{"paper_id":"1ab82cff-1fd6-47bb-84f3-88aea59e26e2","body_text":"Neutrophil arrest in myocardial capillaries drives hypoxia and impairs diastolic function in a \nmouse model of heart failure with preserved ejection fraction \n \n \n \nDavid M. Small 1*, Anne E. Buglione 1,2*, Nathaniel H. Allan -Rahill1, Tyler Locke1, Laila M. Abd Elmagid1, \nSoseh Hovasapian 1, Rachel Kim 1, Adina A. Mistry 1, Sofia A. Vaquerano 1, Chris B. Schaffer 1, Nozomi \nNishimura1 \n \n* co-first authors \n \nContact: nn62@cornell.edu \n \n \nAffiliation: \n1. Nancy E. and Peter C. Meinig School of Biomedical Engineering, Cornell University, Ithaca, New York, \nUnited States. \n2. College of Veterinary Medicine, Cornell University, Ithaca, New York, United States \n \n \n \nABSTRACT \nMyocardial blood flow deficits in heart failure with preserved ejection fraction (HFpEF) patients and related \nanimal models have been recognized for decades, but the underlying mechanisms and resulting \nconsequences for HFpEF pathogenesis remain poorly understood. Using intravital cardiac microcopy in a \n‘two-hit’ mouse model of HFpEF, we identified an increased number of neutrophils in capillaries  slow \nmoving orstalled, blocking blood flow, as compared to control mice. Administration of antibodies against the \nneutrophil marker Ly6G reduced the number of arrested neutrophils in myocardial capillaries, leading to \nboth a reduction in myocardial tissue hypoxia and improvement in diastolic function and exercise tolerance. \nThis study identifies a previously uncharacterized cellular mechanism that explains myocardial blood flow \ndeficits in mouse models of HFpEF, and demonstrates that improving myocardial blood flow improves heart \nfunction. Restoring myocardial perfusion by preventing neutrophil arrest in coronary capillaries may provide \na strategy for improving heart function in HFpEF patients. \n \n \nINTRODUCTION \n \n Heart failure with preserved ejection fraction ( HFpEF) is primarily characterized by an impaired \nfilling capacity of the heart (diastolic dysfunction), and presents with a complex interplay of factors including \nreduced cardiac reserve, endothelial dysfunction, ventricular stiffening, and hypertension.1 The majority of \ntherapies that have improved morbidity and mortality in heart failure with reduced ejection fraction (HFrEF) \nhave proven ineffective against HFpEF.2,3 Only one treatment for HFpEF has been shown to reduce the \nrisk of hospitalization, yet does not reduce mortality.4 Development of novel, rationally targeted therapeutic \napproaches in HFpEF mandates the development of a more complete understanding of alterations in \ncellular homeostasis occurring in this condition.  \nMultiple studies show that blood flow is decreased in myocardial tissue in HFpEF even in the \nabsence of major arterial obstruction .5-9 The cause of this  myocardial perfusion deficit  is still unknown. \nReduced coronary flow reserve, defined as the ratio of coronary flow before and after stimulation, is \ncommon in HFpEF patients.5-7 Additionally, studies in HFpEF patients without obstructive coronary artery \ndisease demonstrate increased microvascular resistance under stimulation, 8 decreased volumetric flow,5 \nand reduced myocardial oxygen supply with exercise compared to healthy controls. 9 Structural loss of \ncapillaries via rarefaction is noted in patients10 and in animal models11 of HFpEF, and is thought to be related \nto inflammation.12 The microvasculature is often referred to as a likely culprit, although specifics are unclear. \nHFpEF is thought to have differing etiology than HFrEF, with a greater contribution of systemic \ninflammation and microvascular dysfunction driven by cardiovascular and metabolic risk factors which \ncontribute to low grade, chronic, systemic inflammation.13,14 It is not clear how inflammation leads to HFpEF, \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted February 22, 2025. ; https://doi.org/10.1101/2025.02.17.638717doi: bioRxiv preprint \n\nbut multiple lines of evidence suggest a contribution of dysfunction of the coronary microvasculature. 15 \nEndothelial activation is increased in HFpEF, indicated by the expression of vascular adhesion molecules \nand decreased bioavailability of nitric oxide (NO). 8,16-18 Evidence supports immune and inflammatory cell \ninfiltration of the heart, including neutrophils, in human and animal HFpEF19 and HFrEF studies20 of post-\nmortem tissue. However, the behavior of these cells within the cardiac microcirculation and how their \nrecruitment impacts coronary microcirculatory dynamics have not been defined.  \nThe convergence of the coronary microcirculation, inflammatory cells, and cardiomyocyte function \nin HFpEF phenotype requires in vivo study. Intravital microscopy is the uniquely able to show the dynamic \ninteractions of circulating leukocytes within the microvasculature during inflammation.21-25 Recent technical \nadvances have overcome the unique set of challenges associated with imaging the live beating heart in \nmouse models of disease.26-29,30 This enables direct visualization of dynamic processes such as blood flow, \nwhich are impossible to study in cultured or extracted tissues . These imaging capabilities provide a link \nbetween microvascular dysfunction, inflammation, and reduced blood flow in the HFpEF heart. Using \nintravital cardiac MPM in mouse models of heart failure, we show that neutrophils arrest in coronary \ncapillaries to reduce microvascular blood flow , contributing to cardiac remodeling and dysfunction. The \nelimination of these stalled capillaries drives rapid improvement of hypoxia and diastolic stiffness, even \nwithout altering hypertrophy. \n \n \nRESULTS \n \n Two-photon intravital imaging shows stalled capillaries plugged by neutrophils in HFpEF. Male \nand female C57Bl/6 mice were subjected to the HFpEF induction protocol of a high-fat diet and L-NAME in \ndrinking water for 15 weeks  (‘HFpEF’) and compared to controls that remained on the normal chow diet \nand water. HFpEF mice developed hypertrophy, as measured by the ratio of heart weight to tibia length,  \nwhile left ventricular ( LV) ejection fraction (Fig. 1 a-b) was unchanged as previously reported \n(Supplementary Figure 1). (Supplementary Figure 1). To investigate the interplay between \nmicrovasculature, leukocyte activity, and myocardial capillary blood flow in HFpEF , we used intravital \ncardiac two-photon microscopy to image the left ventricular epicardial microvasculature 26,27,3 Plasma was \nlabeled with retro-orbital injection of dextran-conjugated dyes or quantum dots. Erythrocytes and leukocytes \nare visible as regions of excluded labeling, which appear as dark patches within the vessel. Use of  \nCatchupIVM-Red  mice, which express TdTomato on the surface of neutrophils,31 and heterozygous knock-in \nmice (Cx3Cr1GFP/+ x CCR2RFP/+)30,31, which express green and red fluorescent proteins (GFP and RFP) in \nmonocytes and macrophages, allowed identification of specific classes of leukocytes. Labeled neutrophils \nand monocytes within capillaries were manually tracked in image stacks (100 frames x 30 fps from ~20 to \n200 µm below the left ventricle surface with 2 µm steps). Due to the periodic motion from heartbeat and \nbreathing at approximately 5-6 Hz and 1.6 Hz, capillaries appear repeatedly in the imaging stack and are \nrecognizable over about 1 -2 minutes, allowing the visualization of neutrophils and monocytes as they \ntransited through the capillary beds. While the majority of neutrophils and monocytes were visible for less \nthan ~2 s, many cells took longer to traverse the imaging region. These stalled neutrophils and monocytes  \nappeared in the same location over multiple image frames (Fig. 1c-d). \nIn Catchup mice, we measured the time each neutrophil was visible within the image stack (Fig. \n1e). To reduce the variability due to the differing amount of time a particular capillary is visible in the image \nvolume we sampled over multiple image stacks per mouse. The distribution of times  differed between \nHFpEF and control with a greater proportion of neutrophils in HFpEF persisting for longer times (p<0.0001 \nKolmogorov–Smirnov (KS) test, Fig. 1e). The total amount of time that a neutrophil was present in the image \nstack and the visible time per cell were increased by 1.7x and 3x (Fig. 1f and 1g) in HFpEF relative to \ncontrol mice. Similar experiments using Cx3Cr1GFP/+ x CCR2 RFP/+ mice with labeled monocytes 22,30,32 \nrevealed no difference between the monocytes of HFpEF and control in the distribution of time in the image \nstack (p= 0.73 KS test, Fig. 1f). Cells expressing only CCR2-RFP were exceedingly rare (<1/min) and the \nobserved circulating cells predominantly expressed both Cx3Cr1-GFP and CCR2-RFP, so we tracked cells \nexpressing GFP as a measure of monocytes . Observed circulating monocytes were on average less \ncommon than neutrophils, slower, and associated with a longer dwell time, regardless of treatment group \n(Figure 1g-h).  The frequency of slow-moving leukocytes was similarly elevated in 26 week-old ApoE-/- mice \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted February 22, 2025. ; https://doi.org/10.1101/2025.02.17.638717doi: bioRxiv preprint \n\nfed a HFD for 20 weeks and we found the number of stalled capillaries was doubled compared to control \n(age-matched wildtype mice on standard diet (Supplementary Figure 2 and 3)).  \nFigure 1. \nMouse models of heart failure with preserved ejection fraction (HFpEF) exhibited increased neutrophil \nresidence in myocardial capillaries compared to control animals in C57Bl/6 mice. \n(a) Schematic of HFpEF induction protocol (top). Example images (bottom left) of extracted hearts and plotted \n(bottom right) heart weight (HW) to tibial length (TL) ratios in control and HFpEF animals, showing cardiac \nhypertrophy \n(b) Example echocardiography (left) and plots showing diastolic function (E/e’) (top right) and left ventricular \nejection fraction (bottom right) in HFpEF and control mice.  Graphs include mice with labeled neutrophils or \nmonocytes, and wild type littermates.  \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted February 22, 2025. ; https://doi.org/10.1101/2025.02.17.638717doi: bioRxiv preprint \n\n(c) In vivo, two-photon (2P) microscopy images of the ventricular wall in “Catchup” mice expressing tdTomato \nwith the Ly6G promoter, a neutrophil surface marker. Intravascular FITC dextran in blood plasma (green), \nLy6G+ neutrophils (red). Scale bar = 100 µm. \n(d) 2P microscopy in mice expressing red fluorescent protein driven by the CCR2 promoter and green \nfluorescent protein with the Cx3Cr1 promoter in monocytes. Blood plasma labeled with quantum dots \n(Qtrack. white). Scale bar = 100 µm. \n(e) Distributions of time spent in visible region of image stack by labeled neutrophils visulalized by Ly6G -\ntdTomato, Kolmorov-Smirnov test. \n(f) Accumulated time in which a neutrophil or monocyte was visible in image stack summed over all cells in \neach stack visualized by Cx3Cr1-GFPm Kolmorov-Smirnov test.  \n(g) Average time in visible region for individual neutrophils or (h) monocytes. \nControl/neutrophils n = 6 mice, HFpEF/neutrophils n = 4, Control/monocytes n = 4, HFpEF/monocytes n = 6. \nEach dot in (e-g) represents an image stack. Indicated p-values are for Student’s t-test unless otherwise \nnoted. \n \n \nWe focused on neutrophil stalls because of the larger accumulated time in capillaries compared to \nmonocytes and clear effect of HFpEF on the distribution of visible cells. Although the average visible time \nduration of each individual monocyte was greater than that of an individual neutrophil, monocyte stalls \noccurred less frequently. The number of neutrophils visible for ≥ 100 sec onds, or “stalled”, doubled in \nHFpEF versus control (Fig. 2a).  Correspondingly, the incidence of “flowing” neutrophils with times < 5s was \nlower in HFpEF mice because HFpEF transit times were shifted towards slower time (Fig. 2b-c).  Although \nthere are reports of loss of capillaries in HFpEF, we found total capillary length measured in z-projections \nspanning 20-µm of the imaging volumes to be similar in HFpEF and control (Fig 2 d-e). Total neutrophil \nnumber per time normalized by capillary lengths were also similar in HFpEF and control (Fig. 2f). The total \ntime for which neutrophils were visible was greater in HFpEF than controls and there were fewer cells per \nminute, suggesting HFpEF neutrophils obstruct capillaries for longer durations than in healthy mice. On the \nwhole, neutrophils were stalled and slowed in capillaries more in HFpEF than in control mice One interstitial \nneutrophil was identified during in vivo imaging out of all neutrophils counted. To further identify interstitial \nneutrophils, we collected tissues and used immunolabeling against tdTomatoand quantified neutrophils in \nventricular tissue sections. Cryoinfarction was used as a positive control. While there was a trend towards \nan increase in the number of total and interstitial neutrophils  HFpEF mice, this elevation is negligible \ncompared to the number of interstitial neutrophils adjacent to an infarction29,33-36 (Fig. 2g-i). \n \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted February 22, 2025. ; https://doi.org/10.1101/2025.02.17.638717doi: bioRxiv preprint \n\n \nFigure 2. Characterization of neutrophils in capillaries of Catchup mice expressing tdTomato in \nneutrophils with HFpEF. \n(a) Number of neutrophils per time visible in two-photon microscopy image stacks for 100 s or longer \n(categorized as long stalls) or (b) visible for <5s (categorized as flowing).  \n(c)   Distributions of time visible of flowing neutrophils.  \n(d)   Capillaries were traced in 20 µm projections of images from two photon image stacks to calculate (e) \nsummed linear length of capillaries.  \n(f)    Incidence of neutrophils observed in an image stack normalized by total capillary length. \n(g)   Example images of neutrophils imaged in sectioned tissue detected by an antibody against tdTomato (anti-\ntdTom). Vessel wall labeled with isolectin-B4 and nuclei with DAPI. Neutrophils were categorized as vascular \nwhen they were adjacent to isolectinB4 labeled vasculature and as interstitial when no contact was observed   \n(h)   Left ventricular tissue collected 7-days after a cyroinfarct served as a positive control for histological \nneutrophil labeling.  \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted February 22, 2025. ; https://doi.org/10.1101/2025.02.17.638717doi: bioRxiv preprint \n\n(i) Quantification of all visible, vascular and interstitial neutrophils with anti-tdTom in tissue sections in control and \nHFpEF mice. \n \nHypoxia, diastolic function, and fibrosis are rescued by depleting neutrophils for 4 weeks in late \nstages of HFpEF. We tested the effects of long-term removal of stalls via neutrophil depletion (a-Ly6G, 2-\nevery 3 days)37 at late stages of the disease (last 4 weeks of the 15-week HFpEF protocol) and compared \nto isotype antibodies (Iso) as control with same timing and dosage ( Fig. 3, control in Supplementary Fig. \n4). a-Ly6G treatment did not improve hypertrophy, as assayed by heart weight/tibia length, left ventricular \nejection fraction and indicators of pulmonary edema compared to control (Fig. 3c-e). However, diastolic \nfunction measured by E/e’, recovered to levels near that of sham animals (Fig. 3f), while a-Ly6G treatment \nin control animals had no impact . Myocardial hypoxia, measured by the pimo nidazole immunoassay, \nincreased in HFpEF hearts to 150% of control (Figure 3g; Supplementary Figure 4). We found that 4-week \ntreatment with a-Ly6G reduced hypoxia compared to control treatment and nearly attained the levels of \nhealthy mice (Fig. 3g-h). Collagen, as assessed by Masson’s Trichrome, was increased in HFpEF animals \nrelative to control, and reduced with a-Ly6G, although not quite to the level of control (Fig. 3i-j).  \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted February 22, 2025. ; https://doi.org/10.1101/2025.02.17.638717doi: bioRxiv preprint \n\n \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted February 22, 2025. ; https://doi.org/10.1101/2025.02.17.638717doi: bioRxiv preprint \n\nFigure 3. Extended neutrophil depletion improves diastolic function and fibrosis, but not hypertrophy.  \n(a) Animals received high fat diet (HFD) and L-NAME for HFpEF induction (HFpEF), while control animals received \nnormal chow and water ( healthy control) for 15 weeks . Starting at week 11, animals were injected with \nantibodies against Ly6G (a-Ly6G) or a nonspecific isotype control antibody (Iso) every three days until imaging \nand tissue collection. \n(b) Animal weights starting from initiation of HFpEF or control protocols.  \n(c) Heart weight and (d) dry lung weight normalized to tibial length at the end of treatment period HFpEF and Chow \nmice treated with a-Ly6G or Iso antibodies.  \n(e) Echocardiography for left ventricular (LV) ejection fraction and (f) mitral E/e’ at the end of treatment time  for \neach group.   \n(g) Example fluorescence microscopy of sections from animals injected with a marker of hypoxia (pimonidazole) \nand h) quantification of labeling with antibodies against pimonidazole.  \n(h) Masson’s Trichrome staining in tissue sections showing collagen with j) quantification of labeling at the end of \nthe treatment period.   \n(Student’s t-test, * p<0.05, ** p<0.005, ***p<0.00050 \n \nAcute neutrophil depletion in late stage HFpEF rescues hypoxia, diastolic function, and reduces \nstalled capillaries. We tested the effects of acute depletion for only 1-day at an advanced stage of HFpEF. \nUsing flow cytometry, we found that aLy6G (4 mg kg-1, Fig. 4a) led to a 54% reduction in systemic circulating \nneutrophils by 24h compared to mice administered an isotype control antibody (Figure 4b-c). We took \nadvantage of this rapidity to isolate the effects of neutrophil depletion from possible slower changes at the \nend of the 15-week HFpEF induction (or control) in Catchup mice. Repeated echocardiography before and \n1 day after aLy6G treatment revealed that treatment reduced E/e’ in HFpEF to the same value as in control \nmice, while HFpEF mice receiving isotype and Chow mice receiving aLy6G did not change (Fig. 4d). \nMyocardial hypoxia, measured by the pimonidazole immunoassay, increased by 1.5 fold in HFpEF hearts \ncompared to controls (Figure 4e-f). One day following aLy6G administration, pimonidazole labeling \ndecreased in HFpEF mice compared to control treatment , approaching the levels found in mice without \nHFpEF induction. Intravital cardiac two-photon microscopy in Catchup mice with 15-week HFpEF induction \nshowed that 1 day after aLy6G the rate of neutrophils appearing in image stacks was reduced to less than \nhalf the rate with the isotype antibody (Fig. 4h).  \n \nFigure 4. Acute, 1 -day neutrophil depletion improves diastolic function and hypoxia and reduces \nneutrophil obstruction in capillaries.  \n(a) Animals received high fat diet and L-NAME for HFpEF induction (HFpEF), while control animals received \nnormal chow and water for 15 weeks. One day before imaging and tissue collection animals were injected \nwith antibodies against Ly6G (aLy6G ) or an isotope, nonspecific antibody as control (Iso). (n = 3-4 per group) \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted February 22, 2025. ; https://doi.org/10.1101/2025.02.17.638717doi: bioRxiv preprint \n\n(b) Flow cytometry gated on tdTomato (tdTom) was used to (c) quantify depletion of circulating neutrophils in \nblood from Catchup mice after 1 day of treatment with anti-Ly6G or isotype antibodies. \n(d) Echocardiography performed before and 1 day after anti -Ly6G or isotype control in HFpEF and healthy \nmice to measure diastolic function with E/e’. Paired t-tests, healthy: n = 6, HFpEF-aly6G: n = 7, HFpEF-Iso: n \n= 6.  \n(f) Example fluorescence microscopy of sections from animals injected with a marker of hypoxia \n(pimonidazole) and (g) quantification of labeling with antibodies against pimonidazole.  \n(g) In vivo two-photon microscopy of Catchup mice was used to quantify the incidence of tdTomato-expressing \n(tdTom) neutrophils and (h) the sum of times a neutrophil was visible in image stacks 1 day after anti-Ly6G or \nisotype injection.  (n = 5-6 mice). Each dot represents an image stack. \n \n \nDISCUSSION \n \n Capillary occlusions drive hypoxia in HFpEF. While many have speculated that microvascular \ndysfunction contributes to the heart failure, the mechanism has been elusive in part because it was difficult \nto visualize. Recently developed intravital two -photon microscopy enables imaging of blood flow in \nmyocardial capillaries. In a mouse model of HFpEF, this imaging revealed that a subset of capillaries were \ntransiently occluded primarily by arrested neutrophils that stall or slow blood flow (Fig. 1). This dynamic \nphenomenon was not reflected in histological counts of neutrophils in vessels and involves a tiny fraction \nof the number of neutrophils recruited to tissues after acute injuries such as infarctions (Fig. 2), explaining \nwhy it may not have been recognized previously. In HFpEF, the improvement of a marker of hypoxia just \n24 hours after a single treatment of anti -Ly6G (Fig. 4) suggests that the presence of stalled and slowed \nneutrophils is detrimental.  \n \nDiastolic function is rescued independently of remodeling and hypertrophy. Diastolic \nfunction also improved with the acute, 1-day treatment to reduce neutrophil plugging of capillaries, \nproviding benefits similar in magnitude to an extended 4-week treatment. Both long and acute treatment \nimproved E/e’ to the levels of healthy, sham-treatment mice, but the 4-week treatment also had an \nadditional benefit of reducing fibrosis, as measured by collagen staining.  Although conventional wisdom \nsuggests that remodeling and the accumulation of fibrosis drives the inability to relax by increasing tissue \nstiffness, the success of the acute treatment suggests a myocardial relaxation mechanism related to \nblood flow improvement. One possibility that relates diastolic stiffness to hypoxia on a fast time scale is \nthe fact that low oxygenation can impair the conversion from ADP to ATP and high ADP levels can impair \ncrossbridge detachment, prevent acti-myosin unbinding38. Reminiscent of the results here, we recently \ndiscovered using similar two-photon microscopy to this work, that a tiny fraction of capillaries (1-2%) in \nthe brains of Alzheimer’s mouse models are stalled by arrested neutrophils, and this leads to a 25% \ndecrement in brain blood flow39. This flow reduction was sufficient to contribute to cognitive dysfunction, \ndemonstrated by the extremely rapid (3 hrs) onset39 of short-term memory rescue caused by reducing the \ncapillary stalls with a-Ly6G (same as used in this study). In both the Alzheimer’s and HFpEF models, \nlong-term treatment even at late stages resulted in sustained functional improvement 40,41. The two \nparallel results suggests that the cumulative effects of distributed microscale occlusions can have large \nimpacts on overall organ function and that at least some of effects can be rapidly reversed without \nrestructuring of tissue. Blood flow changes from capillary dysfunction is sufficient to drive disease \nsymptoms.  \n \nEvidence from patients and experimental work provide support for neutrophil plugging of \ncapillaries. Patient samples show elevated expression of the adhesion markers ICAM1, VCAM1, and E-\nselectin16,17,42, which could contribute to the arrest and slow flow of white blood cells in myocardial \ncapillaries. HFpEF patient samples also show capillary rarefication12, suggesting endothelial cell damage \nthat could drive white cell arrest43. Molecular methods implicate reactive oxygen species (ROS) and \ndownstream signaling in cardiomyocytes and fibroblasts15,44. ROS also affect endothelial cells and drive \nincreased inflammation as seen through increased expression of adhesion molecules17,42 and changes in \nbarrier function45.  In HFpEF patients, the neutrophil to lymphocyte ratio, a marker of inflammation, \ncorrelated with poor prognosis46. Myeloperoxidase (MPO) levels were increased in HFpEF patients, also \nsuggesting an involvement of neutrophils47. Neutrophil depletion can have deleterious consequences \nafter a myocardial infarction48, suggesting that after acute injury, neutrophil \"clean up\" roles may be \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted February 22, 2025. ; https://doi.org/10.1101/2025.02.17.638717doi: bioRxiv preprint \n\ncritical49. However, it has also been shown that depleting neutrophils in the pressure overload model of \nHFrEF can prevent heart failure20. The link between neutrophils and fibrosis is supported by additional \nwork in mice. Zhang, et al. found similarly improved E/e’ and fibrosis without rescue of myocardial size in \na different model of HFpEF after SGLT2 inhibition with dapagliflozin50. They implicated neutrophils in this \nstudy, by interfering with neutrophil extracellular traps (NETs), inhibition of high mobility group protein B1 \n(HMGB1), and by directly degrading NETs with deoxyribonuclease 1 and finding the same changes.  \n \nFuture work. Follow up studies exploring the idea that perfusion deficits caused by capillary \nstalls can drive key symptoms are needed to address some of the limitations in this study. While a-Ly6G \nis useful as experimental tool, because it is not a human protein, further exploration for better translational \ntargets is critical. While there is some concern that depletion by this antibody was not a complete \nelimination of all types of neutrophils51, our results in both the heart and the Alzheimer’s disease brain39-41 \nshow substantial improvements in multiple physiological parameters suggesting that this manipulation is \nsufficient to motivate further work on better strategies. A similar phenomenon of stalled and slowed \ncapillaries in an alternative model of heart failure ApoE-/- with high fat diet, suggests that this phenomenon \nis not an artifact of the HFpEF model (Supplementary Figs. 2-3). We did not find notable sex dependence \nin the incidence of neutrophil stalls or recovery with depletion, but this study was not powered to resolve \nsex differences.  Further study, especially using menopause models for females, will be critical in \nassessing the translational importance of these results.  \n \n \nCONCLUSION  \nThis work points to a novel mechanism implicating capillaries that drives the symptoms of \ndiastolic dysfunction in HFpEF. Elimination of neutrophils that obstructed capillaries decreased hypoxia, \nprevented fibrosis, and decreased ventricular stiffness to improve diastolic function. Because neutrophil \nadherence to endothelium is the first step in inflammation, such capillary stalling with little neutrophil \ntissue extravasation is likely in many diseases, especially those involving chronic inflammation. While this \nsuggests new targets for future drugs, it also is possible that many drugs either directly or indirectly \nimprove capillary stalling so that the therapeutic effect is largely pleiotropic.  \n \n \nMETHODS \nSee separate online methods \n \n \n \nREFERENCES \n \n \n1 Borlaug, B. A. The pathophysiology of heart failure with preserved ejection fraction. Nat Rev \nCardiol 11, 507-515, doi:10.1038/nrcardio.2014.83 (2014). \n2 Massie, B. M. et al. Irbesartan in patients with heart failure and preserved ejection fraction. N \nEngl J Med 359, 2456-2467, doi:10.1056/NEJMoa0805450 (2008). \n3 Yusuf, S. et al. Effects of candesartan in patients with chronic heart failure and preserved left-\nventricular ejection fraction: the CHARM-Preserved Trial. Lancet 362, 777-781, \ndoi:10.1016/S0140-6736(03)14285-7 (2003). \n4 Anker, S. D. et al. Empagliflozin in Heart Failure with a Preserved Ejection Fraction. The New \nEngland journal of medicine 385, 1451-1461, doi:10.1056/NEJMoa2107038 (2021). \n5 Kato, S. et al. Impairment of Coronary Flow Reserve Evaluated by Phase Contrast Cine-Magnetic \nResonance Imaging in Patients With Heart Failure With Preserved Ejection Fraction. J Am Heart \nAssoc 5, doi:10.1161/JAHA.115.002649 (2016). \n6 Srivaratharajah, K. et al. Reduced Myocardial Flow in Heart Failure Patients With Preserved \nEjection Fraction. Circ Heart Fail 9, doi:10.1161/CIRCHEARTFAILURE.115.002562 (2016). \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted February 22, 2025. ; https://doi.org/10.1101/2025.02.17.638717doi: bioRxiv preprint \n\n7 Shah, S. J. et al. Prevalence and correlates of coronary microvascular dysfunction in heart failure \nwith preserved ejection fraction: PROMIS-HFpEF. European heart journal 39, 3439-3450, \ndoi:10.1093/eurheartj/ehy531 (2018). \n8 Dryer, K. et al. Coronary microvascular dysfunction in patients with heart failure with preserved \nejection fraction. American journal of physiology. Heart and circulatory physiology 314, H1033-\nH1042, doi:10.1152/ajpheart.00680.2017 (2018). \n9 Obokata, M. et al. Myocardial Injury and Cardiac Reserve in Patients With Heart Failure and \nPreserved Ejection Fraction. J Am Coll Cardiol 72, 29-40, doi:10.1016/j.jacc.2018.04.039 (2018). \n10 Mohammed, S. F. et al. Coronary microvascular rarefaction and myocardial fibrosis in heart \nfailure with preserved ejection fraction. Circulation 131, 550-559, \ndoi:10.1161/CIRCULATIONAHA.114.009625 (2015). \n11 Schiattarella, G. G. et al. Nitrosative stress drives heart failure with preserved ejection fraction. \nNature 568, 351-356, doi:10.1038/s41586-019-1100-z (2019). \n12 Zeng, H. & Chen, J. X. Microvascular Rarefaction and Heart Failure With Preserved Ejection \nFraction. Front Cardiovasc Med 6, 15, doi:10.3389/fcvm.2019.00015 (2019). \n13 AbouEzzeddine, O. F. et al. Myocardial Energetics in Heart Failure With Preserved Ejection \nFraction. Circ Heart Fail 12, e006240, doi:10.1161/CIRCHEARTFAILURE.119.006240 (2019). \n14 Lyle, M. A., Alabdaljabar, M. S., Han, Y. S. & Brozovich, F. V. The vasculature in HFpEF vs \nHFrEF: differences in contractile protein expression produce distinct phenotypes. Heliyon 6, \ne03129, doi:10.1016/j.heliyon.2019.e03129 (2020). \n15 D'Amario, D. et al. Microvascular Dysfunction in Heart Failure With Preserved Ejection Fraction. \nFront Physiol 10, 1347, doi:10.3389/fphys.2019.01347 (2019). \n16 Franssen, C. et al. Myocardial Microvascular Inflammatory Endothelial Activation in Heart Failure \nWith Preserved Ejection Fraction. JACC Heart Fail 4, 312-324, doi:10.1016/j.jchf.2015.10.007 \n(2016). \n17 Westermann, D. et al. Cardiac inflammation contributes to changes in the extracellular matrix in \npatients with heart failure and normal ejection fraction. Circ Heart Fail 4, 44-52, \ndoi:10.1161/CIRCHEARTFAILURE.109.931451 (2011). \n18 Paulus, W. J. & Tschope, C. A novel paradigm for heart failure with preserved ejection fraction: \ncomorbidities drive myocardial dysfunction and remodeling through coronary microvascular \nendothelial inflammation. J Am Coll Cardiol 62, 263-271, doi:10.1016/j.jacc.2013.02.092 (2013). \n19 Hulsmans, M. et al. Cardiac macrophages promote diastolic dysfunction. J Exp Med 215, 423-\n440, doi:10.1084/jem.20171274 (2018). \n20 Wang, Y. et al. Wnt5a-Mediated Neutrophil Recruitment Has an Obligatory Role in Pressure \nOverload-Induced Cardiac Dysfunction. Circulation 140, 487-499, \ndoi:10.1161/CIRCULATIONAHA.118.038820 (2019). \n21 Devi, S. et al. Multiphoton imaging reveals a new leukocyte recruitment paradigm in the \nglomerulus. Nat Med 19, 107-112, doi:10.1038/nm.3024 (2013). \n22 Ahn, S. J., Anrather, J., Nishimura, N. & Schaffer, C. B. Diverse Inflammatory Response After \nCerebral Microbleeds Includes Coordinated Microglial Migration and Proliferation. Stroke; a \njournal of cerebral circulation 49, 1719-1726, doi:10.1161/STROKEAHA.117.020461 (2018). \n23 McDonald, B. et al. Intravascular danger signals guide neutrophils to sites of sterile inflammation. \nScience 330, 362-366, doi:10.1126/science.1195491 (2010). \n24 Cruz Hernandez, J. C. et al. Neutrophil adhesion in brain capillaries reduces cortical blood flow \nand impairs memory function in Alzheimer's disease mouse models. Nat Neurosci 22, 413-420, \ndoi:10.1038/s41593-018-0329-4 (2019). \n25 Neupane, A. S. & Kubes, P. Imaging reveals novel innate immune responses in lung, liver, and \nbeyond. Immunol Rev 306, 244-257, doi:10.1111/imr.13040 (2022). \n26 Allan-Rahill, N. H., Lamont, M. R. E., Chilian, W. M., Nishimura, N. & Small, D. M. Intravital \nMicroscopy of the Beating Murine Heart to Understand Cardiac Leukocyte Dynamics. Front \nImmunol 11, 92, doi:10.3389/fimmu.2020.00092 (2020). \n27 Jones, J. S., Small, D. M. & Nishimura, N. In Vivo Calcium Imaging of Cardiomyocytes in the \nBeating Mouse Heart With Multiphoton Microscopy. Front Physiol 9, 969, \ndoi:10.3389/fphys.2018.00969 (2018). \n28 Lee, S. et al. Real-time in vivo imaging of the beating mouse heart at microscopic resolution. \nNature communications 3, 1054, doi:10.1038/ncomms2060 (2012). \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted February 22, 2025. ; https://doi.org/10.1101/2025.02.17.638717doi: bioRxiv preprint \n\n29 Kavanagh, D. P. J., Lokman, A. B., Neag, G., Colley, A. & Kalia, N. Imaging the injured beating \nheart intravitally and the vasculoprotection afforded by haematopoietic stem cells. Cardiovascular \nresearch 115, 1918-1932, doi:10.1093/cvr/cvz118 (2019). \n30 Jung, S. et al. Analysis of fractalkine receptor CX(3)CR1 function by targeted deletion and green \nfluorescent protein reporter gene insertion. Molecular and cellular biology 20, 4106-4114 (2000). \n31 Saederup, N. et al. Selective chemokine receptor usage by central nervous system myeloid cells \nin CCR2-red fluorescent protein knock-in mice. PloS one 5, e13693, \ndoi:10.1371/journal.pone.0013693 (2010). \n32 Auffray, C. et al. Monitoring of blood vessels and tissues by a population of monocytes with \npatrolling behavior. Science 317, 666-670, doi:10.1126/science.1142883 (2007). \n33 Bajpai, G. et al. Tissue Resident CCR2- and CCR2+ Cardiac Macrophages Differentially \nOrchestrate Monocyte Recruitment and Fate Specification Following Myocardial Injury. \nCirculation research 124, 263-278, doi:10.1161/CIRCRESAHA.118.314028 (2019). \n34 Jung, K. et al. Endoscopic time-lapse imaging of immune cells in infarcted mouse hearts. \nCirculation research 112, 891-899, doi:10.1161/CIRCRESAHA.111.300484 (2013). \n35 Li, W. et al. Intravital 2-photon imaging of leukocyte trafficking in beating heart. The Journal of \nclinical investigation 122, 2499-2508, doi:10.1172/JCI62970 (2012). \n36 Matsuura, R. et al. Intravital imaging with two-photon microscopy reveals cellular dynamics in the \nischeamia-reperfused rat heart. Scientific reports 8, doi:ARTN 15991 \n10.1038/s41598-018-34295-w (2018). \n37 Erdener, S. E. et al. Dynamic capillary stalls in reperfused ischemic penumbra contribute to injury: \nA hyperacute role for neutrophils in persistent traffic jams. Journal of cerebral blood flow and \nmetabolism : official journal of the International Society of Cerebral Blood Flow and Metabolism \n41, 236-252, doi:10.1177/0271678X20914179 (2021). \n38 Sequeira, V. et al. Synergistic role of ADP and Ca(2+) in diastolic myocardial stiffness. J Physiol \n593, 3899-3916, doi:10.1113/JP270354 (2015). \n39 Cruz Hernandez, J. C. et al. Neutrophil adhesion in brain capillaries reduces cortical blood flow \nand impairs memory function in Alzheimer's disease mouse models. Nature neuroscience, \ndoi:10.1038/s41593-018-0329-4 (2019). \n40 Ali, M. et al. Inhibition of peripheral VEGF signaling rapidly reduces leucocyte obstructions in \nbrain capillaries and increases cortical blood flow in an Alzheimer’s disease mouse model. \nbioRxiv, 2021.2003.2005.433976, doi:10.1101/2021.03.05.433976 (2021). \n41 Bracko, O. et al. Increasing cerebral blood flow improves cognition into late stages in Alzheimer's \ndisease mice. Journal of cerebral blood flow and metabolism : official journal of the International \nSociety of Cerebral Blood Flow and Metabolism 40, 1441-1452, doi:10.1177/0271678X19873658 \n(2020). \n42 van Heerebeek, L. et al. Diastolic stiffness of the failing diabetic heart: importance of fibrosis, \nadvanced glycation end products, and myocyte resting tension. Circulation 117, 43-51, \ndoi:10.1161/CIRCULATIONAHA.107.728550 (2008). \n43 Reeson, P., Choi, K. & Brown, C. E. VEGF signaling regulates the fate of obstructed capillaries in \nmouse cortex. Elife 7, doi:10.7554/eLife.33670 (2018). \n44 Camici, P. G., Tschope, C., Di Carli, M. F., Rimoldi, O. & Van Linthout, S. Coronary microvascular \ndysfunction in hypertrophy and heart failure. Cardiovascular research 116, 806-816, \ndoi:10.1093/cvr/cvaa023 (2020). \n45 Zhu, L. & He, P. fMLP-stimulated release of reactive oxygen species from adherent leukocytes \nincreases microvessel permeability. American journal of physiology. Heart and circulatory \nphysiology 290, H365-372, doi:10.1152/ajpheart.00812.2005 (2006). \n46 Boralkar, K. A. et al. Value of Neutrophil to Lymphocyte Ratio and Its Trajectory in Patients \nHospitalized With Acute Heart Failure and Preserved Ejection Fraction. The American journal of \ncardiology 125, 229-235, doi:10.1016/j.amjcard.2019.10.020 (2020). \n47 Hage, C. et al. Myeloperoxidase and related biomarkers are suggestive footprints of endothelial \nmicrovascular inflammation in HFpEF patients. ESC Heart Fail 7, 1534-1546, \ndoi:10.1002/ehf2.12700 (2020). \n48 Kain, V. & Halade, G. V. Role of neutrophils in ischemic heart failure. Pharmacol Ther 205, \n107424, doi:10.1016/j.pharmthera.2019.107424 (2020). \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted February 22, 2025. ; https://doi.org/10.1101/2025.02.17.638717doi: bioRxiv preprint \n\n49 Kolaczkowska, E. & Kubes, P. Neutrophil recruitment and function in health and inflammation. Nat \nRev Immunol 13, 159-175, doi:10.1038/nri3399 (2013). \n50 Zhang, X. L. et al. HMGB1-Promoted Neutrophil Extracellular Traps Contribute to Cardiac \nDiastolic Dysfunction in Mice. J Am Heart Assoc 11, e023800, doi:10.1161/JAHA.121.023800 \n(2022). \n51 Iliakis, C. S. & Wack, A. Never trust a single myeloid marker: Ly6G on repair-promoting lung \nmacrophages. Sci Immunol 9, eadq7306, doi:10.1126/sciimmunol.adq7306 (2024). \n \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted February 22, 2025. ; https://doi.org/10.1101/2025.02.17.638717doi: bioRxiv preprint","source_license":"CC-BY-4.0","license_restricted":false}