Macromolecule sensing and tumor biomarker detection by harnessing terminal size and hydrophobicity of viral DNA packaging motor channels into membranes and flow cells

Abstract

Biological nanopores for single-pore sensing have the advantage of size homogeneity, structural reproducibility, and channel amenability. In order to translate this to clinical applications, the functional biological nanopore must be inserted into a stable system for high-throughput analysis. Here we report factors that control the rate of pore insertion into polymer membrane and analyte translocation through the channel of viral DNA packaging motors of Phi29, T3 and T7. The hydrophobicity of aminol or carboxyl terminals and their relation to the analyte translocation were investigated. It was found that both the size and the hydrophobicity of the pore terminus are critical factors for direct membrane insertion. An N-terminus or C-terminus hydrophobic mutation is crucial for governing insertion orientation and subsequent macromolecule translocation due to the one-way traffic property. The N- or C-modification led to two different modes of application. The C-terminal insertion permits translocation of analytes such as peptides to enter the channel through the N terminus, while N-terminus insertion prevents translocation but offers the measurement of gating as a sensing parameter, thus generating a tool for detection of markers. A urokinase-type Plasminogen Activator Receptor (uPAR) binding peptide was fused into the C-terminal of Phi29 nanopore to serve as a probe for uPAR protein detection. The uPAR has proven to be a predictive biomarker in several types of cancer, including breast cancer. With an N-terminal insertion, the binding of the uPAR antigen to individual peptide probe induced discretive steps of current reduction due to the induction of channel gating. The distinctive current signatures enabled us to distinguish uPAR positive and negative tumor cell lines. This finding provides a theoretical basis for a robust biological nanopore sensing system for high-throughput macromolecular sensing and tumor biomarker detection.

---

1.Introduction

Nanopore technology is a promising platform for various applications, such as gene sequencing, personalized medicine, and biomedicine.^1–12 Nanopore technology is proving itself capable of detecting and analyzing multiple chemical compounds^13–15 and biomolecules (e.g., RNA,^16–19 DNA,^19–25 peptides,^26–32 and proteins^33–38) at a single-molecule level. The most relevant success of nanopore sensing is DNA sequencing, which was firstly proposed by Deamer et al. in 1996.³⁹ Nanopore technology presents various advantages over other detection modalities, including low cost, high throughput, and label-free sample analysis. Biological and Solid-State nanopores are the two main categories of nanopores in use today.^40,41 This categorization is heavily dependent on the materials composing the nanopore and channel, with biological pores being composed of proteins, whereas solid-state pores are derived from almost entirely inorganic components. However, hybrid nanopores involve a protein pore lodged within a solid-state membrane.^42–44 The biological protein pores for single-pore sensing have the advantage of size homogeneity, structural reproducibility, and flexibility for chemical and structural modification via specific site-directed mutagenesis.^45,46

It is clear that biological nanopore sensing is a very powerful technique; the most important aspect of biological pore sensing is its medical and industrial applicants in disease diagnosis and molecular detection. The fragility of the lipid bilayer membrane (BLM) and surfactant bilayers remains a primary weakness of the biological system that has precluded it from being used more frequently.⁴⁷ A wide range of lipid bilayers has been tested in biomolecule translocation experiments.⁴⁸ However, this application is limited by a lifespan of only a few hours and mechanical instability of the membrane.⁴⁹ Additionally, the lipid membrane has a large area and capacitance, leading to increased noise.^50–53 One of the important areas is the translation of this technology.⁵⁴ That is, how to adapt and incoordinate the biological pore into the industrial instrument. To expand nanopore sensing, several techniques have been used for enhancement of mechanical stability and durability of the membrane, such as solid supports for membrane stabilization,^55,56 porous material,⁵⁷ and soft matter (e.g., hydrogels and polymeric matrices). Among them, polymeric materials have been highly considered due to their increased molecular affinity, showing their application for high-throughput biosensing platforms.⁵⁸

Protein pore in the patch-clamp system remains challenging to translate into clinical applications. Oxford Nanopore Technologies Ltd provides a portable single-molecule sequencer (MinION™) as a novel third-generation sequencing technology using biological nanopores that permit direct DNA and RNA sequencing. Compared with the lipid bilayer, the MinION™ Flow Cell polymeric membrane has shown itself to be more mechanically stable and resistant to high voltage.⁵⁹ The advantages of numbers (2048 regions for pore insertion) and stability permit the MinION™ Flow Cell to be a promising platform for high-throughput peptide sensing.^60,61 A critical step in using the protein pore is to incorporate the protein nanopore into the robust instrument to detect and characterize proteins.⁶² This goal has proven to be more challenging than expected.^63,64 Proteins and peptides offer complexity from their size and structure that is not typically associated with synthetic polymers and nucleic acids.⁶⁴ Ji et al. tested the peptide translocation through the inserted Phi29 channel, revealing the potential of applying the Phi29 channel for high-throughput peptide sensing.⁶¹ Nevertheless, an increasing number of researchers have developed methods for nanopore sensing to enable better characterization of proteins and peptides at the single-molecule limit.⁶⁵ Reading and differentiating 20 amino acids is much more demanding than reading four DNA bases.⁶⁶ Ouldali et al., provide a route to detect all 20 proteinogenic amino acids with a biological nanopore, bringing single-protein sequencing with nanopores closer to fruition.⁶⁷ Nanopore technology has achieved real-time measurement of protein–protein interactions,⁶⁸ detection of post-translational modification,^69–72 measurements of protein size, fluctuations, and conformational changes⁷³ and sequencing biopolymers,⁷⁴etc.

In this study, we investigate the factors relevant to the insertion of protein nanopore into industrial instrumentation. We found that the pore orientation and terminal size of viral DNA packaging motors are essential factors controlling their channel's direct insertion into membranes and MinION™ Flow Cell for analyte translocation. The introduction of hydrophobic amino acid to the N-terminus of the connector channel enhanced the rate of direct membrane insertion, but no translocation of the analyte occurred. By increasing the overall hydrophobicity of the C-terminal, the rate of direct insertion increased, and analyte translocation occurred. By introducing a uPAR probe to the C-terminal of the Phi29 connector, the N-terminal inserted channel could be used for tumor biomarker detection. The finding that orientation and terminal size directly control peptide translocation provides a theoretical basis for designing a nanopore-based sensor for analyte detection. The application of controllable insertion into MinION™ Flow Cell channels provides a stable and high-throughput sensing platform for macromolecule sensing and tumor biomarkers detection.

2. Materials and methods

2.1. Materials

1,2-Diphytanoyl-sn-glycero-3-phosphocholine (DPhPC) was obtained from Avanti Polar Lipids. Brj-58 was obtained from Sigma Aldrich. MinION™, blank MinION™ Flow Cells without pre-inserted biological pores, and Flow Cell buffer were obtained from Oxford Nanopore Technologies Ltd. The uPAR and BSA standard protein was purchased from Abcam. Peptides were synthesized and purified by Genescript. The sequences of peptides used in this study is R12: RRR RRR RRR RRR.

2.2. Cloning, expression and purification of mutant Phi29 and T7 channels

The construction of the plasmid harboring the ORF of Phi29 GP10, and T7 GP8 have been reported recently.^32,75 The mutant ORFs with specific restriction sites were produced by PCR by using a specific oligo (Table S1†). To construct the C-Δ25 channel, nest PCR was performed by using three oligos. The final PCR products were digested with the restriction endonuclease and ligated into restriction endonuclease corresponding sites of the vector pET-21a (Novagen). Recombinant plasmids with correct sequencing were transformed into HMS174 (DE3) cells for inducing protein expression.

For cloning of the engineered uPAR Phi29 connector channel (N-His GP10-C-uPAR), the recombinant plasmid was constructed by introducing a uPAR probe (LWXX(Ar)XFXXYLW;⁷⁶ (Ar = Y/W/F/H)) to the C-terminal of the connector, just downstream of Phi29 GP10 gene; a His tag was inserted into the N-terminal for purification.

The connector mutants constructed were expressed and then purified with Nickel affinity chromatography. Cells were resuspended with His Binding Buffer (Tris-HCl 0.1 M, NaCl 0.5 M, ATP 50 μM, glycerol 14.4%, imidazole 5 mM), and the cleared lysate was loaded onto a HisBind Resin Column (Novagen) and washed with His Washing Buffer (Tris-HCl 0.1 M, NaCl 0.5 M, ATP 50 μM, glycerol 14.4%, imidazole 50 mM). The C-terminal His-tagged connector was eluted by His Elution Buffer (Tris-HCl 0.1 M, NaCl 0.5 M, ATP 50 μM, glycerol 14.4%, imidazole 1 M).

2.3. Insertion of Phi29, T7 and T3 channel into liposome

The procedure for the preparation of Phi29, T7 and T3 proteoliposome was modified based on the previous protocol.⁷⁷ Briefly, chloroform containing 1 mg DPhPC will be evaporated off by a rotary evaporator (∼5 min), and then the connector protein and liposome rehydration buffer (1 M KCl, 250 mM sucrose, 5 mM HEPES, pH 7.4) will be added and vortexed to fully dissolve the membrane (the final connector protein concentration should be ∼200–500 μg mL⁻¹). To get homogenous proteoliposomes, prepared proteoliposomes were filtered by 0.4 μm polycarbonate membrane using the extruder (Avanti Polar Lipids).

2.4. Direct insertion of Phi29, T7, and T3 channel into the copolymer membrane of MinION™ flow cell

The MinKNOW software, developed and provided by Oxford Nanopore Technologies Ltd, ran software for pore insertion and data collection. The connector protein solution, once mixed with a separate solution comprised of 0.1% Brj58 and Flow Cell buffer C13 (25 mM potassium phosphate, 150 mM potassium ferrocyanide, 150 mM potassium ferricyanide, pH 8), is able to be added to the MinION™ Flow Cell directly for pore insertion. The final pore concentration with being between 0.01 to 0.03 mg mL⁻¹; this allows for flexibility between connectors of various concentrations.

For direct insertion of the channel into the Flow Cell membrane, 300 μL C13 buffer with pore samples will be loaded into the priming port. MinKNOW will then apply a voltage profile to facilitate direct channel insertion (−150 mV to −400 mV). 500 μL C13 buffer should be added from the priming port to flush away excess pores and subsequently evaluated the number of inserted channels by Platform Quality Check (PQC) script. A custom program provided by Oxford Nanopore Technologies Ltd, aided in data analysis.

For the insertion of proteoliposomes into the Flow Cell, all steps are the same as those described above, except that the maximum pore insertion voltage is −350 mV, and 20 μL of proteoliposomes are mixed with 280 μL C13 buffer (w/o Brj58 solution).

2.5. Peptide translocation and tumor biomarker detection

R12 peptide solution (2 μL, 1 mg mL⁻¹) mixed with Flow Cell buffer (298 μL) was then loaded into the Flow Cell via priming port. The MinKNOW™ software allowed for observation of the peptide translocation under different potential differences. To detect uPAR protein on the MinION™ Flow Cell, BSA solution (2 μL, 1 mg mL⁻¹) in Flow Cell buffer (298 μL) was loaded into the Flow Cell via priming port pore insertion, and then the current signal was recorded by PQC script. After that, uPAR solution (2 μL, 1 mg mL⁻¹) was added, and the signal was evaluated by Traceviewer2 software. To test uPAR binding to reengineered Phi29 connector channel, uPAR protein solution (0.6 μL, 1 mg mL⁻¹) was added to the cis chamber of Axon Clamp after single pore insertion. The Mosaic software was used for data analysis.

2.6. Electrophysiology assay

The freestanding lipid bilayer membrane in the Axon Clamp, formed over the Teflon partition membrane (pore size: 200 μm). Bilayer Clamp Amplifier BC-535 (Warner Instruments), once connected to the Axon DigiData 1440A analog-digital converter (Molecular Devices) allowed for electrical current flow. Data recording took place at 1k Hz bandwidth, a sampling frequency 20 kHz, and a voltage of ±50 mV. In single pore sensing, the current is going to be the main driving force for pulling the peptides through the pore. Thus, constant current with variable voltage is a general practice. When different buffers (mainly the salt) were applied to the system, the voltage was adjudged to ensure a similar current. With a 1 M KCl, 5 mM HEPES, pH7.4 buffer on the patch-clamp system at 50 mV voltage, the channel current is about 200 pA for the phi29 GP10 connector. While in the MinION™ Flow Cell system, a 100 mV voltage was applied to reach a similar current. The Clampex 10 (Molecular Devices) and Clampfit 10 (Molecular Devices) collected and allowed for viewing of data.

3.Results and discussion

3.1. Introduction of hydrophobic amino acids to the N-terminus of Phi29 connector channel to enhance the rate of direct membrane insertion

The orientation of connector insertion into the lipid bilayer via liposome has previously been investigated.⁷⁸ The direction of connector insertion into the polymer membrane is an important subject, but this has not been reported. To increase the direct insertion rate of the Phi29 connector channel for high-throughput analyte sensing, the hydrophilic amino acids (Fig. 1a, blue) at the N-terminal of Phi29 connector channels were replaced by hydrophobic amino acids. Four recombinant plasmids containing a mutant open reading frame (ORF) of the Phi29 connector channels were constructed to express the connector proteins (Fig. 1b–e, Table S1†). The purified proteins were confirmed by 10% SDS-PAGE (Fig. 1f). Six hydrophobic amino acids were introduced into the connector as N181R, P183L, E189I, D192F S193I, D194 (Fig. 1b). The other three mutations were constructed by introducing hydrophobic AA into the connectors with its first positive charged 14-aa peptide deleted (Fig. 1c–e). As expected, all the N-terminal hydrophobic mutated channels showed higher direct insertion efficiency than the wild-type Phi29 connector channel (Table 1).

Fig. 1 Construction of N-terminal mutants of Phi29 connector channel for direct membrane insertion. (a) PDB structure shows the hydrophobic and hydrophilic amino acids distribution of Phi29 connect channels. (Blue, hydrophilic amino acid; red, hydrophobic amino acids). (b–e) PDB structure derived images showing the sites of mutation of four different constructions of Phi29 connector. (f) The PAGE gel showing the molecular weight of the purified N-terminal mutated connector proteins (lane 1–4), C-Δ25 (lane 5), wild type GP10 (lane 6), N-terminal mutant T7 connector GP8 (lane 7), wild type T7 GP8 (lane 8). (g) Schematic diagram of the MinION™ system comprised of the Flow Cell nanopore array and holes (Mux). ASIC: Application-Specific Integrated Circuit. (h) Visualization of data from direct channel insertion. Mux1, 3, and 4 show the insertion in red, while the applied voltage is in blue (−100 mV).

Table 1 Statistical results of the number of direct insertions after the N-terminal hydrophobic mutation

Nanopore	Mutation sites	Average direct insertions	Peptide translocation (Y/N)
Phi29 GP10		0	N
Phi29 GP10	N181R, P183L, E189I, D192F, S193I, D194L	35	N
Phi29 GP10	N-Terminal 14AA deletion and N181I, P183L, E189I, D192F, S193I, D194L	40	N
Phi29 GP10	N-Terminal 14AA deletion and N166L, N167I,K172F,N176W	57	N
Phi29 GP10	N-Terminal 14AA deletion and N166I, N167I, E189I, D192F	127	N
T7-GP8		5	N
T7-GP8	P308L, R309I, R310I	16	N
T3-GP8		71	Y

---

The successful insertion of N-terminal modified Phi29 connector channels into copolymer membranes of MinION™ Flow Cell inspired its application for high-throughput analysis. MinION™ Flow Cell is an array that has 512 sensors (Fig. 1g), each has four micro-wells (mux, 1–4 in Fig. 1h). That is, up to 2048 channel areas are available for nanopore insertion into the MinION™ Flow Cells, generating a high-throughput sensing platform.

3.2. Assessment of terminal size and hydrophobicity for direct insertion and, in turn, the analyte translocation

A positively charged peptide, R-12 (RRR RRR RRR RRR), was used as a model to determine the application of the system for peptide and protein high-throughput sensing. Application of a solution containing peptides (1 mg ml⁻¹) and Flow Cell buffer (C13 buffer) occurred through the Flow Cell priming port (Fig. 1g). Confirmation of the Phi29 connector channels inserted into the Flow Cell was performed by observing a current at or above ±200 pA (Fig. 1h, Mux 1, 3, and 4, red arrow). A program for Platform Quality Check (QC) for peptide translocation was set at −100 mV, unless otherwise stated. Fig. S1† depicts the method of measuring current blockage. In this study, the blockage (%) is defined as (I_o − I_b) ÷ I_o × 100%. Where I_b represents the current after blocking during peptide translocation, and Io represents the open channel current.

We began by testing the peptide translocation ability of the N-terminal modified Phi29 channel using proteoliposomes on both MinION™ Flow Cell and patch-clamps systems. The N-terminal mutant proteoliposome Phi29 connector channels could translocate peptides (R12, 1 mg ml⁻¹) on the MinION™ Flow Cell system (Fig. 2a–c); this can only occur when the nanopore is placed inside of a liposome, thereby increasing the likelihood of C-terminal insertion regardless of the mutations done to the terminals. The current blockage of R12 on the N-terminal mutant Phi29 connector channels was 32% (Fig. 2c). Peptide translocation through the N-terminal mutant Phi29 proteoliposomes was also tested on the patch-clamp system (Fig. 2d–f) at 50 mV in 0.15 M KCl, 5Mm HEPES, pH7.4 buffer, showing a current blockage by R12 as 32% (Fig. 2f).

Fig. 2 Assessment of the effect of the size and hydrophobicity of the terminus on the analyte translocation via the Phi29 connector channel. a, d, g, and j illustrate the different channel insertion methods and orientations. (b and e) Current trace showing R12 translocation of N-terminal mutant Phi29 connector on the MinION™ Flow Cell (b), or the patch-clamp via liposome directly (e). (h and k) Current trace showing R12 translocation on MinION™ Flow Cell using the Phi29 connector with hydrophobic mutations on either N- or C-terminus for h and k, respectively. c, f, i, and l are the histograms of current blockage for each experimental design.

N-terminal modification could significantly enhance the direct insertion rate (Table 1). We then evaluated the peptide translocation ability in both the patch-clamp and MinION™ Flow Cells system. Unfortunately, after the peptide was added, no peptide translocation events were observed in both the MinION™ Flow Cell system (Fig. 2g–i) and patch-clamp system (Fig. S2†). This data confirmed that Phi29 connector channels with N-terminal insertion could not translocate peptides. Following this, dsDNA (50 bp), ssDNA(50 nt) and siRNA underwent translocation were tested. Again no translocation could be observed in any of the inserted channels (data not shown).

To test the hypothesis that the orientation of insertion determines translocation, the Phi29 connector was subjected to various modifications to enforce C-terminal direct insertion. Three hydrophilic amino acids at the C-terminus (Fig. S3a†) were mutated into hydrophobic ones (E266L, E270L, D283L) (Fig. S3b†). Twenty-five amino acids at the C-terminal were removed, leading to a C-Δ25. Reduction in the size of the C-terminus led to the reduced conductance of the channels from 1.00 nS for wild type connector into 0.65 nS for C-Δ25 in conducting buffer of 0.15 M KCl, 5 mM HEPES, pH 7.4 (Fig. S3c†). Excitingly, with the C-terminal direct insertion, a robust R12 peptide translocation took place (Fig. 2j–l). The current blockage caused by the addition of R12 to the C-Δ25 channel was 30% (Fig. 2l, see Fig. S3d† for no R12 control).

3.3. The conclusion of terminal size as a factor to control insertion rate was upheld by the T3 and T7 connector channel systems

Compared to the cone-shaped Phi29 connector channel (Fig. 3a), the olive-shaped T7 and T3 connectors have a much narrower C-terminus (Fig. 3b and c). To determine the influence of terminus size on insertion rate, we tested the direct insertion ability of wild-type T3 and T7 connectors on the MinION™ Flow Cell system. As shown in Table 1, both wild-type T3 and T7 connector channels underwent efficient insertion into the Flow Cell, unlike the wide-type Phi29 connector channel (Table 1). The connector of bacteriophages T3 and T7 is very similar; however, the T3 connector has a much narrower C-terminal than the T7 connector. Analysis of multiple trials showed that the T3 channel with a narrower C-terminus has a higher direct insertion efficiency (Table 1). For a trial on one flow cell, a total of 71 direct insertions were observed in the T3 connector, while the T7 only had 5 (Table 1). On the Clamp system, fifteen T3 connector channel direct insertions into lipid bilayer were also observed within 30 min in 0.15 M KCl, 5 mM HEPES, pH7.4 conductance buffer (data not shown).

Fig. 3 PDB structure shows the terminal size and hydrophobicity controlling the direct insertion rate and analyte translocation. (a–c) PDB structure derived images show the terminal size of the (a) Phi29 (b) T7 connector and (c) T3 connector. The color represents the group of residues. (d and e) PDB structure shows the hydrophobic and hydrophilic amino acid distribution at the C-terminus (left) and N-terminus (right) of the Phi29 (d), T7, and T3 connector (e), respectively. (Blue, hydrophilic amino acid; red, hydrophobic amino acids).

3.4. The conclusion of orientation as a factor for controlling peptide translocation was supported by the T3 and T7 connector channel systems

The hydrophobicity of the two terminuses for T3 and T7 connectors are entirely different. The T7 connector has a hydrophobic N-terminus, while T3 has a hydrophobic C-terminus, meaning the C-terminal insertion of the T3 connector is a favorable orientation (Fig. 3e). The peptide (R12, 1 mg ml⁻¹) translocation was further tested on an N-terminal mutant T7 connector (P308L, R309I, R310I) (Fig. S4†) and a wild-type T3 connector channel. As shown in Fig. 4g–i, no peptide translocation was observed in all 16 directly inserted T7 connectors. However, after proteoliposome insertion, peptide translocation occurred on both the patch-clamp at −50 mV (Fig. 4a–c) and MinION™ Flow Cell systems (Fig. 4d–f) with an observed current blockage of ∼60%.

Fig. 4 Assessment of the effect of the size and hydrophobicity of the terminus on the analyte translocation via the T3 and T7 connector channels. a, d, g, and j illustrate the different channel insertion methods and orientations. (b and e) Current trace showing R12 translocation of N-terminal mutant T7 connector on the patch-clamp (b) or the MinION™ Flow Cell (e) via liposome directly. (h and k) Current trace showing R12 translocation on MinION™ Flow Cell using the T7 connector with hydrophobic mutations on N-terminus (h) and wild-type T3 connector (k). c, f, i, and l are the histograms of current blockage for each experimental design.

Excitingly, peptide translocation signals were detected on the directly inserted T3 connector (Fig. 4j and k). The current blockage caused by the addition of R12 to the T3 channel was 29% (Fig. 4l, see Fig. S5† for R12 free control tests) at 100 mV. Interestingly, all the directly inserted T3 connector channels could translocate peptides, quite possibly due to their hydrophobicity of C-terminals and the similarity in termini that allowed for a higher chance of C-terminal insertion (data not shown).

3.5. Elucidation of the one-way traffic property of the Phi29 connector channel to explain the analyte blockage by N-terminal insertion

Previous studies demonstrated the ‘one-way traffic’ property of bacteriophage Phi29 DNA packaging motor; analytes unidirectionally flow from the N- to C-terminal.^79,80 This study revealed that an N-terminal hydrophobic mutation enhanced the N-terminal direct insertion into the polymer membrane (Table 1). However, when peptides were added to the MinION™ Flow Cell membrane, no translocation was detected (Fig. 5a). In contrast, C-terminal 25-AA truncation and three hydrophobic mutations facilitate C-terminal direct insertion (Fig. 5b), allowing peptides to pass through the connector channel from the N-terminus that was exposed to the peptides in the buffer. As for proteoliposomes, the connector can insert into the lipid membrane via both orientations (Fig. 5c). This explains why the orientation challenge for the liposome chamber has not been a major concern, as reported previously.⁷⁸

Fig. 5 Schematic diagram to explain the orientation-controlled analyte translocation. Channel insertion orientation is a determining factor for peptide translocation on MinION™ Flow Cell. (a) The peptide cannot translocate due to one-way traffic after N-direct terminal insertion. (b) The peptide can translocate after C-terminal insertion. (c) The peptide can translocate after the C-terminal insertion of proteoliposome, but not N-terminal insertion.

3.6. Application of N-terminal insertion of the Phi29 connector channel for tumor biomarker detection

When the C-terminal of the Phi29 connector was inserted, translocation of the peptide on both the patch-clamp and MinION™ Flow Cell system was observed, but none took place during N-terminal insertion. To expand the application of N-terminal insertion, a re-engineered Phi29 connector channel (N-His GP10-C-uPAR) was constructed with a urokinase-type plasminogen activator receptor (uPAR) binding peptide (probe) fused at the C-terminus (Fig. 6a). uPAR has proven to be predictive biomarkers in several types of cancer, including breast cancer, pancreatic cancer, soft-tissue sarcoma, and pulmonary adenocarcinoma.⁸¹ Once inserted by the N-terminal of N-His GP10-C-uPAR channel and incubated with the uPAR protein, specific binding events characterized by a longer dwell time and a current blockage of roughly 32% were produced (Fig. 6b–d). Interestingly, two uPAR proteins binding to the reengineered Phi29 connector channel would result in a two-steps current change and cause a current blockage of about 64% (Fig. 6b and Fig. S6†). The current blockage change is due to a conformation change of the connector channel, which is similar to the three-steps gating caused by high voltage. It should be pointed out that if three uPAR proteins bind to the C-terminal probe simultaneously, the channel can undergo three-step gating (Fig. S7†). To test the specific binding, the cell lysates (1 mg ml⁻¹) from a uPAR negative breast cancer cell line (SKBR3) were used.⁸² No specific binding events could be observed at either 50 mV or −50 mV after one hour of addition of SKBR3 cell lysates to the 0.15 M KCl, 5 mM HEPES, pH 7.4 buffer (Fig. 6e). In contrast, when MDA-MB-231 cell (uPAR high-expression) lysates⁸³ (1 mg ml⁻¹) were added, permanent binding events appeared (Fig. 6g). Interestingly, besides permanent binding events, three-step gating events were also observed after adding MDA-MB-231 cell lysates (Fig. 6h).

Fig. 6 Application of N-terminal inserted Phi29 connector channel for tumor biomarker detection. (a and i) Schematic diagram of real-time detection of cancer biomarker (uPAR) using N-terminal inserted Phi29 connector channel on patch-clamp (a) and MinION™ Flow Cell (i) systems. uPAR probe was fused at the C-terminal of the Phi29 connector. (b) Current trace showing the specific binding events and noise events (non-specific background events). Green: current signature due to specific binding. Red: non-specific background current signature. Blue arrow: two uPAR proteins binding to the C-terminal probe simultaneously. (c) Histogram plotting of current transition events of specific (green) and non-specific (red) binding to the probe at the C-terminal. (d) Scatter plot of specific (green) and non-specific binding populations. (e) Current trace result after one hour of the addition of SKBR3 cell lysates. (f) Current trace result before the addition of MDA-MB-231 cell lysates. (g) Current trace showing the permanent binding events (green) after the addition of MDA-MB-231 cell lysates. (h) Current trace showing the three-step gating due to the addition of MDA-MB-231 cell lysates. (j and k) Current trace result after the addition of BSA (j) and BSA/uPAR (k) to N-terminal inserted Phi29 channel with uPAR probe on MinION™ Flow Cell. (l) Current trace showing current signature due to specific binding on the MinION™ Flow Cell (red color signals).

The potential of tumor biomarker detection using the N-terminal inserted re-engineered Phi29 connector channel was also evaluated on the MinION™ Flow Cell system (Fig. 6i). As shown in Fig. 6j, the addition of BSA protein to the N-terminal inserted channel could not induce channel gating. In contrast, gating was appeared with specific binding signals (Fig. 6k and l) after adding uPAR protein to the channel in the presence of BSA protein. The gating was only observed at −100 mV, but not 100 mV (Fig. S8a†). The current trace result without protein as control was also recorded (Fig. S8b†).

4. Conclusion

By constructing various mutant Phi29 and T7 connector channels, it was concluded that the size (narrowness) and hydrophobicity of the N-or C-terminus of the connector of the viral DNA packaging motors are essential factors concerning direct insertion into membranes or Flow Cell and their ability for analyte translocation. The introduction of hydrophobic amino acids to the N-terminus of the connector channel enhanced the rate of direct membrane insertion via the N-terminal, but no translocation of analyte took place due to the ‘one-way traffic’ property. The C- and N-terminals of T3 and T7 connectors display similarity in width or diameter, while in phi29, the width of the C- and N terminals are very different. An N- or C-terminus hydrophobic modification governs insertion orientation and subsequent macromolecule translocation, thus leading to two different modes of application. The C-terminal insertion permits translocation of analytes such as peptides to enter the channel through the N terminus, while N-terminus insertion prevents translocation but offers the measurement of gating as a sensing parameter, thus generating a tool for detection of markers. The uPAR probe fused to the Phi29 connector offers a promising platform for tumor biomarker detection. The application of orientation controllable channel insertion into the MinION™ Flow Cell and membranes offers a high-throughput sensing platform for macromolecule sensing and tumor biomarkers detection with potential clinical applications.

Author contributions

L. Z. and N. B. performed experiments, L. Z. prepared the manuscript. M. J. and L. J. provided the instruction for MinION™ setting in the lab; M. J. modified script in MinKnow software; P. G. designed the project, supervised, acquired funding and prepared the manuscript. All authors wrote the manuscript.

Data availability

The raw data required for these findings are available upon request by email to guo.1091@osu.edu.

Conflicts of interest

P. G. is the consultant and licensor of Oxford Nanopore Technologies; the cofounder of Shenzhen P&Z Bio-medical Co. Ltd, as well as the cofounder and the chairman of the board of ExonanoRNA, LLC.

Acknowledgements

The research was supported by a Sponsor Research Contract from Oxford Nanopore Technologies Ltd to The Ohio State University (to P.G.).

References

1. T. Ding, J. Yang, V. Pan, N. Zhao, Z. Lu, Y. Ke and C. Zhang, Nucleic Acids Res., 2020, 48, 2791–2806.
2. F. Haque, J. Li, H. C. Wu, X. J. Liang and P. Guo, Nano Today, 2013, 8, 56–74.
3. L. Liu and H. C. Wu, Angew. Chem., Int. Ed., 2016, 55, 15216–15222.
4. N. Varongchayakul, J. Song, A. Meller and M. W. Grinstaff, Chem. Soc. Rev., 2018, 47, 8512–8524.
5. J. D. Spitzberg, A. Zrehen, X. F. van Kooten and A. Meller, Adv. Mater., 2019, 31, e1900422.
6. J. Wang, J. Yang, Y. L. Ying and Y. T. Long, Chem.– Asian J., 2019, 14, 389–397.
7. L. Q. Gu and J. W. Shim, Analyst, 2010, 135, 441–451.
8. Y. Wang, L. Q. Gu and K. Tian, Nanoscale, 2018, 10, 13857–13866.
9. B. N. Miles, A. P. Ivanov, K. A. Wilson, F. Dogan, D. Japrung and J. B. Edel, Chem. Soc. Rev., 2013, 42, 15–28.
10. J. Houghtaling, J. List and M. Mayer, Small, 2018, 14, e1802412.
11. S. Majd, E. C. Yusko, Y. N. Billeh, M. X. Macrae, J. Yang and M. Mayer, Curr. Opin. Biotechnol, 2010, 21, 439–476.
12. G. Perez-Mitta, M. E. Toimil-Molares, C. Trautmann, W. A. Marmisolle and O. Azzaroni, Adv. Mater., 2019, 31, e1901483.
13. F. Haque, J. Lunn, H. Fang, D. Smithrud and P. Guo, ACS Nano, 2012, 6, 3251–3261.
14. N. S. Galenkamp, M. Soskine, J. Hermans, C. Wloka and G. Maglia, Nat. Commun., 2018, 9, 4085.
15. W. J. Ramsay and H. Bayley, Angew. Chem., Int. Ed., 2018, 57, 2841–2845.
16. D. R. Garalde, E. A. Snell, D. Jachimowicz, B. Sipos, J. H. Lloyd, M. Bruce, N. Pantic, T. Admassu, P. James, A. Warland, M. Jordan, J. Ciccone, S. Serra, J. Keenan, S. Martin, L. McNeill, E. J. Wallace, L. Jayasinghe, C. Wright, J. Blasco, S. Young, D. Brocklebank, S. Juul, J. Clarke, A. J. Heron and D. J. Turner, Nat. Methods, 2018, 15, 201–206.
17. M. Ayub, S. W. Hardwick, B. F. Luisi and H. Bayley, Nano Lett., 2013, 13, 6144–6150.
18. M. Wanunu, T. Dadosh, V. Ray, J. Jin, L. McReynolds and M. Drndic, Nat. Nanotechnol., 2010, 5, 807–814.
19. Y. Wang, D. Zheng, Q. Tan, M. X. Wang and L. Q. Gu, Nat. Nanotechnol., 2011, 6, 668–674.
20. D. Wendell, P. Jing, J. Geng, V. Subramaniam, T. J. Lee, C. Montemagno and P. Guo, Nat. Nanotechnol., 2009, 4, 765–772.
21. J. Clarke, H. C. Wu, L. Jayasinghe, A. Patel, S. Reid and H. Bayley, Nat. Nanotechnol., 2009, 4, 265–270.
22. E. A. Manrao, I. M. Derrington, A. H. Laszlo, K. W. Langford, M. K. Hopper, N. Gillgren, M. Pavlenok, M. Niederweis and J. H. Gundlach, Nat. Biotechnol., 2012, 30, 349–353.
23. D. Deamer, M. Akeson and D. Branton, Nat. Biotechnol., 2016, 34, 518–524.
24. M. Jain, S. Koren, K. H. Miga, J. Quick, A. C. Rand, T. A. Sasani, J. R. Tyson, A. D. Beggs, A. T. Dilthey, I. T. Fiddes, S. Malla, H. Marriott, T. Nieto, J. O'Grady, H. E. Olsen, B. S. Pedersen, A. Rhie, H. Richardson, A. R. Quinlan, T. P. Snutch, L. Tee, B. Paten, A. M. Phillippy, J. T. Simpson, N. J. Loman and M. Loose, Nat. Biotechnol., 2018, 36, 338–345.
25. G. M. Cherf, K. R. Lieberman, H. Rashid, C. E. Lam, K. Karplus and M. Akeson, Nat. Biotechnol., 2012, 30, 344–348.
26. M. G. Larimi, L. A. Mayse and L. Movileanu, ACS Nano, 2019, 13, 4469–4477.
27. F. Piguet, H. Ouldali, M. Pastoriza-Gallego, P. Manivet, J. Pelta and A. Oukhaled, Nat. Commun., 2018, 9, 966.
28. A. Asandei, I. Schiopu, M. Chinappi, C. H. Seo, Y. Park and T. Luchian, ACS Appl. Mater. Interfaces, 2016, 8, 13166–13179.
29. G. Huang, K. Willems, M. Soskine, C. Wloka and G. Maglia, Nat. Commun., 2017, 8, 935.
30. Z. Ji, S. Wang, Z. Zhao, Z. Zhou, F. Haque and P. Guo, Small, 2016, 12, 4572–4578.
31. S. Wang, Z. Zhou, Z. Zhao, H. Zhang, F. Haque and P. Guo, Biomaterials, 2017, 126, 10–17.
32. Z. Ji, X. Kang, S. Wang and P. Guo, Biomaterials, 2018, 182, 227–233.
33. B. Cressiot, A. Oukhaled, G. Patriarche, M. Pastoriza-Gallego, J. M. Betton, L. Auvray, M. Muthukumar, L. Bacri and J. Pelta, ACS Nano, 2012, 6, 6236–6243.
34. J. Su, H. Esmaeilzadeh, F. Zhang, Q. Yu, G. Cernigliaro, J. Xu and H. Sun, Biosens. Bioelectron., 2018, 99, 325–331.
35. J. Nivala, L. Mulroney, G. Li, J. Schreiber and M. Akeson, ACS Nano, 2014, 8, 12365–12375.
36. C. B. Rosen, D. Rodriguez-Larrea and H. Bayley, Nat. Biotechnol., 2014, 32, 179–181.
37. M. Soskine, A. Biesemans, B. Moeyaert, S. Cheley, H. Bayley and G. Maglia, Nano Lett., 2012, 12, 4895–4900.
38. E. C. Yusko, J. M. Johnson, S. Majd, P. Prangkio, R. C. Rollings, J. Li, J. Yang and M. Mayer, Nat. Nanotechnol., 2011, 6, 253–260.
39. J. J. Kasianowicz, E. Brandin, D. Branton and D. W. Deamer, Proc. Natl. Acad. Sci. U. S. A., 1996, 93, 13770–13773.
40. R. Kawano, T. Osaki, H. Sasaki, M. Takinoue, S. Yoshizawa and S. Takeuchi, J. Am. Chem. Soc., 2011, 133, 8474–8477.
41. U. F. Keyser, Nat. Nanotechnol., 2016, 11, 106–108.
42. A. R. Hall, A. Scott, D. Rotem, K. K. Mehta, H. Bayley and C. Dekker, Nat. Nanotechnol., 2010, 5, 874–877.
43. S. Cabello-Aguilar, S. Balme, A. A. Chaaya, M. Bechelany, E. Balanzat, J. M. Janot, C. Pochat-Bohatier, P. Miele and P. Dejardin, Nanoscale, 2013, 5, 9582–9586.
44. B. Cressiot, S. J. Greive, M. Mojtabavi, A. A. Antson and M. Wanunu, Nat. Commun., 2018, 9, 4652.
45. W. Shi, A. K. Friedman and L. A. Baker, Anal. Chem., 2017, 89, 157–188.
46. S. Wang, Z. Zhao, F. Haque and P. Guo, Curr. Opin. Biotechnol, 2018, 51, 80–89.
47. S. Huang, M. Romero-Ruiz, O. K. Castell, H. Bayley and M. I. Wallace, Nat. Nanotechnol., 2015, 10, 986–991.
48. P. Mueller, D. O. Rudin, H. T. Tien and W. C. Wescott, Nature, 1962, 194, 979–980.
49. C. Cao, D. F. Liao, J. Yu, H. Tian and Y. T. Long, Nat. Protoc., 2017, 12, 1901–1911.
50. A. Balijepalli, J. Ettedgui, A. T. Cornio, J. W. Robertson, K. P. Cheung, J. J. Kasianowicz and C. Vaz, ACS Nano, 2014, 8, 1547–1553.
51. J. W. Shim and L. Q. Gu, Anal. Chem., 2007, 79, 2207–2213.
52. P. B. Stranges, M. Palla, S. Kalachikov, J. Nivala, M. Dorwart, A. Trans, S. Kumar, M. Porel, M. Chien, C. Tao, I. Morozova, Z. Li, S. Shi, A. Aberra, C. Arnold, A. Yang, A. Aguirre, E. T. Harada, D. Korenblum, J. Pollard, A. Bhat, D. Gremyachinskiy, A. Bibillo, R. Chen, R. Davis, J. J. Russo, C. W. Fuller, S. Roever, J. Ju and G. M. Church, Proc. Natl. Acad. Sci. U. S. A., 2016, 113, E6749–E6756.
53. A. Fragasso, S. Schmid and C. Dekker, ACS Nano, 2020, 14, 1338–1349.
54. J. Wang, Y. L. Ying, C. B. Zhong, L. M. Zhang, F. Yan and Y. T. Long, Analyst, 2021, 146, 4111–4120.
55. W. Knoll, C. W. Frank, C. Heibel, R. Naumann, A. Offenhausser, J. Ruhe, E. K. Schmidt, W. W. Shen and A. Sinner, J. Biotechnol., 2000, 74, 137–158.
56. J. Bentin, S. Balme and F. Picaud, Soft Matter, 2020, 16, 1002–1010.
57. W. Romer and C. Steinem, Biophys. J., 2004, 86, 955–965.
58. J. Andersson and I. Koper, Membranes, 2016, 6, 30.
59. D. Morton, S. Mortezaei, S. Yemenicioglu, M. J. Isaacman, I. C. Nova, J. H. Gundlach and L. Theogarajan, J. Mater. Chem. B, 2015, 3, 5080–5086.
60. L. Zhang, M. L. Gardner, L. Jayasinghe, M. Jordan, J. Aldana, N. Burns, M. A. Freitas and P. Guo, Biomaterials, 2021, 276, 121022.
61. Z. Ji, M. Jordan, L. Jayasinghe and P. Guo, Nanomedicine, 2020, 25, 102170.
62. Z. L. Hu, M. Z. Huo, Y. L. Ying and Y. T. Long, Angew. Chem., Int. Ed., 2020, 60, 14738–14749.
63. M. Chinappi and F. Cecconi, J. Phys.: Condens. Matter, 2018, 30, 204002.
64. J. W. F. Robertson and J. E. Reiner, Proteomics, 2018, 18, e1800026.
65. J. J. Kasianowicz and S. M. Bezrukov, Nat. Biotechnol., 2016, 34, 481–482.
66. S. Howorka and Z. S. Siwy, Nat. Biotechnol., 2020, 38, 159–160.
67. H. Ouldali, K. Sarthak, T. Ensslen, F. Piguet, P. Manivet, J. Pelta, J. C. Behrends, A. Aksimentiev and A. Oukhaled, Nat. Biotechnol., 2020, 38, 176–181.
68. A. K. Thakur and L. Movileanu, Nat. Biotechnol., 2018, 37, 96–101.
69. L. Harrington, L. T. Alexander, S. Knapp and H. Bayley, ACS Nano, 2019, 13, 633–641.
70. C. Wloka, V. Van Meervelt, D. van Gelder, N. Danda, N. Jager, C. P. Williams and G. Maglia, ACS Nano, 2017, 11, 4387–4394.
71. E. L. Bonome, F. Cecconi and M. Chinappi, Nanoscale, 2019, 11, 9920–9930.
72. L. Restrepo-Perez, C. H. Wong, G. Maglia, C. Dekker and C. Joo, Nano Lett., 2019, 19, 7957–7964.
73. P. Waduge, R. Hu, P. Bandarkar, H. Yamazaki, B. Cressiot, Q. Zhao, P. C. Whitford and M. Wanunu, ACS Nano, 2017, 11, 5706–5716.
74. J. Nivala, D. B. Marks and M. Akeson, Nat. Biotechnol., 2013, 31, 247–250.
75. J. Geng, H. Fang, F. Haque, L. Zhang and P. Guo, Biomaterials, 2011, 32, 8234–8242.
76. H. Gårdsvoll, A. J. van Zonneveld, A. Holm, E. Eldering, M. van Meijer, K. Danø and H. Pannekoek, FEBS Lett., 1998, 431, 170–174.
77. F. Haque, J. Geng, C. Montemagno and P. Guo, Nat. Protoc., 2013, 8, 373–392.
78. Z. Zhou, Z. Ji, S. Wang, F. Haque and P. Guo, Biomaterials, 2016, 105, 222–227.
79. Z. Zhao, E. Khisamutdinov, C. Schwartz and P. Guo, ACS Nano, 2013, 7, 4082–4092.
80. P. Jing, F. Haque, D. Shu, C. Montemagno and P. Guo, Nano Lett., 2010, 10, 3620–3627.
81. N. Mahmood, C. Mihalcioiu and S. A. Rabbani, Front. Oncol., 2018, 8, 24.
82. M. C. Huber, R. Mall, H. Braselmann, A. Feuchtinger, S. Molatore, K. Lindner, A. Walch, E. Gross, M. Schmitt, N. Falkenberg and M. Aubele, BMC Cancer, 2016, 16, 615.
83. S. Zuppone, C. Assalini, C. Minici, S. Bertagnoli, P. Branduardi, M. Degano, M. S. Fabbrini, F. Montorsi, A. Salonia and R. Vago, Sci. Rep., 2020, 10, 2521.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/d1bm01264a

---