All experimental procedures were reviewed and approved by the local authorities (permission number: 19-2024; State Office for Consumer Protection, Saarbrücken, Germany) and were conducted in strict accordance with the European Directive 2010/63/EU, the ARRIVE guidelines and the National Institutes of Health (NIH) Guidelines for the Care and Use of Laboratory Animals (NIH publication No. 85-23, Rev. 1985).

The animals were maintained under standardized laboratory conditions, including a controlled temperature (22 ± 2 °C) and relative humidity (55 ± 10 %), with wood-chip bedding and unrestricted access to water and standard pellet chow (ssniff Spezialdiäten GmbH, Soest, Germany). The animals were either bred in-house at the Institute for Clinical and Experimental Surgery (Saarland University, Homburg, Germany) or obtained from Charles River Laboratories (Sulzfeld, Germany).

To obtain sufficient adipose tissue for nanofat preparation, the inguinal fat pads from eight green fluorescent protein (GFP)⁺ C57BL/6-Tg (CAG-EGFP)131Osb/LeySopJ donor mice of both sexes (age: 6-12 months; body weight: > 30 g; The Jackson Laboratory, Bar Harbor, ME, USA) were harvested. For in vivo analyses, dorsal skinfold chambers were surgically implanted into sixteen GFP^- C57BL/6J wild-type recipient mice of either sex (age: 2-6 months; body weight: 22-30 g).

For all surgical interventions and intravital microscopic examinations, anesthesia was induced through intraperitoneal injection of ketamine hydrochloride (100 mg/kg body weight; Ketabel^®, Bela-pharm GmbH & Co. KG, Vechta, Germany) combined with xylazine (12 mg/kg body weight; Rompun^®, Bayer, Leverkusen, Germany). Perioperative analgesia was achieved through subcutaneous injection of carprofen (10 mg/kg body weight; Rimadyl^®, Zoetis Deutschland GmbH, Berlin, Germany). To protect the cornea during anesthesia, a lubricating ophthalmic ointment (Bepanthen^®, Bayer Vital GmbH, Leverkusen, Germany) was applied.

White subcutaneous adipose tissue obtained from GFP⁺ donor mice was processed into nanofat by a standardized mechanical emulsification, as previously described in detail (Tonnard et al., 2013[38]; Bonomi et al., 2024[4]). Following harvest, the tissue was rinsed in physiological saline and minced into small fragments (approximately 1 × 1 × 1 mm) with a McIlwain Tissue Chopper (CLE Co. Ltd., Gomshall, UK). The minced adipose tissue was subsequently subjected to repeated bidirectional transfer between two syringes connected via sequential female-to-female Luer lock connectors with decreasing inner diameters (2.4, 1.4, and 1.2 mm), with 30 passes performed for each connector size. Following mechanical emulsification, the resulting suspension was passed through a 500 µm mesh filter to eliminate residual larger tissue fragments and debris. The filtered nanofat was then evenly distributed into two microcentrifuge tubes (Eppendorf, Hamburg, Germany) and incubated for 1 h at room temperature in a 1:1 volume ratio with either Hank's Balanced Salt Solution (HBSS; Gibco, Waltham, MA, USA) alone (vehicle control; n = 8) or HBSS supplemented with DFO mesylate (1 mM; Sigma-Aldrich, St. Louis, MO, USA; n = 8).

The bilayer collagen-glycosaminoglycan-based dermal substitute Integra^® (Integra LifeSciences, Ghent, Belgium; thickness: 1.3 mm) was used to generate circular implants with a diameter of 4 mm using a sterile biopsy punch (Kai Europe GmbH, Solingen, Germany). The resulting discs were immediately immersed in either vehicle-pretreated nanofat (control) or DFO-pretreated nanofat and incubated at room temperature for 10 minutes. As previously demonstrated by histological analysis, this short incubation period and seeding technique was sufficient to allow uniform adsorption of the nanofat suspension into the porous matrix of the dermal substitutes, thereby enabling stable loading of the implants (Bonomi et al., 2024[4]).

Dermal substitute discs loaded with either vehicle- or DFO-pretreated nanofat were implanted into the observation window of dorsal skinfold chambers in GFP⁻ recipient mice. This experimental setup enabled repeated intravital fluorescence microscopy of the implants at predefined time points following implantation (days 0, 3, 6, 10 and 14). To visualize microvessels and circulating leukocytes, the animals received retrobulbar intravenous injections of fluorescein isothiocyanate (FITC)-labeled dextran (50 µL of a 5 % solution; molecular weight: 150,000 Da; Sigma-Aldrich, Taufkirchen, Germany) for plasma labeling along with rhodamine 6G (50 µL of a 0.1 % solution; Sigma-Aldrich) to stain leukocytes. Fluorescence imaging was performed using an epi-illumination microscope (Axiotech; Zeiss, Oberkochen, Germany) equipped with a monochrome digital camera (Axiocam 702 mono; Carl Zeiss Microscopy GmbH, Oberkochen, Germany). Quantitative image analyses were performed using the CapImage analysis system (version 8.10.1; Dr. Zeintl Software, Heidelberg, Germany). For quantitative evaluation of implant vascularization, eight predefined regions of interest (ROIs) per implant were analyzed, comprising four ROIs at the periphery and four within the center of the implants. Newly formed microvessels were identified based on the presence of red blood cell (RBC) perfusion. The proportion of perfused ROIs was calculated as the percentage of all analyzed ROIs. The functional microvessel density was determined by measuring the cumulative length of RBC-perfused microvessels within each ROI and expressed as cm/cm². In addition, five microvessels per ROI were randomly selected to determine vessel diameter (µm) and centerline RBC velocity (µm/s). These parameters were subsequently used to calculate shear rate (s⁻¹) and volumetric blood flow (pL/s), as previously described (Koutsiaris et al., 2007[21]). The inflammatory response induced by the implants was assessed in four peri-implant postcapillary or collecting venules. Microhemodynamic parameters, including vessel diameter, centerline RBC velocity, shear rate and volumetric blood flow, were quantified alongside leukocyte-endothelial interactions. Leukocytes were categorized as free-flowing, rolling or adherent. Rolling leukocytes were defined by reduced transluminal velocity and transient endothelial contact, whereas adherent leukocytes were identified by stable attachment to the endothelial surface for a minimum duration of 30s. Leukocyte counts were normalized to the endothelial surface area, which was calculated assuming a cylindrical vessel geometry.

Following completion of the final intravital microscopy session, the animals were euthanized by administration of an anesthetic overdose followed by cervical dislocation. The dorsal skin encompassing the implanted dermal substitutes was subsequently harvested and processed for histological and immunohistochemical evaluation. Routine hematoxylin and eosin (HE) staining was performed to assess general tissue morphology. In addition, immunohistochemical stainings were carried out to detect endothelial cells (CD31), extracellular matrix components (collagen I and collagen III) and inflammatory cell populations, including T lymphocytes (CD3), macrophages (CD68) and neutrophilic granulocytes (myeloperoxidase, MPO), according to established protocols as described previously (Bonomi et al., 2024[4]).

Microscopic imaging and quantitative analyses were performed using a BX53 microscope (Olympus, Hamburg, Germany) in combination with the cellSens Dimension software (version 1.11; Olympus). Microvessel density was determined by quantifying CD31⁺ vessels within predefined ROIs located at the implant border and center and expressed as the number of vessels per mm². In addition, the fraction of GFP⁺ microvessels was calculated as the percentage of the total vessel count in each group, allowing discrimination between host- and donor-derived vascular structures. Extracellular matrix remodeling was assessed by quantifying collagen I and collagen III content relative to normal skin. Inflammatory cell infiltration was evaluated by counting CD3⁺ lymphocytes, CD68⁺ macrophages and MPO⁺ neutrophils, expressed as cells per mm², in two central and two peripheral ROIs per implant.

The sample size per experimental group (n = 8) was determined based on prior experience with the same experimental model and outcome measures (Pruzzo et al., 2025[30]). No animals were excluded from the analysis and subjects were allocated to experimental groups using random assignment.

All datasets were evaluated for normality and homogeneity of variance using GraphPad Prism software (version 10.1.2; GraphPad Software, San Diego, CA, USA). Depending on the distribution of the data, intergroup comparisons were performed using either an unpaired Student's t-test for normally distributed variables or the Mann-Whitney U test for non-normally distributed data. Data are presented as mean ± standard error of the mean (SEM), and statistical significance was defined as a two-sided p value < 0.05.

Repeated intravital fluorescence microscopy enabled the detailed evaluation of the onset of blood perfusion and the progressive development of new microvascular networks in the nanofat-seeded dermal substitutes that had been implanted into dorsal skinfold chambers of recipient mice. Notably, implants seeded with DFO-pretreated nanofat exhibited a much more pronounced microvascular ingrowth both in the border and center zones over time compared to controls (Figure 2A, B(Fig. 2)). This was reflected by a significantly higher number of perfused ROIs and functional microvessel density already at day 6 (Figure 2C-F(Fig. 2)). Microhemodynamic parameters of individual microvessels, such as diameter, centerline RBC velocity, shear rate and volumetric blood flow, were comparable in the two groups (Table 1(Tab. 1)).

To evaluate the inflammatory response to the implants, leukocyte-endothelial cell interactions were examined in peri-implant postcapillary or collecting venules of the surrounding subcutaneous host tissue. For this purpose, leukocytes were labeled in situ with 0.1 % rhodamine 6G for their visualization under green-light epi-illumination, while the microvessels were visualized under blue-light epi-illumination following injection of the plasma marker 5 % FITC-labeled dextran (Figure 3A(Fig. 3)). Notably, the number of both rolling and adherent leukocytes was significantly lower on days 3, 6, 10 and 14 in venules around implants seeded with DFO-pretreated nanofat (Figure 3B, C(Fig. 3)). However, these microvessels did not differ in terms of diameter, centerline RBC velocity, shear rate or volumetric blood flow between the two groups, indicating comparable microhemodynamic conditions (Table 2(Tab. 2)).

The implanted dermal substitutes were further examined by histology and immunohistochemistry at the end of the in vivo experiments on day 14. The vascularization of the implants was assessed by analyzing CD31-stained sections. Implants seeded with DFO-pretreated nanofat exhibited a markedly higher microvessel density in both the border and center zones compared to controls (Figure 4A, B(Fig. 4)). Furthermore, CD31/GFP co-stainings revealed that more than 90 % of microvessels in this group expressed GFP, indicating that the newly formed vessels originated from the seeded GFP⁺ nanofat. In contrast, the fraction of GFP⁺ microvessels was markedly lower in dermal substitutes seeded with vehicle-pretreated nanofat (Figure 4C, D(Fig. 4)).

The analysis of HE-stained sections revealed an improved tissue integration of dermal substitutes seeded with DFO-pretreated nanofat when compared to controls (Figure 5A, B(Fig. 5)). This was indicated by a more pronounced granulation tissue formation in the border zones of the implants. In addition, these implants contained more residual adipocytes originating from the seeded nanofat (Figure 5B(Fig. 5)). To further investigate tissue integration, the collagen content of the dermal substitutes was quantified, thereby distinguishing between Col I and Col III (Figure 6A-D(Fig. 6)). This analysis revealed a comparable total Col I ratio in the border and center zones of the dermal substitutes in both groups (Figure 6A, B(Fig. 6)). However, the overall Col III ratio was significantly higher in the implants seeded with DFO-pretreated nanofat (Figure 6C, D(Fig. 6)).

To assess the immune cell infiltration of the implants, histological sections were stained with antibodies against CD68⁺ macrophages, MPO⁺ neutrophilic granulocytes and CD3⁺ lymphocytes (Figure 7A-F(Fig. 7)). Quantitative analyses of these stained sections showed no marked differences in the lymphocyte content of implants seeded with vehicle- or DFO-pretreated nanofat (Figure 7F(Fig. 7)). However, implants seeded with DFO-pretreated nanofat exhibited a significantly lower density of CD68⁺ and MPO⁺ cells (Figure 7B, D(Fig. 7)), indicating an attenuated early innate inflammatory response.

See also the supplementary information.

We are grateful for the excellent technical assistance of Janine Becker (Institute for Clinical and Experimental Surgery, Saarland University, Homburg, Germany). Moreover, the authors thank Servier Medical Art for providing access to designed medical elements (https://smart.servier.com/), supporting the generation of graphical items in this publication.

The authors have no conflicts of interest to declare.

Artificial intelligence was not used in the preparation of this manuscript.

V.P. contributed to the conceptualization of the study, conducted the methodology and investigation, performed formal analysis and data curation, and was responsible for writing - original draft preparation. F.B. contributed to the conceptualization of the study, the methodology, and writing - review and editing. E.L. contributed to the methodology and writing - review and editing. A.W. contributed to the methodology and writing - review and editing. Y.H. contributed to the conceptualization of the study, provided resources, and participated in writing - review and editing. M.W.L. contributed to the conceptualization of the study, the methodology, provided resources, and was responsible for writing - original draft preparation as well as writing - review and editing. All authors have read and agreed to the published version of the manuscript.

The present study received no external funding.