Undergraduate Research Experience Award Program Report

The Impact of Nonsteroidal Anti-inflammatory Drugs
on Endothelial Cell and Platelet Production of
Microvesicles During Strenuous Exercise in Humans
Jesse Biddlecombe

August 2019

Supervisor: Dr. Mark Rakobowchuk

Introduction
Microvesicles (MVs) are anucleate extracellular vesicles derived from plasma membrane
as a result of cytoskeletal reorganization and in response to stimuli4,6 such as exercise or disease6,8.
Interestingly, the concentration of circulating MVs is altered following strenuous exercise7-9 and
the relatively large diameter of MVs enables lipids, proteins, mRNA, and miRNA transport from
where they are produced to distant recipient cells3,6,8. This suggests MVs play a role in physiology
by facilitating cell-to-cell communication that leads to structural and functional changes4-8. MVs
are taken up by the cells lining the walls of blood vessels, allowing MVs to modulate vessel
functions including proliferation and migration which result in angiogenesis3-6,8,9. This capability,
along with regulation of thrombosis and inflammation, demonstrate a key role of MVs in vascular
adaptation3-5. The functions MVs alter in target cells depends on the cell type from which they are
derived, endothelial, platelet, smooth muscle, leukocyte, or erythrocyte, and the conditions under
which they are formed3,5-8. A wide variety of MVs, and substantial differences in concentrations,
exist naturally in circulation with platelet derived microvesicles (PMVs) being most abundant3-6,8.
Similarly, the dynamics of MV response to exercise is also cell specific, with platelets producing
the most, particularly following strenuous exercise6-9. This response mirrors the increases in
vascular stress induced by exercise, as sympathetic nervous system activation and increases in
circulating metabolites like adenosine-diphosphate have been suggested agonists of PMV
formation5,7,8. This pro-angiogenic response in addition to the absence of an increase in endothelial
MVs associated with vascular damage, suggests MVs play an integral role in vascular adaption to
the stress responses of exercise6-8. Therefore, any drug affecting platelet function may also affect
MV formation and influence a person’s ability to adapt to exercise.
Non-steroidal anti-inflammatory drugs (NSAIDs), such as ibuprofen or aspirin, inhibit
platelet function1,2 by interacting with cyclooxygenase (COX), an enzyme responsible for
production of thromboxanes and prostaglandins, which are critical for platelet-endothelium
interaction and hemostasis1,2. This inhibition actively supresses COX in both endothelial cells and
platelets, however, the lack of synthetic capability in platelets8 results in greater functional impact2.
Further, in vitro investigations have shown some COX inhibitors specifically inhibit MV formation
following stimuli similar to that of exercise3. The aim of the proposed research is to determine
whether taking NSAIDs blunts MV concentrations that are normally elevated following strenuous
exercise. We will focus on changes in PMV concentrations, however, as NSAIDs effect both
platelets and endothelial cells, endothelial MVs will also be monitored. The proposed research is
an important step in connecting MV involvement in vascular adaptation to exercise with the effects
of NSAIDs. Additionally, the present study may provide novel insight into the potential
determinantal effects of NSAID use on an individual’s ability to adapt to exercise training. This is
particularly important in the rehabilitation setting where pain relievers like Ibuprofen are used
routinely. The ubiquitous accessibility of NSAIDs as over-the-counter medications, demonstrates
further importance of investigation into potential negative effects.
Project Aims
This research aimed to address the hypothesis that using non-steroidal anti-inflammatory
drugs prior to strenuous exercise blunts circulating concentrations of endothelial and platelet
Biddlecombe, J. UREAP Report.

Page 2 of 7

derived microvesicles during and following the exercise period. Specifically, MV production
throughout and after exercise was to be compared in young and healthy individuals prior to and
following the use of NSAIDs.
Exercise trials were to be completed by participants 18-35 years of age in good health and
include three conditions, on one occasion, a control trial, during which the participant would
receive a placebo (sugar pill), on another occasion, an experimental trial, involving a standard 600
mg dosage of ibuprofen two hours prior to the exercise period, and on a third occasion, an
experimental trial, involving a 325 mg dosage of aspirin one hour before exercise. Before these
experimental and control trials, an exercise tolerance test was to be performed by each participant
to establish the intensities of all three experimental and control trials that involve 45 minutes of
intense cycling. Analysis was to be performed on venous blood samples taken at various time
points following exercise (Dr. Rakobowchuk or affiliated nurse).
Plasma sample analysis includes quantification of MVs, blood lactate and glucose,
hormones, and other metabolites. MV concentration was to be determined using the Cytoflex flow
cytometer with antibody treatment for derivation-specific quantification. In combination with flow
cytometry, MVs may also be quantified through nanoparticle tracking using a custom-built
microscope. Portable analyzers, specifically lactate plus, were to be used for measurement of
lactate and glucose concentration. Finally, enzyme-linked immunosorbent assays for
catecholamines were to be used to quantify hormones present in the samples.
Project Progress
Upon receipt of the Undergraduate Research Experience Award, application for ethical
approval of the project was submitted through the Thompson Rivers University ROMEO online
portal. While waiting for ethical approval, participant recruitment began along with preliminary
method development for sample analysis. Challenges arose during the recruiting process as many
of the target age were unavailable during summer months. This in conjunction with the delayed
approval of project ethics resulted in an inability to move past the recruitment phase into exercise
trials. As a result, the remainder of the project timeline consisted of method development in
preparation for sample analysis at a later point. Specifically, the methods developed include sample
preparation, total antioxidant capacity (TAC) determination, and quantification of
epinephrine/norepinephrine, thiobarbituric acid reactive substances (TBARS), and total nitric
oxide (NO) concentrations. These assays all measure different circulating factors thought to
modulate the dynamics of MV concentrations in circulation. To develop these assays, samples
obtained from another study involving exercise and MVs were used.
Over the course of the summer, I analyzed plasma samples as described above (TAC,
catecholamines and NO). The NO assay was a unique aspect of my development since this assay
needed to be developed from scratch using published literature methodologies. The first aspect
that I worked on was determining the best method to use when preparing plasma samples. Through
consulting the published preparation methods and the requirements of each analysis technique, I
determined that the anti-coagulant used during blood collection could influence the accuracy of
some quantification methods, with the NO assay being the most sensitive in the present study
(Griess Assay for total nitric oxide determination)10. I determined that a serial centrifugation
approach would be needed to first isolate plasma from the venous blood sample, obtain plateletBiddlecombe, J. UREAP Report.

Page 3 of 7

free plasma, and isolate MVs. For all the above-mentioned analytical methods, samples would
then need to be stored as platelet-free plasma at -80oC and thawed only immediately prior to
analysis.
The TAC of plasma as well as concentrations of epinephrine/ norepinephrine and TBARS
within samples was determined using commercially available kits. To further evaluate the sample
preparation method chosen, plasma samples collected in a similar manner during a previous study
were analyzed using the following kits: Epinephrine/ Norepinephrine ELISA Kit (Abnova
KA1877), Parameter TBARS Assay (R&D Systems KGE013), and Antioxidant Assay Kit
(Cayman Chemical 709001). Both the assays for determination of epinephrine/norepinephrine and
TBARS concentrations worked well with the sample preparation used (results have been
summarized and are being incorporated into a manuscript). The antioxidant assay kit did not give
reproducible results, and upon further investigation it was found that EDTA can interfere with the
assay11. This is problematic as EDTA is a common anti-coagulant used in blood collection, and
sensitivities in the Griess Assay deter use of the other common anti-coagulant, heparin. However,
repeated freeze/thaw of plasma samples can interfere with the antioxidant assay and therefore must
also be considered.
Development and Assessment of NO in Plasma Samples
The total nitric oxide concentration in plasma samples is generally determined indirectly
from nitrite and nitrate concentrations using a commercially available kit. Initially, Dr.
Rakobowchuk had envisaged using a commercially available kit, but upon reading the details of
the assay it was decided that developing the assay from raw reagents would be the more financially
viable route in the long-term. As such, I developed a protocol for quantifying nitrates and nitrites
in plasma without using an expensive commercial assay kit. I found that current research generally
used a diazotization assay based upon the Griess reaction to complete this quantitative analysis10.
Specifically, the Griess reaction involves the interaction of nitrite with sulfanilamide and N-(1napthyl)-ethylenediamine to produce a diazonium product10. I also determined that prior to this
reaction a nitrate reduction step is required to enable spectrophotometric measurement of nitrite
concentration and estimation of in vivo nitric oxide formation12. Use of the Griess reaction on
plasma samples, as in the present study, is limited in sensitivity due to the presence of proteins13.
As a result, I incorporated a deproteination step prior to the reduction of nitrate, to eliminate
inflated measurements due to high background absorbance13. There are ten commonly used
methods of protein precipitation, each of which can influence the Griess assay differently12. In
accordance with the findings of Ghasemi et al., I decided to use zinc sulphate to deproteinate
plasma samples, as it has a similar influence on the assay as the ideal deproteination method,
ultrafiltration, without the exorbitant costs (2007). Currently, intra- and inter-assay precision tests
as well as recovery assessments have been performed in replicate and results have been consistent
with assay validations previously published. Experimental trials have also been performed on
plasma samples collected in a previous study to demonstrate appropriateness for the type of sample
preparation employed in the present study.
Developing an Assay to Assess Movement of Insulin and the Influence of MVs

Biddlecombe, J. UREAP Report.

Page 4 of 7

Over the summer, I also developed an assay to measure the movement of insulin across
endothelial cells. This assay will be used in the present study as a functional bioassay of how MVs
could influence glucose regulation in people. We hypothesize that exercise-induced alterations to
circulating MV concentrations may contribute to the efficacy of exercise as an insulin resistance
countermeasure. If an influence was observed than the hypothesized decrease in concentrations of
circulating endothelial cell and platelet derived MVs as a result of NSAID use prior to exercise,
must also be considered when suggesting exercise as a corrective. To directly assess whether MVs
impact the transcytosis of insulin, a transcytosis assay was performed using human umbilical vein
endothelial cells (Ea.hy926 cell line) and transwells containing a polyester membrane with a pore
size of 0.4 µm. The confluent monolayer of endothelial cells was representative of the endothelium
lining microvasculature of muscle tissue and was exposed to either MV-containing or platelet-poor
plasma from pre- or post-high intensity exercise blood draws completed in a previous study14.
Exposure occurred for one hour and 45 minutes in accordance with the duration of major MV
concentration changes following exercise documented in previous studies8. Following this,
incubation with dextran and insulin conjugated to fluorescent markers allowed quantification of
the two molecules movement across the monolayer over a four-hour period. FitC-insulin
fluorescence was compared to determine both quantity and rate of insulin transcytosis, whereas
observation of TritC-dextran was used as a normalization method for cell monolayer integrity.
Previous analysis of the plasma samples using flow-cytometry demonstrated that high intensity
exercise increased concentrations of platelet (Pre-cycling: 2.94 ± 1.16 x 105/mL [mean ± SEM];
Post-cycling: 6.03 ± 2.41 x 105/mL; p = 0.01) and not endothelial cell microvesicles (Pre-cycling:
2.62 ± 1.8 x 105/mL; Post-cycling: 1.79 ± 1.51 x 105/mL; p = 0.16)14. However, in the transcytosis
assays performed thus far, the presence of microvesicles did not alter endothelial insulin
transcytosis (total insulin transcytosed: 125.0 ± 11.5 ng and 116.3 ± 8.4 ng for MV-containing and
platelet-poor plasma, respectively), nor did alterations in MV concentration as a result of highintensity exercise (121.7 ± 8.5 ng and 125.0 ± 11.5 ng for pre- and post-cycling, respectively).
Overall, the method development completed thus far will allow the project to continue in
a timely manner once participant recruitment can be completed. The plans to disseminate research
findings described in the funding application will be carried out upon project completion at a later
date. Presently, research findings from the inquiry into the role of MVs in insulin transcytosis by
endothelial cells have been submitted in the form of an abstract and will be presented at the
Canadian Society for Exercise Physiology Conference in November of 2019.

Biddlecombe, J. UREAP Report.

Page 5 of 7

References
1. Curvers, J., Dielis, A. W. J. H., Heeremans, J. & Van Wersch, J. W. J. Platelet function in
whole-blood donors is impaired: The effects of painkillers. Transfusion 47, 67–73 (2007).
2. Schafer, A. Effects of nonsteroidal anti-inflammatory drugs on platelet function and systemic
hemostasis. Journal of Clinical Pharmacology. 35(3): 209-219 (1995).
3. Minuz, P., Meneguzzi, A., Degan, M., Giacomazzi, A. & Calabria, S. Antiplatelet Agents
Inhibit the Generation of Platelet-Derived Microparticles. Front. Pharmacol. 7, (2016).
4. Morel, O., Jesel, L., Freyssinet, J. M. & Toti, F. Cellular mechanisms underlying the
formation of circulating microparticles. Arterioscler. Thromb. Vasc. Biol. 31, 15–26
(2011).
5. VanWijk, M. J., VanBavel, E., Sturk, A. & Nieuwland, R. Microparticles in cardiovascular
diseases. Cardiovasc. Res. 59, 277–287 (2003).
6. Shantsila, E., Kamphuisen, P. W. & Lip, G. Y. H. Circulating microparticles in cardiovascular
disease: Implications for atherogenesis and atherothrombosis. J. Thromb. Haemost. 8,
2358–2368 (2010).

Biddlecombe, J. UREAP Report.

Page 6 of 7

7. Wilhelm, E. N., González-Alonso, J., Parris, C. & Rakobowchuk, M. Exercise intensity
modulates the appearance of circulating microvesicles with proangiogenic potential upon
endothelial cells. Am. J. Physiol. Circ. Physiol. 311, H1297–H1310 (2016).
8. Wilhelm, E. N., Mourot, L. & Rakobowchuk, M. Exercise-Derived Microvesicles: A Review
of the Literature. Sport. Med. 48, 2025–2039 (2018).
9. Sossdorf, M., Otto, G. P., Claus, R. A., Gabriel, H. H. & Lösche, W. Release of pro-coagulant
microparticles after moderate endurance exercise. Platelets 21, 389–391 (2010).
10. Miranda, K. M., Espey, M. G. & Wink, D. A. A rapid, simple spectrophotometric method for
simultaneous detection of nitrate and nitrite. Nitric Oxide – Biology and Chemistry 5(1),
62-71 (2001).
11. Antioxidant Assay Kit. (Cayman Chemical Company, 2016).
12. Ghasemi, A., Hedayati, M. & Biabani, H. Protein precipitation methods evaluated for
determination of serum nitric oxide end products by the griess assay. JMSR 2(15), 29-32
(2007).
13. Moshage, H., Kok, B., Huizenga., J. R., & Jansen, P. L. M. Nitrite and nitrate determinations
in plasma: a critical evaluation. Clin. Chem. 41(6), 892-896 (1995).
14. Lauriente T.J. The Impact of High Intensity Exercise on the Anti-angiogenic Profile of
Platelet Derived Microvesicles. (Thompson Rivers University, 2018).

Biddlecombe, J. UREAP Report.

Page 7 of 7