Liposomal delivery of azithromycin enhances its immunotherapeutic efficacy and reduces toxicity in myocardial infarction

A growing body of evidence shows that altering the inflammatory response by alternative macrophage polarization is protective against complications related to acute myocardial infarction (MI). We have previously shown that oral azithromycin (AZM), initiated prior to MI, reduces inflammation and its negative sequelae on the myocardium. Here, we investigated the immunomodulatory role of a liposomal AZM formulation (L-AZM) in a clinically relevant model to enhance its therapeutic potency and avoid off-target effects. L-AZM (40 or 10 mg/kg, IV) was administered immediately post-MI and compared to free AZM (F-AZM). L-AZM reduced cardiac toxicity and associated mortality by 50% in mice. We observed a significant shift favoring reparatory/anti-inflammatory macrophages with L-AZM formulation. L-AZM use resulted in a remarkable decrease in cardiac inflammatory neutrophils and the infiltration of inflammatory monocytes. Immune cell modulation was associated with the downregulation of pro-inflammatory genes and the upregulation of anti-inflammatory genes. The immunomodulatory effects of L-AZM were associated with a reduction in cardiac cell death and scar size as well as enhanced angiogenesis. Overall, L-AZM use enhanced cardiac recovery and survival after MI. Importantly, L-AZM was protective from F-AZM cardiac off-target effects. We demonstrate that the liposomal formulation of AZM enhances the drug’s efficacy and safety in an animal model of acute myocardial injury. This is the first study to establish the immunomodulatory properties of liposomal AZM formulations. Our findings strongly support clinical trials using L-AZM as a novel and clinically relevant therapeutic target to improve cardiac recovery and reduce heart failure post-MI in humans.

Liposomal formulations were prepared according to a previously established protocol 6,7 , using the thin film hydration method with equal molar ratios of distearyl phosphatidylcholine (DSPC), distearyl phosphatidylglycerol (DSPG) and cholesterol (Avanti polar lipids). AZM was included at 10 and 30 mol% based on phospholipid content. Formulations containing fluorophores were prepared in the same manner with 0.5 mol% of the lipophilic dye, 1,1'-dioctadecyl-3,3,3',3'tetramethylindodicarbocyanine (DiD). Lipids and AZM were mixed in chloroform and methanol at the desired ratio and the organic solvents removed by rotary evaporation to yield a thin film. After being placed under a vacuum overnight, the lipid/drug films were hydrated in phosphate buffered saline (pH 7.4) and sonicated at 65°C for 30 min. The resulting unilamellar vesicles were extruded through 200 nm polycarbonate membranes to obtain uniform particle size. Liposome size, polydispersity and stability over time were determined by dynamic light scatter testing.

Flow cytometry
Peripheral blood (PB). Cell phenotype, in addition to gene analyses, of monocytes and neutrophils was examined in peripheral blood as previously described 5 . Briefly, blood was collected in tubes with a 1:5 ratio of ethylene diaminetetraacetic acid (EDTA)/citrate-theophyllineadenosinedipyridamole (CTAD) on days 1, 3 and 7 post-MI. Whole blood was centrifuged for 5 minutes at 500 x g, and the plasma layer was separated and reserved at -80 0 C. To lyse red blood cells, the residual cell pellet was incubated with 0.5 ml of diluted red blood lysing buffer (BD pharm lyse) for 10 minutes with mild agitation. To end the lysis process, 0.5 ml of staining buffer (5% goat serum, 0.05% sodium azide in phosphate-buffered saline) was added to the suspension and centrifuged at 400 x g for 5 minutes, and the supernatant removed. This step was repeated if the red blood cells were still detectable in the cell pellet. Cells were then washed twice in staining buffer to remove any residual lysis buffer. After supernatant removal, the pellet was resuspended in a pre-determined volume of staining buffer and counted. Cells for staining were incubated immediately with conjugated primary antibodies against PERCPCY5.5-conjugated Ly6G/C (BD Pharmingen), PECY7-conjugated F4/80 (Biolegend), brilliant violet 421-conjugated CD11b (Biolegend), APC-CY7-conjugated CD45 (Biolegend), and PE-conjugated CD115 (Biolegend) for 30 minutes on ice. Following incubation, cells were washed twice with staining buffer and analyzed by an LSR II (Becton Dickinson) in the University of Kentucky Flow Cytometry Core.
Using unstained cells and single fluorescent controls, laser calibration and compensations were performed for all experiments. Monocytes with the CD45 hi /CD115 hi /Ly6-C hi profile were identified as classical (pro-inflammatory), while cells with these markers CD45 hi /CD115 hi /Ly6-C lo were identified as non-classical (anti-inflammatory). Neutrophils were identified as CD45 hi /CD115 lo /Ly6-C/G lo .
Heart. Phenotypic cell analysis of macrophages and neutrophils and gene expression analyses waere conducted in the heart as previously described 5 . Briefly, mice were sacrificed, and hearts were rapidly isolated and placed in ice cold PBS (VWR International). Using a razor blade, the heart was minced manually. After mincing, tissue was incubated with a Collagenase B (Roche, Indianapolis, IN) and Dispase II (Roche, Indianapolis, IN) mixture at 37°C for 30 minutes, with gentle agitation every 5-10 minutes. Tissue digestion ceased using cold staining buffer and the suspension was placed on ice and filtered using a 100 μm cell strainer, followed by centrifugation at 400 x g for 5 mins at 4°C. Supernatant was discarded and the pellet was resuspended in 0. CD11c. Neutrophils were defined as CD45 hi /CD115 lo /Ly6-C/G lo and CD206 was used to further identify N1 neutrophils (CD206 lo ).

Histology
Histological analysis was performed on deparaffinized and rehydrated sections as previously described 5 . Briefly, at 30 days post-MI, mice (N = 6-10/treatment group) were sacrificed and hearts were isolated and perfused with PBS (VWR International) then by 10% buffered formalin (VWR International) at 75 mmHg. Perfused hearts were placed in 10% neutral buffered formalin (VWR) overnight at room temperature, then sectioned into 2-mm cross-sections starting at the level of coronary ligation then imbedded in paraffin and sectioned into 4-μm sections.
Sections were stained with Masson's trichrome to evaluate scar size as previously described 5 .
Digital images of stained sections were acquired, and areas of interest were assessed using NIH ImageJ (version 7) software. The LV area, LV cavity area, and infarct area were measured as previously described 5 . Scar size was presented as a percentage of the total LV volume.

Immunohistochemistry
Immunostaining of heart sections was performed on deparaffinized and rehydrated sections as previously described 5 . Briefly, after deparaffinization, rehydration, and antigen retrieval, sections were incubated with primary antibodies: rabbit anti-mouse IBA1 (Wako), goat Findings are presented as total number of positive cells per high power field in the peri-infract region. All measurements were obtained in the peri-infarcted areas only and analyzed by blinded observer.

Reverse Transcription Polymerase Chain Reaction (RT-PCR).
After RBC lysis (see above), aliquots of 1x10 6 cells were incubated with lysis buffer (Life technologies) for 10 minutes with multiple aggressive agitations, and then kept at -80 0 C for further gene expression analysis. PureLink RNA Mini Kit (ThermoFisher Scientific) was used to isolate total mRNA from heart and blood cell lysates according to manufacturer protocol. Isolated RNA was quantified using NanoDrop 8000 spectrophotometer (Thermofisher). Next, cDNA was generated using SuperScript VILO cDNA synthesis kit (Invitrogen). Using a QuantaStudio 7 Flex Appropriate negative control reactions (template free controls) were used, (c) Careful examination of the melting curve of augmented products (dissociation graphs) for consistency was performed, and (d) The probe Tm was at least 10°C more than the primer Tm, while the melting temperature (Tm) was 57°C-60°C.

Echocardiography
Given the fact that the left ventricular appearance in histological sections is affected by multiple technical issues such as perfusion pressure and imbedding techniques in paraffin, we opted to use echocardiography for all left ventricular internal diameter and volume measurements.
Mice were anaesthetized using 1%-3% isoflurane during echocardiography to maintain heart rate of 450-500 BPM during imaging. A Vevo 3100 system coupled with a 15-7-MHz linear broadband transducer and a 12-5-MHz phased array transducer was used to perform Echocardiogram analyses. Heart function was examined at baseline (before cardiac surgery) then at 48 hours and at 37°C, continuously assessed by a rectal temperature probe. Using modified parasternal longaxis and short-axis, two-dimensional and Doppler echocardiography was used to assess the LV function and volume in M-mode. Internal dimension tracing was performed at the mid-papillary level to calculate the systolic and diastolic parameters and Teichholz formula were used to quantify LV function. Echocardiography imaging and analyses were carried out by a blinded investigator.

In vivo fluorescence imaging
Mice were placed inside the Maestro imaging system under isoflurane anesthesia. They were positioned so that the chest and abdomen of the animal were in the field of view. We used an orange filter (excitation 605nm, emission 675nm long pass) to capture the APC signal that comes from liposomes. Acquisition settings were in range from 640 to 820nm in 10-nm steps. To enhance the quality of the images we adjusted variables including stage height, focus, and exposure length within time points. Images were taken with side-by-side LAZM and vehicleinjected control animals and were analyzed using Maestro software (Cambridge