Simulation

Process

Design of Linker Sequences Connecting Actin and Thymosin β4

Examination of Linker Sequences Connecting Actin and Thymosin β4

Protein Structure Analysis

Modification of Thymosin β4 by RFdiffusion

Design and exzamination of amino acid sequence of Modified Thymosin

Design and Examination of linkers conecting Actin and Modified Thymosin

Conclusion

Data from Simulation and Experimental Results

To access any information that is not directly available on this page, please check the documentation here.

Process

fig 1. simulation process

Access to the Details of Software Used

According to fig1, we conducted simulations and aimed to design new proteins that satisfy the following two functional requirements to achieve our goal:

Molecules floating freely in solution do not spontaneously polymerize.
Paths formed by the polymerization of molecules extending from BASE and TARGET must connect with each other.

Design of Linker Sequences Connecting Actin and Thymosin β4

First, we considered designing molecules that do not polymerize in solution by linking Actin and Thymosin β4 with a linker to create a single fusion molecule. To achieve this, we worked on designing the amino acid sequence of the linker connecting Actin and Thymosin β4. Specifically, we aimed for a design that incorporates amino acids such as glycine (G) and serine (S) for flexibility, proline (P) for structural rigidity, and alanine (A) for its simple structure. The goal was to achieve high flexibility and low polarity while minimizing any influence on each molecule. Additionally, we referred to linkers used related research to connect proteins and made improvements to optimize our design.

fig 2. List of 20 amino acids

Source：What Are The Two Rare Amino Acids?

Examination of Linker Sequences Connecting Actin and Thymosin β4

Using the designed linker sequence, we connected Actin and Thymosin β4 and employed AlphaFold3 to predict molecular structures and intermolecular interactions. Specifically, we compared the output results from AlphaFold3 when inputting actin - thymosin β4 (linker) with the results obtained when inputting Actin and Thymosin β4 individually. This approach aimed to design a linker sequence that ensures Thymosin β4 inhibits the polymerization of Actin molecules.

Below is a summary of how to interpret the output results from AlphaFold3.

Summary of how to interpret AlphaFold3 output results

fig 3. AlphaFold3 output results

When running AlphaFold3, output results like those shown in the image above are obtained. The following provides a brief explanation for points ①, ②, and ③:

① ipTM and pTM Scores

The ipTM score is an indicator used to evaluate the strength of interactions between proteins. When the interaction between proteins is strong, the ipTM score is high; conversely, it is low when the interaction is weak. A score close to 1 indicates high interaction accuracy, and a value of 0.5 or higher suggests a high probability of molecular interaction. The pTM score indicates the prediction accuracy of the overall output structure.
In this simulation, we focused on the ipTM score, which quantifies intermolecular interactions, as the goal is to inhibit Actin polymerization via Thymosin β4.

② Predicted Protein Structures

The predicted structure on the left is color-coded based on the plDDT score. Regions with scores of 90 or higher are shown in blue, scores between 70 and 90 are light blue, scores between 50 and 70 are yellow, and scores below 50 are orange. The plDDT score ranges from 0 to 100 and indicates the confidence level of the predicted structure. Blue and light blue regions (plDDT ≥ 70) are considered structurally stable and likely fixed by interactions. In contrast, yellow and orange regions (plDDT < 70) are expected to be more flexible, indicating weak interactions in those areas.

③ Graph of Relative Position Errors Between Amino Acids

The graph on the right shows the sequential numbering of the input amino acid sequences on the vertical and horizontal axes, with the pae score represented as a gradient from green to white. The pae score indicates the relative position error between two amino acids. Green regions represent low pae scores and thus small relative position errors, while white regions indicate larger errors.
When multiple molecules are input, the relative position errors between amino acids within a single molecule are small and displayed in green due to stable structures. In contrast, the relative position errors between amino acids of different molecules depend on the strength of their interactions. Weak interactions result in larger errors, shown in white, while strong interactions result in smaller errors, displayed in green.

AlphaFold3 results did not reveal an optimal linker connecting Actin and Thymosin β4

Below, we have summarized the output results for the following four cases: Actin only, Actin and Thymosin β4, and Actin and Thymosin β4 linked by the linker sequences "ASSGGSGSGGSGGA" and "GGGS." For each case, outputs were generated for molecule counts of 1, 3, 5, 8, and 10. The AlphaFold3 output results and ipTM scores for each molecule count were organized into a table for comparison. Notably, the ipTM scores for molecule counts 3 and 5, which showed significant changes, were highlighted in blue.

Actin (Click here to see the amino acid sequence)

DEDETTALVCDNGSGLVKAGFAGDDAPRAVFPSIVGRPRHQGVMVGMGQKDSYVGDEAQSKRGILTLKYPIEHGIITNWDDMEKIWHHTFYNELRVAPEEHPTLLTEAPLNPKANREKMTQIMFETFNVPAMYVAIQAVLSLYASGRTTGIVLDSGDGVTHNVPIYEGYALPHAIMRLDLAGRDLTDYLMKILTERGYSFVTTAEREIVRDIKEKLCYVALDFENEMATAASSSSLEKSYELPDGQVITIGNERFRCPETLFQPSFIGMESAGIHETTYNSIMKCDIDIRKDLYANNVMSGGTTMYPGIADRMQKEITALAPSTMKIKIIAPPERKYSVWIGGSILASLSTFQQMWITKQEYDEAGPSIVHRKCF

fig 4. AlphaFold3 output result of Actin

Five images in fig.4 represent the AlphaFold3 output results for Actin only, with molecule counts of 1, 3, 5, 8, and 10. First, focusing on the protein structures and graphs, it is evident that for molecule counts of 3 and 5, the graphs are predominantly green, indicating strong interactions between Actin molecules and polymerization into fibrous structures. Additionally, for molecule counts of 8 and 10, the green regions in the graphs split into two distinct blocks, confirming that the Actin molecules polymerize into two separate fibrous structures (highlighted with red circles).

Furthermore, examining the ipTM scores reveals values of 0.62, 0.75, 0.43, and 0.47 for molecule counts of 3, 5, 8, and 10, respectively. Notably, the higher values for molecule counts of 3 and 5 indicate strong interactions leading to fibrous polymerization. On the other hand, the relatively lower scores of 0.43 for molecule count 8 and 0.47 for molecule count 10 can be attributed to the division of Actin into two separate fibrous structures (highlighted with red circles).

Actin and Thymosin β4 (Click here to see the amino acid sequence)

Actin :

SDKPDMAEIEKFDKSKLKKTETQEKNPLPSKETIEQEKQAGES

fig 5. AlphaFold3 output result of Actin and Thymosin β4

Five images in fig 5 represent the AlphaFold3 output results for Actin and Thymosin β4 with molecule counts of 1, 3, 5, 8, and 10. To verify whether Thymosin β4 effectively inhibits Actin polymerization, we compared the results obtained when only Actin was input with those when Actin and Thymosin β4 were input separately.

Focusing on the graph on the left, the large relative positional errors between amino acids of different molecules are displayed in white, indicating weakened interactions between Actin molecules. Additionally, examining the ipTM scores for molecule counts of 3, 5, 8, and 10 reveals values of 0.33, 0.25, 0.47, and 0.34, respectively. Notably, for molecule counts of 3 and 5, the scores are less than half of those when only Actin was input.

Furthermore, for a molecule count of 10, Actin forms a circular structure, while for a molecule count of 8, Actin polymerizes into two separate fibrous structures (highlighted with red circles), similar to the case when only Actin was input. These results confirm that adding Thymosin β4 weakens interactions between Actin molecules and partially inhibits polymerization.

Actin-Thymosin β4 (ASSGGSGSGGSGGA) (Click here to see the amino acid sequence)

fig 6. AlphaFold3 output result of Actin-Thymosin β4(ASSGGSGSGGSGGA)

Five images in fig 6 represent the AlphaFold3 output results for molecules where Actin and Thymosin β4 are linked by the sequence "ASSGGSGSGGSGGA" (Actin-Thymosin β4 (ASSGGSGSGGSGGA)), with molecule counts of 1, 3, 5, 8, and 10. To verify whether Thymosin β4 can inhibit Actin polymerization even when linked via the linker, we compared these results with those obtained when Actin and Thymosin β4 were input separately.

First, focusing on the protein structures and graphs, for molecule counts of 3 and 5, the graphs are predominantly green, indicating strong interactions between Actin molecules and polymerization into fibrous structures. For molecule counts of 8 and 10, the green regions on the graphs split into two blocks, confirming that Actin polymerizes into two separate fibrous structures (highlighted with red circles).

Additionally, examining the ipTM scores for molecule counts of 3, 5, 8, and 10 reveals values of 0.55, 0.7, 0.44, and 0.52, respectively. Compared to the case where Actin and Thymosin β4 were input separately, the scores are notably higher, particularly for molecule counts of 3 and 5.

These findings indicate that when Actin and Thymosin β4 are linked via the "ASSGGSGSGGSGGA" linker sequence, Thymosin β4 is insufficiently effective at inhibiting Actin polymerization.

Actin-Thymosin β4 (GGGS) (Click here to see the amino acid sequence)

fig 7. AlphaFold3 output result of Actin-Thymosin β4(GGGS)

Five images in fig 7 represent the AlphaFold3 output results for molecules where Actin and Thymosin β4 are linked by the sequence "GGGS" (Actin-Thymosin β4 (GGGS)), with molecule counts of 1, 3, 5, 8, and 10. To verify whether Thymosin β4 can inhibit Actin polymerization even when linked via the linker, we compared these results with those obtained when Actin and Thymosin β4 were input separately.

First, focusing on the protein structures and graphs, it was observed that, similar to the case where the linker sequence "ASSGGSGSGGSGGA" was used, interactions between Actin molecules occurred, resulting in polymerization into fibrous structures.

Additionally, examining the ipTM scores for molecule counts of 3, 5, 8, and 10 reveals values of 0.6, 0.58, 0.45, and 0.34, respectively. Compared to the case where Actin and Thymosin β4 were input separately, the scores are notably higher, particularly for molecule counts of 3 and 5.

These findings indicate that when Actin and Thymosin β4 are linked via the "GGGS" linker sequence, Thymosin β4 is insufficiently effective at inhibiting Actin polymerization.

From these results, comparing the ipTM scores of molecules where Actin and Thymosinβ4 are linked via a linker with those where Actin and Thymosin β4 are input separately reveals that linking them significantly increases the ipTM score, with the entire graph displayed in green.

Furthermore, in addition to the linkers listed on the wiki, we designed approximately 100 different linker sequences and input them into AlphaFold3 to predict interactions. In all cases, Actin polymerization occurred.

These findings indicate that when Actin and Thymosin β4 are linked via a linker, Thymosin β4 is insufficiently effective at inhibiting Actin polymerization.

Protein Structure Analysis

We investigated the reason why Thymosinβ4 becomes nonfunctional when Actin and Thymosin β4 are linked via a linker. The results output by AlphaFold3 were downloaded, and detailed structural analysis was performed using PyMOL.

In AlphaFold3, the structure's color corresponds to the plDDT score. However, note that in PyMOL, to improve visibility, Actin is colored green and Thymosinβ4 is colored cyan.

The C-terminus of Actin and the N-terminus of Thymosin β4 are in close proximity

fig 8. Overall view of
Actin and Thymosin β4

fig 9. The C-terminus of Actin(red) and
the N-terminus of Thymosin β4(blue)

fig 10. The N-terminus of Actin(blue) and
the C-terminus of Thymosin β4(red)

Fig8,9,10 show the AlphaFold3 output results observed in PyMOL when Actin and Thymosin β4 were input separately as single molecules. To improve clarity, the C-terminus of each molecule is colored red, and the N-terminus is colored blue.

Using PyMOL for structural analysis of Actin and Thymosin β4, the distance between the C-terminus of Actin and the N-terminus of Thymosin β4 was measured to be 8.7 Å, while the distance between the N-terminus of Actin and the C-terminus of Thymosin β4 was 61.8 Å. Based on this, it was determined that linking the relatively closer C-terminus of Actin and the N-terminus of Thymosin β4 would result in a linker that is less likely to interfere with the interactions between Actin and Thymosin β4. Consequently, a new protein was designed by connecting Actin and Thymosin β4 in that order using a linker.

When Actin and Thymosin β4 are connected by a linker, the α-helix at the C-terminus of Thymosin β4 is removed from actin.

fig 11. Actin + Thymosin β4

fig 12. Actin-Thymosin β4 (ASSGGSGSGGSGGA)

fig 13. Actin-Thymosin β4
(GGGS)

Fig 11,12,13 represent the AlphaFold3 output results observed in PyMOL for "actin and thymosin β4," "actin-thymosin β4 (ASSGGSGSGGSGGA)," and "actin-thymosin β4 (GGGS)" with a molecule count of 3. To improve clarity, the linker regions are displayed in red, and the α-helix at the C-terminus of Thymosinβ4 is highlighted with a red circle.

From each figure, it can be observed that compared to "actin and thymosin β4," the α-helix at the C-terminus of Thymosin β4 detaches from Actin when Actin and Thymosin β4 are connected via a linker. This suggests that Thymosin β4 does not function adequately on Actin, leading to continued polymerization of Actin.

Modification of Thymosin β4 by RFdiffusion

Therefore, we utilized RFdiffusion to generate a new peptide designed to replace the α-helix at the C-terminus of Thymosin β4, aiming to create stronger interactions with Actin.

The length of peptides generated by RFdiffusion was set to 10-15 residues

fig 14. Alpha helix (red) at the C-terminus of thymosin

First, to generate a new peptide to replace the α-helix at the C-terminus of Thymosin β4 using RFdiffusion, we determined the length of the peptide to be generated.

To decide the length of the protein generated by RFdiffusion, we used PyMOL to observe the α-helix on the C-terminal side of Thymosin β4, coloring it red for clarity. The amino acid sequence of Thymosin β4 is shown below, with the structure's color corresponding to the text color of the sequence. From this image, it was confirmed that the α-helix at the C-terminus of Thymosin β4 consists of 13 residues. Based on this, the length of the protein generated by RFdiffusion was set to 10–15 residues.

Peptides generated by RFdiffusion were set to interact with residues 199, 200, 207, and 208 of actin

fig 15. Overall hydrophobic amino acids ( red)

fig 16. Directed hotspot (red)

Next, we determined the amino acids on Actin that interact with the protein generated by RFdiffusion (hotspots). Generally, when generating a protein that interacts with a specific protein using RFdiffusion, 3–6 hydrophobic amino acids are specified as hotspots.

To identify the hotspots, we used PyMOL to highlight the hydrophobic amino acids on Actin and Thymosin β4 (valine, isoleucine, leucine, methionine, phenylalanine, and tryptophan) in red. Observing the hydrophobic amino acids on Actin around the C-terminus of Thymosin β4, we set the hotspots to residues 199, 200, 207, and 208 of Actin (A199, A200, A207, A208). Fig 16 shows Actin with the specified hotspots highlighted in red.

RFdiffusion was performed and amino acid sequences were extracted
(Click here to see detailed input settings for RFdiffusion)

fig 17. RFdiffusion output result

For the results output by RFdiffusion, the i_pae score, which indicates the relative error between two amino acids, and the plDDT score, which represents the confidence level of the structure, were summarized in a graph. Amino acid sequences within the range of pae ≤ 10 and plddt ≥ 0.8 (highlighted by the red rectangle) were extracted. RFdiffusion was run multiple times with the same and other condition settings as above, generating a total of about 100,000 peptides for examination.

Design and exzamination of amino acid sequence of Modified Thymosin

The amino acid sequence of Modified Thymosin was designed from the extracted amino acid sequence

fig 18. The best sequence of amino acid sequences generated

fig 19. AlphaFold3 output result of the alpha helix on the C-terminal side of thymosin was replaced with the generated amino acid sequence(red)

Using the peptides extracted from RFdiffusion, we input them along with Actin into AlphaFold3 and evaluated the interactions based on the resulting ipTM scores. Among the eight amino acid sequences analyzed, "KEEEERRLREEIEE" was found to be the most effective sequence.

The structure of this sequence was observed in detail using PyMOL. Fig 18 shows the hydrophobic amino acids highlighted in red, revealing that hydrophobic amino acids are located in the middle and latter parts of the sequence.

This amino acid sequence was then used to replace the α-helix at the C-terminus of Thymosin β4, and the modified sequence was input into AlphaFold3. However, from fig 19, it was found that simply replacing the α-helix at the C-terminus of Thymosin β4 with the generated peptide was not effective.

fig 20. Structure analysis of Thymosin β4

To address this, we aimed to design a Modified Thymosin that combines the C-terminus of Thymosin β4 with the generated peptide to strengthen interactions with Actin.

For this purpose, the structure of Thymosin β4 was observed using PyMOL. From fig 20, it was found that the α-helix (red) at the C-terminus contains hydrophobic amino acids in the front section, while the middle and latter sections lack hydrophobic amino acids. Based on this observation, we designed an amino acid sequence that retains the front section of the α-helix at the C-terminus of Thymosin β4 as it is, while replacing the middle and latter sections with the generated sequence.

We named this Modified Thymosin.

The following sequence represents the amino acid sequence of the designed Modified Thymosin (the red text indicates the parts modified from Thymosin β4):
"SDKPDMAEIEKFDKSKLKKTETQEKNPLPSKETIRRLREEIEE"

Designed Modified Thymosin inhibits actin polymerization better than Thymosin

　The designed Modified Thymosin was input into AlphaFold3 along with Actin to verify whether it can inhibit Actin polymerization more effectively than Thymosin β4. The results are summarized below.

Actin and Modified Thymosin (Click here to see the amino acid sequence)

Actin :
DEDETTALVCDNGSGLVKAGFAGDDAPRAVFPSIVGRPRHQGVMVGMGQKDSYVGDEAQSKRGILTLKYPIEHGIITNWDDMEKIWHHTFYNELRVAPEEHPTLLTEAPLNPKANREKMTQIMFETFNVPAMYVAIQAVLSLYASGRTTGIVLDSGDGVTHNVPIYEGYALPHAIMRLDLAGRDLTDYLMKILTERGYSFVTTAEREIVRDIKEKLCYVALDFENEMATAASSSSLEKSYELPDGQVITIGNERFRCPETLFQPSFIGMESAGIHETTYNSIMKCDIDIRKDLYANNVMSGGTTMYPGIADRMQKEITALAPSTMKIKIIAPPERKYSVWIGGSILASLSTFQQMWITKQEYDEAGPSIVHRKCF

Modified Thymosin :
SDKPDMAEIEKFDKSKLKKTETQEKNPLPSKETIRRLREEIEE

fig 21. AlphaFold3 output result of Actin and Modified Thymosin

The above five images in fig 21 represent the AlphaFold3 output results for Actin and Modified Thymosin with molecule counts of 1, 3, 5, 8, and 10.Focusing on the graph on the left in fig 21, the large relative positional errors between amino acids of different molecules, displayed in white, indicate weakened interactions between Actin molecules. Additionally, examining the ipTM scores for molecule counts of 3, 5, 8, and 10 reveals values of 0.33, 0.26, 0.29, and 0.29, respectively. These scores suggest that Actin polymerization can be sufficiently inhibited.

Next, a comparison with Thymosin β4 was conducted.

table 1. Difficulty of polymerization at each molecular number of "Actin and Modified Thymosin" and "Actin and Thymosin"

Table 1 summarizes the ipTM scores obtained when "Actin and Modified Thymosin" and "Actin and Thymosin β4" were each input with molecule counts of 3, 5, 8, and 10; AlphaFold3 was run 10 times for each case. The ipTM score represents the strength of interactions between molecules. Lower score suggests a greater inhibition towards Actin polymerization.

Blue bars represent "Actin and Modified Thymosin", while orange bars represent "Actin and Thymosin β4". The red line indicates an ipTM score of 0.5, above which significant intermolecular interactions may occur. For molecule counts of 3 and 5, the structural predictions are simple have small error range, shown in the enlarged view.

As a result, the ipTM scores for both cases were nearly identical for molecule counts of 3 and 5. However, for molecule counts of 8 and 10, the blue bars (Actin and Modified Thymosin) generally showed smaller ipTM scores. This indicated that Modified Thymosin has more effectivenes in inhibiting Actin polymerization. For the specific case of molecule count of 8, the scores for Modified Thymosin rarely exceeded the red boundary line, whereas almost half of the Thymosin β4's scores did.

In conclusion, an amino acid sequence for Modified Thymosin, capable of inhibiting Actin polymerization more effectively than Thymosin β4, was successfully designed.

Modified Thymosin

amino acid sequence ：SDKPDMAEIEKFDKSKLKKTETQEKNPLPSKETIRRLREEIEE

fig 22. Modified Thymosin Structure

Design and Examination of linkers conecting Actin and Modified Thymosin

Next, we examined linkers connecting the designed Modified Thymosin with Actin. The linker design used was similar to the one previously employed to connect Actin and Thymosin β4, and the analysis was conducted using AlphaFold3.

Actin-Modified Thymosin (ASSGGSGSGGSGGA) (Click here to see the amino acid sequence)

fig 23. AlphaFold3 output result of Actin-Modified Thymosin (ASSGGSGSGGSGGA)

The five images fig 23 represent the AlphaFold3 output results for Actin and Modified Thymosin connected using the linker sequence "ASSGGSGSGGSGGA" with molecule counts of 1, 3, 5, 8, and 10. In the left-hand graph, a green dominant indication meant that intermolecular interactions were occurring, leading to Actin polymerization.

Additionally, examining the ipTM scores for molecule counts of 3, 5, 8, and 10 reveals values of 0.56, 0.71, 0.44, and 0.34, respectively. These results suggest that Modified Thymosin is not effectively inhibiting Actin polymerization. These results suggest that the presence of this linker eliminate the ability of Modified Thymosin to inhibit Actin polymerization.

Next, Actin and Thymosin β4 with the linker sequence "ASSGGSGSGGSGGA" was compared.

table 2. Difficulty of polymerization at each molecular number of "Actin-Modified Thymosin (ASSGGSGSGGSGGA)" and "Actin-Thymosin β4 (ASSGGSGSGGSGGA)"

Table 2 summarizes the ipTM scores obtained from executing AlphaFold3 10 times for each case of "Actin-Modified Thymosin (ASSGGSGSGGSGGA)" and "Actin-Thymosin β4 (ASSGGSGSGGSGGA)" at molecule counts of 3, 5, 8, and 10. Blue bars represent "Actin-Modified Thymosin (ASSGGSGSGGSGGA)"m while orange bars represent"Actin-Thymosin β4 (ASSGGSGSGGSGGA)".

It was found that the scores for both cases were nearly equivalent, with more than half of the obtained data points for molecule counts of 3 and 5 surpassing the red boundary line. This suggests that Modified Thymosin was unable to inhibit Actin polymerization when Actin and Modified Thymosin were connected by a linker.

Conclusion

From all the simulations, a Modified Thymosin with higher inhibition capability than Thymosin β4 through improving the amino acid sequence was succesfully designed. However, the design of molecules in which Actin and Modified Thymosin are linked via a linker, but still maintain the inhibition capability has not been developed within the timeframe of BIOMOD.

Nevertheless, one found demonstrated that modifying the amino acid sequences of currently existing proteins may enhance their intermolecular interactions. This suggests that further improvements in amino acid sequence modification is possible to provide a succesful design of a molecule linked via linker similar to our goals. In addition, amino acid sequence modification also offers the potential to introduce various new properties into the molecules.

REFERENCE

[1] Josh Abramson, Jonas Adler, Jack Dunger, Richard Evans, Tim Green, Alexander Pritzel, Olaf Ronneberger, Lindsay Willmore, Andrew J. Ballard, Joshua Bambrick, Sebastian W. Bodenstein, David A. Evans, Chia-Chun Hung, Michael O’Neill, David Reiman, Kathryn Tunyasuvunakool, Zachary Wu, Akvilė Žemgulytė, Eirini Arvaniti, Charles Beattie, Ottavia Bertolli, Alex Bridgland, Alexey Cherepanov, Miles Congreve, Alexander I. Cowen-Rivers, Andrew Cowie, Michael Figurnov, Fabian B. Fuchs, Hannah Gladman, Rishub Jain, Yousuf A. Khan, Caroline M. R. Low, Kuba Perlin, Anna Potapenko, Pascal Savy, Sukhdeep Singh, Adrian Stecula, Ashok Thillaisundaram, Catherine Tong, Sergei Yakneen, Ellen D. Zhong, Michal Zielinski, Augustin Žídek, Victor Bapst, Pushmeet Kohli, Max Jaderberg, Demis Hassabis & John M. Jumper (2024) "Accurate structure prediction of biomolecular interactions with AlphaFold 3" Nature 630:493-500

[2]Rustam Ali 1, Jacob A. Zahm & Michael K. Rosen (2022) "Bound nucleotide can control the dynamic architecture of monomeric actin" Nature Structural & Molecular Biology, 29 : 320–328

[3]Kelli L. Hvorecny, Thomas E. Sladewski, Enrique M. De La Cruz, Justin M. Kollman & Aoife T. Heaslip (2024) "Toxoplasma gondii actin filaments are tuned for rapid disassembly and turnover" Nature Communications, 15 : 1840

[4]Xiaoying Chen, Jennica L. Zaro & Wei-Chiang Shen(2013) "Fusion protein linkers: Property, design and functionality" Advanced Drug Delivery Reviews 65 : 1357-1369

Google Sites

Report abuse

Simulation

Data from Simulation and Experimental Results

Process

Design of Linker Sequences Connecting Actin and Thymosin β4

Examination of Linker Sequences Connecting Actin and Thymosin β4

Summary of how to interpret AlphaFold3 output results

AlphaFold3 results did not reveal an optimal linker connecting Actin and Thymosin β4

Protein Structure Analysis

The C-terminus of Actin and the N-terminus of Thymosin β4 are in close proximity

When Actin and Thymosin β4 are connected by a linker, the α-helix at the C-terminus of Thymosin β4 is removed from actin.

Modification of Thymosin β4 by RFdiffusion

The length of peptides generated by RFdiffusion was set to 10-15 residues

Peptides generated by RFdiffusion were set to interact with residues 199, 200, 207, and 208 of actin

RFdiffusion was performed and amino acid sequences were extracted(Click here to see detailed input settings for RFdiffusion)

Design and exzamination of amino acid sequence of Modified Thymosin

The amino acid sequence of Modified Thymosin was designed from the extracted amino acid sequence

Designed Modified Thymosin inhibits actin polymerization better than Thymosin

Design and Examination of linkers conecting Actin and Modified Thymosin

Conclusion

REFERENCE

RFdiffusion was performed and amino acid sequences were extracted
(Click here to see detailed input settings for RFdiffusion)