The Effect of Pro-Inflammatory Conditions on Neutrophil Rolling Adhesion CURRENT

Objective: Type 2 diabetes mellitus (T2D) is the result of a dysregulation of insulin concentrations and signaling, leading to an increase in both glucose concentration and proinflammatory cytokines such as interleukin (IL)-6 and tumor necrosis factor (TNF)-α. Previous work showed that T2D patients exhibited immune dysfunction associated with increased adhesion molecule expression on endothelial cell surfaces, accompanied by decreased neutrophil rolling velocity on the endothelial cell surface. Changes in cell rolling adhesion have direct vascular and immune complications such as atherosclerosis and decreased healing time seen in T2D patients. While previous studies focused primarily on how endothelial cells affect neutrophil rolling under T2D conditions, little is known on changes to neutrophils that affect their rolling. In this study, we aim to show how the rolling behaviour of neutrophils are affected by T2D conditions on a controlled substrate. Results: We found that neutrophils cultured in T2D-serum mimicking media showed an increase in cell rolling velocity compared to neutrophils under normal conditions. Specifically, glucose alone is responsible for higher rolling velocity. While cytokines further increase the rolling velocity, they also reduce the cell size. It is likely that both glucose and cytokines reduce the PSGL-1 expression level on neutrophils.


Introduction
Neutrophils are the most numerous immune cells, and some of the first responders to infection. Under physiological conditions, neutrophils flow freely in the blood stream.
However, during the inflammatory response, neutrophils are captured out of the blood stream and roll on the endothelial wall of postcapillary venules. [1] This is mediated by the rapid formation and dissociation of interactions between P-selectin glycoprotein ligand 1 (PSGL-1) on neutrophil membranes and P-selectin expressed on endothelial surfaces under inflammatory conditions. [1,2] This tightly regulated process is essential for immune responses mounted by neutrophils. Therefore, any changes in rolling adhesion can have profound effects on the immune system.
Diabetes is a condition resulting from insufficient insulin production and/or impaired insulin response with associated hyperglycemia (elevated blood glucose concentration).
T2D patients represent 90.9% of the 23.1 million adults with diabetes according to the National Health Interview Survey. [3] T2D is associated with high basal cytokine levels and a wide range of innate immune responses. [4] Hyperglycemia has been linked to a proinflammatory state leading to increased production of interleukin-6 (IL-6) and tumor necrosis factor α (TNF-α). [5,6] Hyperglycemia can also cause insulin resistance from chronic exposure to glucose and reactive oxygen species formation. [5,6] Patients with T2D can experience several complications such as cardiovascular disease, atherosclerosis, blindness, and kidney failure. [7,8] Atherosclerosis is associated with chronic inflammation possibly due to the increased recruitment of immune cells including neutrophils. [9] Previous studies have shown patients with T2D to exhibit immune dysfunction related to neutrophil chemotaxis, rolling, and adhesion. [7,10,11] Specifically, research has shown PMNs (polymorphonuclear neutrophils) taken from T2D patients demonstrated a decrease in cell rolling velocity compared to a control population. [7,11] These studies examined cell rolling on an activated human umbilical vein endothelial cells (HUVEC) surface where variable concentrations of P-selectin and other adhesion molecules can be present. [7,9,11] This surface receptor heterogeneity may have significant effects on cell rolling behaviour. [9,12] Additionally, these studies used a singular wall shear stress value of 0.07 Pa which may not be sufficient to demonstrate the full range of effects imparted by diabetic growth conditions. [7,11] To better understand to role of the PSGL-1 in T2D, we investigated neutrophil rolling on a surface with a controlled P-selectin density.
Additionally, we chose a range of shear stress levels to mimic conditions found in both arterial (1-7 Pa) and venous (0.1-0.6 Pa) blood flow. [13] PSGL-1 is both a key ligand for rolling adhesion and a receptor to enable intracellular signaling, inducing cytokine secretion, activation of SRC family kinase, β2-integrin, and potential activation of Phosphoinositide 3-kinase (PI3K) signaling in either neutrophils or the endothelial cells they are rolling on. [14] In this paper, rolling adhesion is used as an assay to directly probe how PSGL-1 expression is affected by T2D conditions using controlled substrate P-selectin concentration. Our aim is to investigate how neutrophil rolling is affected by T2D-like proinflammatory conditions, such as hyperglycemia, and supraphysiological TNF-α, IL-6, and insulin concentrations, related solely to the interaction between PSGL-1 and P-selectin.

Flow chamber construction
The flow chamber construction has been reported previously [15]. Briefly, the flow chamber consisted of a coverslip for the bottom (rolling surface) and a glass slide for the top. The cover glass was passivated by polyethylene glycol (PEG) and functionalized by PEG-biotin (Laysan Bio) at 20:1 ratio. Once the flow chamber was assembled, a protein scaffold made up of 1 mg/mL streptavidin (Thermofisher), 100 μg/mL protein G (Thermofisher) and 10.6 μg/mL of P-selectin-Fc (R&D System) was functionalized through PEG-biotin onto the surface.
Cell growth and rolling experiment HL-60 cells were cultured in 25 cm 2 culture flasks (VWR) containing glucose-free RPMI-1640 media (Thermofisher), 10% FBS (VWR), and 1% pen/strep (Gibco). Cell cultures were maintained in 5% CO 2 at 37 o C. HL-60 cells were differentiated into neutrophils using 1.5% dimethyl sulfoxide (DMSO). To make up the final concentrations of pro-inflammatory conditions, 40% glucose (VWR), 1000 pM insulin (Humulin), 10 ng/mL TNF-α (PeproTech), and 10 ng/mL IL-6 (PeproTech) were used. Studies have demonstrated the proinflammatory effects of these concentrations of peptides [16][17][18]. Glucose concentration was measured using a commercial glucose monitor (Contour). [19] Two mL of cell culture were centrifuged for 3.5 minutes at 300 rcf. Approximately 1.7 mL of liquid was removed, and cells were re-suspended in rolling buffer and added to the flow chamber. [15] Darkfield microscopy at 10x magnification was used to capture cell rolling movies at 30 fps. A syringe pump (Harvard Apparatus) was used to control the flow rate. Lastly, we examine whether chronic exposure to non-metabolic sugar such as mannitol induces similar effects as glucose. We replaced glucose with mannitol under the exact cell culturing and rolling conditions. The cell rolling velocity increased only slightly (<10%) at high mannitol concentration (Figure 1e) compared to an increase of up to 400% in the case of glucose (Figure 1a). Furthermore, different concentrations of mannitol induced no change in the cell size across the whole range of shear stress (Figure 1f). This result shows that chronic exposure specifically to glucose has a significant effect on cell rolling behaviour. Exposure to mannitol also serves as a control for osmotic pressure on the cell caused by the increase of sugar concentration, which in our case, did not change the size of the cell and hence, insignificant effect on the cell rolling velocity. From these evidences, it is more likely that such large effect of glucose on cell rolling velocity is the result of changes to cell surface protein (i.e. PSGL-1) expression than physical size changes. Such change may be the result of increased growth rate under high glucose concentration. Indeed, the rate of cell growth increases with glucose in the culturing media, which is particularly true when considering that the insulin resistant glucose transporter GLUT1 is expressed in neutrophils. [21] Effect of Pro-inflammatory Cytokines on Cell rolling Cells cultured in media containing the diabetic cocktail (hyperglycemic media plus TNF-α, IL-6 and insulin, see Table 1) were compared to those cultured under only hyperglycemic conditions in rolling experiments. We observed a small increase in cell rolling velocity over a range of shear stresses for neutrophils cultured under these conditions (Figure 2a).
However, the effect of the diabetic cocktail compounds on cell size is quite significant (Figure 2b). Here, we observe a significant decrease in cell projection area (up to 25%) upon chronic exposure to the diabetic cocktail compared to hyperglycemic exposure alone. When plotting the cell size against mean rolling velocity (Figure 2c), two clusters corresponding to hyperglycemic and diabetic cocktail conditions clearly emerge. The decrease of cell size is generally accompanied by a decrease in rolling velocity, as the shear force acting on a smaller cell is lower. However, the fact that the smaller cells under diabetic cocktail condition roll faster than larger cells under just hyperglycemic condition came as a surprise. Because of our surface passivation and functionalization, only Pselectin are presented on the surface, enabling only adhesive interactions between Pselectin and PSGL-1. We are directly probing the rolling adhesion behaviour due to the Pselectin, PSGL-1 interaction. Hence a further decrease of PSGL-1 expression under the diabetic cocktail condition comparing to hyperglycemic condition is the most likely cause for the observed behaviour. This is also supported by the need for chronic exposure to hyperglycemia and the lack of effect upon acute exposure which does not allow for the time to affect surface protein expression.

Limitations
There are several limitations in this work: The rolling velocity of neutrophils are dictated by both the phenotype of the neutrophil and the endothelial cells. Our study focuses exclusively on how neutrophils are affected given the substrate they roll on remain unchanged. Hence, the rolling velocity changes should be considered only within the context of the current experiment. It would be difficult to draw direct extrapolations towards an in vivo system in terms of the neutrophil rolling velocity.
Although differentiated HL-60 cells are terminally differentiated and do not proliferate, they still metabolize glucose in the media. We notice the glucose concentration drop up to 20% over the 48 hours. The values reported are average glucose concentration.
Neutrophils were obtained by differentiating HL-60 cells. The differentiation process does not produce 100% neutrophil, as up to 20% cells remain undifferentiated. This may cause bias in the observation, but we do not have means to separate the differentiated cells from undifferentiated cells.
Lastly, we found that it is challenging to precisely control the surface density of P-selectin in our flow chamber. We observed up to 15% differences in rolling velocities on different functionalized coverslip surfaces while keeping other experimental conditions constant. Additionally, the high shear flow also removes P-selectin functionalized on the surface over time (~ 30-60 mins). This makes it challenging to assign statistical significance if the true rolling velocity is less than 15% different. In order to be able to quantitatively compare rolling velocity with the same surface density, we created multiple channels on the same functionalized surface and recorded data only in the first 30 mins. However, replicates across different functionalized surface is simply averaged for the lack of a better normalization method.

Declarations
Ethics approval and consent to participate Not applicable Consent for publication Not applicable Availability of data and materials The datasets used in the current study are available from the corresponding author on request.

Competing interests
The authors declare that they have no competing interests