Inhibition of MicroRNA-92a Protects Against Ischemia/Reperfusion Injury in a Large-Animal Model

The researchers tested whether inhibiting a small non-coding RNA molecule called microRNA-92a (miR-92a) could protect heart tissue from damage caused by a heart attack followed by restoration of blood flow (ischemia/reperfusion injury). They used a synthetic inhibitor called LNA-92a — a locked nucleic acid antisense oligonucleotide that binds to and neutralises miR-92a.

The intervention was LNA-92a delivered via three different routes:

**Antegrade delivery** (into the coronary artery, the normal direction of blood flow) — mimicking how a cardiologist would deliver drugs during a stent procedure

**Retrograde delivery** (into the coronary vein, against blood flow) — an alternative catheter approach

**Intravenous delivery** (systemic, into a vein) — the simplest method

The comparator was a control LNA oligonucleotide (a scrambled sequence that does not target any known pig or human microRNA) or phosphate-buffered saline (PBS, the vehicle solution).

Primary outcome measures were:

Infarct size (percentage of the area at risk that actually died)

Cardiac function (ejection fraction and left ventricular end-diastolic pressure)

miR-92a expression levels in heart tissue

Secondary outcomes included:

Capillary density (new blood vessel growth)

Leukocyte influx (inflammatory cell infiltration)

Cardiomyocyte cell death (apoptosis)

miR-92a expression in other organs (liver, spleen, kidney, lung)

**Pig study:** 5 pigs per experimental group (exact total N not stated, but approximately 25–30 pigs across all groups). Pigs were adult, breed not specified, purchased from Veterinary Medicine, LMU Munich. All experiments conducted at the Walter-Brendel Center for Experimental Medicine at the University of Munich.

**Mouse study:** Constitutive miR-92a knockout mice (miR-92a−/−) and cardiomyocyte-specific miR-92a knockout mice (miR-92a fl/fl, αMHC-Cre), plus wild-type controls. Exact numbers per group not stated in the abstract, but the paper reports n=5–8 per group for functional measurements.

**Cell culture:** HL-1 cardiomyocytes (a mouse atrial cardiomyocyte cell line) and human umbilical vein endothelial cells (HUVECs).

**Infarct size:** Dual staining method — methylene blue exclusion to define the area at risk (the heart region deprived of blood), followed by tetrazolium red viability staining to distinguish living from dead tissue within that area. Infarct size expressed as percentage of area at risk.

**Cardiac function:** Left ventricular ejection fraction (LVEF, percentage of blood pumped out per beat) and left ventricular end-diastolic pressure (LVEDP, pressure in the heart's main pumping chamber when filled, measured in mmHg). Measured before ischemia and after 72 hours or 7 days of reperfusion.

**miR-92a expression:** Real-time polymerase chain reaction (qRT-PCR) from tissue samples, normalised to control small RNAs. Fold-change relative to non-ischemic tissue.

**Capillary density:** Immunohistochemistry using platelet endothelial cell adhesion molecule-1 (PECAM-1/CD31) staining to count blood vessels in the ischemic border zone.

**Leukocyte influx:** Myeloperoxidase (MPO) assay — MPO is an enzyme released by neutrophils (a type of white blood cell) during inflammation.

**Cell death:** Terminal deoxynucleotidyl transferase dUTP nick-end labeling (TUNEL) staining, which labels fragmented DNA in dying cells.

**Leukocyte adhesion (in vitro):** Flow chamber assay using HUVECs stimulated with tumour necrosis factor-alpha (TNF-α, 15 ng/mL) at a shear stress of 1 dyne/cm², mimicking blood flow conditions.

**Study design:** Randomised controlled trial in a large-animal model (pigs), with additional mechanistic studies in genetically modified mice and cell culture experiments.

**Randomisation:** Pigs were assigned to groups (n=5 per group), but the paper does not describe the randomisation method in detail. Given the small group sizes, randomisation likely used simple allocation.

**Blinding:** Not explicitly stated. The paper does not mention whether the researchers measuring infarct size, cardiac function, or histological endpoints were blinded to treatment group. This is a significant methodological concern — lack of blinding in animal studies can inflate effect sizes by 30% or more.

**Duration:** Ischemia was induced for 60 minutes (balloon inflation in the left anterior descending coronary artery). Reperfusion was allowed for either 72 hours (3 days) or 7 days before measurements were taken. The LNA-92a or control was administered 5 minutes before reperfusion began.

**Dose:** 0.03 mg/kg body weight for the standard dose, or 5 mg per 1 kg heart weight (with a 5× higher dose of 25 mg per 1 kg heart weight tested in one antegrade group).

**Statistical approach:** One-way ANOVA with Student-Newman-Keuls post-hoc test for multiple comparisons. P<0.05 considered significant. Results reported as mean±SEM (standard error of the mean). Sample size of n=5 per group is very small — this limits statistical power and increases the risk of false positives or false negatives.

**What this design can prove:**

Causal relationship between miR-92a inhibition and reduced infarct size in pigs

Differential efficacy of delivery routes (catheter-based vs. intravenous)

Mechanistic insights from knockout mice (constitutive vs. cardiomyocyte-specific deletion)

**What this design cannot prove:**

Long-term safety or efficacy beyond 7 days

Effectiveness in humans (pigs are anatomically and physiologically similar to humans for cardiac studies, but not identical)

Whether the same dose would work in humans (dose scaling across species is complex)

Whether the effect is durable after a single dose (only one administration was tested)

Whether there are off-target effects in humans (only a limited panel of related microRNAs was checked)

**Major methodological weaknesses:**

Very small sample size (n=5 per group) — underpowered to detect moderate effects reliably

No explicit mention of blinding — potential for observer bias in infarct size measurement and functional assessment

Only one time point for most measurements (72 hours or 7 days, not both for all groups)

The 5× higher dose group had only one delivery route tested (antegrade), not all routes

The control LNA sequence targets a C. elegans microRNA not present in pigs — this is appropriate but the paper does not confirm that the control LNA has no biological effects in pigs

No power analysis reported to justify the sample size

**Primary outcomes:**

**miR-92a expression:** All three delivery routes (antegrade, retrograde, intravenous) significantly reduced miR-92a levels in both infarcted and non-infarcted heart tissue compared to PBS and control LNA (P<0.01). The 5× higher antegrade dose did not produce further reduction.

**Infarct size:** Catheter-based delivery significantly reduced infarct size:

- Antegrade LNA-92a: 32±5% of area at risk (vs. control LNA ~50–55%, P<0.05)

- Retrograde LNA-92a: 36±4% of area at risk (P<0.05)

- Intravenous LNA-92a: No significant reduction compared to controls

- High-dose antegrade: 34±2% (no additional benefit)

**Cardiac function:** Improved ejection fraction and left ventricular end-diastolic pressure with catheter-based LNA-92a delivery (P<0.05 for both). Specific numbers not fully reported in the abstract but the paper states significant improvement.

**Secondary outcomes:**

**Capillary density:** LNA-92a increased capillary density in the ischemic border zone (P<0.05)

**Leukocyte influx:** LNA-92a decreased myeloperoxidase activity (marker of neutrophil infiltration) (P<0.05)

**Cell death:** LNA-92a reduced TUNEL-positive cardiomyocytes (P<0.05)

**Off-target effects:** LNA-92a did not affect other members of the miR-17-92 cluster or miR-363, but did reduce miR-25 (a closely related microRNA) at high doses

**Tissue distribution:** Intravenous delivery reduced miR-92a in liver, spleen, kidney, and lung. Catheter-based delivery reduced miR-92a in liver and spleen but spared the lung (unless high dose was used)

**Mouse knockout studies:** Complete loss of miR-92a (constitutive knockout) protected cardiac function after infarction more than cardiomyocyte-specific deletion alone, suggesting that miR-92a in non-cardiomyocyte cells (e.g., endothelial cells, immune cells) also contributes to injury

**In vitro:** LNA-92a protected HL-1 cardiomyocytes from hypoxia/reoxygenation-induced cell death

**Infarct size reduction:** Catheter-based LNA-92a reduced infarct size from approximately 50–55% of the area at risk to approximately 32–36% — a relative reduction of roughly 35–40%. In absolute terms, this means about 18–20% more of the endangered heart muscle survived.

**Functional improvement:** The paper reports improved ejection fraction and left ventricular end-diastolic pressure, but does not provide exact numerical values in the abstract. Based on similar studies, a 5–10 percentage point improvement in ejection fraction would be clinically meaningful.

**Capillary density:** Increased, but the paper does not report the fold-change in the abstract. Histological images suggest a visible increase in vessel number.

**Leukocyte influx:** Reduced, but the magnitude is not quantified in the abstract.

To put this in perspective: In human heart attacks, every 5% reduction in infarct size translates to approximately a 10–15% reduction in mortality over 1 year. A 35–40% relative reduction in infarct size would be considered a major therapeutic advance — if it translates to humans.

**Acknowledged by authors:**

The study was conducted in a pig model, not humans — translation to clinical practice requires further validation

Only one dose was tested for most delivery routes (except the 5× antegrade group)

The 7-day time point was only assessed for infarct size, not for all endpoints

The mechanism of action was explored in mice and cell culture, not fully in pigs

**Critical reader observations:**

**Sample size:** n=5 per group is very small. With only 5 animals per group, a single outlier can dramatically influence results. The study would need approximately 10–15 animals per group to have 80% power to detect a 20% difference in infarct size with reasonable confidence.

**Blinding:** Not mentioned. In animal studies of cardiac ischemia, lack of blinding is a well-documented source of bias — researchers who know which animals received the active treatment may unconsciously measure infarct size differently.

**Sex:** The paper does not specify whether pigs were male or female. Sex differences in cardiac ischemia/reperfusion injury are well established. If only one sex was used, generalisability is limited.

**Duration:** 7 days is short for assessing cardiac remodelling and long-term functional recovery. Heart attacks trigger weeks to months of remodelling. A 30-day or 60-day endpoint would be more clinically relevant.

**Single dose:** Only one administration was tested (given 5 minutes before reperfusion). Whether repeated dosing or later administration would be beneficial is unknown.

**Off-target effects:** LNA-92a reduced miR-25 at high doses. miR-25 is involved in cardiac contractility and calcium handling — reducing it could have unintended consequences.

**Funding:** One author (E. van Rooij) was affiliated with miRagen Therapeutics, a company developing microRNA-based therapeutics. This is a potential conflict of interest. The paper does not state whether miRagen funded the study.

**Translation to humans:** Pigs have similar heart anatomy and physiology to humans, but the immune response, drug metabolism, and microRNA biology may differ. The LNA chemistry used may have different pharmacokinetics in humans.

**Intravenous failure:** The fact that intravenous delivery did not reduce infarct size despite reducing cardiac miR-92a levels is puzzling. This suggests that either the dose was insufficient locally, or that the catheter-based delivery provides additional benefits (e.g., pressure effects, local drug concentration) beyond simple miR-92a inhibition.

For someone running their own n=1 experiment (this is NOT something you should attempt at home — LNA drugs are experimental and require medical supervision):

**What to test:**

This is not a self-experiment you can safely conduct. LNA-92a is an experimental drug not approved for human use. However, the principles of local vs. systemic delivery and timing of intervention are relevant to anyone interested in cardiac protection.

**If you were designing a clinical trial:**

Test LNA-92a delivered via intracoronary catheter during percutaneous coronary intervention (stent placement) for acute myocardial infarction

Dose: 0.03 mg/kg body weight (scaled from pigs to humans using body surface area — approximately 0.01 mg/kg in humans)

Timing: Administer 5 minutes before balloon deflation (reperfusion)

**Minimum meaningful duration:**

For infarct size: 72 hours (3 days) is sufficient for the infarct to mature

For functional recovery: 30 days minimum, 6 months preferred

For safety: 12 months of follow-up for potential off-target effects

**What to measure:**

Primary: Infarct size as percentage of area at risk (by cardiac MRI or SPECT imaging)

Secondary: Left ventricular ejection fraction (by echocardiography or MRI), left ventricular end-diastolic pressure (by catheterisation), cardiac biomarkers (troponin I or T, CK-MB)

Safety: miR-92a levels in blood and tissue (if accessible), liver and kidney function tests, complete blood count, ECG monitoring for arrhythmias

Exploratory: Capillary density (if tissue available), inflammatory markers (CRP, IL-6)

**Key confounds to control for:**

Time from symptom onset to reperfusion (shorter is better — must be standardised)

Collateral blood flow (varies between individuals and affects infarct size)

Pre-existing cardiac medications (beta-blockers, ACE inhibitors, statins all affect infarct size)

Age, sex, and cardiovascular risk factors (diabetes, hypertension, smoking)

The dose and type of contrast agent used during catheterisation (some contrast agents are cardioprotective)

Body temperature (hypothermia is cardioprotective — must be standardised)

**What a positive result would look like:**

Infarct size reduced by at least 30% (relative) compared to placebo, measured by cardiac MRI at 72 hours

Ejection fraction improved by ≥5 percentage points at 30 days

Troponin release reduced by ≥40% over the first 48 hours

No increase in serious adverse events (bleeding, arrhythmias, organ toxicity)

miR-92a levels in cardiac tissue (if biopsied) reduced by ≥50% compared to non-ischemic tissue

**Important caveat for self-experimenters:**

Do not attempt to obtain or self-administer LNA-92a. This is a potent experimental drug with unknown long-term effects. The pig study showed that even local delivery reduced miR-92a in the liver and spleen, and high doses reduced miR-25 (a microRNA important for heart function). The risk of unintended consequences — including potential harm to the