This assignment was an extremely informative foray into 3D printing. We used Fused Deposition Modeling (FDM) and Stereolithography (SLA) to print protein models. Specifically, we decided to print the tau protein, a critical protein in research of Alzheimer’s and other tauopathy diseases such as Argyrophilic grain disease, Pick’s disease and even Chronic Traumatic Encephalopathy (CTE) (Zhang et al., 2022).
The function of Tau is to bind to and stabilize microtubules; when this function is impeded, the microtubules destabilize and the dysfunctional tau ends up binding to other tau. A large number of these can bind together and precipitate, leading to the infamous “neurofibrillary tangles” seen in cases of Alzheimer’s disease (Zhang et al., 2022).
Tau is a small protein, composed of a single peptide chain with 758 residues. However, we decided to limit our analysis even further to residues 267-312, which are most important for tau’s binding to microtubules (Kadavath et al., 2015). This sequence also constitutes the bulk of a sequence recently (this week recently!) found to be a potential therapeutic target for antibodies to soluble pre-tangle tau (Islam et al., 2025).
First, we find the protein segment on the Research Collaboratory for Structural Bioinformatics Protein Data Bank (RCSB PDB). This particular protein has the PDB code 2MZ7. We download the .pdb file, then render it in PyMol:
Ribbon model of tau protein residues binding to microtubule (PDB 2MZ7).
We then change the display settings to show the protein as a space-filled model by selecting the “Surface” option:
Space-fill model of the protein. This creates a larger 3D structure for the protein.
We export this as a .stl file from PyMol and port it to a USB, which we connect to a computer in the OEDK 3D printing section. We open up Bambu Studio and PreForm for FDM and SLA, respectively.
Screen capture of sliced file for the protein in Bambu Studio.
Screen capture of sliced model for SLA printing via PreForm.
In both environments, we scale the .stl by a factor of 2 to increase the size of the print without over-resolving it –– there are not very many residues in this model and having a large model would risk overemphasizing structural aspects of the space-filled representation that are probably not very informative / revealing about the protein function.
FDM print in progress. It took roughly two and a half hours to complete.
SLA print in progress. It took roughly 7 hours for the printing, followed by 20 minutes for washing and 30 minutes for curing.
The FDM print took 2 hours 37 minutes, while the SLA print took 6 hours 58 minutes (followed by 20 minutes for washing and 30 minutes for curing).
The first FDM print had hard-to-remove supports, leaving rough edges that needed to be sanded; we hypothesized that changing the print orientation slightl would allow for better auto-support positioning, and consequently less roughness in a second FDM print.
Second FDM print with support. The two FDM prints were oriented slightly differently for the purpose of determining whether it would be easier to remove the support and have less roughness for the second one.
SLA print post-washing, pre-curing. The print was then placed in the curing machine for 30 minutes to solidify the resin.
Cured SLA print with supports removed.
We then sanded down the FDM print. Given the relatively small size of the print, we decided to print it a second time to assess whether changing the print orientation slightly and consequently auto-support placing would lead to a smaller “rough area” and be something worth considering for future prints. This was a roughly 2.5 degree rotation along the y-axis of the protein.
The difference between the two FDM prints was not significant but removing the supports was considerably easier the second time, which may have led to reduced roughness / tiny pieces of extra support getting stuck to the model.
Sanding the rough edges of the FDM prints before painting.
We then spray painted the proteins with grape spray paint.
Spray painting the FDM proteins with grape purple color.
Spray-painted SLA print.
After two coats of this, we gave the prints a matte coat.
Dried matte FDM prints.
The cost analysis for this project is below:
| Cost Type | Cost | Price | Source | Quantity | Total |
| Materials | PLA filament | $19.99 / kg | Bambu Lab | 2.12 oz | $1.20 |
| Clear Resin V4 | $149.99/L | formlabs | 62.36 mL | $9.35 | |
| Rust-Oleum Spray Paint (Grape Gloss) | $5.98 | Home Depot | 6.60 g (Based on a rate of 0.33 g/sec) | $0.12 | |
| Rust-Oleum Spray paint (Matte Clear) | $5.98 | Home Depot | 1.65 g (Based on a rate of 0.33 g/sec) | $0.03 | |
| Labor | Prototyping Engineer | $36 / hr | ZipRecruiter | 6 hours | $216 |
| Overhead | Bambu 3D printer (rental price) | $19/day | Fat Llama | 1 day | $19 |
| Facility Time | $85/day | The Ion | 1 day | $85 | |
| Post-processing tool (Sandpaper) | $12.68/20 pcs of 9×11 320 grit paper | Amazon | ⅛ sheet | $0.08 | |
| Post-processing tool (Tweezers) | $5.47 | Amazon | 1 set | $5.47 | |
| Design | Engineering and Development | $25.75/hr | ZipRecruiter | 1 hour | $25.75 |
| TOTAL COST | $398.00 |
References
Islam, T., Hill, E., Abrahamson, E.E. et al. Phospho-tau serine-262 and serine-356 as biomarkers of pre-tangle soluble tau assemblies in Alzheimer’s disease. Nat Med (2025). https://doi.org/10.1038/s41591-024-03400-0
Kadavath, H., Jaremko, M., Jaremko, Ł., Biernat, J., Mandelkow, E., & Zweckstetter, M. (2015). Folding of the Tau Protein on Microtubules. Angewandte Chemie (International ed. in English), 54(35), 10347–10351. https://doi.org/10.1002/anie.201501714
Zhang, Y., Wu, KM., Yang, L. et al. Tauopathies: new perspectives and challenges. Mol Neurodegeneration 17, 28 (2022). https://doi.org/10.1186/s13024-022-00533-z
