Peripheral Levels of EPCs, T Cells, and Macrophages’ as 2 Possible Predictive Biomarkers of Acute Myocardial 3 Infarction 4

Acute myocardial infarction (AMI), with a very relevant global disease burden, remains the major mortality 15 and morbidity cause among all cardiovascular diseases. Patient prognosis is strictly dependent on early 16 diagnosis and the adoption of adequate interventions. AMI diagnosis requires constant optimization, 17 particularly considering the individuals at higher risk (or more vulnerable to worse outcomes) such as patients 18 with diabetes mellitus and atherosclerosis. Herein, we investigated the levels of peripheral blood EPCs and 19 immune cell-subsets from myeloid and lymphoid lineages, as well as their temporal dynamics, in the quest for 20 new prognostic biomarkers of AMI. We collected blood from 18 hospitalized patients (days 3 and 7 after AMI 21 onset) and 16 healthy volunteers, and resolved their circulating PBMC populations via flow cytometry. 22 Overall, our data demonstrate a significant decrease in peripheral EPCs and CD8+ T cells, three days following 23 an AMI. EPCs appear to be functionally impaired in AMI patients, and their circulating numbers associate 24 with cardiac vessel lesions. Furthermore, CD8+ T cells (and even M1-macrophages) in the periphery, in 25 combination with the classical laboratory determinations, may serve as high accuracy biomarkers of AMI, potentially aiding to prevent worse AMI outcomes. macrophage infiltration into the myocardium and their subsequent activation to express pro-inflammatory cytokines and chemokines promoting acute cardiac 25 . In a previous study in vivo , using a murine model, we have shown that infiltration of M1-macrophages to the myocardium starts from day one and peaks at day five after AMI onset, while M2-macrophages gradually increase from day three and peak at day seven post-AMI onset 9 . However, the same was not observed when we induced AMI in obese mice (elevated glucose and cholesterol levels): macrophages (M1 and M2) and T cells (CD4+, CD8+ and Tregs) dramatically decreased by 3-fold in peripheral blood compared to the non-obese group immediately after the onset of myocardial infarction (unpublished data). This drop in macrophage and T cell subsets continued up to day three after AMI onset, gradually increasing afterward (unpublished data). Our ROC curve analysis showed that at day seven (but not day three) post-AMI M1-macrophages biomarker potential (in the context of AMI) was significant, while for CD8 T cells this was true at both three (AUC = 0.94) and seven (AUC = 0.86) days post- AMI onset. Several clinical and preclinical studies have shown that T cell recruitment to the ischemic myocardium starts immediately following the onset of AMI. In fact, this was clearly demonstrated in a porcine AMI model 26 . Remarkably, this is particularly true for CD8+ T cells in a murine AMI model, where their early emergence to the ischemic myocardium has been previously demonstrated 9,26 . Recently, Forteza et al. demonstrated that in STEMI patients, T cells experienced a sharp decrease within 24 hours post-PCI, indicting the early response of T cells to the aseptic necrotic myocardium 26 . Herein, we have appropriately dissected T cells for their down subsets, and subsequently as outcomes revealed CD8 cells to potentially be reliable biomarkers of AMI. The combination of CD8 cell absolute numbers and myocardium necrosis markers is


32
Acute myocardial infarction (AMI) is a pathological trigger that amplifies medullary hematopoiesis via 33 sympathetic nervous system or damage-associated molecular pattern (DAMP) signaling 1,2 . Early mobilization 34 and trafficking of stem cells to the site of injury seems to be important for an adequate cardiac repair response 1,2 .

35
However, in patients with concomitant diseases/comorbidities or other important risk factors such as advanced 36 age, the bone marrow stem cell niches become exhausted, potentially leading to worse clinical outcomes 3,4,5 . In

51
In the present study, we investigated the levels of peripheral blood EPCs (CD34+) and immune cell-

Subjects' Enrollment (Patients and Controls) 58
Eighteen patients admitted to the Tokai University Hospital (Kanagawa, Japan), from September 2016 to 59 December 2017, with the diagnosis of AMI were included in this study: 12 patients with ST-elevation 60 myocardial infarction (STEMI), two patients with non-ST elevation myocardial infarction (NSTEMI) and two 61 patients with unstable angina pectoris. All other patients with a history of chronic inflammatory disease and 62 under anti-inflammatory medication were excluded from this study. Myocardial infarction was defined by the 63 detection of "rise and fall" of cardiac troponin T (cTnT) and at least one of the following: symptoms of ischemia, 64 new ST-T changes, or development of pathological Q-waves in the electrocardiogram. Immediately after 65 hospitalization, patients underwent emergency coronary angiography and drug-eluting stents were deployed to 66 their culprit lesions. The percutaneous coronary intervention (PCI), performed via the radial artery, was guided 67 by optical coherence tomography (OCT) which allowed for the choice of stent size and length as well as proper 68 stent positioning, and the pathological assessment of acute coronary syndromes. The OCT was performed prior 69 to and post stent deployment. Non-culprit vessels were also screened using OCT to evaluate atherosclerotic pathophysiology. All patients received the standard AMI pharmacotherapy, including aspirin, P2Y 12 71 inhibitors, and low-molecular-weight heparin, according to usual hospital practice. Beta-blockers and 72 angiotensin-converting enzyme inhibitors / angiotensin 2 receptor blockers were initiated post PCI, unless 73 contraindicated, according to the AHA guidelines.

74
The control group consisted of sixteen healthy human volunteers aged under 60 years old, without overt 75 cardiovascular diseases and DM, and were screened at least twice in the Tokai University Hospital outpatient 76 clinic. Individuals with family history of dyslipidemia and AMI were not included in this study.

77
Clinical features were carefully recorded, including classical cardiovascular risk factors, age at diagnosis 78 of DM (and duration, when applicable), body mass index (BMI), and glycosylated hemoglobin A1c (HbA1c),

Reduction of EPCs is a Potential Early Biomarker of Cardiac Vessel-Lesions 130
Flow cytometry analysis revealed different patterns with respect to the peripheral blood cells of AMI

134
77.5 ± 2, P> 0.08, respectively). As shown in Fig. 1E, the lowest frequency of total CD34+ cells were seen in 135 the group of AMI patients, three days after AMI onset (AMI-D3). Furthermore, and as expected, at day seven 136 after AMI (AMI-D7) CD34+ frequency peaked to values similar or slightly higher (without statistical 137 significance) than those of the control group (Fig. 1E). To address which subsets of CD34+ cells were increased

CD8+ T Cells in Circulation Decrease Early After AMI 202
As shown in the illustration graph obtained using UMAP, T cells, and the relative distribution their subsets 203 undoubtedly declined in AMI patients compared with controls ( Fig. 3A-C). Additionally, this distribution also 204 slightly changed over time in AMI patients (Fig. 3A, B). Remarkably, total T cell frequency was significantly 205 lower in AMI-D3 and AMI-D7 patients compared to healthy controls ( Fig. 3D;

211
Furthermore, when a correlation rank test was applied to understand the relation between T cells and the cardiac 212 necrosis marker creatine kinase myocardial band (CK-MB), a tendential positive correlation was observed for 213 CD8+ T cells, three days after AMI onset ( Fig. 3H; r = 0.58; P < 0.054). However, this correlation was lost at 214 day seven post-AMI (Fig. 3I). Additionally, no relevant association was found for CD4+ T cells at both three-215 and seven-days post-AMI (Fig. 3H, I). Next, we addressed if CD8+ T cells could be used as an early AMI 216 predictor, considering their relevant association with the cardiac necrosis marker at day three post-AMI.

Reduction of EPCs Colony-Forming Capability in AMI Patients 234
The endothelial progenitor cells in vitro colony-forming assay (EPC-CFA) revealed that colony-forming 235 ability of cells from AMI patients collected 3-and 7-days post-AMI onset was tendentially lower compared 236 with that from healthy controls (Fig. 4A-B). This was true considering either primitive EPC-CFU or definitive 237 EPC-CFU (Fig. 4A-B). Particularly, the AMI-D3 group had the lowest DEPC-CFU (however non-significant 238 compared to the control condition; P < 0.05), indicating a potentially lower vasculogenic EPC number, in line 239 with the data obtained for CD34+ cells (Fig. 1E-F and Fig. 4B). This result may also suggest that EPCs were 240 functionally impaired in our patients with risk factors. Furthermore, there was an increasing trend of EPC colony 241 formation at day 7 post-AMI onset (compared with day 3), but without statistical relevance (Fig. 4B).   In our study, we observed a marked reduction in the viability of total PBMCs in AMI patients, both at 257 days three and seven after AMI onset, compared to healthy controls. Previous studies have revealed that AMI 258 patients show altered intracellular glutathione levels and an imbalance in the glutathione-system key enzymes 259 glutamyl transferase and glutathione-S-transferase 16 . This, in conjunction with glycated end-products and "bad" cholesterol accumulation, may aggravate mitochondrial DNA damage and lead to telomere shortening, 261 promoting PBMC senescence 16,17,19 ; our phenotype aligns with this idea. Furthermore, our correlation data 262 suggests that absolute EPC numbers inversely correlated with glycated albumin and MAGE, suggesting that 263 high glucose levels (and concomitant diseases or risk factors) may reduce EPCs in AMI patients (Fig. S1C).

264
Even the elevated frequency of immature EPCs we observed in AMI patients seven days after AMI onset was 265 unable to generate EPC colonies to the same extent as healthy individuals. This may suggest that EPCs from 266 AMI patients were functionally impaired. In our study, we focused on both primitive and definitive colonies.

267
While PEPC are defined as very immature and highly proliferative colonies, DEPC are believed to derive from 268 PEPC and represent cells prone to differentiation and promotion of vasculogenesis. Curiously, with respect to 269 EPCs from our AMI-patients, along with the lower potential of colony formation, we observed a prevalence of

299
Interestingly, this was mainly due to a significant increase in circulating M1-macrophages, seven days after 300 AMI onset. These observations may align with the notion that T cells (especially CD8+ cells) promote/need macrophage infiltration into the myocardium and their subsequent activation to express pro-inflammatory 302 cytokines and chemokines promoting acute cardiac inflammation 25 . In a previous study in vivo, using a murine 303 AMI model, we have shown that infiltration of M1-macrophages to the myocardium starts from day one and 304 peaks at day five after AMI onset, while M2-macrophages gradually increase from day three and peak at day 305 seven post-AMI onset 9 . However, the same was not observed when we induced AMI in obese mice (elevated  mortality, 6 months after AMI, and suggested the contribution of these specific CD8+ T cell subsets to acute 334 coronary events via their pro-inflammatory and high cytotoxic capacities 30,31 . In our study, other cell subsets 335 also increased in patients PB, 3 days after AMI onset (28% compared to controls). On the other hand, compared 336 to healthy individuals, T cell populations decreased more than 12% and 16%, three and seven days after AMI 337 onset, respectively (aligning with some abovementioned studies). Taken together, our results-consistent with 338 classical laboratory findings-indicate that mainly CD8+ T cells, but also M1-macrophage cell counts in PB 339 can be used as high accuracy biomarkers of AMI.

340
A few limitations of our study need to be acknowledged. First, our sample-size is relatively small, which 341 might have conditioned our statistical relevance, due to the intrinsic variations detected by flow cytometry analysis (method we used to quantify EPCs, macrophages, and T cell subsets). Second, patients who received 343 statins and anti-diabetic treatments prior to an AMI were not analyzed separately (precisely because of our small 344 sample-size), which may result in some type II errors: previous studies documented that DPP4-inhibitors and 345 statins may affect hematopoiesis, as well as EPC number, kinetics, and mobilization 32,33,34,35 . Indeed, a large 346 sample-size study is required to answer several critical points such as severe vs. mild MI, diabetes vs. non-347 diabetes, and gender differences. Nevertheless, our study attempts to show the significance of circulating blood 348 cell dynamics following an onset of MI (acute and subacute phases) as a biomarker of cardiac vessel injury and 349 myocardial injuries. Third, the average age of our control group was younger than the study groups, which may