Real-Time Sampling of Rat Femoral Arteriovenous Adventitial Interstitial Fluid by Placement of a Microdialysis Probe

Tissue fluid, or interstitial fluid (ISF), constitutes a crucial element of the cellular microenvironment, with scholarly investigation dating back to the late 19th century. In 1896, Starling introduced the capillary filtration-reabsorption theory, which first characterized ISF as a homogeneous fluid generated through plasma ultrafiltration, playing a vital role innutrient transport and metabolic waste removal¹. In the past two decades, advancements in microscopic imaging technology and molecular biology have enhanced researchers' understanding of the considerable diversity in the anatomical distribution, composition, and function of ISF².

Research on interstitial fluid (ISF) across various anatomical sites has elucidated its distinct functions within different tissues. For instance, in cerebral tissue, ISF facilitates the clearance of neurotoxic metabolites, such as β-amyloid, during sleep via the glymphatic system³. Conversely, tumor interstitial fluid (TIF) contributes to tumor invasion and immune evasion by promoting an acidic microenvironment through the accumulation of lactic acid⁴. These findings suggest that ISF functions not only as a mediator of metabolic exchange but also as an active regulator of tissue-specific pathophysiological processes. In this context, Adventitial Interstitial Fluid (AISF) presents a unique research challenge due to its specific anatomical localization and functional complexity. AISF is located at the interface between the adventitia and surrounding tissues⁵, playing a crucial role in the interaction between the vascular wall and the nervous and immune systems. However, its deep anatomical location and proximity to blood and lymphatic fluids have posed significant technical challenges for in situ sampling and compositional analysis.

Traditional techniques for sampling ISF, including the wick method and surgical drainage, encounter significant challenges in obtaining high-purity AISF samples due to the risk of blood contamination and tissue damage. For example, the surgical removal of the vascular epithelium may cause injury to peripheral nerves or lymphatic vessels, resulting in the admixture of non-target biomolecules⁶. Furthermore, these methods are limited by low sampling rates, instability, susceptibility to interference, and poor reproducibility, often attributable to inflammation induced by subcutaneous wick insertion⁷ and cellular damage during muscle implantation. The development of microdialysis presents a more effective approach for ISF studies. This technique allows for the in situ and dynamic capture of small molecules, such as metabolites and cytokines, within the ISF without compromising the vascular structure by utilizing a semi-permeable membrane probe. The probe can be precisely positioned within the vascular epithelial space, thereby reducing the risk of contamination from blood or lymphatic fluids. Additionally, it facilitates continuous sampling, which enables real-time monitoring of dynamic changes in ISF metabolism under various conditions. Ensure continuous and stable monitoring of these metabolic processes during the experiment.

The animal experiments were conducted following approval from the Animal Ethics Committee of Peking Union Medical College. All procedures strictly adhered to established protocols. The reagents and the equipment used are listed in the Table of Materials.

1. Experimental preparation

1. Illustrate the microdialysis sampling system and associated surgical instruments as shown in Figure 1.
2. Prepare artificial extracellular fluid (aECF) containing 153.5 mM Na, 4.3 mM K, 0.41 mM Mg, 0.71 mM Ca, 139.4 mM Cl, and 1.25 mM glucose, and adjust the pH to 7.4.
3. Prepare two 20-kDa mD probes and designate them as Probe 1 (tissue fluid probe) and Probe 2 (blood probe). Place Probe 1 in sterile filtered deionized water (dH₂O) and Probe 2 in 1% sodium heparin, ensuring that the probe tips remain continuously moist.
4. Establish two pathway pairs (Pathways 1 and 2) for "inflow" and "outflow." Connect two 1-m FEP tubing segments using PE50 tubing, minimizing dead space. Attach each inflow tube to a 1-mL Exmire syringe: fill Syringe 1 with dH₂O and Syringe 2 with 1% sodium heparin. Connect the microsyringes to the probes and prepare for aECF infusion after placing the probes in the jugular vein.
5. Activate the mD pump and set the flow rate to 1.5 µL/min until fluid emerges from the outflow tube. Connect the outflow tube to the sample collection line.
6. Fill two microsyringes with filtered aECF and initiate pumping at 1 µL/min.

2. Animal preparation

1. Select adult Sprague-Dawley rats weighing 280 ± 20 g (n = 6).
2. Induce anesthesia using 3% isoflurane and maintain it with 1.8%-2.2% isoflurane in an air/oxygen mixture while monitoring cardiac activity (following institutionally approved protocols). Confirm deep anesthesia by assessing the absence of pain response and noting a reduced heart rate (340-390 bpm). Place the rat on a thermostatic blanket and immobilize the limbs.
3. Apply hair-removal cream to the neck and left hindlimb, extending to the left lower abdomen. Disinfect the depilated areas twice with iodophor and remove residual iodine using 75% medical alcohol.

3. Double-site probe placement (femoral adventitia and jugular vein)

1. Use a probe with a 10-mm active dialysis window (CMA-20, 20 kDa cut-off, 0.5 mm shaft diameter) as shown in Figure 2A.
2. Prepare the guide needle and tear-off tube (Figure 2B).
3. Extend the rat's hindlimb so that the line connecting the knee joint and the midpoint of the inguinal crease is perpendicular to the body's longitudinal axis. Measure a 2 cm segment along the skin of the inguinal region in the direction of the hindlimb to define the surgical landmark area (Figure 2C). Identify the midpoint of this inguinal region as the microwindow positioning point. Mark the puncture site 2 cm lateral to this positioning point (Figure 2D).
4. Place the animal under a stereomicroscope.
5. Make a 0.5-0.8 cm incision at the midpoint of the inguinal crease to facilitate visualization and ensure accurate placement of the probe within the perivascular membrane. Bluntly dissect the subcutaneous tissue to expose the femoral artery and vein, the saphenous vein, and the superficial abdominal wall vein (Figure 2E).
6. Encase the outer layer of the puncture needle in a tearable tube to create a guide needle.
7. Confirm that no blood vessels are present beneath the skin at the puncture site. Insert the needle vertically through the skin, then adjust the angle to 45° to enter the tissue space surrounding the saphenous vein.
8. Advance the needle in the direction of blood flow within the great saphenous vein, maintaining close contact with the vessel wall throughout the procedure. Use a stereomicroscope to monitor the needle position in real time and prevent vascular injury or blood contamination. Subsequently, guide the needle into the adventitia of the femoral vein (Figure 2F).
9. Withdraw the puncture needle while leaving the tear-away tube in place within the tissue (Figure 2G).
10. Insert the tissue fluid probe into the tearable catheter, and carefully open the catheter to expose the sampling site and retrieve the tissue sample (Figure 2H). Ensure that the probe is precisely positioned within the adventitia of the femoral artery and vein (Figure 2I). Secure the probe to the skin before suturing (Figure 2J). Perform all procedures under strict aseptic conditions, taking care to avoid disturbance of the vascular adventitia and to prevent tissue inflammation or foreign body reactions that could compromise tissue fluid collection.
11. Verify the probe position by photographing a surgical ruler together with the fixed probe to document the exposed length, and present this image as shown in Figure 2I.
12. Make a 2-cm skin incision approximately 0.5 cm lateral to the midline of the neck, beginning 1 cm superior to the sternal notch.
13. Bluntly dissect through the sternocleidomastoid muscle to expose the jugular vein, which lies lateral to the carotid artery and is identified by its dark-blue coloration. Use a moistened cotton swab to gently separate the fascia, taking care to avoid contact with the vagus nerve.
14. Using microscopic forceps, gently elevate the outer membrane of the jugular vein and carefully remove surrounding connective tissue to obtain at least 1 cm of free vessel length. Place a 4-0 silk suture proximally to serve as a backup ligature (Figure 2K).
15. For venous puncture, make an oblique incision at a 45° angle, approximately one-third of the vessel diameter. Use microforceps to gently open the incision and create an entry for probe insertion.
16. Carefully insert the probe tip into the vein and advance it slowly to a depth of 3.5-4.0 cm, ensuring the tip reaches the entrance of the right atrium (Figure 2L).
17. Secure the probe by tightening the proximal silk ligature with a double knot. Close the muscle layer using 4-0 Vicryl sutures, then suture the skin with interrupted stitches. Fix the external portion of the probe to the skin surface to prevent displacement.

4. Sample collection and probe recovery

1. Sample collection: Evaluate the relationship between tissue dialysis and sample transfer into the collection tube. Maintain a total volume of 20 µL between the probe and the collector.
   1. Allow a 20 min period for the sample to travel through the probe and tubing and reach the end of the tubing for collection at a flow rate of 1 µL/min. Collect an adequate volume to achieve the required concentration for the assay. Collect 30 µL of dialysate per collection tube.
2. Sample storage: Pre-chill 0.5 mL polypropylene collection tubes to 4 °C. Immediately after collection, cap the tubes, snap-freeze them in liquid nitrogen for 30 s, and transfer them to a -80 °C freezer within 2 min.
3. Baseline sample: Complete sample equilibration prior to initiating sample collection. Collect three samples per rat under anesthesia to establish a stable measurement baseline.
4. Probe removal: Remove the probe from the animal's head upon completion of sample collection and place it in a storage vial containing 1% sodium heparin. Euthanize the animal by inhalation of excess isoflurane.
5. Probe preservation: Rinse the probe thoroughly with saline and store it in a vial containing saline solution. Flush the tubes with saline using a syringe to prevent microbial growth. Reuse the probe only if it maintains sufficient flow and the membrane remains intact.

5. Data analysis

1. Perform Principal Component Analysis (PCA), Orthogonal Partial Least Squares Discriminant Analysis (OPLS-DA), and additional statistical analyses using MetaboAnalyst 6.0. Use OPLS-DA specifically to determine the Variable Importance in Projection (VIP).
2. Apply non-parametric tests, including P-value assessments and fold change analysis, to examine differential metabolites. Set critical thresholds for screening differential metabolites as VIP > 1, FC ≥ 2, or FC ≤ 0.5, and P < 0.05.
3. Annotate the identified metabolites using the KEGG Compound Database.

6. HPLC - MS/MS

1. Perform Metabolite quantification using an AB Sciex 6500+ triple quadrupole mass spectrometer equipped with an electrospray ionization (ESI) source operated in positive ion mode.
2. Set the instrument parameters as follows: curtain gas, 40 psi; collision gas, 10 psi; ion source gas 1 (GS1), 50 psi; ion source gas 2 (GS2), 55 psi; ion spray voltage, 4500 V; and source temperature, 550 °C.
3. Operate the system in multiple reaction monitoring (MRM) mode. Optimize MRM transitions, declustering potentials, and collision energies individually for each metabolite. Perform data acquisition using Analyst 1.7.1 software and quantitative analysis using MultiQuant 3.0.3.
4. For chromatographic separation, use a C18 column (2.1 × 100 mm, 1.7 µm) maintained at 45 °C. Prepare the mobile phases as (A) 0.1% formic acid in water and (B) 0.1% formic acid in acetonitrile. Set the flow rate to 0.35 mL/min and the injection volume to 6 µL. Maintain the autosampler temperature at 4 °C before analysis.
   NOTE: For this study, each run lasted approximately 25 min in total.

The dataset was normalized by applying summation and logarithmic transformations, and then proceeded to undergo a comprehensive set of statistical analyses. Principal Component Analysis (PCA) was conducted (Figure 3A), which revealed distinct separations and highlighted metabolic differences. Subsequently, an Orthogonal Partial Least Squares Discriminant Analysis (OPLS-DA) model was developed, with the resulting scores depicted in Figure 3B, demonstrating the model's discriminatory effectiveness. In the evaluation of 618 metabolites, volcano plots comparing AISF and blood samples identified 146 differential metabolites, with tissue fluid exhibiting both down-regulated and up-regulated metabolites (Figure 3C). Additionally, a heatmap was created to provide a comparative overview of the top 50 differential metabolites (Figure 3D).

To assess the reliability of the sampling method, serotonin and lactic acid were utilized as positive controls representing tissue fluid components and blood, respectively. The results aligned with the established distribution patterns of these components across various body fluids, as illustrated in Figure 4. Supplementary Figure 1 shows the stability of lactate and serotonin in microdialysate aliquots stored at -80 °C.

The 146 differential metabolites were systematically classified into several categories: nucleosides, nucleotides, and their analogues; fatty acids and their derivatives; amino acids, peptides, and their analogues; carbohydrates and carbohydrate conjugates; organic acids and their derivatives; biogenic amines; purines　and pyrimidines; vitamins and steroids; as well as other compounds (Figure 5A). Supplementary Table 1 shows differential metabolites detected in blood.Heatmap analyses further elucidate the enrichment patterns of biogenic amines and amino acids, peptides, and their analogues across various body fluids (Figure 5B,C). Differential metabolite enrichment analysis is presented in Figure 6. Supplementary Table 2 lists the coefficient of variation (CV) for fifteen randomly selected metabolites in three consecutive perivascular tissue-fluid samples.

Figure 1: Images of the experimental apparatus. (A) Refrigerated fraction collector. (B) Syringe pump. (C) Microsyringes. (D) Surgical instruments, including vascular scissors and tweezers, etc. (E) Anesthesia machine. (F) Microdialysis probes. (G) Microdialysis probes, tissue guide needle, and a tearable tube. (H) Tissue guide needle assemblies. Please click here to view a larger version of this figure.

Figure 2: The key steps in the study. (A) A schematic representation depicting a diagram illustrating the positioning of the surgical site and the orientation of the probe. (B) Identification of critically vascular junctions to ensure precise observation of probe placement sites. (C) Insertion of the guide needle that penetrates the skin and advances into the tissue space. (D) Retraction of the guide pin needles is withdrawn, leaving the tearable tube within the tissue. (E) The microdialysis probe is introduced into the tissue via the established channel. (F) Gentle retraction of the tearable tube. The tearable tube is gently retracted after securing the probe in place. (G) Visualization of probes positioned in the epithelium of the femoral arterial and venous vessels. (H) Separation of tissue from the jugular vein. (I) Placement of the probe in the jugular vein. (J) The microdialysis probe is fixed on the skin surface. (K) isolation of the jugular vein. (L) Insertion of the probe into the jugular vein. Please click here to view a larger version of this figure.

Figure 3: Multivariate statistical analysis of interstitial fluid (ISF) and blood. (A) The data have been downscaled to emphasize significant separations, demonstrating distinct differences between ISF and blood components. (B) The orthogonal partial least squares discriminant analysis (OPLS-DA) model scores indicate R2Y and Q2 values of 0.925 and 0.811, respectively. (C) The volcano plot depicts 146 differential metabolites, with 85 metabolites down-regulated and 61 up-regulated in ISF. (D) Correspondingly, heatmaps contrasting the top 30 metabolites in ISF and blood reveal a clear distinction between the two sets of components. Please click here to view a larger version of this figure.

Figure 4: The presentation of a series of box plots elucidated the enrichment levels of lactate and serotonin in ISF and blood. All data points represent mean ± SEM; error bars are shown for all quantitative panels. Panel (A) illustrates that ISF samples exhibit a significantly greater enrichment of lactate compared to blood samples. In contrast, panel (B) demonstrates that serotonin is more enriched in blood samples, which concurrently display lower lactate levels. Please click here to view a larger version of this figure.

Figure 5: The classification ring chart of differential metabolites and the heat map of amino acids, peptides, and analogues, as well as biogenic amines, are depicted. (A) Each color block corresponds to a distinct class of metabolites. Amino acids, peptides, and analogues represent the largest category, comprising 19.29% of the total, followed by nucleosides, nucleotides, and analogues (15%), fatty acids and derivatives (12.86%), carbohydrates and carbohydrate conjugates (7.86%), organic acids and derivatives (10%), biogenic amines (7.14%), purines/pyrimidines (5.71%), vitamins and steroids (8.57%), and other compounds, including indoles, azepines, pyridines, and benzenes, at 13.57%. (B) In interstitial fluid (ISF), the enrichment levels of compounds such as Deoxycarnitine and Spermidine within the biogenic amines category are significantly elevated compared to those in blood, whereas the levels of Serotonin (D4), Serotonin, Tryptamine, and related compounds are significantly reduced. (C) In ISF, the majority of metabolites associated with amino acids, peptides, and analogues are markedly lower than those observed in blood. Please click here to view a larger version of this figure.

Figure 6: Metabolite set enrichment analysis (MSEA) of normal interstitial fluid revealed that metabolic activity is predominantly distributed across pathways related to energy production, redox homeostasis, and amino acid turnover. The top enriched pathways include butanoate metabolism, purine metabolism, and glutathione metabolism, reflecting the maintenance of mitochondrial energy supply and antioxidant capacity under physiological conditions. Additionally, several amino acid-related pathways, such as arginine and proline metabolism, tryptophan metabolism, and cysteine and methionine metabolism, are represented, indicating active amino acid interconversion and nitrogen balance regulation in the extracellular microenvironment. These findings suggest that the normal interstitial metabolome maintains a dynamic equilibrium between bioenergetic processes, oxidative defense, and amino acid metabolism, supporting tissue homeostasis and microcirculatory stability. Please click here to view a larger version of this figure.

Supplementary Figure 1: Stability of lactate and serotonin in microdialysate aliquots stored at -80 °C. Data are mean ± SEM (n = 6 replicates per time point). No analyte deviated >5% from baseline over 72 h. Please click here to download this File.

Supplementary Table 1: Differential metabolites detected in blood and adventitial interstitial fluid microdialysate. This includes metabolite name, fold-change (FC), Student's t-test P value, and VIP (OPLS-DA). Please click here to download this File.

Supplementary Table 2: Coefficient of variation (CV) for fifteen randomly selected metabolites in three consecutive perivascular tissue-fluid samples. All CVs are <20%. Please click here to download this File.

Interstitial fluid (ISF) is the aqueous micro-environment that envelops every tissue cell and constitutes roughly 75% of the extracellular fluid compartment^8,9. Generated by capillary filtration of plasma, ISF lacks the cellular and high-protein content of blood but retains electrolytes, small solutes, and water⁹. This composition positions ISF as the pivotal interface between the vascular and cellular worlds. Functionally, ISF is a dynamic transport medium. It conveys oxygen, glucose, amino acids, and micronutrients from the bloodstream to cells while simultaneously collecting carbon dioxide, urea, and other metabolic wastes for return to the circulation¹⁰. The fluid also serves as a lubricant and buffer, stabilizing interstitial osmotic pressure, pH, and ionic balance¹¹. Beyond metabolism, ISF supports immune surveillance by ferrying leukocytes, antibodies, and cytokines through tissue spaces and acts as a communication highway for hormones and paracrine signals^12,13. Compared with blood, ISF is essentially a cell-free, low-protein filtrate¹⁴. Blood, confined to the vascular tree, contains red and white cells, platelets, and high concentrations of albumin and clotting factors. Oxygen and nutrient levels are highest in blood and progressively lower in ISF because of cellular uptake, whereas waste products accumulate in ISF before entering the lymphatic or venous return¹⁰. The absence of large proteins and cells in normal ISF minimizes oncotic pressure and allows rapid diffusion, but inflammatory states can transiently alter this profile¹⁵.

However, the study of ISF remains limited due to the absence of straightforward, rapid, and reliable sampling methodologies. Current methods include microneedle (MN) patches, suction blisters, reverse iontophoresis, thermal ablation, and microdialysis. MN patches create microchannels for ISF extraction through the skin, with core-shell variants extracting ISF in under 30 s and osmotic hydrogel MNs in about 3 min. These can be combined with sensors for on-patch analysis, but are limited by volume and require a balance between strength and swelling. Suction blisters provide pure ISF with minimal blood contamination but are invasive, painful, and slow^16,17,18. Reverse iontophoresis (RI) utilizes low electrical currents to transport charged analytes, such as glucose and lithium ions, to the skin surface for non-invasive monitoring. However, it faces challenges such as low throughput, variability between individuals, and potential skin irritation^19,20. Transdermal administration (TA) disrupts the stratum corneum with localized heating to allow ISF to passively enter hydrophilic microchannels. Although simpler in device design, it collects less than 5 µL of fluid, requiring highly sensitive detection and careful thermal management to avoid burns^21,22. In contrast, microdialysis (MD), which relies on diffusion across implanted semi-permeable probes, offers unique molecular selectivity through adjustable molecular weight cut-off (MWCO) membranes, such as 10 kDa for glucose and 100 kDa for proteins and cytokines^23,24. This technique supports continuous and minimally invasive sampling. However, MD is operationally complex, requires in vivo calibration, and shows variable recovery rates for hydrophobic species. Notably, MD is the only technique capable of size-selective isolation of large biomolecules²⁵. Comparative studies of dialysate collected adjacent to vascular adventitia versus blood demonstrate distinct lactate and serotonin profiles, thereby validating the stability and specificity of the MD approach.

Animal studies have shown that, synergistically regulated by the heart and lungs, the ISF can flow in the adventitia associated with large arteries, veins, and peripheral nerves^5,9. Usually, ISF contains metabolites that are derived directly from plasma as well as synthesized locally within tissues^26,27,28. It is plausible, therefore, that the identified overlaps and differences of the metabolites in AISF and plasma represent the involvement of the ISF circulatory system in the transport and clearance of nutrients and metabolites in tissues along the vasculature. The microdialysis-based technique introduced here for the sampling of the AISF will significantly enhance the studies of the systemic ISF circulation in regulating metabolism in multiple organs and tissues throughout the body.

The 20-kDa molecular-weight cut-off (MWCO) membrane utilized in this study is optimally designed for the quantitative recovery of small metabolites, including glucose, lactate, pyruvate, and amino acids. However, it consistently under-samples or entirely excludes numerous biologically significant macromolecules, such as cytokines, chemokines, growth factors, and extracellular enzymes. Clinical and methodological reviews have highlighted that standard 20-kDa catheters are effective in collecting small hydrophilic metabolites, yet they exhibit poor recovery rates for proteins and inflammatory mediators²⁹. Consequently, membranes with higher MWCOs (e.g., 100 kDa) or alternative sampling strategies are necessary for accurate cytokine quantification^30,31. Therefore, future research conducted by our group will employ membranes with larger MWCOs to characterize the macromolecular composition of peripheral interstitial fluid. In this study, dialysates were collected under general anesthesia. Microdialysis has been previously validated in awake, freely moving animals, and several "open-flow" micro-perfusion techniques have been adapted for use in conscious models³². However, adapting our probe design and implantation protocol for use in awake animals will necessitate addressing challenges related to tethering, animal stress, probe stability, motion artifacts, and long-term patency. We are currently in the process of developing a lightweight tether-swivel system and plan to undertake conscious sampling in our subsequent study.