Microfluidic Separation of Adipocytes

This paper presents a deterministic lateral displacement microfluidic device that successfully overcomes the fragility of mature adipocytes to achieve high-purity, high-recovery size-based separation while preserving cell viability and hormonal responsiveness for metabolic research.

The Gist

Imagine your body's fat tissue not as a uniform blob, but as a bustling city made up of houses of all different sizes. Some are tiny cottages, while others are massive mansions. Scientists have long known that the size of these "fat houses" (adipocytes) matters a lot for your health. The giant ones are often the troublemakers, linked to diabetes and heart disease, while the smaller ones are usually the good citizens.

The problem? Trying to sort these fat cells by size is like trying to separate a pile of fragile, water-filled balloons without popping them. Traditional methods are too rough, causing the cells to burst or clump together, leaving researchers with a messy, unusable mix.

The Solution: A Microscopic Traffic Cop

In this paper, a team of scientists built a tiny, high-tech sorting machine called a microfluidic device. Think of it as a microscopic highway system designed specifically for these delicate fat balloons.

Here's how it works, using a few simple analogies:

1. The "Deterministic Lateral Displacement" (The Obstacle Course)

Imagine a hallway filled with a perfectly arranged grid of pillars (like a forest of trees).

• The Small Cells: If you roll a marble (a small fat cell) through this forest, it can weave easily between the trees, following the straight path of the wind (the water flow). It exits the hallway at the bottom.
• The Large Cells: If you try to roll a beach ball (a large fat cell) through the same forest, it's too big to squeeze between the trees. Instead of weaving, it gets bumped by the pillars and forced to zig-zag sideways, taking a completely different path to exit at the top.

This is the core magic of the device. It doesn't use electricity or sticky labels to sort the cells; it just uses the laws of physics and the size of the cells to guide them into different lanes.

2. The "Creaming" Problem (The Stirrer)

Fat cells are lighter than water, so they naturally float to the top, like cream rising in milk. If you just pour them into a machine, they would all stick to the ceiling and never enter the sorting area.

• The Fix: The scientists put a tiny magnetic stirrer in the tank, like a gentle spoon stirring a cup of coffee. This keeps the fat cells floating evenly in the liquid so they can all enter the sorting highway fairly.

3. The "Guard" Fraction (The Security Check)

Sometimes, a cell is right on the borderline between "small" and "large." If you try to force it into one lane, you might get it wrong.

• The Fix: The machine has a third, middle exit lane. Any cell that is unsure of its size gets sent there. This acts like a security guard at an airport, catching the "maybe" travelers so that the "definitely small" and "definitely large" lanes remain perfectly pure.

The Results: Gentle but Effective

The team tested this machine and found it was a huge success:

• It's Gentle: The cells survived the trip. They didn't pop or get damaged.
• It's Smart: They successfully separated the cells into two distinct groups: a "small" group (average size 47 micrometers) and a "large" group (average size 82 micrometers).
• It Works: After sorting, they tested the cells with insulin (a hormone that tells fat cells to store energy). The cells responded perfectly, proving they were still alive and healthy.

Why Does This Matter?

Before this, studying fat cells by size was like trying to study a specific type of fish by catching them with a net that tears them apart. Now, scientists have a gentle, high-speed conveyor belt that can sort thousands of these delicate cells without hurting them.

This opens the door to understanding exactly why big fat cells cause diabetes and how to fix them. It's a new tool that lets us look at the "fat city" with a much clearer, more detailed map, potentially leading to better treatments for obesity and metabolic diseases in the future.

Technical Summary

1. Problem Statement

Adipocyte (fat cell) size is a critical independent predictor of metabolic disorders, including type 2 diabetes, liver disease, and cardiovascular conditions. Large, hypertrophic adipocytes are generally less insulin-sensitive and pro-inflammatory, while small adipocytes are more insulin-responsive.

• The Challenge: Despite the biological importance of size-based separation, isolating intact, mature adipocytes by size has been technically difficult due to their fragile nature and buoyancy.
• Limitations of Current Methods: Conventional techniques like Fluorescence-Activated Cell Sorting (FACS), dialysis tubing (buoyancy-based), and nylon mesh filtration suffer from:
  • High rates of cell breakage and lysis.
  • Cell aggregation (small cells sticking to large ones).
  • Low recovery rates and poor purity.
  • Requirement for large input volumes of tissue to yield small amounts of sorted cells.

2. Methodology

The authors developed a microfluidic device utilizing Deterministic Lateral Displacement (DLD) to sort intact, mature adipocytes based on size without labels.

• Device Design:
  • Mechanism: A DLD array consisting of an ordered pillar structure. Particles larger than a critical diameter ($D_c$) are displaced laterally, while smaller particles follow the fluid streamlines.
  • Geometry: Designed with a gap size ($G$) of 122 µm and 13 pillars per row ($N=13$), theoretically targeting a critical diameter ($D_c$) of 50 µm.
  • Outlets: Three outlets were designed:
    1. Small Fraction: Collects particles smaller than $D_c$.
    2. Large Fraction: Collects particles larger than $D_c$.
    3. Guard (Intermediate) Fraction: Collects particles near the critical diameter to minimize cross-contamination and improve the purity of the extreme fractions.
• Fabrication:
  • Master molds were created using SU-8 photoresist and a 200 µm dry film resist (SUEX) on silicon wafers.
  • Microfluidic chips were fabricated via replica molding using Polydimethylsiloxane (PDMS).
  • Surface treatment (Aluminum Oxide + FDTS) was applied to the mold to prevent PDMS adhesion.
• Operational Conditions:
  • Sample Handling: To counteract the buoyancy of adipocytes (which float and separate from the buffer), a magnetic stir bar was placed in the sample reservoir to ensure continuous, gentle mixing.
  • Flow Control: Pressure-driven flow (20 mbar) generated laminar flow with a Reynolds number of ~0.1.
  • Shear Stress: Calculated wall shear stress was ~0.04 Pa, orders of magnitude lower than deformability assays, ensuring cell integrity.
• Validation & Analysis:
  • Imaging: Time-lapse microscopy (35,000 cells) analyzed cell trajectories and size distribution in real-time.
  • Coulter Counter: Osmium-fixed adipocytes were analyzed to determine precise size distributions and mean diameters.
  • Viability Assay: Western blotting was performed to assess insulin signaling pathways (phosphorylation of PKB and AS160) to confirm cells remained metabolically active post-sorting.

3. Key Contributions

• First Successful DLD Sorting of Mature Adipocytes: This study demonstrates the first application of DLD microfluidics specifically for sorting fragile, buoyant, mature adipocytes, a cell type previously considered too delicate for such continuous flow methods.
• Optimized Handling of Buoyant Cells: The integration of magnetic stirring in the reservoir solved the critical issue of cell creaming, ensuring uniform entry into the microchannel.
• High Purity and Recovery: The inclusion of a "guard" fraction allowed for high-purity separation of small and large populations, overcoming the trade-off usually seen in size-based sorting.

4. Results

• Separation Efficiency:
  • Purity: The small fraction achieved 97% purity, and the large fraction achieved 93% purity.
  • Yield: The recovery rate was 66% for the small fraction and 42% for the large fraction. (Note: The lower yield in the large fraction is partly due to the guard fraction capturing cells near the cutoff).
  • Size Distribution:
    • Small Fraction: Mean diameter ~47–56 µm.
    • Large Fraction: Mean diameter ~76–82 µm.
    • Volume Difference: This represents an approximate 2.5-fold difference in mean cellular volume between the two fractions.
• Discrepancy in Critical Diameter: While the device was designed for a $D_c$ of 50 µm, the experimental separation threshold was observed at ~70 µm. The authors attribute this to:
  • Osmium fixation swelling cells for Coulter counting.
  • Optical blur in image segmentation.
  • The deformability of adipocytes allowing them to pass through gaps smaller than their static diameter.
• Cell Viability:
  • Western blot analysis confirmed that sorted adipocytes retained their ability to respond to insulin. Both small and large fractions showed clear induction of PKB and AS160 phosphorylation upon insulin stimulation, proving the cells were viable and functionally intact.
• Throughput: The method processed ~500 µL of packed adipocytes to yield ~150 µL (small) and ~100 µL (large) fractions, a significant improvement over traditional methods requiring grams of tissue for similar outputs.

5. Significance

This work establishes microfluidics as a superior alternative to traditional methods for adipocyte research.

• Functional Analysis: By preserving cell integrity and hormonal responsiveness, this method enables new functional studies linking adipocyte size to metabolic health, which was previously hindered by cell damage during sorting.
• Scalability: The platform offers high throughput and the potential for parallelization to process larger volumes, addressing the input/output limitations of buoyancy-based methods.
• Future Applications: The success of this approach opens the door for more sophisticated "adipocyte-on-a-chip" systems and high-throughput screening of metabolic drugs on size-separated fat cells.

In conclusion, the authors successfully overcame the technical barriers of sorting fragile adipocytes, providing a robust, label-free tool to explore the relationship between adipocyte size and metabolic disease.