Reconstitution and dual-path screening method of anti-freezing and anti-bacterial bifunctional short peptide

Abstract

The application belongs to the technical field of bioactive peptide design and screening, and discloses a reconstruction and double-path screening method of an anti-freezing and anti-bacterial bifunctional short peptide; based on the Janus-like radial hydrophilic / hydrophobic partitioning characteristics shared by alpha-helix anti-freezing peptides and alpha-helix anti-bacterial peptides, one is to use the Janus alpha-helix structure characteristics to rationally reconstruct the existing anti-bacterial peptides, so as to obtain the bifunctional short peptide with anti-freezing and anti-bacterial activities; and the other is to combine the Janus alpha-helix structure constraint with the anti-freezing activity prediction, and establish a virtual screening process for screening potential anti-freezing activity sequences from the natural anti-bacterial peptide space.

Description

Technical Field

[0001] This invention relates to the field of bioactive peptide design and screening technology, and in particular to a method for reconstructing and screening a short peptide with both antifreeze and antibacterial functions via a dual-pathway approach. Background Technology

[0002] During cryopreservation, ice crystal formation and recrystallization can cause mechanical and osmotic damage to biological samples such as cells and tissues. Furthermore, there is a risk of microbial contamination during sample preparation, storage, and transportation. Existing systems typically use cryoprotectants such as dimethyl sulfoxide (DMSO) and antibiotics for control, but the former has certain cytotoxicity, while the latter is limited by drug resistance issues in long-term use. Therefore, short peptides with both antifreeze and antibacterial functions, and simple, designable structures, have potential application value.

[0003] Previous studies have shown that some natural ice-binding proteins or antimicrobial peptides can simultaneously exhibit antifreeze and antimicrobial activities. For example, Mitchell and Gibson reported that the antimicrobial peptide Nisin can form an amphiphilic structure similar to antifreeze proteins under pH changes or metal ion-induced conditions, and exhibits detectable ice recrystallization inhibition activity (Biomacromolecules, 2015, 16, 3411–3416). In recent years, de novo design work targeting antifreeze peptide sequences has also been conducted, but antifreeze peptides and antimicrobial peptides are generally still designed as two relatively independent functional systems.

[0004] Current technologies lack a common structural framework that can simultaneously link ice interface recognition and bacterial membrane interactions at the short α-helical peptide level, and also lack an efficient screening process that combines this structural feature with data-driven prediction. The main problem this invention aims to solve is: identifying common structural features between antifreeze peptides and antimicrobial peptides that can be used for bifunctional short peptide design, and applying these features to the rational reconstruction of existing antimicrobial peptides and the screening of potential antifreeze active sequences in the natural antimicrobial peptide space. Summary of the Invention

[0005] The purpose of this invention is to address the technical deficiencies in existing technologies by providing a method for reconstructing and screening short peptides with both antifreeze and antibacterial properties via a dual-pathway approach. More specifically, this invention provides a method for reconstructing and screening short peptides with both ice recrystallization inhibition and antibacterial activity based on the radial partitioning characteristics of Janus α-helical hydrophilic / hydrophobic surfaces.

[0006] The technical solution adopted to achieve the purpose of this invention is: A method for reconstructing an antifreeze and antibacterial bifunctional peptide based on the Janus α-helix structure includes the following steps: Step s1: Using the antimicrobial peptide as the parent sequence, the spatial distribution of hydrophilic or cationic residues and hydrophobic residues on its α-helix surface is analyzed, and the radial separation of the hydrophilic and hydrophobic surfaces is enhanced by adjusting the residue arrangement, thereby obtaining candidate peptides with both antifreeze and antibacterial functions. Step s2 involves experimentally verifying the candidate peptide with both antifreeze and antibacterial functions obtained in step s1, thus obtaining the antifreeze and antibacterial dual-function peptide.

[0007] In the above technical solution, in step s1, the residues are adjusted by introducing lysine onto the antimicrobial short peptide to enhance the hydrophilic surface of the cationic peptide, retaining the leucine-rich region to form a hydrophobic surface, and using glycine as a non-hydrophobic spacer residue.

[0008] In the above technical solution, step s2 includes experimental verification: The ice-water interface molecular dynamics simulation was used to evaluate the ice interface adsorption stability, interfacial orientation, and ability to inhibit ice growth of candidate peptides. The membrane perturbation behavior of candidate peptides was evaluated using bacterial membrane molecular dynamics simulations. The antifreeze activity, antibacterial activity, membrane disruption behavior, secondary structure, and hemolytic activity of candidate peptides were evaluated using ice recrystallization inhibition assay, minimum inhibitory concentration assay, scanning electron microscopy, circular dichroism spectroscopy, and hemolysis assay.

[0009] Another aspect of the present invention includes an antifreeze and antibacterial bifunctional peptide obtained by the reconstruction method, the amino acid sequence of which is shown in SEQ ID NO.2.

[0010] Another aspect of the present invention includes a dual-pathway screening method for antifreeze and antibacterial bifunctional peptides based on Janus α-helix structure constraints and antifreeze activity prediction, comprising the following steps: Step 1: Obtain a library of natural antimicrobial peptide sequences with known antimicrobial activity; Step 2: Based on the Janus α-helix structure parameters, the first antifreeze and antibacterial bifunctional peptide candidate sequence is screened from the natural antimicrobial peptide sequence library in Step 1. Step 3: Use the antifreeze activity prediction model to score each sequence in the natural antimicrobial peptide sequence library from Step 1, and obtain the prediction score S. AFP Filter S AFP The second qualified antifreeze and antibacterial bifunctional peptide candidate sequence; Step 4: Take the intersection of the first and second antifreeze and antibacterial bifunctional peptide candidate sequences obtained in Step 2 and Step 3 to obtain the consensus sequence of the antifreeze and antibacterial bifunctional peptide. Step 5: The consensus sequence of the antifreeze and antibacterial bifunctional peptide obtained in Step 4 is experimentally verified to obtain the antifreeze and antibacterial bifunctional peptide.

[0011] In the above technical solution, the Janus α-helix structural parameters in step 2 include those based on peptide chain length, α-helix tendency, and P. α Hydrophobic moment μ H Hydrophobicity F of hydrophobic surface H .

[0012] In the above technical solution, the screening principle of step 2 is that the peptide chain length is 10-40 amino acid residues and the α-helix tends to be P α >0.85, hydrophobic moment μ H >0.20 and hydrophobicity F of the hydrophobic surface H >1.5.

[0013] In the above technical solution, step 3 uses the AFProPred antifreeze activity prediction model for scoring, and the predicted score S AFP The standard is met when the value is not lower than 0.60.

[0014] In the above technical solution, the experimental verification in step 5 includes ice-water interface molecular dynamics simulation and ice recrystallization inhibition experiment.

[0015] Another aspect of the present invention includes an antifreeze and antibacterial bifunctional peptide obtained based on the screening method, wherein the amino acid sequence is SEQ ID NO.3, SEQ ID NO.4, SEQ ID NO.5, SEQ ID NO.6 or SEQ ID NO.7.

[0016] Compared with the prior art, the beneficial effects of the present invention are: 1. This invention establishes a method for designing and discovering bifunctional short peptides with both antifreeze and antimicrobial properties, based on the Janus-like radial hydrophilic / hydrophobic partitioning characteristic shared by α-helical antifreeze peptides and α-helical antimicrobial peptides. Based on this structural framework, this invention provides two complementary technical approaches: first, rationally reconstructing existing antimicrobial peptides using Janus α-helical structural features to obtain bifunctional short peptides with both antifreeze and antimicrobial activities; second, combining Janus α-helical structural constraints with antifreeze activity prediction to establish a virtual screening process for screening potential antifreeze-active sequences from the natural antimicrobial peptide space.

[0017] 2. In a representative embodiment of the present invention, the hydrophobic moment of P-Janus μ H The concentration of the compound was increased from 0.828 for the parent compound P-Parent to 1.531. Ice-water interface simulation showed that its adsorption orientation was more stable. In the experiment, its ice recrystallization inhibition activity was higher than that of P-Parent, and the inhibition activity could be detected at 0.1 mg / mL.

[0018] The minimum inhibitory concentration (MIC) of P-Janus against Escherichia coli ATCC 25922 decreased from 35 μM to 17.5 μM, and the MIC against Staphylococcus aureus ATCC 25923 was 70 μM; the hemolysis rate was less than 2% in the test range of 2.1875–280 μM.

[0019] 3. Using the dual-path screening process of this invention, 12 consensus candidates were obtained from 3306 natural antimicrobial peptides, and further experimental verification revealed 5 antimicrobial peptide source sequences with significant ice recrystallization inhibitory activity, indicating that this method can be used to narrow the experimental screening scope. Attached Figure Description

[0020] Figure 1 The image shows the hydrophilic / hydrophobic spatial characteristics of different antifreeze proteins and antimicrobial peptides reported in the literature.

[0021] Figure 2 The figures show the sequence composition statistics of the parent peptide P-Parent and the reconstructed peptide P-Janus, the helical wheel structure analysis, and the molecular dynamics simulation of the ice-water interface and bacterial membrane system.

[0022] Figure 3 The image shows the performance verification of the reconstructed peptide P-Janus.

[0023] Figure 4 The diagram shows the dual-path screening process for the antifreeze and antimicrobial peptides of this invention. Detailed Implementation

[0024] The present invention will be further described in detail below with reference to specific embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

[0025] Example 1 Validation of the common interface structure framework coupling antifreeze and antibacterial functions: This embodiment compares the hydrophilic / hydrophobic spatial organization characteristics of different antifreeze and antibacterial molecules reported in the literature. Figure 1The study compared representative antifreeze and antibacterial molecules. Antifreeze glycoproteins (AFGP), snow flea antifreeze protein (sfAFP), facially amphipathic glycopolymers, and α-helical type I antifreeze protein (AFP I) all share a common feature: the hydrophilic and hydrophobic regions are separated from each other on the molecular surface or radially. Mode 2A amphipathic α-helical antimicrobial peptide (AMP) also exhibits a similar radial separation of the hydrophilic and hydrophobic surfaces. This structural convergence suggests that Janus-like amphipathic topology with bifacial characteristics may constitute a common basis for both antifreeze and antibacterial functions. All three types of molecules exhibit varying degrees of spatial separation between hydrophilic and hydrophobic regions, suggesting that both types of functional molecules may share a common bifacial structural feature.

[0026] This invention retrieved 35 α-helical antifreeze peptides and 30 α-helical antimicrobial peptides from the RCSB PDB database and published literature, and analyzed their sequences and structures, such as... Figure 1 As shown in B and C, the hydrophobic moment μ of each sequence was calculated under the standard α-helix conformation assumption. H The helical projection was generated using HeliQuest. A randomized sequence control was also constructed to determine whether the observed amphiphilicity originated from a specific sequence arrangement. Results showed that the μ-coefficients of the α-helical antifreeze peptide and the α-helical antimicrobial peptide... H The distributions show significant overlap and share μ. H The range is approximately 0.3 to 1.6; randomly shuffled sequences generally trend towards lower μ. H Regional movement, such as Figure 1 In C, for α-helical antifreeze peptides AFPTs, μ after AFP-2 random scrambled sequence. H The value is 0.141, while the μ values ​​of AFP-2 and NK-AFP obtained from the literature are... H The values ​​were 0.909 and 1.554, respectively; for α-helical antimicrobial peptides AMPs, the μ value after 2PC0 random scrambled sequence was... H The value is 0.035, while the μ values ​​of 2PC0 and 2JPY obtained in the literature are... H The values ​​are 0.911 and 1.782, respectively. The spiral wheel further demonstrates that the higher μ... H Sequences typically exhibit a clearer separation between hydrophilic and hydrophobic surfaces, while random scrambling disrupts this continuous surface partitioning.

[0027] In conclusion, Figure 1Comparative structural analysis of the collected AFPTs and AMPs revealed that they share a common amphiphilic α-helical structure, namely, the hydrophilic and hydrophobic surfaces are radially oriented and separated. This anisotropic structure is called Janus α-helix. This feature is related to the specific amino acid arrangement, rather than being determined solely by the amino acid composition. Therefore, it can serve as a common interfacial structural framework for coupling antifreeze and antibacterial functions.

[0028] Example 2 Reconstruction and validation of P-Janus bifunctional short peptides: Guided by the common interface structure framework of Example 1, this example uses a short peptide with antibacterial activity as the parent sequence, analyzes the spatial distribution of hydrophilic / cationic residues and hydrophobic residues on its α-helix surface, and enhances the radial separation of the hydrophilic and hydrophobic surfaces by adjusting the residue arrangement, thereby obtaining candidate short peptides with more obvious Janus α-helix characteristics.

[0029] In this embodiment, an AMP identified through deep learning is redesigned into a P-Janus peptide to enhance its Janus amphiphilicity. Specifically, the antimicrobial peptide AMP-48, obtained through deep learning, is used as the parent peptide P-Parent. Based on the compositional characteristics of highly active antimicrobial peptides, P-Parent is reconstructed to achieve amphiphilicity, resulting in the reconstructed peptide P-Janus. The reconstructed P-Janus exhibits a more pronounced Janus-like radial hydrophilic / hydrophobic partitioning.

[0030] The reconstructed short peptides can be evaluated by ice-water interface molecular dynamics simulation to assess their ice interface adsorption stability, interface orientation, and ability to inhibit ice growth. Their membrane perturbation behavior can be evaluated by bacterial membrane molecular dynamics simulation. Furthermore, their antifreeze activity, antibacterial activity, membrane disruption behavior, secondary structure, and hemolysis can be evaluated by ice recrystallization inhibition experiment, minimum inhibitory concentration experiment, scanning electron microscopy experiment, circular dichroism spectroscopy, and hemolysis experiment.

[0031] Specifically as follows: Using the 20-peptide AMP-48 (HTWLLMMRRLLLMMMNLLMG, amino acid sequence as shown in SEQ ID NO.1) as the parent peptide P-Parent, this peptide exhibits certain antibacterial activity, but the radial separation of the hydrophilic and hydrophobic surfaces is insufficient. Based on the Janus α-helix design principle, lysine distribution was enhanced on the hydrophilic / cationic surface, and the arrangement of residues on the hydrophobic surface was reconstructed. Lysine was introduced during the reconstruction process to enhance the hydrophilic cationic surface, while the leucine-rich region was retained to form the hydrophobic surface. Glycine was used as a non-hydrophobic spacer residue. After reconstruction, P-Janus (HTWLKMMGRLLKGMMKLLMG, amino acid sequence as shown in SEQ ID NO.2) was obtained.

[0032] P-Janus was prepared using the conventional Fmoc solid-phase synthesis method, purified by reversed-phase high-performance liquid chromatography, and its molecular weight was confirmed by mass spectrometry. Ice recrystallization inhibition assays were performed at concentrations of 0.1, 0.5, 1.0, and 5.0 mg / mL. The results showed that P-Janus exhibited strong ice recrystallization inhibition ability across the tested concentration range, with activity observed as low as 0.1 mg / mL.

[0033] In antibacterial experiments, the minimum inhibitory concentrations (MICs) of P-Janus against Escherichia coli ATCC 25922 and Staphylococcus aureus ATCC 25923 were 17.5 μM and 70 μM, respectively; the hemolysis rate of sheep erythrocytes in the range of 2.1875–280 μM was less than 2%. These results indicate that enhancing Janus-like radial partitioning can simultaneously preserve or enhance antifreeze and antibacterial functions in short peptides.

[0034] like Figure 2 Helical wheel analysis of A and D reveals the amino acid spatial distribution of the parent peptide P-Parent and the reconstructed peptide P-Janus, respectively. The hydrophobic moment μ of P-Parent... H The value is 0.828, while the μ value of P-Janus after sequence reconstruction is... H The value was increased to 1.531, resulting in a clearer separation between the hydrophilic / cationic and hydrophobic surfaces. For example... Figure 2 B and C are statistical analyses of the composition of highly active antimicrobial peptides from the Deep-ABPpred database, used to analyze the proportion of positively charged and hydrophobic amino acids, and to assist in the sequence reconstruction of P-Parent. Figure 2 Information regarding E and F was obtained through molecular dynamics simulations at the ice-water interface. In the simulation, the peptide was placed at the ice-water interface. The P-Janus system was simulated for 600 ns with four independent replicates, and a P-Parent and corresponding control system were also included. Common nearest neighbor analysis was used to distinguish between ice-water and liquid water, thereby statistically analyzing the change in the number of ice molecules during the simulation. Simultaneously, the angle θ between the instantaneous hydrophobic moment direction of the peptide and the interface normal was calculated to analyze the orientation stability of the peptide on the ice surface. Figure 2 The results showed that the continuous growth of ice molecules in the P-Janus system was more significantly inhibited, indicating that P-Janus has a stronger ability to inhibit ice growth than P-Parent. Figure 2 The middle figure shows that P-Janus maintains a more stable ice surface orientation and adsorbs at the ice-water interface in a near-parallel manner for a longer period, while the orientation of P-Parent fluctuates more significantly. Therefore, these two subfigures illustrate that enhancing the radial partitioning of Janus-like hydrophilic and hydrophobic components is beneficial for improving the stable adsorption of peptides at the ice-water interface and the inhibition of ice growth. Figure 2Information on G and H was obtained through molecular dynamics simulations of the bacterial membrane-water interface. The simulation used a 3:1 DMPC / DMPG lipid bilayer as the bacterial membrane model, placing the peptide on top of the membrane and observing the interaction between the peptide and the lipid membrane over a long period. Membrane structural changes were observed through simulated snapshots, and the number of water molecules inside the membrane was counted to evaluate whether the peptide induced the formation of transmembrane water channels and the persistence of these channels. The results showed that P-Janus was able to form more persistent transmembrane water channels than P-Parent, with a higher percentage of the time the number of water molecules inside the membrane exceeded a set threshold. Therefore, Figure 2 The values ​​of G and H in the equation demonstrate that enhancing the radial dihedrality of Janus-like structures not only benefits adsorption at the ice interface but also enhances the peptide's ability to perturb bacterial membranes and form pores. In summary, from... Figure 2 It can be seen that after the reconstruction guided by Janus α-helix, the P-Parent's ability to interact with ice interfaces and its ability to disrupt bacterial membranes are both enhanced.

[0035] Figure 3 The information was mainly obtained through ice recrystallization inhibition experiments, optical microscopy, scanning electron microscopy, circular dichroism spectroscopy, and hemolysis experiments, and was used to experimentally verify the antifreeze activity, antibacterial membrane disruption, secondary structure, and biocompatibility of P-Janus. Figure 3 Peptides A and B were obtained using a modified splat ice recrystallization inhibition assay. Peptides were prepared at concentrations of 0.1, 0.5, 1.0, and 5.0 mg / mL, with PBS as a blank control and TisAFP as a positive control. 10 μL of sample was dropped from a height of approximately 1.5 m onto a -60 °C coverslip to form a polycrystalline ice layer, which was then annealed at -8 °C for 30 min, and the ice crystals were observed using a polarizing microscope. The mean crystal area (MGA) was used and normalized relative to PBS; a lower MGA indicated stronger ice recrystallization inhibition activity. The results showed that P-Janus exhibited stronger ice recrystallization inhibition activity than P-Parent within the tested concentration range, and significant activity was detectable at 0.1 mg / mL. Therefore, Figure 3 Experiments A and B demonstrate that P-Janus has good ability to inhibit ice recrystallization. Figure 3The C-type peptide was obtained through scanning electron microscopy. *Escherichia coli* and *Staphylococcus aureus* were cultured to approximately OD600 = 0.5. After adjusting the bacterial concentration, the bacteria were treated with the corresponding peptide, and the bacterial surface and membrane morphology were observed using scanning electron microscopy. *P-Janus* was treated at 4×MIC; since no significant antibacterial activity was detected in *P-Scrambled* (randomized *P-Parent*), a fixed concentration of 280 μM was used for comparison. The results showed that bacteria treated with *P-Janus* exhibited significant membrane structure damage, while bacteria treated with *P-Scrambled* maintained a relatively intact morphology. Therefore, Figure 3 The results from C provide experimental support for the significant bacterial membrane disruption effect of P-Janus, consistent with the membrane pore formation behavior observed in molecular dynamics simulations. Figure 3 The secondary structures of P-Janus and P-Scrambled were obtained by circular dichroism spectroscopy. The CD spectra of P-Janus and P-Scrambled in PBS at 25 °C and 4 °C were determined using a circular dichroism spectroscopy instrument to evaluate their secondary structures. The results showed that P-Janus maintained a distinct α-helical structure at both 25 °C and 4 °C, while P-Scrambled mainly exhibited a random coil structure. Therefore, these results demonstrate that P-Janus can form and maintain the designed α-helical conformation and support the correlation between the ordered Janus α-helical structure and its functional performance. Figure 3 The hemolysis rate of P-Janus was obtained through a sheep erythrocyte hemolysis experiment. Different concentrations of P-Janus were incubated with a 4% sheep erythrocyte suspension. Triton X-100 was used as a positive control for complete hemolysis, and PBS as a negative control. The hemolysis rate was calculated by measuring the absorbance of released hemoglobin at 540 nm. The final test concentration range was 2.1875–280 μM. The experimental results showed that the hemolysis rate of P-Janus within the concentration range of 2.1875–280 μM was less than 2%. Therefore, Figure 3 The results from the E-test indicate that P-Janus, while possessing antifreeze and antibacterial activities, exhibits low destructive effect on sheep erythrocytes and good blood compatibility.

[0036] Example 3 Design and screening of bifunctional short peptides with antifreeze and antibacterial properties: Based on the results verified in Examples 1 and 2, this example provides a method for reconstructing and dual-pathway screening of antifreeze and antibacterial bifunctional short peptides.

[0037] like Figure 4 As shown, using a library of 3306 natural antimicrobial peptides with known antimicrobial activity as starting sequences, one pathway is based on peptide chain length, α-helix tendency, and P... α Hydrophobic moment μ H Hydrophobicity F of hydrophobic surface HThe Janus α-helix structure parameters were screened; another approach used the AFProPred antifreeze activity prediction model (see N. Kumar, S. Choudhury, N. Bajiya, S. Patiyal, GPS Raghava, Prediction of anti-freezing proteins from their evolutionary profile, Proteomics 25 (2025) e202400157) for scoring, and the prediction score S of the antifreeze activity prediction model was obtained. AFP The intersection of the candidate sequences obtained from the two paths yielded 12 consensus candidate sequences, which were further screened and verified using ice-water interface molecular dynamics simulations and ice recrystallization inhibition experiments. In summary, the Janus α-helix structural features proposed in this invention can not only be used for the rational reconstruction of existing antimicrobial peptides, but also serve as a structural constraint combined with antifreeze activity prediction to screen potential antifreeze active sequences from the natural antimicrobial peptide space.

[0038] Specifically, when screening based on Janus α-helix structural parameters, short peptides of 10–40 amino acid residues are first screened, and then selected according to α-helix tendency P α >0.85, hydrophobic moment μ H >0.20 and hydrophobicity F of the hydrophobic surface H >1.5 Perform structural path screening; when using the antifreeze activity prediction model for data-driven screening, retain the prediction score S. AFP Sequences with a value of at least 0.60 were selected. After cross-fertilizing two pathways, 12 consensus candidates were obtained, which were then further refined using ice-water interface molecular dynamics simulations and ice recrystallization inhibition experiments.

[0039] The representative sequences that were finally verified include: NILNTIINLAKKIL, as shown in SEQ ID NO.3; RICRIIFLRVCR, as shown in SEQ ID NO.4; RLCRIVVIRVCR, as shown in SEQ ID NO.5; LLSLALAALPKLFCLIFKKC, as shown in SEQ ID NO.6; DWTAWSALVAAACSVELL, as shown in SEQ ID NO.7.

[0040] All five antimicrobial peptide source sequences exhibited significant ice recrystallization inhibition activity, indicating that the screening method combining structural constraints and data-driven approaches of this invention is feasible.

[0041] Specifically, the five antimicrobial peptides were validated experimentally using IRI experiments. Within a concentration range of 0.1–5.0 mg / mL, the normalized mean crystallite area (MGA) of all five peptides was significantly lower than that of PBS, generally falling within the range of approximately 45–65%, and they formed smaller and more uniform ice crystals compared to PBS. Under low concentration conditions, the performance of all five peptides was comparable to or better than that of TisAFP (the natural antifreeze protein of Typhula ishikariensis). However, their concentration-response curves showed significant differences, suggesting that different candidate peptides exhibit different interfacial behaviors.

[0042] The above description is only a preferred embodiment of the present invention. It should be noted that, for those skilled in the art, several improvements and modifications can be made without departing from the principle of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A method for reconstructing an antifreeze and antibacterial bifunctional peptide based on the Janus α-helix structure, characterized in that, Includes the following steps: Step s1: Using the antimicrobial peptide as the parent sequence, the spatial distribution of hydrophilic or cationic residues and hydrophobic residues on its α-helix surface is analyzed, and the radial separation of the hydrophilic and hydrophobic surfaces is enhanced by adjusting the residue arrangement, thereby obtaining candidate peptides with both antifreeze and antibacterial functions. Step s2 involves experimentally verifying the candidate peptide with both antifreeze and antibacterial functions obtained in step s1, thus obtaining the antifreeze and antibacterial dual-function peptide.

2. The reconstruction method as described in claim 1, characterized in that, In step s1, the residues are adjusted by introducing lysine onto the antimicrobial short peptide to enhance the hydrophilic surface of the cationic peptide, retaining the leucine-rich region to form a hydrophobic surface, and using glycine as a non-hydrophobic spacer residue.

3. The reconstruction method as described in claim 1, characterized in that, In step s2, the experimental verification includes: The ice-water interface molecular dynamics simulation was used to evaluate the ice interface adsorption stability, interfacial orientation, and ability to inhibit ice growth of candidate peptides. The membrane perturbation behavior of candidate peptides was evaluated using bacterial membrane molecular dynamics simulations. The antifreeze activity, antibacterial activity, membrane disruption behavior, secondary structure, and hemolytic activity of candidate peptides were evaluated using ice recrystallization inhibition assay, minimum inhibitory concentration assay, scanning electron microscopy, circular dichroism spectroscopy, and hemolysis assay.

4. The antifreeze and antibacterial bifunctional peptide obtained by the reconstruction method according to any one of claims 1-3, characterized in that, The amino acid sequence is shown in SEQ ID NO.

2.

5. A dual-pathway screening method for antifreeze and antibacterial bifunctional peptides based on Janus α-helix structure constraint and antifreeze activity prediction, characterized in that, Includes the following steps: Step 1: Obtain a library of natural antimicrobial peptide sequences with known antimicrobial activity; Step 2: Based on the Janus α-helix structure parameters, the first antifreeze and antibacterial bifunctional peptide candidate sequence is screened from the natural antimicrobial peptide sequence library in Step 1. Step 3: Use the antifreeze activity prediction model to score each sequence in the natural antimicrobial peptide sequence library from Step 1, and obtain the prediction score S. AFP Filter S AFP The second qualified antifreeze and antibacterial bifunctional peptide candidate sequence; Step 4: Take the intersection of the first and second antifreeze and antibacterial bifunctional peptide candidate sequences obtained in Step 2 and Step 3 to obtain the consensus sequence of the antifreeze and antibacterial bifunctional peptide. Step 5: The consensus sequence of the antifreeze and antibacterial bifunctional peptide obtained in Step 4 is experimentally verified to obtain the antifreeze and antibacterial bifunctional peptide.

6. The dual-path filtering method as described in claim 5, characterized in that, In step 2, the Janus α-helix structural parameters include those based on peptide chain length and α-helix tendency P. α Hydrophobic moment μ H Hydrophobicity F of hydrophobic surface H .

7. The dual-path filtering method as described in claim 5, characterized in that, The screening criteria in step 2 are: peptide chain length of 10–40 amino acid residues, α-helix tendency P α >0.85, hydrophobic moment μ H >0.20 and hydrophobicity F of the hydrophobic surface H >1.

5.

8. The dual-path filtering method as described in claim 5, characterized in that, In step 3, the AFProPred antifreeze activity prediction model is used for scoring, and the predicted score S is obtained. AFP The standard is met when the value is not lower than 0.

60.

9. The dual-path filtering method as described in claim 5, characterized in that, The experimental verification in step 5 includes molecular dynamics simulation of the ice-water interface and experiments to inhibit ice recrystallization.

10. The antifreeze and antibacterial bifunctional peptide obtained based on the dual-path screening method as described in any one of claims 5-9, characterized in that, The amino acid sequences are SEQ ID NO.3, SEQ ID NO.4, SEQ ID NO.5, SEQ ID NO.6 or SEQ ID NO.7.

Images