Dendrimer Functionalized Metal Oxide Nanoparticle mediated Self-Assembled Collagen Scaffold for Skin Regenerative Application: Function of Metal in Metal oxides

Functionalized metal oxide nanoparticles cross-linked collagen scaffolds are widely used in skin regenerative applications because of their enhanced physico-chemical and biocompatibility properties. From the safety clinical trials point of view, there are no reports that have compared the effects of functionalized metal oxide nanoparticles mediated collagen scaffolds for in-vivo skin regenerative applications. In this work, Triethoxysilane - Poly (amido amine) dendrimer generation 3 (TES-PAMAM -G 3 or G 3 ) functionalized spherical shape metal oxide nanoparticles (MO NPs: ZnO, TiO 2 , Fe 3 O 4 , CeO 2, and SiO 2 , Size: 12 -25 nm) cross-linked collagen scaffolds were prepared by using a self-assembly method. Triple helical conformation, pore size, mechanical strength and in-vitro cell viability of MO-TES-PAMAM-G 3 - collagen scaffolds were studied through different methods. The in-vivo skin regenerative prociency of MO-TES-PAMAM-G 3 - collagen scaffolds were analysed by implanting the scaffold on wounds in Wistar Albino rats. The results demonstrated that MO-TES-PAMAM-G 3 - collagen scaffold showed superior skin regeneration properties than other scaffolds. The skin regenerative eciency of MO NPs followed order: ZnO> TiO 2 > CeO 2 > SiO 2 > Fe 3 O 4 NPs. This result can be attributed to higher mechanical strength, cell – viability and better antibacterial activity of ZnO-TES-PAMAM-G 3 -collagen scaffold lead to accelerate the skin regenerative properties in comparison to other metal oxide based collagen scaffolds.


Introduction
Fabrication of scaffold material with versatile properties including desirable mechanical strength, antibacterial, antioxidant, controlled biodegradability and biocompatibility is the forefront of research in skin regenerative application [1][2][3][4][5]. The collagen based scaffolds are widely used in the skin regenerative process due to their precious biological properties such as biodegradability, non-toxic, low immunogenicity and good cell adhesibility [6][7][8][9]. However, the pure form of collagen scaffold exhibits lower mechanical strength and fast biodegradation, which limits their application in-vivo studies. To conquer this issue, different strategies include using cross-linkers (physical, chemical, enzymatic), metal/metal oxide nanoparticles (Ag, Au, ZnO, TiO 2 ,SiO 2 NPs) and surface functionalized nanoparticles (metal/ metal oxide) have been used to stabilize the collagen by numerous scientists [10][11][12][13][14][15]. Among the various methods, functionalized metal oxide nanoparticle cross-linked collagen scaffold exhibited higher mechanical properties and cell viability compared to bar nanoparticles [13,16]. This result can be due to the smaller size combined with the larger surface area, resulting in many cross-linking points with functional molecules and protein. In addition to this, metal oxides based scaffold may have unpredictable physical (lighter and more porous), optical (tunable optical emission), magnetic (super paramagnetic), antibacterial property and antioxidant properties, which are unavailable at micro -or macroscales [13,[16][17][18][19][20][21]. Recently, Huang et al have been reported that functionalized metal oxide nanoparticle mediated scaffold accelerated wound healing process [22]. The functionalized metal oxide nanoparticle forms an adhesive layer because of adsorption of protein molecules onto the nanoparticle surface. The resulting bridging layer can support high amounts of stress due to rearrangement of adsorbed proteins on the nanoparticle surface, leading to the use of scaffold in regenerative medicine [23]. For example, orange essential oil (EO) functionalized ZnO nanoparticle incorporated collagen based bioresorbable scaffold improve the skin regeneration process due to their higher antibacterial and biocompatibility [24]. Collagen integrated with polycaprolactone or polyvinyl pyrrolidone (PVP) coated titanium dioxide (TiO 2 ) based scaffold can be useful as skin substitute [25][26]. Mertens et al., have demonstrated that (3-aminopropyl) Triethoxy silane (APTES) functionalized Fe 3 O 4 nanoparticle labelled collagen scaffold can be useful for non-invasive MR imaging in tissue regenerative applications [16]. Inspire of these reports, our group have studied an effect of various morphology of TES-PAMAM-G 1 dendrimer functionalized ZnO nanoparticle cross-linked collagen scaffold in skin regenerative application. The result obtained in that study indicated the sphere shape of ZnO nanoparticle exhibited a better-wound healing process than other shapes [27]. The use of PAMAM based dendrimer in that study is due to their well-de ned nanostructured macromolecules with relatively low toxicity and large number of surface functional groups leading to higher cross-linking density with collagen through EDC-NHS treatment [28][29].This cross-linking method is highly e cient, nontoxic (EDC-NHS used to activate the carboxylic groups of collagen and do not take part in cross-linking reaction) and the resultant by product can be easily removed by washing process [30].
Cell culture equipments were procured from Nunc, Denmark. Polyethylene tray with six compartments was procured from Science Ware Pvt. Ltd India. All these chemicals and other general reagents were directly used without additional puri cation. In this research, Milli-Q-water (18MΩ.cm) was used in all the experimental analysis.

Type I Collagen Extraction from Tendon of a Rat
Type I collagen solution was excerpted from rat tail tendon by using salt precipitation method [31][32]. In brief, type I collagen bers were teased out from rat tail tendons (Four month-old male Wistar Albino Rat, 3 Numbers) and placed in the ether -chloroform mixture to remove lipid content. Subsequently, tendons were immersed in 0.5 N acetic acid for 12 h at 4 o C to swell completely leads to tendons were ground and centrifuged for 30 minutes at 16000 rpm to obtain the supernatant solution. To this, 5% sodium chloride (NaCl) was added slowly until white precipitate was formed and centrifuged to collect the precipitated collagen. To concentrate crude collagen, the precipitated was dialyzed toward 0.5M acetic acid. To obtain a pure collagen solution, this solution was further dialyzed against 0.05M acetic acid to eliminate the excess salts. As a measure of hydroxyproline content, the concentration of collagen solution was determined [12].

Synthesis of Metal Oxide Nanoparticle -Coprecipitation Method
Metal oxide nanoparticles were synthesized by using the reported method with minor modi cation [33][34]. For ZnO nanoparticles, Zinc acetate dihydrate was dissolved in 100 ml ethanol/ water (80: 20) mixture and heated to 90 o C for 30 minutes. To this, the calculated amount of Lithium hydroxide solution was added drop by drop and the resulting solution was allowed to stir at the same temperature for 30 minutes. Subsequently, sodium oleate was added to the above solution and heated at 180 o C for 6h. The nal precipitate was separated and centrifuged in the presence of hydrochloric acid -methanol mixture (1:5 ratio) and water alternatively to remove the excess sodium oleate. The nal product was dried at 65 o C in a hot air oven to obtain ZnO nanoparticles. A similar procedure was adopted for the preparation of TiO 2 , Fe 3 O 4 , CeO 2 and SiO 2 nanoparticle from the respective precursor. The Fe 3 O 4 nanoparticle was prepared from a mixture of Iron (II) chloride and Iron (III) chloride precursors.

Preparation of collagen scaffold -Self-Assembly Process
Collagen bril formation (Self-assembly process) was carried out according to the previously described method [12,35]. The bril formation processes were initiated by mixing 300 µL of collagen (3 µM) solution with 2560 µL of Phosphate buffer (50 mM PBS, NaCl ionic strength 120 mM) and the pH was

Characterization of Nanoparticle
The X-ray diffraction pattern was recorded using a Mini ux 11, Rigaku diffractometer (λ = 0.1548, CuKα radiation), with samples scanned along 2θ at a speed of 4 o min − 1 in the range of 5 to 80 o . Analysis of amorphous and crystalline phases was carried out using standards in the JCPDS data le. HR-TEM and SAED images were obtained using a JEOL 3010 instrument. For this, synthesized nanoparticles were dispersed in methanol in an ultra-sonication bath, and two drops of it were deposited on carbon-coated copper (Cu) grid and analyzed under an accelerating voltage of 300 kV. Image J software was used to calculate the particle size from obtained HR-TEM image. The molecular mass of the TES-PAMAM-G 3 dendrimer was con rmed by using a Bruker MALDI-TOF (Matrix-assisted laser desorption ionization-time of ight and equipped with a pulsed nitrogen laser (337nm), operating in positive ion re ector mode, using 19 kV acceleration voltage and 2,5-dihydroxybenzoic acid in 0.1 M tri uoro acetic acid solution was used as a matrix (Fig. 2, Supplementary le). FT-IR spectrum (JASCO spectrometer) of the nanoparticles was recorded in the range 400-4000 cm − 1 and 600-4000 cm − 1 (ATR mode) at 4 cm − 1 resolution, averaged over 100 scans for powder and lm samples respectively. A transparent pellet was prepared with potassium bromide in the ratio-sample: KBr (2:98) for the powder samples, and the nal spectrum was obtained after subtracting the KBr background spectra. In the case of the lm sample, the air was used as a background. Differential scanning calorimetry (DSC) was used to assess the thermal stability of scaffolds (Q200, TA instruments). In a nitrogen atmosphere, collagen -based scaffold (∼5 mg) was uniformly mounted in aluminium pans with lids and scanned over a temperature range of 30-180 o C at a heating rate of 10 o C per minute. The denaturation temperature (Td) was estimated at the transition peaks midpoint. For each sample, DSC measurements were performed ve times and Td values were measured as the average of ve replicates.

Circular Dichroism spectroscopy -Collagen-Nanoparticle Interaction
The effect of TES-PAMAM-G 3 and its functionalized MO nanoparticle on the triple helical structure of collagen was investigated through circular dichroism spectro polarimeter (JASCO-J.815). In the nitrogen atmosphere, CD spectra were recorded for every 0.2nm by averaging ve scans in the far UV region (190-260nm) with a 1 nm bandwidth and a scan speed of 100 nm min -1 . The results of the CD spectra were expressed in molar ellipticity. The ration of positive to negative pinnacle statures (Rpn) can be used as a measure of triple -helix content, with values near 0.12 indicating a completely triple helical structure.

Characterization of Collagen-Nanoparticle Scaffolds
Tensile strength analysis was performed on collagen scaffolds using a Universal testing machine (INSTRON model 1405) at 23oC and 50% relative humidity. The rectangular shaped (1.5 x 7 cm, thickness 0.03-0.09) scaffolds were tensile tested at a crosshead speed of 5mm min -1 before the collagen scaffold ruptured. The linear portion of tensile stress-tensile strain curve was used to calculate the Young's modulus of collagen scaffold. The tensile strength, elongation at break and Young's modulus of scaffolds were measured in triplicate. High-resolution scanning electron microscopy was used to investigate the morphology of collagen scaffold (HR-SEM, Quanta 200 FEG). Before placing the sample on the stub, scaffolds were sputter -coated with gold to prevent contamination. Image J software was used to examine the collagen scaffold pore size, ber diameter and D-periodicity. where the sample weight before immersion in PBS is W i , and the sample nal weight after freeze drying is W t . Three samples were analysed in parallel, and the average outcome was plotted.

Swelling Degree of MOs-TES-PAMAM-G 3 -Collagen Scaffold
The gravimetric approach was used to assess the swelling degree (Sd) or water absorption of collagenbased scaffolds. The swelling ration, as de ned by Eq. (2), was used to assess the Sd.
Where W d is the initial weight of dry collagen, and W s is the weight of swollen collagen at equilibrium.
Every 5 hours, the dried scaffolds were weighed and immersed in ultrapure water until they reached a swelling equilibrium, and readings were taken. After that, the scaffolds were weighed again after being impaired with lter paper to extract excess water from their surface. Each value was calculated as the mean of ve separate measurements and expressed as a standard deviation (SD).

Evaluation of Cross linking degree in Collagen Scaffold
The percentage of cross-linked carboxylic group in collagen-dendrimer composite was calculated through reported methods [29]. The reaction of collagen with the acylating -acetic acid ester (HAc-NHS) in the ratio of (1:3) congested the unrestricted amine groups in collagen, and the reaction mixture was stirred for 5 h at 25 o C.This amine group congested collagen samples were added to a solution of EDC and NHS reagent, and the pH of the resulting solution was maintained at 5.5.Subsequently, 0.2 M of NaH 2 PO 4 buffer was slowly added to the above solution and centrifuged at 15000 rpm for 45 minutes to eliminate the excess NHS, and then 1 mL of 0.1M Na 2 HPO 4 buffer (pH 9.1) was added and stirred for 2 h.
The quantity of NHS released in the sample was determined through absorbance at 260 nm by consideringε = 9700 M − 1 cm − 1 .

Bio-Compatibility
The biocompatibility of collagen-based scaffolds was analysed by using MTT assay [36]. Brie y, the scaffolds viz., collagen, TES-PAMAM-G 3 -collagen, MO-TES -PAMAM G 3 -collagen (MO: ZnO, TiO 2 , Fe 3 O 4 , CeO 2 and SiO 2 NPs) were placed in the cell culture plate, and UV sterilized for 3 hours. After UV sterilization, 3x10 4 human immortalized keratinocyte (HaCaT) cells were seeded onto the 48 well tissue culture plates comprising scaffolds and allowed to grow in 5% CO 2 . After 24 hours of incubation, the culture medium was separated, and the cells were treated with 0.5 mgmL − 1 of MTT reagent in PBS and incubated in the dark at 37°C for 4 h. The purple-colored formazan crystals produced were solubilized with dimethyl sulphoxide (DMSO) after incubation and the absorbance was measured at 570nm in a Bio-Red ELISA plate reader. The following Eq. (3) was used to measure the percentage of viable cells. Group-VIII (wound enclosed with scaffold of SiO 2 -TES-PAMAM -G 3 -collagen). With the aid of scissor and a surgical needle, a 4 cm 2 (2 x2 cm 2 ) incision was made on the dorsal part of all the rats, and the wound region was immediately sterilised with surgical spirit. Animals were dressed and gauze was used to close the dressing scaffold. Each alternate day, all of the rats had their dressings changed. At every 7th day cycle, wound contraction was calculated as a percentage reduction in wound size and photographed. The region of the wound was measured using image J software, and the size of the wound was monitored on a regular basis by tracing the boundary.
The wound healing percentage was estimated by using Eq. (4)

Histopathology and Masson's Trichrome Analysis
On the 7th, 14th, and 21st post-operative days, the healed wound region was excised along with scar tissue from various groups and xed with a 12% formalin solution for 24h at room temperature to understand the healing process. Different grades of alcohol and xylene were used to treat the tissues. The tissue was then embedded in para n wax and cut into 5µm thick sections. Hematoxylin-Eosin (H&E) and Masson's trichrome reagent were used to stain tissue parts, which were then examined under an optical microscope (10X, Zeiss Axioscope microscope). c  e  a  t  5  7  0  n  m  i  n  t  r  e  a  t  e  d  c  e  l  l  s   A  b  s  o  r  b  a  n  c  e  a  t  5  7  0  n  m  i  n  t  h  e  c  o  n  t  r  o  l  c  e

Metal ion quanti cation by ICP-OES Analysis
ICP-OES was used to test the quanti cation of metal oxide nanoparticles in wound-healed skin samples (Perkin Elmer Optima 5300DV). The healed skin samples having an area of 4 cm 2 were digested in conc.
HNO 3 and HCl (7:3 ratio) mixture in a hot air oven at 65 o C for 15 minutes. The reagent blank used in the analysis was a mixture of HNO3 and HCl solution, and a calibration plot was generated by examining high purity ICP standards. The results were recorded and processed using Win Lab 32 software.

Statistical Evaluation
The experimental data from all the studies were repetitive ve times and indicated as means ± standard deviation (SD). Graph pad prism software versions 5.00 (San Diego California USA, trial version) was used for statistical analysis. Statistical signi cance was set to a p-value ≤ 0.05. suggesting that the synthesized nanomaterial exhibits an amorphous nature [37]. As seen in gure, the width of peak expansions in all metal oxide nanoparticles, a clear indication that the particle size of nanoparticle decreases [38][39]. The crystallite size of nanoparticles was estimated by using the Debye-Scherrer formula. The crystal structure, phase group, lattice constants, and crystallite size are presented in Table.1, Supplementary le. The results reveal that all the synthesized metal oxide nanoparticles exhibit a crystallite size of about ZnO (4 nm), TiO 2 (7 nm), Fe 3 O 4 (11 nm), CeO 2 (9 nm), and SiO 2 (7 nm).

Results
Absence of impurity peaks in the XRD spectrum, implying that prepared nanoparticles exhibit high purity.

HR-TEM-Morphology of Metal Oxide Nanoparticles
The morphology of different metal oxide nanoparticles was assessed through HR-TEM analysis. Figure 3 shows that the synthesized MO nanoparticles (ZnO, TiO 2 , Fe 3 O 4 , CeO 2, and SiO 2 ) exhibit spherical morphology with uniform size through the area. The diameter of ZnO, TiO 2 , Fe 3 O 4 , CeO 2, and SiO 2 nanoparticles were in the range of 3-5 nm, 3-7 nm, 5-12 nm, 4-9 nm and 7-14 nm respectively. As seen in gure, the SAED pattern reveals that synthesized nanoparticle exhibit polycrystalline nature except for SiO 2 nanoparticles, which exhibits an amorphous character.

FTIR-Functionalization of TES-PAMAM-G 3 dendrimer on Metal Oxide Nanoparticles
. Metal oxide nanoparticles were evaluated using FT-IR spectroscopic techniques to ensure the functionalization of the TES-PAMAM-G 3 dendrimer. Figure 4 shows the FTIR spectrum of the TES-

DLS-Hydrodynamic Size of MO-TES-PAMAM-G 3 NPs
Hydrodynamic diameter, polydispersity index, and colloidal stability (measured in terms of zeta potential) of TES-PAMAM-G 3 dendrimer and its functionalized metal oxide nanoparticles were analyzed and results are presented in Table 2 TES-PAMAM-G 3 functionalized metal oxide nanoparticles were higher than pure metal oxide nanoparticles owing to existence of dendrimer molecules on metal oxide nanoparticle surface. The observed difference in size of metal oxide nanoparticle by XRD, TEM and DLS analysis due to difference principles employed in each techniques during the measurement of nanoparticle size. The DLS analysis is based on the Brownian motion of the nanoparticle in a suspension that provide mean hydrodynamic diameter, which is typically larger than XRD and TEM techniques because it contains a few solvent layers on surface. In general, colloidal stability of nanoparticle has been playing an important role in proteinnanoparticle interaction for various biomedical applications. TES-PAMAM -G 3 dendrimer functionalized metal oxide nanoparticles exhibited zeta potential in the rage of + 12 to + 30 mV and polydispersity index (PDI) ∼0.3, and these results indicate moderate to high colloidal stability. Both FT-IR and zeta potential measurements clearly reveal that TES-PAMAM -G 3 dendrimer was present on metal oxide nanoparticles, providing them with better colloidal stability.

CD-Structural Changes of Collagen in presence of MO-TES-PAMAM-G 3 NPs
Collagen triple helical structure is critical for preserving the mechanical strength and cell viability of the protein in biomedical applications [40]. The effects of TES-PAMAM-G 3 and MO-TES-PAMAM-G 3 on the triple helical structure of collagen were studied using a CD spectrum (S 2, Supplementary le). Collagen has a positive peak at 220 nm (n→π*) and a negative peak at 198 nm (π→π*) in the CD spectrum, with a cross-over at 210 nm in the far UV-region. Collagen CD spectrum was found to be identical to previously recorded spectra [41]. collagen is preserved. It has been reported that during the protein -nanoparticle interactions, the conformation changes of protein depending on the size and shape of the nanoparticle [42][43]. In addition, the surface curvature nanoparticle has substantial in uence on adsorption of protein molecules, which can cause various conformational changes structure adsorbed to at surface of the same material [44]. In this study, sphere shape of dendrimer functionalized metal oxide nanoparticle upon interaction with collagen does not change its triple helical structure due to smaller size (12-25 nm) with no surface curvature leads to retain more native like structure.

Self-Assembly Process-Rate of Fibril Formation
The effect of MO-TES-PAMAM-G 3 -NPs on the collagen self-assembly process can be analysed by  Fig. S3, Supplementary le The collagen displayed a typical selfassembly or turbidity curve, similar to a sigmoidal pro le, as seen in the gure, which is thought to have resulted from the three different stages of the brillogenesis process, namely nucleation, growth, and gelation [35,45]. The pseudo kinetic parameters lag (t) (nucleation lag time) and slope (growth rate) were extracted from the absorbance pro le by tting the experiment results. The presence of TES-PAMAM -G 3 and TES-PAMAM -G 3 functionalized ZnO nanoparticle reduced the lag (t) value gradually with increasing concentration of nanoparticle, indicating that functionalized metal oxide nanoparticles accelerated the nucleation process of collagen. Experiments with additions of the other metal oxides such as TiO 2 , Fe 3 O 4 , CeO 2, and SiO 2 nanoparticles to collagen also followed the same trend in lag (t) values (Table 3 scaffold has 28 times higher denaturation temperature [27]. This observation suggested that ZnO-TES-PAMAM-G 3 nanoparticle-mediated collagen scaffold had higher denaturation temperature leading to higher order stability than other scaffolds.

Mechanical Properties
The mechanical strength of the collagen based scaffold was assessed using a tensile strength test and the results are shown in Fig. 6b. As can be seen in the gure, all collagen-based scaffolds exhibited a typical stress-strain curve. The pure collagen scaffold exhibited higher elongation than other scaffolds, which implies that the collagen scaffold possesses higher exibility than other scaffolds. In the case of metal oxide nanoparticle mediated collagen scaffold a distinctive yield point is observed when compared to collagen, EDC-NHS, and TES-PAMAM -G 3 cross-linked collagen scaffold indicating a transition from an elastomeric to a tough plastic nature. The linear component of the stress-strain curves was used to measure the Young's modulus. The tensile strength, percentage of strain, and Young's modulus of scaffolds are presented in Table 4

Biodegradation of MO-TES-PAMAM-G 3 -Collagen Scaffold
The in vitro biodegradation pro le of dendrimer cross-linked collagen scaffold as a function of degradation time is shown in Fig. 6c. The obtained results showed that all collagen-based scaffolds prepared in this work had higher biodegradation. However, in contrast to the collagen scaffold (98% after 30th day), the rate of weight loss was lower in the case of nanoparticle-mediated collagen scaffolds (82-90 % ± 0.32-2.5), indicating that the degradation trend is affected by the degree of cross-linking and the existence of the cross-linker. The cross-linked scaffold initially had low weight loss (up to one week) and thereafter the scaffold underwent steady degradation indicated by a decrease in weight with time. As compared to the collagen scaffold, the metal oxide nanoparticle mediated collagen scaffold has a higher degree of cross-linking and is more resistance to degradation in PBS medium at 37 o C. The formation of covalent bonds between collagen and MO-TES-PAMAM-G3 NPs may be possible reason for higher scaffold strength and for a longer period of time.

Swelling Degree of MO-TES-PAMAM-G 3 -Collagen Scaffold
The swelling degree or water uptake is a signi cant parameter that represents the e ciency of oxygen and nutrient transfer inside the scaffold. Swelling degrees of collagen, EDC-NHS-collagen, TES-PAMAM-G 3 -collagen, and MO-TES-PAMAM-G 3 -collagen scaffolds in water medium are showed in Fig. 6d.
Generally, highly porous collagen scaffolds have a very high swelling degree, compared to less porous scaffold [49]. As shown in the gure, collagen, TES-PAMAM-G 3 -collagen, and MO -TES-PAMAM-G 3collagen scaffolds possess higher swelling capability. However, the swelling degree of the collagen scaffold was much higher than the metal oxide nanoparticle mediated collagen scaffold because of the hydrophilic properties of the carboxylic group and its porous structure, which can greatly in uence the scaffold swelling degree. Among the different metal oxide nanoparticle mediated collagen scaffolds, there is no signi cant difference in swelling behaviour. This nding was corroborated with reports of Ullah et al. have reported that the swelling ratio of collagen was decreased in the presence of ZnO nanoparticle [14].This effect may be explained by the metal oxide nanoparticle interacting with the collagen bre and serving as a ller to ll the gap space inside the scaffold network, resulting in a more compact scaffold that does not swell as much as a pure collagen scaffold.

Wound healing study of MO-TES-PAMAM-G 3 -Collagen Scaffolds
The wound healing e ciency of the collagen-based scaffold was investigated through in vivo animal test on Wistar Albino rats. For this collagen, TES-PAMAM-G 3 -collagen, and MO-TES-PAMAM-G 3 -collagen scaffolds were implanted on the wounded area and wound healing e ciency monitored periodically (on days 7, 14, 21, 28) and percentage of wound closure was measured ( Fig. 7a-b). For the comparison of wound healing process, saline water treated open wound was kept as a control. As seen in the gure, the groups that were treated with metal oxide nanoparticles had stronger wound healing properties than the other groups all of the time.
On day 7, there was 62-79% wound contraction in the metal oxide nanoparticle treated groups, whereas in control, collagen, TES-PAMAM-G 3 -collagen scaffold treated groups wound contraction was observed to be 28%, 46% and 54% respectively. On day 9, wound contraction (90%) was observed in ZnO mediated collagen scaffold compared to other metal oxide nanoparticles. The wound contraction was enhanced in all the groups with no sign of in ammation on day 14. In presence of metal oxide nanoparticle treated groups, the wound contraction nearly reached 71-96%, while the wounds of the control and other groups were not healing well. On day 21 and 28, the same tendency was observed with metal oxide nanoparticle treated groups. The increase in wound healing e cacy observed with metal oxide nanoparticle treated scaffolds due to their smaller size with better antibacterial activity through electrostatic interaction of positively charged metal oxide with negatively charged bacterial membrane leads to release the reactive oxygen species (ROS) on surfaces of the nanoparticles, which result in damage bacterial cell (Fig. S7  Although the nanoparticle-mediated collagen scaffold treated wound parts still had a few in ammatory cells, they also showed broblast and blood vessel formation, as well as collagen synthesis. As can be seen in the gure on day 14, the wound contraction was greatly improved in all groups, in particular, ZnO nanoparticle treated group showed higher wound contraction than other groups.
On day 21st, a continuous nascent epithelial layer and some collagen bers began to grow over the nanoparticle-mediated collagen scaffolds-treated wounds, indicating complete healing. Furthermore, histopathological images of nanoparticle-mediated collagen scaffold revealed more matured dermis and epidermis layers when compared to control groups. The non-treated group had a slower epithelialization rate and less collagen bundle development, as well as irregular collagen bre packing. The average gap length between newly formed epithelium in TES-PAMAM-G 3 -collagen scaffolds treated group was lower than collagen and control groups. On day 21, metal oxide nanoparticle treated groups showed complete wound contraction than other scaffolds, especially ZnO nanoparticle treated group exhibited completely closed wound. The enhanced wound healing behavior of nanoparticle -mediated collagen scaffold treated groups may be attributed to a combination of factors including a favourable bioactive environment for re-epithelialization, antimicrobial growth inhibition, uid handling properties, and moisture vapour permeability of the scaffold, all of which could provide an ideal moist environment around the wound site for accelerated healing [50].
On the other hand, Masson's trichrome staining of TES-PAMAM-G 3 and ZnO -TES-PAMAM-G 3 treated collagen scaffolds were indicated (Fig. 8b) the formation of fair and typical collagen layer with large macrophage and broblast density. This observation inferring that wounds treated with ZnO-TES-PAMAM-G 3 NPs healed faster in comparison to all other groups due to more number of broblast, keratinocyte migration and collagen deposition in wound site leading to acceleration in the healing process.

Quanti cation of released metal ion from scaffold -ICP-OES Analysis
The metal ion content of TES-PAMAM-G 3 functionalized MO nanoparticle mediated collagen scaffolds implanted skin is presented in Table 5