding e.g. adjustment of platelet count, incubation temperatures or buffer pH, which likely affects obtained data and complicates comparability of studies.
Therefore, we investigated the influence of platelet count, blood pH and temperature on platelet responsiveness, PLA formation and platelet-mediated leukocyte activation in vitro. We found that thrombocytopenia and pH influenced platelet activation and PLA formation, but not innate leukocyte activation. In contrast, incubation temperature affected both platelet responsiveness and platelet-mediated activation of neutrophils and monocytes.
2. Materials and methods
2.1 Blood collection and ethics approval
Blood was drawn from healthy volunteers devoid of any medication for 14 days from the antecubital vein using 21G needles and anticoagulated with 3.8 % sodium citrate or K3EDTA (9 mL or 3.5 mL tubes, all Greiner bio-one). The study was approved by the Ethics Board of the Medical University of Vienna (EK1548/2020) and conformed to institutional guidelines and the Declaration of Helsinki. Volunteers gave written informed consent before blood draw.
2.2 Isolation of platelet-rich plasma (PRP) and buffy coat
Citrated blood was centrifuged for 15 min (3.5 mL tube) or 20 min (9 mL tube) at 120 ×g without brake. The upper PRP phase was collected and platelets counted using an automated hematology analyzer (Sysmex). The leukocyte-enriched buffy coat interphase between PRP and erythrocytes was collected separately.
2.3 Isolation of platelet-poor plasma (PPP)
PPP was generated by centrifuging citrated blood at 1000 ×g for 10 min.
2.4 Isolation of leukocytes
EDTA-anticoagulated blood was layered over a density gradient consisting of 4 mL Histopaque 1119 overlaid with 2 mL Histopaque 1077 (Sigma-Aldrich) and centrifuged for 30 min at 700 ×g (acceleration: 3, brake: 3). Layers containing monocytes and neutrophils were pooled and washed twice with 50 mL PBS (400 ×g, 10 min). Leukocytes were resuspended in Tyrode-HEPES buffer (140 mM NaCl, 3 mM KCl, 1 mM MgCl2, 16.63 mM NaHCO3, 10 mM HEPES, pH 7.4) and counted using an automated hematology analyzer (Sysmex).
2.5 Adjustment of platelet count
Thrombocytopenic conditions were mimicked by mixing isolated PRP, PPP and leukocytes at controlled cell counts. Final platelet count was adjusted to 200,000 platelets/μL (100 %), 100,000 platelets/μL (50 %) or 20,000 platelets/μL (10 %), while keeping leukocytes constant at 10,000/μL.
2.6 Adjustment of pH value
Citrated whole blood or PRP (pH 7.2) was adjusted to pH 6.9 with 0.1 M HCl before stimulation.
2.7 Adjustment of temperature
All incubation steps were carried out at either 22 °C or 37 °C. Platelet activation, PLA formation and leukocyte activation were determined in whole blood, monocyte subsets were determined in buffy coat.
2.8 Platelet activation
Platelets in the presence of other blood components were activated for 15 min with 3 μM or 6 μM of either thrombin receptor activating peptide 6 (TRAP-6) or adenosine di-phosphate (ADP). For detection of platelet activation, PLA formation and leukocyte activation, samples were first stained for 20 min with labelled antibodies before cells were fixed and erythrocytes lysed using fix & lyse reagent (Invitrogen). In order to exclude potential impact of temperature or pH on antibody binding, activated samples were alternatively first fixed in 1 % formaldehyde (10 min) before erythrocytes were lysed using lysis buffer (150 mM NH4Cl, 10 mM KHCO3, 0.1 mM Na2EDTA, pH 7.4). Cells were pelleted (600 ×g, 5 min), resuspended in PBS and stained for 20 min using labelled antibodies. For detection of platelet effects on monocyte subsets, activated samples were also fixed and erythrocytes lysed before antibody labelling as described above. All samples were measured on a Cytoflex S flow cytometer and analyzed with CytExpert 2.4 software (both Beckman Coulter).
2.9 Antibodies
Platelet activation was measured using the following antibodies: anti-CD42b-PerCP, anti-CD62P-BV605, anti-CD63-PB, anti-CD40L-PE (BioLegend) and PAC1-FITC recognizing activated GPIIb/IIIa (BD Bioscience). Leukocytes were analyzed using anti-CD42b-PerCP, anti-CD16-PC7, anti-CD66b-PB, anti-CD11b activated-FITC, anti-CD62L-BV650 (BioLegend) and anti-CD14-APC (BD Biosciences).
2.10 Flow cytometric gating strategies
2.10.1 Platelet panel
Single cells were analyzed for CD42b, identifying platelets. Platelets were further analyzed regarding the expression of activated GPIIb/IIIa (PAC1 antibody binding), CD62P, CD40L and CD63, which were quantified as percent of platelets (Sup. Fig. 1).
2.10.2 Leukocyte panel
Single cells were discriminated based on the expression of either CD66b (neutrophils) or CD14 (monocytes). Both cell types were further analyzed regarding the formation of platelet-neutrophil/monocyte aggregates by identifying CD42b-positive neutrophils/monocytes as well as regarding CD11b activation and CD62L shedding. PLA and CD11b activation were analyzed as percent of neutrophils/monocytes, CD62L was analyzed as mean fluorescence intensity (MFI) (Sup. Fig. 2).
2.10.3 Monocyte subset panel
Single events were discriminated based on the expression of CD42b and characteristic size (FSC) and granularity (SSC). Monocytes were identified as CD66b-negative cells and further subdivided by the expression of CD14 and CD16 into classical (CD14++CD16−), intermediate (CD14++CD16+) and non-classical (CD14+CD16++) subsets. The fraction of individual monocyte subsets was quantified as percent of all monocytes (Sup. Fig. 3).
2.11 Release of CXCL4
PRP was mixed with PPP to final platelet counts of 200,000/μL (100 %), 100,000/μL (50 %) or 20,000/μL (10 %). Alternatively, PRP was adjusted to pH 6.9 with 0.1 M HCl. Platelets were stimulated with 6 μM TRAP-6 for 15 min at 22 °C or 37 °C before pelleting (3000 ×g, 90 s) in the presence of 0.3 μg/mL prostacyclin. Supernatant was collected and analyzed for CXCL4 (platelet factor 4) by ELISA according to the manufacturer's instructions (R&D Systems).
2.12 Statistical analysis
Statistical analyses were performed with GraphPad Prism 9.1.1. Samples were measured in duplicates and mean values were used for statistical analysis. All data sets were analyzed for Gaussian distribution using Shapiro-Wilk test. Differences between platelet counts, pH values and temperatures were assessed using matched two-way ANOVA with Geissner-Greenhouse correction and a p-value of <0.05 was considered nominally significant. Data are depicted in boxplots with whiskers showing median, quartiles, minimum and maximum values.
We investigated the impact of platelet count on platelet activation and reactivity. Therefore, isolated platelets, plasma and leukocytes were mixed to obtain platelet counts of 200,000 platelets/μL (100 %), 100,000 platelets/μL (50 %) or 20,000 platelets/μL (10 %) before stimulation with 3 μM or 6 μM agonist to induce medium or strong activation (Fig. 1A ).
Fig. 1Thrombocytopenia impairs platelet degranulation. (A) Scheme of experiment: Platelet-rich plasma was mixed with plasma and leukocytes to 200,000 (100 %), 100,000 (50 %) and 20,000 (10 %) platelets/μL while keeping leukocytes constant at 10,000 cells/μL. Platelet activation in response to 3 μM or 6 μM thrombin activating peptide 6 (TRAP-6), adenosine di-phosphate (ADP) or control (PBS) was measured by flow cytometry. (B-G) Platelet activation upon TRAP-6 stimulation was measured by expression of degranulation markers (B, C) CD62P, (D) CD40L and (E) CD63 as well as by (F, G) GPIIb/IIIa activation. Representative flow cytometry plots with mean values shown in B and F. (H, I) Platelet activation upon ADP stimulation was evaluated by (H) CD62P expression and (I) GPIIb/IIIa activation. N = 6. *p < 0.05, **p < 0.01.
Degranulation processes upon 6 μM TRAP-6 measured by CD62P, CD40L and CD63 (Fig. 1B-E) were significantly reduced at 10 % platelets, and CD63 and CD40L expression were also reduced at 50 % platelets. However, platelet count did not affect degranulation upon medium stimulation. In contrast, no differences in activated GPIIb/IIIa were detected between 100 %, 50 % and 10 % platelet counts, regardless of TRAP-6 concentration (Fig. 1F-G).
In response to ADP stimulation, CD62P expression differed only between 100 % and 10 % platelet count after stimulation with medium ADP concentration (Fig. 1H) while GPIIb/IIIa activation was again unaffected by platelet count (Fig. 1I).
These results indicate that already a mild reduction of platelets (50 %) as well as thrombocytopenic conditions (10 % platelet count) result in impaired degranulation processes while activation of GPIIb/IIIa is not affected by thrombocytopenia.
3.2 Platelet degranulation increases upon acidification
Next, we investigated how pH affects platelet activation and reactivity by adjusting citrated whole blood (pH 7.2) to acidic pH (pH 6.9) (Fig. 2A ).
Fig. 2Platelet degranulation increases upon acidification. (A) Scheme of experiment: Citrate-anticoagulated whole blood (pH 7.2) or platelet-rich plasma was adjusted with 0.1 N HCl to pH 6.9. Platelet activation in response to 3 μM or 6 μM thrombin activating peptide 6 (TRAP-6), adenosine di-phosphate (ADP) or control (PBS) was measured by flow cytometry or ELISA of platelet supernatant. (B-H) Platelet activation upon TRAP-6 stimulation was measured by expression of degranulation markers (B, C) CD62P, (D) CD40L and (E) CD63 or (F) release of CXCL4 as well as by (G, H) GPIIb/IIIa activation. Representative flow cytometry plots with mean values shown in B and G. (I, J) Platelet activation upon ADP stimulation was evaluated by (I) CD62P expression and (J) GPIIb/IIIa activation. N = 6. *p < 0.05, **p < 0.01.
Platelet degranulation upon TRAP-6 stimulation was slightly higher at pH 6.9 relative to pH 7.2, as demonstrated by elevated expression of CD62P, CD40L and CD63, as well as augmented release of α-granule-derived CXCL4 (platelet factor 4) (Fig. 2B-F). In contrast, acidosis did not affect TRAP-6-induced activation of GPIIb/IIIa (Fig. 2G-H). Further, ADP-induced CD62P expression and GPIIb/IIIa activation were also independent of pH (Fig. 2I-J).
As acidic pH could potentially affect antigen/antibody interaction and thus influence obtained data, we also determined the effect of pH on platelet degranulation in a different methodological approach by fixing cells before performing antibody labelling at homogenous conditions. Again, platelets exhibited enhanced TRAP-6-induced expression of degranulation markers CD62P, CD40L and CD63 in acidic microenvironment relative to normal citrated blood (Sup. Fig. 4A).
These results demonstrate that platelet degranulation but not GPIIb/IIIa activation is increased upon acidification.
3.3 Ambient temperature increases degranulation and GPIIb/IIIa activation
We also investigated the impact of incubation temperatures commonly used in laboratory assays on platelet activation and reactivity by stimulating whole blood at 22 °C or 37 °C (Fig. 3A ).
Fig. 3Ambient temperature increases degranulation and GPIIb/IIIa activation. (A) Scheme of experiment: Whole blood or platelet-rich plasma was aliquoted, and the experiment was performed at 22 °C or 37 °C. Platelet activation in response to 3 μM or 6 μM thrombin activating peptide 6 (TRAP-6), adenosine di-phosphate (ADP) or control (PBS) was measured by flow cytometry or ELISA of platelet supernatant. (B-H) Platelet activation upon TRAP-6 stimulation was measured by expression of degranulation markers (B, C) CD62P, (D) CD40L and (E) CD63 or (F) release of CXCL4 as well as by (G, H) GPIIb/IIIa. Representative flow cytometry plots with mean values shown in B and G. (I, J) Platelet activation upon ADP stimulation was evaluated by (I) CD62P expression and (J) GPIIb/IIIa activation. N = 6. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001.
Platelet degranulation in response to TRAP-6 stimulation was slightly elevated at 22 °C relative to 37 °C, with small but significant differences in CD62P and CD40L expression upon strong stimulation (Fig. 3B-E). In line, ambient temperature also facilitated TRAP-6-induced release of CXCL4 (Fig. 3F). In contrast, activation of GPIIb/IIIa was strongly increased at 22 °C compared to 37 °C at both basal condition and upon TRAP-6 stimulation (Fig. 3G-H).
Further, both CD62P expression and activation of GPIIb/IIIa in response to ADP stimulation were also strongly increased at 22 °C relative to 37 °C, though only activation of GPIIb/IIIa showed basal differences while none were observed for CD62P (Fig. 3I-J). Of note, the observed results were independent of temperature during antibody labelling (Sup. Fig. 4B).
Our data show that hypothermic condition (22 °C) increases activation of GPIIb/IIIa as well as platelet degranulation relative to 37 °C.
3.4 PLA formation depends on platelet count and pH but is unaffected by hypothermia
As our results show that platelet activation is affected by platelet count, temperature and pH, we investigated the impact of these parameters on the formation of platelet-monocyte (PMoA) and platelet-neutrophil aggregates (PNA).
PMoA formation was markedly reduced at 10 % platelets upon TRAP-6 stimulation (Fig. 4A ) while no effects upon ADP stimulation were observed (Sup. Fig. 5A). PNA formation upon TRAP-6 stimulation was also significantly reduced at 10 % platelets (Fig. 4B), but also ADP-induced PNA formation was dependent on platelet count at medium stimulation (Sup. Fig. 5B).
Fig. 4PLA formation depends on platelet count and pH but is unaffected by hypothermia. Experiments with mixed blood components or whole blood were performed at different (A-B) platelet count, (C-D) pH or (E-F) temperature as described in Fig. 1, Fig. 2, Fig. 3. PLA formation with (A, C, E) monocytes and (B, D, F) neutrophils in response to 3 μM or 6 μM thrombin activating peptide 6 (TRAP-6) or control (PBS) was measured by flow cytometry and quantified as the percentage of platelet positive leukocyte subtypes. Representative flow cytometry plots of TRAP-6-stimulated samples with mean values are given. N = 6. *p < 0.05, **p < 0.01, ***p < 0.001.
In accordance with the effects of pH on platelet degranulation, acidification induced a significant increase in PNA formation compared to pH 7.2 at strong TRAP-6 stimulation, though PMoA formation did not reach statistical significance (Fig. 4C-D). Of note, the effect of acidosis on PMoA and PNA formation became, respectively remained, significant when cells were fixed before staining at homogenous conditions (Sup. Fig. 4C). However, no pH-dependent differences in PMoA or PNA formation were observed in response to ADP stimulation (Sup. Fig. 5C-D).
In line with the minor impact of hypothermia on platelet degranulation in response to TRAP-6 (Fig. 3), neither PMoA nor PNA formation was affected by temperature (Fig. 4E-F). This observation was independent of temperature during antibody staining (Sup. Fig. 4D). Interestingly, also ADP-induced platelet-leukocyte interaction was not modulated by temperature despite its effect on ADP-induced degranulation (Sup. Fig. 5E-F).
Overall, these results show that the impact of platelet count, temperature and pH levels on PLA formation reflects the effects on platelet activation, particularly degranulation. While reduced platelet count and pH can influence PMoA and PNA formation, room temperature (clinically relevant hypothermia) had no impact on PLA formation.
3.5 Temperature but not platelet count or pH modulates platelet-mediated monocyte activation
Since reduced platelet count and acidosis but not hypothermia affected platelet-leukocyte interactions, we next investigated the impact of these parameters on platelet-induced monocyte and neutrophil activation by determining activation of CD11b and shedding of CD62L.
Monocyte CD11b activation only showed a minor reduction at 10 % platelets relative to 100 % upon medium TRAP-6 stimulation (Fig. 5A ) while no differences in CD62L shedding were observed upon TRAP-6 stimulation between different platelet counts (Fig. 5B). In contrast, 10 % platelet count diminished CD11b activation relative to 100 % upon ADP stimulation while basal activation was unaffected (Sup. Fig. 6A ), though ADP-mediated CD62L shedding was only reduced in the basal state at 10 % platelets (Sup. Fig. 6B). Reduced platelet counts also showed no effects on monocyte polarization into CD16+ subsets (data not shown).
Fig. 5Temperature but not platelet count or pH modulates platelet-mediated monocyte activation. Experiments with mixed blood components or whole blood were performed at different (A-B) platelet count, (C-D) pH or (E-H) temperature as described in Fig. 1, Fig. 2, Fig. 3. Monocyte activation in response to 3 μM or 6 μM thrombin activating peptide 6 (TRAP-6) or control (PBS) was measured by flow cytometry by assessing (A, C, E) activated CD11b as well as (B, D, F) shedding of CD62L. Representative flow cytometry plots with mean values are given in E and F. (G-H) Monocyte polarization into (H, left) CD14++ CD16- classical, (H, middle) CD14++ CD16+ intermediate and (H, right) CD14+ CD16+ non-classical subsets were measured as the percentage of total monocytes. Representative flow cytometry plots with mean values are given in G. N = 6. *p < 0.05, **p < 0.01, ***p < 0.001. MFI: mean fluorescence intensity.
Fig. 6Platelet-mediated neutrophil activation depends on physiologic temperature, but is independent of platelet count and pH level. Experiments with mixed blood components or whole blood were performed at different (A-B) platelet count, (C-D) pH or (E-F) temperature as described in Fig. 1, Fig. 2, Fig. 3. Neutrophil activation in response to 3 μM or 6 μM thrombin activating peptide 6 (TRAP-6) or control (PBS) was measured by flow cytometry by assessing (A, C, E) activated CD11b as well as (B, D, F) shedding of CD62L. Representative flow cytometry plots with mean values are given in E and F. N = 6. *p < 0.05, **p < 0.01, ***p < 0.001. MFI: mean fluorescence intensity.
Similarly, acidic microenvironment also had no influence on CD11b activation nor CD62L shedding by monocytes upon TRAP-6 (Fig. 5C-D) or ADP (Sup. Fig. 6C-D) stimulation.
Interestingly, although platelet activation was elevated at ambient temperature, activation of CD11b in monocytes was drastically impaired at 22 °C relative to 37 °C in the basal state and upon TRAP-6 (Fig. 5E) or ADP (Sup. Fig. 6E) stimulation. Surface levels of CD62L were also higher at 22 °C compared to 37 °C in the basal state and upon TRAP-6 (Fig. 5F) and ADP (Sup. Fig. 6F) stimulation, which indicates impaired shedding. When we investigated effects of temperature on platelet-mediated monocyte polarization, we further found that after strong platelet stimulation with TRAP-6 the percentage of classical monocytes remained higher at 22 °C than at 37 °C, while intermediate monocytes were less frequent. In contrast, the percentage of non-classical monocytes was higher at 22 °C compared to 37 °C in the basal state but also after stimulation (Fig. 5G-H).
Our results thus indicate that platelet-mediated monocyte activation is largely independent of platelet count and pH but impaired at room temperature.
3.6 Platelet-mediated neutrophil activation depends on physiologic temperature, but is independent of platelet count and pH level
Finally, we also investigated the impact of platelet count, temperature and pH value on platelet-mediated neutrophil activation which we again assessed by CD11b activation and CD62L shedding.
Similar to monocytes, we found that reduced platelet counts had no effects on neutrophil activation of CD11b (Fig. 6A) and CD62L shedding (Fig. 6B) upon TRAP-6 stimulation. In contrast, ADP-induced CD11b activation was slightly, but significantly diminished at 10 % platelets relative to 100 % (Sup. Fig. 7A) while CD62L shedding was only affected after strong ADP stimulation at 50 % platelets (Sup. Fig. 7B). Although, PNA formation was significantly increased by acidosis, no alterations were observed for neutrophil CD11b activation upon changes of pH value, neither in response to platelet stimulation with TRAP-6 nor ADP (Fig. 6C, Sup. Fig. 7C). Similarly, both TRAP-6 and ADP-induced CD62L shedding were unaffected by acidic pH (Fig. 6D, Sup. Fig. 7D).
In contrast, temperature changes drastically affected neutrophil activation which was similar to the effects observed for monocyte activation (Fig. 5E-F) as again activation of CD11b and shedding of CD62L were significantly reduced at 22 °C relative to 37 °C in the basal state and upon TRAP-6 or ADP stimulation (Fig. 6E-F, Sup. Fig. 7E-F).
Our results thus show that platelet-induced neutrophil activation is mainly unaffected by changes in platelet count or pH but depends on temperature, showing augmented activation with increasing temperatures.
4. Discussion
In this study, we demonstrate that temperature influences platelet responsiveness and activation of neutrophils and monocytes without affecting direct platelet-leukocyte interactions, whereas platelet count and pH have negligible effect on platelet-mediated innate leukocyte activation. Methods to measure platelet function and platelet-leukocyte interactions are not standardized and temperature conditions vary between different laboratories and assays. Moreover, pH and platelet count are affected by anticoagulants, isolation procedures as well as various diseases.
We found that already mild thrombocytopenia reduced platelet degranulation without affecting GPIIb/IIIa activation. In line with impaired upregulation of CD62P and CD40L, which mediate platelet-leukocyte interaction [
], PMoA and PNA formation were platelet count dependent, whereas leukocyte activation remained unaffected during our observation time. Of note, PAR-1-mediated platelet responses showed higher dependency on platelet count than responses to P2Y1/12 stimulation. Intriguingly, a recent study also observed agonist-specific effects of platelet count on GPIIb/IIIa, but only minor impact on CD62P [
]. Currently, only little is known about the effect of thrombocytopenia on platelet activation as previous in vitro and clinical studies focused on platelet aggregation, with platelet counts ≤100,000/μL being associated with significantly reduced platelet aggregation in in vitro experiments [
]. The association of platelet count and aggregation may thus at least partially explain platelet hypo-responsiveness in inflammatory or infectious diseases that are associated with thrombocytopenia [
In line with our in vitro data, circulating PLA correlate with platelet count and CD62P expression in patients with rheumatic heart disease or inflammatory bowel disease, while leukocyte count, CD62L expression and TNF-α levels do not associate with platelet count nor PLA formation [
]. Thrombocytopenia clearly affects pro-thrombotic platelet functions, while immunomodulatory functions are likely maintained at low platelet concentrations. Our findings demonstrate that platelet count should be taken into account when comparing platelet degranulation or PLA levels in patient cohorts.
Further, we observed that acidic conditions enhanced PAR1-induced expression of CD62P, CD40L and CD63 but not GPIIb/IIIa activation, suggesting that acidosis may boost immunomodulatory but not hemostatic platelet functions. This is in contrast with previous findings in isolated platelets, showing curtailed pro-thrombotic platelet functions in acidic environment, including fibrinogen binding, spreading and aggregation [
]. Despite the regulation of platelet-leukocyte interactions by pH, we did not observe any pH-dependent effects on CD11b activation nor CD62L shedding of innate leukocytes. However, acidosis enhances platelet-modulated neutrophil chemotaxis [
], pointing towards differentiated regulation of individual cellular functions. Importantly, monocytes and neutrophils directly respond to intra- and extracellular changes of pH already in the absence of platelets [
], which may overshadow pH-dependent platelet-mediated regulation of leukocytes in our setup.
Local acidosis at sites of active inflammation might thus in part counteract the pro-coagulatory microenvironment caused by immunothrombosis to mitigate organ damage. In addition, our findings underline the importance to prevent (artefact) pH changes, e.g. due to different anticoagulants or prolonged air exposure when handling samples, in order to ensure correct and comparable results of functional platelet studies.
Finally, we also tested the impact of temperature on platelet function, focusing on temperatures routinely used in laboratory settings. We observed that platelet responsiveness was higher at room temperature than at physiologic 37 °C, showing increased integrin activation and degranulation, though differences in platelet activation were not mirrored by PNA or PMoA formation. While our results corroborate previous findings on elevated CD62P expression and fibrinogen binding at room temperature both in human and murine settings [
]. However, mice subjected to mild (31–34 °C) or clinically used moderate (28 °C) hypothermia display a reversible increase of in vivo thrombus formation and stability [
]. However, we found that leukocyte CD11b activation and CD62L shedding correlated with temperature. Intriguingly, this was associated with reduced release of soluble factors from platelets, indicating that platelet-derived inhibitory factors or temperature-dependent responsiveness of leukocytes themselves might play a role. Indeed, previous studies reported mildly increased CD11b expression of innate leukocytes in blood upon rising temperatures [
], though the putative contributions of platelets were not considered.
Our data thus suggest that immunomodulatory platelet functions may be modulated in vivo by hypothermia and in vitro by assay incubation temperatures. Importantly, temperature-dependent immunomodulation appears to occur via indirect rather than direct platelet-leukocyte interactions. Therefore, evaluation of only circulating PLA in patients may not provide an adequate picture of platelet-mediated immunomodulation.
In contrast to hypothermia, fever is associated with decreased hemostatic platelet function, but increased PLA formation and expression of monocytic tissue factor [
], and may thus contribute to immunothrombosis upon infection. Hence, human platelets may favor either pro-thrombotic or immunomodulatory functions depending on the microenvironmental temperature. It is currently unknown if murine platelets are as sensitive to environmental changes. However, previous studies on hypothermia in mice suggest a similar pattern [
]. Indeed, platelets have been proposed to act as thermosensors and augmented responsiveness in mild hypothermia may promote hemostatic functions at the body surface where injuries are likely to occur most frequently [
]. In contrast, enhanced platelet-leukocyte interplay during fever may help to combat pathogens in infection.
5. Conclusion
In summary, we provide evidence that platelet count, pH and temperature, which commonly change during infection but are also affected by unstandardized laboratory tests, differently regulate immunomodulatory platelet functions. Thrombocytopenic platelet count had only little impact on platelet-leukocyte interaction and acidosis was associated with slightly increased PLA formation, but neither regulated immediate activation of innate leukocytes. Contrarily, temperature did not regulate platelet binding to monocytes or neutrophils, whereas platelet-mediated leukocyte activation was highly dependent on physiologic temperature. Our data demonstrate the importance of standardized protocols for sample preparation and assay procedure, optimized to the specific research question and readout, in order to ensure comparability of studies. Furthermore, unspecific host responses of fever and local acidosis may foster platelet-leukocyte interplay in infection.
Declaration of competing interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Acknowledgement
We thank Celine Nemethy for her excellent technical assistance.
Funding
This work was supported by the Austrian Science Fund (FWF) [P-32064, P-34783]; and the Austrian National Bank [OeNB18450].
00033-6&id=ga1.jpg){kind=link}
00033-6&id=gr1.jpg){kind=link}
00033-6&id=gr2.jpg){kind=link}
00033-6&id=gr3.jpg){kind=link}
00033-6&id=gr4.jpg){kind=link}
00033-6&id=gr5.jpg){kind=link}
00033-6&id=gr6.jpg){kind=link}