Pliable, Scalable, and Degradable Scaffolds with Varying Spatial Stiffness and Tunable Compressive Modulus Produced by Adopting a Modular Design Strategy at the Macrolevel

Clinical results obtained when degradable polymer-based medical devices are used in breast reconstruction following mastectomy are promising. However, it remains challenging to develop a large scaffold structure capable of providing both sufficient external mechanical support and an internal cell-like environment to support breast tissue regeneration. We propose an internal-bra-like prototype to solve both challenges. The design combines a 3D-printed scaffold with knitted meshes and electrospun nanofibers and has properties suitable for both breast tissue regeneration and support of a silicone implant. Finite element analysis (FEA) was used to predict the macroscopic and microscopic stiffnesses of the proposed structure. The simulations show that introduction of the mesh leads to a macroscopic scaffold stiffness similar to the stiffness of breast tissue, and mechanical testing confirms that the introduction of more layers of mesh in the modular design results in a lower elastic modulus. The compressive modulus of the scaffold can be tailored within a range from hundreds of kPa to tens of kPa. Biaxial tensile testing reveals stiffening with increasing strain and indicates that rapid strain-induced softening occurs only within the first loading cycle. In addition, the microscopic local stiffness obtained from FEA simulations indicates that cells experience significant heterogeneous mechanical stimuli at different places in the scaffold and that the local mechanical stimulus generated by the strand surface is controlled by the elastic modulus of the polymer, rather than by the scaffold architecture. From in vitro experiments, it was observed that the addition of knitted mesh and an electrospun nanofiber layer to the scaffold significantly increased cell seeding efficiency, cell attachment, and proliferation compared to the 3D-printed scaffold alone. In summary, our results suggest that the proposed design strategy is promising for soft tissue engineering of scaffolds to assist breast reconstruction and regeneration.


Fabrication of textiles, flat-knitting and non-woven
To assess how the mechanical properties of the multi-layered scaffolds can be tailored depending on the mechanical properties of the textiles, an array of textiles was fabricated from PLA, through flat knitting and die punching. The elastic modulus of the textiles was used in simulation of scaffold according to Table S1.
The flat knitted textiles were made using two yarns, in 1a and 1b tubular stiches only have been used however with an increased yarn tension in 1a. Textile 2 is a non-woven mesh produced through die punching using the same web as have been spun into yarn for the other textiles. By combining textile 1a and 2 to yield textiles 3a-3c, the non-woven can reduce the stiffness of the construct depending on how it is attached to the textile and subsequent treatment of the textile.
In 3a the non-woven (2) was placed inside textile 1a, whilst for 3b and c it was attached on the side, 3b was additionally treated with calendaring afterwards. In textile 4 a stitch pattern of 1 rib and 5 tubular was used, then the textile was calendared. For textile 5 the non-woven was attached on the side of textile 4 using die punching and once again calendared. Due to the low mechanical properties and experimental issues related to testing and fabrication the non-woven mesh was only incorporated the simulation only been used in combination with textile 1a and 4 to create textiles 3a-c and 5. Micrographs of the textiles made through different manufacturing techniques and knitting patterns are shown in Figure S1.  Flat knitted made with tubular stiches made of two yarns, the non-woven was placed inside and the whole structure was calendared 3b 20 Flat knitted of tubular stiches using two yarns before calendaring, the non-woven was fixed to the textile through die punching before it was once again calendared.
3c 10 Flat knitted of tubular stiches using two yarns before calendaring, the non-woven was fixed to the textile through die punching.

60
Flat knitted with the pattern of 1 rib and 5 tubular stitches using two yarns before heat treated with calendaring.

30
Flat knitted with the pattern of 1 rib and 5 tubular stitches using two yarns before heat treated with calendaring, non-woven was fixed to the textile through die punching before it was once again calendared.

Simulation Cases
The details regarding the simulation of the combined scaffold for compression and tension using 3D printed structures and mesh are further stated in tables S2-S3. The G15 and GS15 structures have previously been analysed and published. 1 The mechanical properties of the textiles in Table S1 were characterized, and the elastic modulus data were used in the simulations below. Information and data from the mesh characterization can be found in the Table S2 and Fig. S2. The meshes that are attached to the original G15 scaffold in Table S2 are represented by six textiles/meshes with elastic moduli of 10, 20, 30, 40, 60 and 130 MPa and Poisson's ratio of 0.3. The manufacturing details of these textiles/meshes are given in Table S1, among them, TIGR® Matrix is a commercially available product with an elastic modulus of 40 MPa.
To understand the impact of elastic modulus of the textiles used in Table S2 on the three modular scaffold designs, simulations were run on the three cases OS, TS and TSM, Fig. S2. More information regarding the influence of mesh type on the compressional and tensional modulus can be seen in Table S3 and S4 as well as Fig. S3 and S4. Main assumption A prefect adhension between the mesh and the scaffold is assumed, i.e. the adhensive surface is plat, smooth, and no penetrations.
The impact of the selection of meshes from Table S1 used in the simulation case in Table S2 on the compressive moduli if G15 or GS15 are printed using PCLDX are shown in Fig. S3. The G15 and GS15 designs are further described in an earlier publication 1 .

Fig. S3
The influence of mesh type from table S1 on the compressional properties of scaffolds (PCLDX) with different combination designs (the position of mesh to the scaffold) and if the G15 or GS15 were printed.
The impact of the selection of meshes from Table S1 used in the simulation case in Table S3 on the tensional moduli are shown in Fig. S4.

Fig. S4
The influence of mesh type from table S1 on the tensional properties of scaffolds (PCLDX) with different combination designs (the position of mesh to the scaffold) and if the G15 or GS15 were printed.
The simulated case of the effect of the pore size of the printed scaffold to on the tensile moduli from Table S3 is shown in Fig. S5.

Fig. S5
The effect of the pore size (the strand space between two strands in one layer) on mechanical properties of scaffolds (PCLDX) with different combination designs (the position of mesh to the scaffold) under tension.
To explore the future use of these modular scaffolds for larger devices, the simulation case for a device composted of two scaffolds is shown in Fig. S6. Due to the isotropic nature of the TSM and G15 scaffold, the values from the x-direction is showed in Fig. S7.

Fig. S7
The stress arising in the x-direction in cyclic testing during 10 cycles of the TSM and G15 scaffold design in wet conditions at 37 °C are shown.
Cell proliferation and attachment at 7 and 11 days of the HDFs are shown with SEM micrographs in Fig. S8   Fig. S8 The proliferation and attachment HDFs in the modular designs are shown at day 7 and day 11 in SEM micrographs, the scale bar showing 200 micrometer. The yellow arrows are indicating the cells onto the scaffold structures. One mesh layer and one 3-layer 3D-printed G15 structure was gently lifted off to expose the nanofiber network. The red arrows show the nanofibrous structure inside the TSM_NF scaffold, with the scale bar showing 500 micrometers.