Immune cell expression patterns of CD39/CD73 ectonucleotidases in rodent models of cardiac arrest and resuscitation

Background Cardiac arrest (CA) is a significant public health concern. There is the high imminent mortality and survival in those who are resuscitated is substantively compromised by the post-CA syndrome (PCAS), characterized by multiorgan ischemia–reperfusion injury (IRI). The inflammatory response in PCAS is complex and involves various immune cell types, including lymphocytes and myeloid cells that have been shown to exacerbate organ IRI, such as myocardial infarction. Purinergic signaling, as regulated by CD39 and CD73, has emerged as centrally important in the context of organ-specific IRI. Hence, comprehensive understanding of such purinergic responses may be likewise imperative for improving outcomes in PCAS. Methods We have investigated alterations of immune cell populations after CA by utilizing rodent models of PCAS. Blood and spleen were collected after CA and resuscitation and underwent flow cytometry analysis to evaluate shifts in CD3+CD4+ helper T cells, CD3+CD8a+ cytotoxic T cells, and CD4/CD8a ratios. We then examined the expression of CD39 and CD73 across diverse cell types, including myeloid cells, T lymphocytes, and B lymphocytes. Results In both rat and mouse models, there were significant increases in the frequency of CD3+CD4+ T lymphocytes in PCAS (rat, P < 0.01; mouse, P < 0.001), with consequently elevated CD4/CD8a ratios in whole blood (both, P < 0.001). Moreover, CD39 and CD73 expression on blood leukocytes were markedly increased (rat, P < 0.05; mouse, P < 0.01 at 24h). Further analysis in the experimental mouse model revealed that CD11b+ myeloid cells, with significant increase in their population (P < 0.01), had high level of CD39 (88.80 ± 2.05 %) and increased expression of CD73 (P < 0.05). CD19+ B lymphocytes showed slight increases of CD39 (P < 0.05 at 2h) and CD73 (P < 0.05 at 2h), while, CD3+ T lymphocytes had decreased levels of them. These findings suggested a distinct patterns of expression of CD39 and CD73 in these specific immune cell populations after CA. Conclusions These data have provided comprehensive insights into the immune response after CA, highlighting high-level expressions of CD39 and CD73 in myeloid cells.


Introduction
Cardiac arrest (CA) is a major public health issue, afflicting more than 356,500 people in the out-of-hospital setting and 209,000 people in the in-hospital setting in the United States each year (1).Post-CA syndrome (PCAS) is a lethal condition characterized by high mortality and systemic ischemia-reperfusion injury (IRI) (2)(3)(4), which can trigger immune responses that play a significant role in the pathophysiology after CA (5).
Even after successful resuscitation, out-of-hospital CA (OHCA) patients suffer from a hyperinflammatory status, a so called sepsislike syndrome, which is characterized by high levels of circulating pro-inflammatory cytokines and a dysregulation of the immune system (6)(7)(8)(9).Hyporesponsiveness of circulating leukocytes such as decreased production of cytokines in OHCA patients was observed ex vivo (8)(9)(10), indicating that the cell-mediated immunity of OHCA patients is impaired (10).These changes are also similar to those in sepsis, in which the immune system is disordered (11), therefore, understanding the mechanisms underlying this immune disorder may be the key to rescuing OHCA patients after return of spontaneous circulation (ROSC).
The immune response during and after CA and resuscitation is a complex process that involves various immune cell types, and understanding their roles is crucial for improving survival.Clusterof-differentiation 3 (CD3) is a cell surface protein complex found on T cells, which are a type of leukocyte that plays a central role in the immune system.CD4 is a glycoprotein found on the surface of helper T cells.Helper T cells play a vital role in orchestrating the immune response by assisting other immune cells in recognizing and responding to pathogens or damaged tissues.CD8a is found on cytotoxic T cells, which recognize and eliminate infected or damaged cells.Although these lymphocytes assume a central role in the immune response, differential counts of these T cell populations after CA have been scarcely addressed.In addition to T lymphocytes, other major immune cells expressing CD19 or CD11b are also involved in post-CA immune responses.CD19 is found primarily on B cells, a type of leukocyte involved in the immune response that can become relevant following acute phase of diseases that are associated with infection, inflammatory conditions, or other conditions triggering a systemic inflammatory response (12).CD11b is an integrin receptor primarily found on the surface of myeloid cells, specifically neutrophils and monocytes.These myeloid cells are important components of the immune system and play a role in the systemic response from injuries, infection, and inflammation, specifically in the adhesion of these immune cells to the blood vessel wall, their subsequent migration into tissues, and neutrophil activation (13,14).
In addition to aforementioned surface markers characterizing major immune cell populations, we explored the intricate field of purinergic signaling, a novel focal point in the context of IRI (15).Elucidating this pathway holds paramount significance in mitigating hyperinflammatory responses and enhancing post-CA survival.
Adenosine triphosphate (ATP), a fundamental molecule for intracellular energy transfer, assumes a dual role in which, when released from ischemic cells into the extracellular milieu, purines become signaling molecules.Extracellular ATP (eATP) is capable of activating purinergic receptors, specifically P2X and P2Y receptors, thereby triggering cellular injury and hyperinflammatory cascades, leading to cell death and organ dysfunction (16,17).eATP can be converted stepwise to adenosine by cell-surface enzymes such as ectonucleoside triphosphate diphosphohydrolase-1 (ENTPD1, CD39) and ecto-5'-nucleotidase (NT5E, CD73 phosphohydrolysis into adenosine is a multi-step process facilitated by these ectoenzymes.CD39 catalyzes the conversion of adenosine tri-and diphosphates to monophosphate, while CD73 further converts the monophosphate into adenosine.Notably, when adenosine binds to purinergic receptors, it exerts cytoprotective and anti-inflammatory effects (16,17).Both CD39 and CD73 hold the potential to significantly impact the outcomes of CA and resuscitation through the fine-tuned regulation of adenosine levels.
In this study, cytometric analyses to assess alterations in predominant immune cell populations and expression of ectonucleotidases following CA were investigated.We employed both rat and mouse models of CA and PCAS.

Rat cardiac arrest animal preparation
We performed all instrumentation according to the previously described protocol (Figure 1A) (4,18,19).In brief, adult male 12-16-week-old Sprague-Dawley rats (400-500 g, Charles River Laboratories, Wilmington, MA, USA) were anesthetized with 4% isoflurane (Isosthesia, Butler-Schein AHS, Dublin, OH, USA) and intubated with a 14-gauge plastic catheter (Surflo, Terumo Medical Corporation, Tokyo, Japan).We used only male rats to reduce potential confounding among animals that may cause outcome variabilities from hormonal and/or genetic differences.Rats were mechanically ventilated (Ventilator Model 683, Harvard Apparatus, Holliston, MA, USA) at a minute ventilation (MV) volume of 180 mL per minute at a respiratory rate of 45 breaths per minute.In this study, we used one ventilation setting for all animals at all times and did not change the MV or respiratory rate during the experiments.Positive end-expiratory pressure (PEEP) was set at 2 cm H 2 O. Anesthesia was maintained with isoflurane at 2% and at a fraction of inspired O 2 (F I O 2 ) of 0.3.Core temperature was maintained at 36.5 ± 1.0 °C during the surgical procedure.After instrumentation, neuromuscular blockade was achieved by slow intravenous administration of 2 mg/kg of vecuronium bromide (Hospira, Lake Forest, IL, USA) and asphyxia was induced by turning off the ventilator.CA normally occurred three to four minutes after a s p h y x i a s t a r t e d .T h e C A g r o u p a n i m a l s r e c e i v e d cardiopulmonary resuscitation (CPR) after they were subjected to 12 minutes asphyxia.We defined CA as a mean arterial pressure (MAP) of < 20 mmHg; CA was completely untreated during the initial 10 minutes.After 12 minutes of asphyxia, mechanical ventilation was restarted at an F I O 2 of 1.0 and manual chest compression CPR was delivered simultaneously.Chest compressions were performed with two fingers over the sternum at a rate of 260 to 300 per minute.At 30 seconds after beginning of CPR, a 20 µg/kg bolus of epinephrine was given to rats through a Experimental protocol on animal surgery.(A) Animal protocol of rat CA.Rats were subjected to 12 minutes of asphyxia-induced CA and resuscitation.Following observation under mechanical ventilation for 120 minutes after initiating CPR, whole blood was collected.(B) Animal protocol of mouse CA.Mice were subjected to 8 minutes of KCl-induced CA and resuscitation.At 2 hours or 24 hours after initiating CPR, whole blood and the spleen were collected.venous catheter.Following ROSC, defined as MAP > 60 mmHg, CPR was discontinued.At 2 hours after initiating CPR, mechanical ventilation was discontinued, and rats were euthanized for sample collections of whole blood.Whole blood were drawn through the arterial catheter.

Mouse cardiac arrest animal preparation
We performed all instrumentation according to the previously described protocol (Figure 1B) (20).In brief, adult male 14-18week-old C57BL/6 mice (weight 25 to 30 g; Charles River Production, Wilmington, MA, USA) were freely accessed to food and water before experiments.Mice were anesthetized with 4% isoflurane (Isosthesia, Butler-Schein AHS, Dublin, OH, USA) and the trachea was orally intubated with a 20-gauge catheter (Angiocath, Becton Dickinson).Mechanical ventilation (Minivent, Harvard Apparatus, Holliston, MA, USA) was initiated at rate of 110/min, I:E = 1:1, F I O 2 = 1.0, and tidal volume was maintained at 8-12 mL/kg per minute.We used only male mice to reduce potential confounding among animals that may cause outcome variabilities from hormonal and/or genetic differences.In this study, we used one ventilation setting for all animals at all times and did not change the MV or respiratory rate during the experiments.PEEP was set at 1 cm H 2 O. Anesthesia was maintained with isoflurane at 2%.During surgical preparation, body temperature was maintained between 37.0 ± 0.5 °C by an incandescent heating lamp and monitored by a transesophageal thermocouple probe.A microcatheter (EZ-1101, BioTime Inc, Carlsbad, CA, USA) was inserted into the left femoral vein for drug and fluid administration.The other microcatheter was inserted into left femoral artery to monitor the blood pressure.Electrocardiograph (ECG) was recorded by needle probe.The mouse CA was quickly induced by 100 mL KCl (80 mg/kg) followed by 200 mL room temperature saline flushing and weaned off the ventilator at the same time.The CA was confirmed by loss of ECG activity and blood pressure, which usually sharply decreased, below 10 mmHg.The esophageal temperature of the mice was maintained at 37.0 ± 0.5 °C during the whole procedures by heating lamp.Eight minutes after the induction of CA, resuscitation via manual chest compression with one finger over the sternum at a rate of 300 to 400 per minute was initiated together with 1 mg/100 mL epinephrine followed by flushing with 200 mL room temperature saline via left femoral vein, and ventilation was resumed at the rate of 110/min, I:E = 1:1, F I O 2 = 1.0 with tidal volume maintained at 8-12 mL/kg per minute.Chest compression was performed to provide a uniform rate and to maintain maximal MAP.ROSC was confirmed by the return of ECG rhythm and significant increase in MAP from that during chest compression.The mouse was maintained on mechanical ventilation after ROSC.At 2 hours after initiating CPR, mechanical ventilation was discontinued, and mice were euthanized for sample collections of the spleen and whole blood.For the assessment of alterations in predominant immune cell populations and the expression of ectonucleotidases at a later time point, 3 mice were monitored until 24 hours after initiating CPR (Figure 1B).Subsequent to extubation, a subcutaneous injection of 0.05 mL extended-release buprenorphine (Ethiqa XR, Fidelis Pharmaceuticals, North Brunswick Township, NJ, USA) was administered to the animals for analgesia.At 24 hours after initiating CPR, the mice were euthanized for sample collections of the spleen and whole blood.Whole blood samples were obtained from inferior vena cava under inhaled isoflurane anesthesia prior to perfusion and subsequent spleen extraction.Postsurgical care, including animal housing and observation, were provided by the animal facility.

Cell preparation of whole blood for flow cytometry
Whole blood samples were collected into heparinized tubes.In the rat CA experiment, 500 mL of whole blood was collected after catheterization as baseline and again at 2 hours after initiating CPR.In the mouse CA experiment, animals were euthanized 2 hours or 24 hours after initiating CPR and 500 mL of whole blood was collected, and 500 mL of whole blood was collected from a naïve mouse for reference purposes.The collected blood samples were diluted into 5 mL of 1× RBC Lysis Buffer (prepared from 10× RBC Lysis Buffer, BioLegend, San Diego, CA, USA).After an incubation period of 4-5 minutes at room temperature with occasional shaking, the reaction was halted by dilution of the Lysis Buffer with 20-30 mL of 1× PBS.Subsequently, cells were pelleted by centrifugation at 500×g at room temperature for 5 minutes and then resuspended in 2 mL of 1× RBC Lysis Buffer.Following a similar incubation duration with occasional shaking, the reaction was halted by diluting the Lysis Buffer with 10-15 mL of 1× PBS.Cells were pelleted again by centrifugation at 500×g at room temperature for 5 minutes and finally resuspended in 500 mL of Flow Cytometry Staining Buffer (Invitrogen, Waltham, MA, USA).

Cell preparation of the spleen for flow cytometry
In the mouse CA experiment, a spleen was harvested 2 hours or 24 hours after initiating CPR, meanwhile a spleen from a naïve mouse was harvested for comparative reference.The harvested spleen sample was finely minced using sharp scissors and then placed onto a 70 mm Cell Strainer (Corning, NY, USA), which was situated atop a 50 mL tube.The minced spleen was further homogenized using the plunger end of a 5 mL syringe and rinsed with 5 mL of 1× PBS.Subsequently, the strained cells were pelleted by centrifugation at 500×g at room temperature for 5 minutes.The resulting cell pellets were resuspended in 2 mL of 1× RBC Lysis buffer.After an incubation period of 1-2 minutes at room temperature with occasional shaking, the reaction was halted by dilution of the Lysis Buffer with 10-15 mL of 1× PBS.Subsequently, cells were pelleted by centrifugation at 500×g at room temperature for 5 minutes and finally resuspended in 500 mL of Flow Cytometry Staining Buffer.

Microscale protein labeling
A single fluorophore was conjugated to either the anti-rat CD39/ENTPD antibody (Proteintech, cat no.14211-1-AP, Rosemont, IL, USA) or the anti-rat NT5E/CD73 antibody (Proteintech, cat no.12231-1-AP) using the Alexa Fluor ® 594 Microscale Protein Labeling Kit or the Alexa Fluor ® 647 Microscale Protein Labeling Kit (Invitrogen), respectively, following the manufacturer's provided protocols.Briefly, 1 M sodium bicarbonate solution was prepared by adding 1 mL deionized water to the vial of sodium bicarbonate (Component B).Next, 50 µL of antibodies and 5 µL of Component B were transferred to a designated reaction tube (Component C).Separately, 10 µL of deionized water was introduced to each vial containing Alexa Fluor 594 and 647 succinimidyl ester (Component A).Subsequently, 0.616 µL (Alexa Fluor 594) or 0.7 µL (Alexa Fluor 647) of Component A was transferred to Component C, followed by a 15-minute incubation period at room temperature.To purify the Alexa Fluor 594-labeled anti-rat CD39 antibody and the Alexa Fluor 647-labeled anti-rat CD73 antibody, 50 µL of Component C was meticulously dispensed onto the center of the resin bed surface within specialized spin columns.The spin columns were then subjected to centrifugation at 16,000×g for 1 minute, effectively yielding the purified labeled antibodies in the collection tubes.

Statistical analysis
Data are shown as means and standard errors of the means (SEM) for all variables.Two-tailed P-values of Wilcoxon matchedpairs signed rank test and Mann-Whitney test were used for twogroup comparisons: between baseline and 2 hours after CA in the rat model; and between naïve and 2 hours after CA in the mouse model, respectively.Dunn's multiple comparisons test was used for three-group comparisons: between naïve, 2 hours, and 24 hours after CA in the mouse model.P < 0.05 was considered statistically significant.Prism 10.0.3 (GraphPad, San Diego, CA, USA) was used for these statistical analyses.

Discussion
The present study comprehensively explored the expression levels of ectonucleotidases.This investigation represents the inaugural exploration of CD39 and CD73 expression within distinct immune cell populations in our rodent model of CA.It is noteworthy that, while our prior research has demonstrated upregulation of RNA levels of Cd39 and Cd73 in the brain and heart at 2 hours after CA and resuscitation, the expression of these ectonucleotidases on the cellular surface of circulating immune cells has not yet been addressed (19).
Ryzhov and colleagues conducted an important study in which CD39 and CD73 levels in leukocytes were assessed in patients after CA at distinct times over one week after resuscitation.Their findings indicated that CD39 and CD73 expression rapidly increased, peaking at 6 and 12 hours after resuscitation and persisting (24).No B cell alterations were noted.Nevertheless, it is important to acknowledge that drawing comprehensive mechanistic conclusions from their study is challenging due to the relatively small sample size, missing data points, and heterogeneity of patient characteristics, some of whom underwent therapeutic temperature management.More importantly, there has been no data elucidating which population of leukocytes is predominantly involved in purinergic signaling.Our data provides experimental evidence support for the substantive increases of CD39 and CD73 on circulating immune cell populations following CA and in PCAS.
In separate research studies unrelated to CA and resuscitation, Köhler and colleagues demonstrated the protective effects of CD39 in the context of IRI using mouse models of myocardial infarction.Here, the preconditioning therapy enhanced CD39 expression in cardiomyocytes and endothelial cells within 2 hours of initiating the therapy, resulting in a significant reduction in infarct size in the heart.Notably, the protective effects were compromised in mice with Cd39 gene deletion, highlighting the crucial role of CD39 in this process (25).In the peripheral blood, CD39 is constitutively expressed in over 90% of B cells, over 90% of monocytes, and approximately 20-30% of CD4 + T cells (26).CD39 expression can be induced by various factors, including exposure to eATP, proinflammatory cytokines such as interleukin-6 (IL-6), as well as conditions of oxidative stress and hypoxia (17).Within the rodent population, CD11b + leukocytes represent myeloid cells, including monocytes, macrophages, and granulocytes.Purinergic signaling has a notable impact on monocyte functions, as eATP not only serves as a chemoattractant for monocytes but also activates the NF-kB signaling pathway, promoting polarization towards M1 macrophages (17,27).Furthermore, adenosine signaling via A2A and A2B receptors has the capacity to induce anti-inflammatory M2 macrophages.These M2 macrophages are characterized by a reduction in tumor necrosis factor-alpha (TNF-a) levels and an increase in anti-inflammatory IL-10 levels (28-30).
The monocyte-macrophage phagocytic system stands as a principal driver of inflammation in the context of IRI (31)(32)(33)(34)(35). Our published (36) and unpublished data have demonstrated a notable surge in proinflammatory cytokine levels, both within the affected tissues and in systemic circulation, within the initial 6 hours following resuscitation.However, an intriguing question arises concerning whether the integrity of the blood-brain barrier undergoes a transient disruption after CA, potentially facilitating the infiltration of circulating immune cells into the brain parenchyma.Notably, a study by Zhang and colleagues revealed the presence of various immune cell populations, including CD11b + CD11c + dendritic cells, CD11b + CD45 +hi macrophages, CD11b + Ly6G -monocytes, and cytotoxic CD8 + T cells, in brain parenchyma 3 days after resuscitation.This coincided with increased numbers of these cell populations observed in the blood and bone marrow (37).Collectively, these findings emphasize the pivotal role of recruited immune cells, particularly bone-marrowderived circulating monocytes and macrophages, in orchestrating the post-CA immune response within the brain.Notably, our current study provides additional insights by revealing a significant upregulation of ectonucleotidase expression within the circulatory CD11b + myeloid cells, concomitant with a substantial increase in their population.This populational shift of myeloid cells likely constitutes a pivotal factor in the increased levels of ectonucleotidases observed in whole leukocytes.Of particular interest is the remarkable upregulation of CD73, indicating that myeloid cells, which encompass monocytes and macrophages, are likely to play a prominent role in immune responses and purinergic signaling following CA and resuscitation.
In the context of global cerebral ischemia caused by CA and resuscitation, the traditionally immune-privileged brain undergoes rapid infiltration of pro-inflammatory T cells (38).Predominantly, these infiltrating cells are CD4 + T lymphocytes, which not only become activated but also maintain their presence within the cerebral parenchyma for an extended duration, persisting up to 72 hours after resuscitation (39,40).In experimental stroke studies, the timing of immune cell entry into the brain has been welldocumented, with neutrophils being the initial responders, appearing within a few hours of stroke onset, followed by lymphocytes, which typically arrive 12 to 24 hours after the stroke (39)(40)(41)(42).This lymphocytic infiltration coincides with a significant increase in neuronal cell death in the hippocampal CA1 region, a phenomenon possibly mediated by the secretion of pro-inflammatory cytokines like TNFa and IFNg (41,43).These observations suggest that modulating T cell infiltration and the subsequent immune cascade in the aftermath of CA could represent a novel therapeutic approach (44).Our current findings from both rodent models also reveal that it is CD3 + CD4 + helper T lymphocytes, rather than CD3 + CD8a + cytotoxic T lymphocytes, that undergo upregulation, leading to an increased CD4/CD8a ratio within the circulatory system following CA.These results collectively indicate an elevation in circulatory immune activity, primarily driven by an expanded population of CD4 + T lymphocytes after CA.Intriguingly, CD39 and CD73 exhibit upregulation within the B lymphocyte population, while their expression decreases in the T lymphocyte population.These findings shed light on the specific immune cell populations prone to ectonucleotidase expression, with a distinct increase evident in myeloid cells and B lymphocytes after CA.
This study presents a number of limitations.Firstly, our approach utilized cytometric analyses to assess alterations in predominant immune cell populations and the expression of ectonucleotidase following CA.While this method provides valuable insights, the generalizability of our findings as applied to human condition of PCAS may be constrained.Fundamental responses in rodents may not mimic human inflammatory diseases (45).However, the upregulation of purinergic signaling represented by increases of CD39/CD73 found in both humans (24) and our rodent models strongly support the need for further mechanistic studies.
The most applicable and relevant animal model of PCAS has been controversial for decades (46).Of the available models, rodents have the advantage of commercially available bioassays and share many physiological similarities with humans.It is worth noting that we employed both rat and mouse models of CA in the current study, revealing analogous shifts in the population of CD3 + CD4 + helper T lymphocytes and CD3 + CD8a + cytotoxic T lymphocytes after CA across two different species.Due to the limited availability on conjugated antibodies targeting immune cell subpopulations for rats, it was imperative for us to utilize mice in order to obtain the detailed information on ectonucleotidase expressions after CA.Our evaluation was further based on the analysis of whole blood and spleen samples collected at two time points, specifically 2 and 24 hours after CA.Consequently, any potential variations in the immune cell profile occurring at time points other than 2 or 24 hours remain unexplored.We may have failed to detect significant changes that occur at other time points.Therefore, further investigations into the dynamic changes in immune cell profiles over various time intervals are warranted.
In conclusion, our study has revealed an upregulation in circulatory immune activity, primarily attributed to the expanded population of CD4 + T lymphocytes.Furthermore, our findings have elucidated the immune cell populations that are likely to express CD39 and CD73, with notably increased levels observed in blood myeloid cells and B lymphocytes, following CA.These findings provide a foundation for future studies to explore the therapeutic potential of modulating immune responses and purinergic signaling following CA and resuscitation.

Data availability statement
The original contributions presented in the study are included in the article/supplementary material.Further inquiries can be directed to the corresponding author.

5
FIGURE 5 CD39 and CD73 expression in whole blood of rat CA model.(A) Representative images of dot plots through gating process.Leukocytes were gated.(B) Representative overlaid histograms of the CD39 fluorescence intensity.(C) CD39 median fluorescence intensity.(D) Representative overlaid histograms of CD73 fluorescence intensity.(E) CD73 median fluorescence intensity.Data were expressed as means ± SEM (n = 5 rats/group).Two groups were compared by Wilcoxon matched-pairs signed rank test (* P < 0.05).
Animal protocols were approved by The Institutional Animal Care and Use Committees of Feinstein Institutes for Medical Research (#2016-004, and #2022-012).All methods were performed in accordance with the National Institutes of Health Guide for the Care and Use of Laboratory Animals and ARRIVE guidelines.

TABLE 1
Baseline characteristics of the animals.
Data are presented as mean ± standard error.CA, cardiac arrest; MAP, mean arterial pressure; HR, heart rate; RR, respiratory rate; BT, body temperature; ROSC, return of spontaneous circulation.