Loading...

MODEL

vertical line 卡通图片
shrimp

Overview

Our team combined PBPs with nanoflowers to create quadruple-function nanoflowers with targeted specificity and dual signal amplification. To investigate how different PBPs affect bacterial detection, we used Escherichia coli as the target organism. In the NCBI database, we identified two PBPs capable of specifically binding to E. coli:
(1)TFP (tail fiber protein [Escherichia phage T7], Genbank NP_042005.1)
(2)gp17 (tail fiber protein [Enterobacteria phage T3], Genbank NP_523342.1).

We first performed structural prediction and analysis of TFP and gp17 using Alphafold3 (https://alphafoldserver.com/) and Sequence Manipulation Suite 2.0 software. Subsequently, molecular docking with the KDO2 receptor recognition unit was performed using AutoDock Vina to investigate their affinities. Wet-lab results demonstrated that when using both quadruple-combination nanoflowers for bacterial detection, the TFP nanoflowers (HRP-TFP-CaHPO₄@AuPt) exhibited higher binding efficiency than the gp17 nanoflowers (HRP-gp-CaHPO₄@AuPt). This indicates consistency between model predictions and experimental outcomes, providing valuable insights for subsequent PBP screening efforts.

1. Model Construction

We obtained the amino acid sequences of TFP and gp17 from GenBank. Structural predictions for TFP and gp17 were performed using Alphafold3 (https://alphafoldserver.com/). Since the phage tail filament protein is a trimer, the model design required the construction of three copies.

Name Source Genbank
TFP tail fiber protein [Escherichia phage T7] NP_042005.1
gp17 tail fiber protein [Enterobacteria phage T3] NP_523342.1
The structure of TFP and gp17 is shown below:
Figure 1
Figure 1 Protein Prediction Results of (A) TFP, and (B) gp17 using Alphafold3.

According to the literature, the six tail filament proteins of T7 phage specifically bind to the lipopolysaccharide (LPS) of the E. coli outer membrane, while the C-terminal region (371-553 AA) of TFP forms a tip domain (Tip-Domain) that interacts with cell surface receptors (Garcia-Doval & van Raaij, 2012). Below are the alignment results between the Tip-Domain of TFP and the Tip-Domain of gp17:

>gp17-Tip-Domain (188 aa)
GHALYLESASDKAQYILSKDGNRNNWYIGRGSDNNNDCTFHSYVYGTNLTLKPDYAVVNKRFHVGQAVVATDGNIQGTKW
GGKWLDAYLNDTYVKKTMAWTQVWAAASGSYMGGGSQTDTLPQDLRFRNIWIKTRNNYWNFFRTGPDGIYFLSAEGGWLK
FQIHSNGRVFKNIADRDAPPTAIAVEDV

>TFP-Tip-Domain (183 aa)
GHVLQLESASDKAHYILSKDGNRNNWYIGRGSDNNNDCTFHSYVHGTTLTLKQDYAVVNKHFHVGQAVVATDGNIQGTKW
GGKWLDAYLRDSFVAKSKAWTQVWSGSA-----GGGVSVTVSQDLRFRNIWIKCANNSWNFFRTGPDGIYFIASDGGWLR
FQIHSNGLGFKNIADSRSVPNAIMVENE

Using the software Sequence Manipulation Suite 2.0 (Stothard, 2000), we compared the Tip-Domain sequences of TFP and gp17, revealing an amino acid sequence similarity of 75.53%. Furthermore, protein structure comparison via Pymol software showed a root mean square deviation (RMSD) of 0.598 (a lower RMSD indicates greater structural similarity between proteins), further indicating that TFP and gp17 possess similar functions.

The outer membrane of E. coli is depicted in Figure 2. The outermost layer of LPS comprises the lipid A-KDO2 complex. Structural information for lipid A, KDO2, and lipid A-KDO2 can be retrieved from PubChem. Direct docking with lipid A or lipid A-KDO2 fails due to excessive ligand size. KDO serves as the critical bridge connecting Lipid A to the core oligosaccharide region and is one of the minimal units recognized by phage and host receptors. KDO2's small molecular weight facilitates modeling and protein docking. Therefore, an analysis was conducted on the KDO2 fragment of E. coli LPS based on PubChem data.

The structural information for KDO2 is represented by the Simplified Molecular Input Line Entry System (SMILES) formula:
C1[C@H]([C@H]([C@H](O[C@]1(C(=O)O)O[C@@H]2C[C@@](O[C@@H]([C@@H]2O)[C@@H](CO)O)(C(=O)O)O)[C@@H](CO)O)O)O
Figure 2
Figure 2 Extracellular membrane structure of E. coli.

KDO2 has the molecular formula C16H26O15 (PubChem CID: 196322), with its structure shown in Figure 3.
Figure 3
Figure 3 KDO2 Structural Formula

2. Molecular Docking

AutoDock Vina docking was performed between the TFP-Tip-Domain and gp17-Tip-Domain with KDO2 (Figure 4). The docking results reveal that ASN124/129 occupies the center of the hydrogen bond network, serving as the primary hydroxyl recognition site.

The TFP Tip-Domain forms 12 hydrogen bonds with KDO2, distributed across multiple hydroxyl groups of the disaccharide. The two sugar rings are recognized by ASN124 and ARG123/GLU181, respectively, forming a "symmetrical triangular support" structure. Additionally, ARG138 assists in positioning peripherally while participating in weak hydrogen bonds/electrostatic locking.

The gp17 Tip-Domain forms 9 hydrogen bonds with KDO2, with interaction points slightly biased toward one side. The reduced number of hydrogen bonds and incomplete recognition of both sugar ring endpoints result in structurally unbalanced, unilateral recognition.

Based on KDO2 binding performance, TFP exhibits richer hydrogen bonds with more balanced distribution and multiple chain-coordinated recognition, indicating its Tip-Domain is more likely to possess native LPS-binding capability.

Figure 4
Figure 4 Molecular docking results for the (A) TFP-Tip-Domain and (B) gp17-Tip-Domain and KDO2

3.Wet-lab validation

We performed bacterial detection using the synthesized quadruple-functionalized nanoflowers HRP-TFP-CaHPO₄@AuPt and HRP-gp17-CaHPO₄@AuPt. Within a bacterial concentration range of 100–104 CFU/mL, both nanoflowers generated detectable signal values (compared to the negative control PBS). Furthermore, we observed that HRP-TFP-CaHPO₄@AuPt consistently detected higher signal values than HRP-gp17-CaHPO₄@AuPt within the same concentration range. We hypothesize that this difference stems from TFP exhibiting stronger affinity toward E. coli, a finding corroborated by our modeling results.

Figure 5
Figure 5 Bacterial Test Results of (A) HRP-TFP-CaHPO4@AuPt, and (B) HRP-gp17-CaHPO4@AuPt


References

Garcia-Doval, C., & van Raaij, M. J. (2012). Structure of the receptor-binding carboxy-terminal domain of bacteriophage T7 tail fibers. 109(24), 9390-9395. doi:doi:10.1073/pnas.1119719109
Stothard, P. (2000). The sequence manipulation suite: JavaScript programs for analyzing and formatting protein and DNA sequences. Biotechniques, 28(6), 1102, 1104. doi:10.2144/00286ir01

To Top