Article | Open | Published:

# Identification of single amino acid differences in uniformly charged homopolymeric peptides with aerolysin nanopore

## Abstract

There are still unmet needs in finding new technologies for biomedical diagnostic and industrial applications. A technology allowing the analysis of size and sequence of short peptide molecules of only few molecular copies is still challenging. The fast, low-cost and label-free single-molecule nanopore technology could be an alternative for addressing these critical issues. Here, we demonstrate that the wild-type aerolysin nanopore enables the size-discrimination of several short uniformly charged homopeptides, mixed in solution, with a single amino acid resolution. Our system is very sensitive, allowing detecting and characterizing a few dozens of peptide impurities in a high purity commercial peptide sample, while conventional analysis techniques fail to do so.

## Introduction

Recently, the concept of personalized medicine has evolved towards the analysis of rare cells (tumor and endothelial cells, cancer stem cells, etc.) and extracellular vesicles circulating in the blood, containing a few copies of biomarkers. Among the biomarkers there are short signal peptides that are able to modulate the cancer cells aggressiveness more or less upon their ability to be processed by proteolysis driven by N-end rules pathways1. Classical metabolomics and proteomics techniques, such as mass spectrometry (MS), High Performance Liquid Chromatography (HPLC), or Nuclear Magnetic Resonance (NMR), suffer a lack of reproducibility and sensitivity that prohibit their use for medical diagnostic2,3. In addition, these techniques are too expensive, time consuming and heavy to handle in the view of developing a portable medical device application3. In the industry context, pharmaceutical peptides manufacturing requires a high purity level and is still waiting for high-sensitivity analytical methods addressing Good Manufacturing Practice issues such as impurities profile characterization4. Nanopore single-molecule technology5 seems to be an excellent alternative for addressing the above issues. For more than two decades, the nanopore technology has been essentially applied to DNA6. The development of nanopore-based DNA sequencing is now achieved and commercialized. The current challenge is the application of nanopore technology to peptides and proteins identification, fingerprinting and sequencing.

To date, nanopores were successfully used to detect protein or peptides either through biological nanopores7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39 or through artificial nanopores40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59. Protein or peptides were probed either inside a natural nanopore21,29,35 or outside a natural nanopore with an aptamer21,22 or with a ligand7. Pioneering studies were devoted to the dynamics of peptides transport9,10,13,14,16,30, and then to the dynamics of unfolded proteins transport18,31,48 through the nanopore. Others studies focused on protein-antibodies complex32,41,42, on intrinsically disordered proteins49, on enzymatic reactions15,17,23,60,61 and on the aggregation of amyloidogenic peptides36,47,55. The unfolding of proteins was investigated, either in the presence of an unfolding agent12,31,44, or by varying temperature24, or with an electrical force50,62, or with a molecular motor such as an unfoldase27. Nanopores were also used to study the structure of peptides8, conformational changes of proteins12,39,44,54,57, post-translational modification of proteins52,58 and to estimate the shape, volume, charge, rotational diffusion coefficient and dipole moment of individual proteins63.

In the context of nanopore-based protein sizing, glass nanopores were used to determine folded protein sizes ranging from 12 to 480 kDa51,53. The best discrimination resolution obtained with this approach was a difference of 4 kDa in the case of the shortest proteins, and a difference of 50 kDa in the case of the longest proteins51,53. Then solid-state nanopores were demonstrated to enable the detection of proteins as small as 8.5 kDa54 and the discrimination between two different proteins with a resolution of 1.7 kDa57. Regarding biological nanopores, Fragaceatoxin C (FraC) channels were recently successfully used to discriminate peptides and proteins as a model biomarkers ranging from 25 kDa down to 1.3 kDa38. In the same context, α-hemolysin nanopore equiped with Au25(SG)18 clusters was used to increase the on-rate and off-rate of peptides to the pore in order to improve the mass resolution of the nanopore sensor37. Thus, while biological nanopores were previously used for size-discriminating DNA hairpins64, synthetic polymers65,66,67 and oligonucleotides68 with monomeric resolution, the single-amino-acid size-discrimination between several (more than two) peptides mixed in solution that differ by a single amino acid and the characterization of rare peptide impurities were not demonstrated, regardless of the pore used9,10,20,30,37,38,65,66,68. In previous works using biological nanopores, there was no more than two peptides differing by one amino acid among the different peptides used10,37,38.

Here, we demonstrate the proof-of-concept of size-discrimination of short uniformly charged homopeptides, mixed in solution or independently, with a single amino acid resolution by using the recombinant aerolysin nanopore without any physical or chemical modifications. The sensitivity of this pore allows detecting, with a single amino acid resolution, peptide fragments of different lengths present in a high purity commercial peptide sample. We also show the discrimination between two homopolymeric- and one heteropolymeric peptides having the same length. Our system also enables the real-time monitoring of the evolution of the peptide populations with a single amino acid resolution during the enzymatic degradation of a peptide sample. These findings open new perspectives, in particular towards the use of nanopore technology for the size-discrimination and sequence recognition of peptides and proteins.

## Results

### Single amino acid resolution size-discrimination of peptides

A schematics of the experimental setup is shown in Fig. 1a. A single wild-type aerolysin nanopore was inserted in a lipid bilayer separating the cis-compartments and trans-compartments filled with an electrolyte solution. An external voltage was applied on the trans-side of the bilayer, the cis-side being at voltage ground. In absence of analytes, a stable ionic current of mean value I0 flowing through the nanopore was measured. The addition of an equimolar mixture of arginine peptides of 6 different lengths differing by a single amino acid in length (5, 6, 7, 8, 9, 10 amino acids) on the cis-side of the bilayer induced well-resolved current blockades of different amplitudes and durations under a negative applied voltage (Fig. 1a,c), in agreement with the positive electrical charge of arginine peptides under our experimental conditions. In contrast, no current blockades were observed either when the peptides were added on the cis-side under a positive applied voltage, or when the peptides were added on the trans-side regardless of the voltage polarity (Supplementary Fig. 1).

The histogram of the relative blockade current Ib/I0 values induced by the mixture of arginine peptides of 6 different lengths is shown in Fig. 1d. Well-separated peaks corresponding to distinct populations with preferred Ib/I0 values are observed. For each of the 6 main populations, the mean relative blockade current is (mean value of a gaussian fit of the peak +/− one standard deviation): 0.234 +/− 0.001, 0.286 +/− 0.002, 0.353 +/− 0.002, 0.435 +/− 0.002, 0.530 +/− 0.002, 0.631 +/− 0.004. The interval between two consecutive peaks is at least 25-fold greater than the peaks widths. The mean blockade duration of each population increases by almost two orders of magnitude as the relative blockade current value decreases, from 0.9 ms for the shallowest population to 72 ms for the deepest population.

In order to ascertain whether the different populations observed in the Ib/I0 histogram correspond to the different peptide lengths present in the arginine peptide mixture, experiments probing independently the interaction of arginine peptides of each individual length (5, 6, 7, 8, 9, 10 amino acids) with the aerolysin nanopore were performed under the same experimental conditions. The histograms of the relative blockade current Ib/I0 values corresponding to each of the 6 different peptide lengths are shown in Fig. 2b–g. Each histogram shows an overwhelming majority population with a preferred Ib/I0 value, corresponding to the peptide length used. The location of the majority population depends on the peptide length: the mean Ib/I0 value of the majority population decreases as the peptide length increases. Comparing with the Ib/I0 histogram of the peptide mixture (Fig. 2a) reveals that each majority population corresponding to a given peptide length coincides with one of the main populations observed in the mixture histogram. This enables to attribute an individual peptide length to each population observed in the mixture histogram, from 10 amino acids long arginine peptides for the deepest population (Ib/I0 = 0.234) to 5 amino acids long arginine peptides for the shallowest population (Ib/I0 = 0.631). This demonstrates the ability of the wild-type aerolysin nanopore to perform the size-discrimination of arginine peptides with a single amino acid resolution. Furthermore, measurements of the mean relative blockade current value and of the mean blockade duration corresponding to each individual peptide length are highly reproducible between independent experiments (Supplementary Fig. 2), highlighting the high-fidelity resolution power of the nanopore.

The effect of voltage on the interaction of the aerolysin nanopore with a mixture of arginine peptides of 6 different lengths (5 to 10 amino acids) is shown in Supplementary Figs. 3 and 4, for negative voltages ranging from −25 to −80 mV. For each voltage, the populations corresponding to the different peptide lengths in the mixture were identified from the Ib/I0 histogram (Supplementary Fig. 4) and independently analyzed (see the Methods section). Regardless of the applied voltage, the mean blockade duration increases with the peptide length, up to two orders of magnitude between 5 amino acids long and 10 amino acids long peptides (Supplementary Fig. 3b). For each peptide length, the mean blockade duration depends non-monotonically on voltage, first increasing with voltage magnitude, reaching a maximum and then decreasing. The voltage corresponding to the maximum of the mean blockade duration depends only slightly on the peptide length, if at all, and is comprised between −35 and −40 mV. Note that the mean blockade duration corresponding to each peptide length are identical when measured whether in a peptide mixture experiment or in independent experiments for each peptide length.

The qualitative voltage dependence of the mean relative blockade current Ib/I0 depends on the arginine peptide length (Supplementary Figs. 3a and 4): the mean Ib/I0 slightly decreases with voltage magnitude for the longest peptides (10, 9, and 8 amino acids), is voltage-independent for intermediate peptide lengths (7 and 6 amino acids) and even slightly increases with voltage magnitude for the shortest peptides (5 amino acids). As shown in the Ib/I0 histograms of the peptide mixture under different applied voltages (Supplementary Fig. 4), the size-discrimination of arginine peptides of different lengths with a single amino acid resolution is preserved in the whole voltage range explored, even at a voltage as low as −25 mV. Note that maximizing the mean blockade duration for each of the different peptide lengths (at ≈−40 mV) does not maximize the size-discrimination power of the nanopore. Interestingly, the different populations of the Ib/I0 histogram of the peptide mixture become increasingly separated as the negative voltage magnitude increases. For voltage magnitudes greater than or equal to −50 mV (Supplementary Fig. 4 and Fig. 1d), an additional population with a mean Ib/I0 ≈ 0.73 appeared. This additional population is the 7th population detected from an arginine peptide mixture of 6 different lengths, and is located in the short peptide part of the Ib/I0 histogram. Observing this additional population was not possible at low voltages, probably due to the presence of a significant number of current fluctuations from the residual open pore current noise, with Ib/I0 values between ≈0.7 and 0.8. As the voltage magnitude is increased, the signal-to-noise ratio is enhanced ($$\frac{{I_0 - 5\sigma }}{{I_0}} = 0.83$$ at −25 mV and 0.91 at −80 mV, I0 and σ being the mean open pore current and its standard deviation respectively, see the Methods section and Supplementary Fig. 8 for details); current fluctuations due to residual open pore current noise are shifted to greater Ib/I0 values with increasing voltage magnitude, enabling the clear observation of the additional population whose identification is established in the next section.

### Detection and identification of peptide impurities

The histogram of the relative blockade current Ib/I0 values and the scatter plot of blockade duration as a function of the relative blockade current are shown in Fig. 3b in the case of the interaction of a high-purity arginine peptide sample (>98% purity in peptide length) of given length (9 amino acids) with the aerolysin nanopore. As already observed in Fig. 2c, the Ib/I0 histogram shows an overwhelming majority population corresponding to the 9 amino acids peptide length used. In addition to the majority population, at least four minority populations are detected. Each peak of the Ib/I0 histogram corresponds to a distinct line-shaped cluster of dots in the scatter plot of blockade duration as a function of the relative blockade current, showing the different populations, and in particular the minority populations, even more clearly. Comparing with the case of the interaction of an equimolar mixture of arginine peptides of 6 different lengths (5 to 10 amino acids) with the aerolysin nanopore (Fig. 3a) reveals that the minority populations detected in the high-purity 9 amino acids long arginine peptide sample correspond to arginine peptide fragments whose length can be identified (5, 6, 7, 8 amino acids). Interestingly, quantities as small as a few dozens of molecules are detected for each peptide fragment length (see the right axis of the Ib/I0 histogram in Fig. 3b), revealing the high sensitivity of the nanopore system. In contrast, the purity analysis provided by the peptide supplier using conventional techniques (High Purity Liquid Chromatography, HPLC, and Mass Spectrometry, MS) did not identify these peptide fragments, including them among general impurities (Supplementary Fig. 5). This demonstrates that the aerolysin nanopore enables the detection, identification and quantification of the peptide impurities present in a high-purity peptide sample with a higher sensitivity than the conventional purity analysis techniques (HPLC and MS).

### Discrimination between lysine and arginine peptides

In order to ascertain whether the above-described method could be used with other peptides than arginine peptides, experiments probing the interaction of lysine peptides with the aerolysin nanopore were performed.

The histograms of the relative blockade current Ib/I0 values and the scatter plots of blockade duration as a function of the relative blockade current are shown in Fig. 5 in the case of the interaction of a high-purity (>98%) 10 amino acids long lysine peptide sample with the aerolysin nanopore in absence of trypsin enzyme (Fig. 5a), 30 min after trypsin addition (Fig. 5b) and 1 h 45 min after trypsin addition (Fig. 5c). Before trypsin addition (Fig. 5a), the Ib/I0 histogram shows an overwhelming majority population, corresponding to the 10 amino acids lysine peptide length used, and additional minority populations, as in the case of arginine peptides. Moreover, as time increases after trypsin addition (Fig. 5b,c), the number of current blockades corresponding to long lysine peptides decreases while the number of current blockades corresponding to shorter lysine peptides increases, indicating the enzymatic cleavage of long lysine peptides into shorter ones, as in the case of arginine peptides. This observation is an important indicator of the ability of the wild-type aerolysin nanopore to perform the size-discrimination of lysine peptides.

The histograms of the relative blockade current Ib/I0 values and the scatter plots of blockade duration as a function of the relative blockade current are shown in Fig. 6 in the case of the interaction of the aerolysin nanopore with a high-purity (>98%) sample of 10 amino acids long arginine peptides (Fig. 6a) and of 10 amino acids long lysine peptides (Fig. 6c). Interestingly, the observed majority population clearly differs between the two different peptides, both in term of mean Ib/I0 value and in term of mean blockade duration. The mean Ib/I0 value is 0.236 +/− 0.003 in the case of arginine and 0.377 +/− 0.004 in the case of lysine, while the mean blockade duration decreases by almost two orders of magnitude from nearly 100 ms in the case of arginine to nearly 1 ms in the case of lysine (see also Supplementary Fig. 7). This reveals that our method enables the discrimination between arginine and lysine homopeptides of same length, even while arginine and lysine amino acids have the same electrical charge and very similar chemical structures.

Moreover, experiments probing the interaction of the aerolysin nanopore with a high-purity (>98%) sample of 10 amino acids long heteropeptides formed by a sequence of 5 lysine amino acids coupled to 5 arginine amino acids (Fig. 6b) reveal that this peptide can also be discriminated from 10 amino acids long arginine peptides and from 10 amino acids long lysine peptides (Fig. 6a–c and Supplementary Fig. 7), highlighting the high-sensitivity of our method. In the case of 10 amino acids long lysine/argine heteropeptides, the mean Ib/I0 value is 0.297 +/− 0.003 and the mean blockade duration is nearly 10 ms.

## Discussion

The high resolution of aerolysin pore for the highly-sensitive analysis of short peptides is most likely attributed to its geometry, to its highly charged pore wall and to the high affinity between aerolysin and peptides (see Supplementary Fig. 6). Nanopore-based single amino acid size-discrimination of several peptide lengths mixed in solution that differ by a single amino acid is subject to four conditions: i) the whole peptide must be entirely fitted en bloc inside the vicinity of the pore, ii) all amino acids must contribute individually to the current blockades, iii) a high size-sensitivity of the pore conductance change and iv) a high affinity between the pore and the peptide (the life time of the peptide/pore interaction must exceed 2 times the rise time of the bandwidth). In our experiments, observing different residual blockade currents in the case of peptides differing by only one amino acid indicates that each amino acid contributes individually to the current blockade, corresponding to a coil or to a compact conformation of the peptide in the nanopore and not to a rod-like conformation of the peptide. In the hypothesis of a rod-like conformation of the peptide, it is expected that current blockades would only weakly, if at all, depend on the peptide length. A coil or compact conformation of the peptide is not surprising in the high salt conditions of our experiments and considering the short persistence length of peptides (2 amino acids)69. Concerning peptides longer than 10 amino acids, experiments performed with 12 amino acids long arginine homopeptides (Supplementary Fig. 11) display a distinct preferential residual blockade current population corresponding to this peptide length, as observed for shorter peptides. Our perspectives are rather turned towards the analysis of short peptides, as short as a single amino acid, which is of particular importance towards peptide/protein sequencing. Furthermore, short peptides are widely used by cells for their communications in normal and pathologic conditions. These short peptides can be found either in the intracellular or extracellular compartments and some of them are secreted or released into extracellular vesicles, assuming functions both inside and outside the cell. Otherwise, for industrial applications, our proof-of-concept method could be extended to longer homopeptides of different chemical nature by using larger biological nanopores such as ClyA nanopore28. Under appropriate experimental conditions, our proof-of-concept method could be also extended to the size-discrimination and purity analysis of other short homopeptides and heteropeptides. A possible way out would be to use low38 or high pH conditions, where polypeptides might be uniformly charged, or not charged at all.

Ultra-fast protein recognition of peptide and protein with a nanopore is another challenge that already started. The first attempts focused on the detection and discrimination of native proteins either outside7,22,23,56,59 or inside21,29,35,38,39 the nanopore. Then an approach towards protein sequencing was to unfold the native protein and to try to read the polypeptide sequence into the nanopore during the translocation process. An interesting work demonstrated protein unfolding and translocation through a biological nanopore using an unfoldase molecular motor (ClpX variant)27. The ionic current trace during protein translocation did not allow to detect single amino-acid differences along the polypeptide sequence but showed electrical signal differences according to various protein domains with a maximum resolution of ≈663 amino acids70. Up to now, unfolded protein transport through a biological nanopore was too fast to read the protein sequence18,48. Recently a subnanometre diameter pore was manufactured, from a silicon nitride membrane, to indirectly read the primary sequence of an unfolded protein with a resolution of blocks of four amino acids, based on a theoretical excluded volume model71. So far, solid-state nanopores still suffer from a poorly reproducible manufacturing process and from insufficient wetting surface control. Here, in contrast, production of recombinant biological nanopores is standardized and highly reproducible. The electrical signal measurements to detect the size of short peptides with a single amino acid resolution and to discriminate between different chemical natures of homo-and heteropeptides are also highly reproducible. The high current blockade depth resolution -single amino acid- combined with the high temporal resolution –up to 10 ms/amino acid, i.e., a 3 orders of magnitude improvement as compared to previous published data18– of our system pave the road for the construction of a sequence library of all the different amino acids. As a first perspective, coupling the detection and identification of individual amino acids of different chemical nature either with the controlled translocation of an unfolded peptide/protein or with the sequential degradation of a peptide/protein near the nanopore entrance, would provide a new tool for peptide/protein sequencing. As a second perspective, combining the single-molecule sensitivity of the nanopore technology with a nanofluidic device guiding a single cell to the nanopore entrance would allow to quantify very small quantities of peptide biomarkers in a single cell. For instance, recently, Maglia and co-workers analyzed model peptide and protein biomarkers through FraC nanopores38. This might lead to important advances in the field of single-molecule proteomics. Our results and those of more recent works37,38,39,59,72 pave the road to analyse realistic samples, such as those from blood stream, serum, biopsies, or lysate extracts.

## Methods

### Synthesis of the aerolysin nanopore

The aerolysin nanopore is a pore-forming toxin from Aeromonas hydrophila which forms an heptameric β-barrel pore assembly in cell membranes73. The atomistic structure of aerolysin was recently defined74. It has an inner pore diameter of 1.0–1.7 nm and has a high number of charged residues (91 charges) pointing into the pore lumen74. Recombinant wild-type pro-aerolysin was synthetized using the following procedure. We transformed Escherichia coli BL21 strain with pET22b-proAL plasmid containing pro-aerolysin sequence. This plasmid allowed the induction of pro-aerolysin production with IPTG (1 mM final concentration) and the periplasmic localization of the recombinant protein. The periplasm was extracted with an osmotic shock and pro-aerolysin was furthermore purified by affinity chromatography using the C-terminal his-tag of the recombinant protein (His SpinTrap minicolumns, GE Healthcare Life Science). Pro-aerolysin binding to the Ni-Sepharose was made with 100 mM Tris-HCl pH 7.4 and 50 mM imidazole, and after three washing steps, elution was made with 100 mM Tris-HCl pH 7.4 and 500 mM imidazole. The recombinant pro-aerolysin purity was determined by SDS-polyacrylamide gel electrophoresis and Coomassie bleu staining, to 99 ± 1% (w/w). Concentration was calculated by absorbance at 280 nm. The propeptides of pro-aerolysin monomers were stored at 4 °C at a final concentration of 0.1 g/L.

### Nanopore recordings

A classical vertical lipid bilayer setup (Warner Instruments, Hamden, CT, USA or Elements, Cesena (FC), Italy) was used for all the experiments. The lipid bilayer was formed by painting and thinning a film of diphytanoyl-phosphocholine (Avanti Polar Lipids, Alabaster, AL, USA) dissolved in decane (10 mg/mL) over a 150 μm-wide aperture separating the cis and trans chambers. Both chambers were filled with 1 mL of 4 M KCl solution buffered with 5 mM HEPES and set to pH = 7.5. Two Ag/AgCl electrodes were used to apply a transmembrane voltage and to measure the transmembrane ionic current. The cis-chamber was at voltage ground.

In the case of the Warner Instruments setup, single-channel current recordings were performed using an Axopatch 200B patch-clamp amplifier (Molecular Devices, Sunnyvale, CA, USA) in the whole-cell mode with a CV-203BU headstage. The signal was filtered using an internal 4-pole Bessel filter at a cut-off frequency of 5 kHz. Data were acquired with a DigiData 1440A AD-converter (Molecular Devices) controlled by the Clampex 10.2 software (Molecular Devices). Current recordings at constant voltage were performed at a sampling rate of 250 kHz. Current versus voltage recordings were performed from −100 to +100 mV during 4 s and at a sampling rate of 5 kHz. A Peltier device controlled by a bipolar temperature controller (CL-100, Warner Instruments) allowed to set the temperature of the solution via a water circuit (LCS-1, Warner Instruments) around the chambers. The temperature was set to 20.0 ± 2 °C for all the experiments.

In the case of the Elements setup, single-channel current recordings were performed using an eONE amplifier (Elements, Cesena (FC), Italy). The signal was filtered at a cut-off frequency of 5 kHz. Data were acquired using the Elements Data Reader software (Elements, Cesena (FC), Italy), at a sampling rate of 200 kHz.

Recombinant wild-type pro-aerolysin was activated by trypsin digestion (0.6 μM trypsin final concentration) during 15 min at room temperature to eliminate the pro-peptide sequence. After the formation of the lipid bilayer, activated aerolysin was added to the cis-chamber at ≈1 nM final concentration. The insertion of a single nanopore in the lipid bilayer was identified as a step-jump of the transmembrane ionic current.

High-purity (>98%) arginine homopeptide samples (ProteoGenix, Schiltigheim, France) of different lengths (3, 5, 6, 7, 8, 9, 10, 12 amino acids) were used. Peptides were dissolved in aqueous 5 mM HEPES pH = 7.5 buffer. In the case of current recordings at constant voltage, peptides were added to the cis-chamber, either as a 1 μM equimolar mixture of peptides of different lengths or as a 1 μM solution of peptides of a given length. In the case of current versus voltage recordings, 10 amino acids long peptides were added either to the trans-chamber at a 6 μM concentration, or to the cis-chamber at a 1.2 μM concentration (Supplementary Fig. 1). In each case, after peptide addition, a ≈2 mL mixing of the volume of the chamber corresponding to the side of peptide addition was performed, and a ≈10 min waiting time was observed in order to ensure an homogeneous peptide concentration in the chamber.

The enzymatic degradation of arginine homopeptides was performed by adding trypsin at 25 nM final concentration to a 4.4 μM solution of 9 amino acids long peptides. Ten minutes current recordings were performed every 10 min after trypsin addition, during 3 h 30 min.

High-purity (>98%) lysine homopeptide samples (ProteoGenix, Schiltigheim, France) of 10 amino acids were used. Peptides were dissolved in aqueous 5 mM HEPES pH = 7.5 buffer. Peptides were added to the cis-chamber as a 213 μM solution of peptides. After peptide addition, a ≈2 mL mixing of the volume of the chamber corresponding to the side of peptide addition was performed, and a ≈10 min waiting time was observed in order to ensure an homogeneous peptide concentration in the chamber.

The enzymatic degradation of lysine homopeptides was performed by adding trypsin at 266 nM final concentration to a 213 μM solution of 10 amino acids long peptides. 5 min current recordings were performed every 5 min after trypsin addition, during 1 h 45 min.

High-purity (>98%) lysine/arginine heteropeptide samples (ProteoGenix, Schiltigheim, France) of 10 amino acids were used. Peptides were dissolved in aqueous 5 mM HEPES pH = 7.5 buffer. Peptides were added to the cis-chamber as a 1 μM solution of peptides. After peptide addition, a ≈2 mL mixing of the volume of the chamber corresponding to the side of peptide addition was performed, and a ≈10 min waiting time was observed in order to ensure an homogeneous peptide concentration in the chamber.

### Data treatment

The data treatment was performed using the Igor Pro 6.12A software (WaveMetrics, Portland, OR, USA) with homemade procedures. The treatment was based on a statistical analysis of a least 2000 blockade events for each current trace. A two-thresholds method was used to detect peptide-induced current blockades (Supplementary Fig. 8). A first threshold th1 = I0 − 4σ was used to define every possible blockade event, where I0 and σ are respectively the mean open pore current and its standard deviation obtained from a gaussian fit of the open pore current distribution. A possible blockade event begins when the pore current decreases below th1 and ends when the pore current increases above th1. This first threshold allows to eliminate the overwhelming part of the open pore current noisy fluctuations. If the mean blockade current value Ib during the event is inferior to a second threshold th2 = I0 − 5σ, then the event is considered as a peptide-induced blockade event.

Histograms of the relative blockade current Ib/I0 values were obtained using a bin width of 0.002. For each peptide length, the mean relative blockade current value $$\left\langle {I_{\mathrm{b}}{\mathrm{/}}I_0} \right\rangle _i$$ (i = 5, 6, 7, 8, 9, or 10 amino acids long peptides) corresponds to the mean value of a gaussian fit of the Ib/I0 peak corresponding to this peptide length. The error bar of each $$\left\langle {I_{\mathrm{b}}{\mathrm{/}}I_0} \right\rangle _i$$ corresponds to the standard deviation σ i of the gaussian fit of the peak. The current blockades corresponding to each peptide length were identified as the current blockades with Ib/I0 values comprised between $$\left\langle {I_{\mathrm{b}}{\mathrm{/}}I_0} \right\rangle _i - 3\sigma _i$$ and $$\left\langle {I_{\mathrm{b}}{\mathrm{/}}I_0} \right\rangle _i + 3\sigma _i$$. From the current blockades corresponding to each peptide length, the mean blockade duration and frequency were calculated for each peptide length. The mean blockade duration is the mean value of: the calculated mean value of blockade durations, the mean blockade duration obtained from a single exponential fit of the blockade durations distribution and the mean blockade duration obtained from a single exponential fit of the cumulative blockade durations distribution (Supplementary Fig. 9). The error bar of the mean blockade duration is the maximal value among: the sampling time interval of the acquisition system (4 μs at 250 kHz sampling rate), the dispersion of the mean blockade durations obtained from the 3 methods above and the standard deviation of the fit coefficients from the data analysis software. The same procedure applied to the inter-blockade durations provided the mean inter-blockade duration, and thus the mean blockade frequency, and its error bar for each peptide length.

The results were reproduced with a high fidelity from at least 4 independent experiments (Supplementary Fig. 4). The results presented in this article correspond to individual experiments.

### Data availability

The authors declare that the data supporting the findings of this study are available within the article and its Supplementary Information files or from the corresponding authors upon reasonable request.

Publisher's note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

## References

1. 1.

Lee, K.-E., Heo, J.-E., Kim, J.-M. & Hwang, C.-S. N-terminal acetylation-targeted N-end rule proteolytic system: the Ac/N-end rule pathway. Mol. Cells 39, 169–178 (2016).

2. 2.

Shao, S. et al. Minimal sample requirement for highly multiplexed protein quantification in cell lines and tissues by PCT-SWATH mass spectrometry. Proteomics 15, 3711–3721 (2015).

3. 3.

Schubert, O. T., Röst, H. L., Collins, B. C., Rosenberger, G. & Aebersold, R. Quantitative proteomics: challenges and opportunities in basic and applied research. Nat. Protoc. 12, 1289–1294 (2017).

4. 4.

Swietlow, A. & Lax, R. Quality control in peptide manufacturing: specifications for GMP peptides. Chim. Oggi. 22, 22–24 (2004).

5. 5.

Kasianowicz, J. J., Brandin, E., Branton, D. & Deamer, D. W. Characterization of individual polynucleotide molecules using a membrane channel. Proc. Natl Acad. Sci. USA 93, 13770–13773 (1996).

6. 6.

Venkatesan, B. M. & Bashir, R. Nanopore sensors for nucleic acid analysis. Nat. Nanotechnol. 6, 615–624 (2011).

7. 7.

Movileanu, L., Howorka, S., Braha, O. & Bayley, H. Detecting protein analytes that modulate transmembrane movement of a polymer chain within a single protein pore. Nat. Biotechnol. 18, 1091–1095 (2000).

8. 8.

Sutherland, T. C. et al. Structure of peptides investigated by nanopore analysis. Nano. Lett. 4, 1273–1277 (2004).

9. 9.

Movileanu, L., Schmittschmitt, J. P., Scholtz, J. M. & Bayley, H. Interactions of peptides with a protein pore. Biophys. J. 89, 1030–1045 (2005).

10. 10.

Stefureac, R., Long, Y.-T., Kraatz, H.-B., Howard, P. & Lee, J. S. Transport of alpha-helical peptides through alpha-hemolysin and aerolysin pores. Biochemistry 45, 9172–9179 (2006).

11. 11.

Cheley, S., Xie, H. & Bayley, H. A genetically encoded pore for the stochastic detection of a protein kinase. Chembiochem 7, 1923–1927 (2006).

12. 12.

Oukhaled, G. et al. Unfolding of proteins and long transient conformations detected by single nanopore recording. Phys. Rev. Lett. 98, 158101 (2007).

13. 13.

Wolfe, A. J., Mohammad, M. M., Cheley, S., Bayley, H. & Movileanu, L. Catalyzing the translocation of polypeptides through attractive interactions. J. Am. Chem. Soc. 129, 14034–14041 (2007).

14. 14.

Goodrich, C. P. et al. Single-molecule electrophoresis of beta-hairpin peptides by electrical recordings and Langevin dynamics simulations. J. Phys. Chem. B. 111, 3332–3335 (2007).

15. 15.

Zhao, Q., de Zoysa, R. S. S., Wang, D., Jayawardhana, D. A. & Guan, X. Real-time monitoring of peptide cleavage using a nanopore probe. J. Am. Chem. Soc. 131, 6324–6325 (2009).

16. 16.

Bikwemu, R., Wolfe, A. J., Xing, X. & Movileanu, L. Facilitated translocation of polypeptides through a single nanopore. J. Phys. Condens. Matter 22, 454117 (2010).

17. 17.

Kukwikila, M. & Howorka, S. Electrically sensing protease activity with nanopores. J. Phys. Condens. Matter 22, 454103 (2010).

18. 18.

Pastoriza-Gallego, M. et al. Dynamics of unfolded protein transport through an aerolysin pore. J. Am. Chem. Soc. 133, 2923–2931 (2011).

19. 19.

Mahendran, K. R., Romero-Ruiz, M., Schlösinger, A., Winterhalter, M. & Nussberger, S. Protein translocation through Tom40: kinetics of peptide release. Biophys. J. 102, 39–47 (2012).

20. 20.

Singh, P. R. et al. Pulling peptides across nanochannels: resolving peptide binding and translocation through the hetero-oligomeric channel from Nocardia farcinica. ACS Nano 6, 10699–10707 (2012).

21. 21.

Soskine, M. et al. An engineered ClyA nanopore detects folded target proteins by selective external association and pore entry. Nano. Lett. 12, 4895–4900 (2012).

22. 22.

Rotem, D., Jayasinghe, L., Salichou, M. & Bayley, H. Protein detection by nanopores equipped with aptamers. J. Am. Chem. Soc. 134, 2781–2787 (2012).

23. 23.

Mohammad, M. M. et al. Engineering a rigid protein tunnel for biomolecular detection. J. Am. Chem. Soc. 134, 9521–9531 (2012).

24. 24.

Payet, L. et al. Thermal unfolding of proteins probed at the single molecule level using nanopores. Anal. Chem. 84, 4071–4076 (2012).

25. 25.

Lamichhane, U. et al. Peptide translocation through the mesoscopic channel: Binding kinetics at the single molecule level. Eur. Biophys. J. 42, 363–369 (2013).

26. 26.

Mahendran, K. R., Lamichhane, U., Romero-Ruiz, M., Nussberger, S. & Winterhalter, M. Polypeptide translocation through the mitochondrial TOM channel: temperature-dependent rates at the single-molecule level. J. Phys. Chem. Lett. 4, 78–82 (2013).

27. 27.

Nivala, J., Marks, D. B. & Akeson, M. Unfoldase-mediated protein translocation through an α-hemolysin nanopore. Nat. Biotechnol. 31, 247–250 (2013).

28. 28.

Soskine, M., Biesemans, A., De Maeyer, M. & Maglia, G. Tuning the size and properties of clya nanopores assisted by directed evolution. J. Am. Chem. Soc. 135, 13456–13463 (2013).

29. 29.

Van Meervelt, V., Soskine, M. & Maglia, G. Detection of two isomeric binding configurations in a protein-aptamer complex with a biological nanopore. ACS Nano 8, 12826–12835 (2014).

30. 30.

Mereuta, L. et al. Slowing down single-molecule trafficking through a protein nanopore reveals intermediates for peptide translocation. Sci. Rep. 4, 3885 (2014).

31. 31.

Pastoriza-Gallego, M. et al. Evidence of unfolded protein translocation through a protein nanopore. ACS Nano 8, 11350–11360 (2014).

32. 32.

Fahie, M., Chisholm, C. & Chen, M. Resolved single-molecule detection of individual species within a mixture of anti-biotin antibodies using an engineered monomeric nanopore. ACS Nano 9, 1089–1098 (2015).

33. 33.

Biesemans, A., Soskine, M. & Maglia, G. A protein rotaxane controls the translocation of proteins across a ClyA nanopore. Nano. Lett. 15, 6076–6081 (2015).

34. 34.

Soskine, M., Biesemans, A. & Maglia, G. Single-molecule analyte recognition with ClyA nanopores equipped with internal protein adaptors. J. Am. Chem. Soc. 137, 5793–5797 (2015).

35. 35.

Ji, Z. et al. Fingerprinting of peptides with a large channel of bacteriophage Phi29 DNA packaging motor. Small 12, 4572–4578 (2016).

36. 36.

Hu, Y.-X. et al. Single molecule study of initial structural features on the amyloidosis process. Chem. Commun. 52, 5542–5545 (2016).

37. 37.

Chavis, A. E. et al. Single molecule nanopore spectrometry for peptide detection. ACS Sensors 2, (1319–1328 (2017).

38. 38.

Huang, G., Willems, K., Soskine, M., Wloka, C. & Maglia, G. Electro-osmotic capture and ionic discrimination of peptide and protein biomarkers with FraC nanopores. Nat. Commun. 8, 935 (2017).

39. 39.

Van Meervelt, V. et al. Real-time conformational changes and controlled orientation of native proteins inside a protein nanoreactor. J. Am. Chem. Soc. 139, 18640–18646 (2017).

40. 40.

Han, A. et al. Sensing protein molecules using nanofabricated pores. Appl. Phys. Lett. 88, 093901 (2006).

41. 41.

Uram, J. D., Ke, K., Hunt, A. J. & Mayer, M. Label-free affinity assays by rapid detection of immune complexes in submicrometer pores. Angew. Chem., Int. Ed. 45, 2281–2285 (2006).

42. 42.

Sexton, L. T. et al. Resistive-pulse studies of proteins and protein/antibody complexes using a conical nanotube sensor. J. Am. Chem. Soc. 129, 13144–13152 (2007).

43. 43.

Fologea, D., Ledden, B., McNabb, D. S. & Li, J. Electrical characterization of protein molecules by a solid-state nanopore. Appl. Phys. Lett. 91, 053901 (2007).

44. 44.

Talaga, D. S. & Li, J. Single-molecule protein unfolding in solid state nanopores. J. Am. Chem. Soc. 131, 9287–9297 (2009).

45. 45.

Firnkes, M., Pedone, D., Knezevic, J., Döblinger, M. & Rant, U. Electrically facilitated translocations of proteins through silicon nitride nanopores: conjoint and competitive action of diffusion, electrophoresis, and electroosmosis. Nano Lett. 10, 2162–2167 (2010).

46. 46.

Oukhaled, A. et al. Dynamics of completely unfolded and native proteins through solid-state nanopores as a function of electric driving force. ACS Nano 5, 3628–3638 (2011).

47. 47.

Yusko, E. C. et al. Controlling protein translocation through nanopores with bio-inspired fluid walls. Nat. Nanotechnol. 6, 253–260 (2011).

48. 48.

Cressiot, B. et al. Protein transport through a narrow solid-state nanopore at high voltage: experiments and theory. ACS Nano 6, 6236–6243 (2012).

49. 49.

Japrung, D. et al. Single-molecule studies of intrinsically disordered proteins using solid-state nanopores. Anal. Chem. 85, 2449–2456 (2013).

50. 50.

Freedman, K. J., Haq, S. R., Edel, J. B., Jemth, P. & Kim, M. J. Single molecule unfolding and stretching of protein domains inside a solid-state nanopore by electric field. Sci. Rep. 3, 1638 (2013).

51. 51.

Li, W. et al. Single protein molecule detection by glass nanopores. ACS Nano 7, 4129–4134 (2013).

52. 52.

Rosen, C. B., Rodriguez-Larrea, D. & Bayley, H. Single-molecule site-specific detection of protein phosphorylation with a nanopore. Nat. Biotechnol. 32, 179–181 (2014).

53. 53.

Steinbock, L. J. et al. Probing the size of proteins with glass nanopores. Nanoscale 6, 14380–14387 (2014).

54. 54.

Nir, I., Huttner, D. & Meller, A. Direct sensing and discrimination among ubiquitin and ubiquitin chains using solid-state nanopores. Biophys. J. 108, 2340–2349 (2015).

55. 55.

Balme, S. et al. Influence of adsorption on proteins and amyloids detection by silicon nitride nanopore. Langmuir 32, 8916–8925 (2016).

56. 56.

Bell, N. A. W. & Keyser, U. F. Digitally encoded DNA nanostructures for multiplexed, single-molecule protein sensing with nanopores. Nat. Nanotechnol. 11, 645–651 (2016).

57. 57.

Waduge, P. et al. Nanopore-based measurements of protein size, fluctuations, and conformational changes. ACS Nano 11, 5706–5716 (2017).

58. 58.

Wloka, C. et al. Label-free and real-time detection of protein ubiquitination with a biological nanopore. ACS Nano 11, 4387–4394 (2017).

59. 59.

Sze, J. Y. Y., Ivanov, A. P., Cass, A. E. G. & Edel, J. B. Single molecule multiplexed nanopore protein screening in human serum using aptamer modified DNA carriers. Nat. Commun. 8, 1552 (2017).

60. 60.

Xie, H., Braha, O., Gu, L.-Q., Cheley, S. & Bayley, H. Single-molecule observation of the catalytic subunit of camp-dependent protein kinase binding to an inhibitor peptide. Chem. Biol. 12, 109–120 (2005).

61. 61.

Kukwikila, M. & Howorka, S. Nanopore-based electrical and label-free sensing of enzyme activity in blood serum. Anal. Chem. 87, 9149–9154 (2015).

62. 62.

Rodriguez-Larrea, D. & Bayley, H. Multistep protein unfolding during nanopore translocation. Nat. Nanotechnol. 8, 288–295 (2013).

63. 63.

Yusko, E. C. et al. Real-time shape approximation and fingerprinting of single proteins using a nanopore. Nat. Nanotechnol. 12, 360–367 (2016).

64. 64.

Vercoutere, W. et al. Rapid discrimination among individual DNA hairpin molecules at single-nucleotide resolution using an ion channel. Nat. Biotechnol. 19, 248–252 (2001).

65. 65.

Robertson, J. W. F. et al. Single-molecule mass spectrometry in solution using a solitary nanopore. Proc. Natl Acad. Sci. USA 104, 8207–8211 (2007).

66. 66.

Baaken, G. et al. High-resolution size-discrimination of single nonionic synthetic polymers with a highly charged biological nanopore. ACS Nano 9, 6443–6449 (2015).

67. 67.

Piguet, F. et al. High temperature extends the range of size discrimination of nonionic polymers by a biological nanopore. Sci. Rep. 6, 38675 (2016).

68. 68.

Cao, C. et al. Discrimination of oligonucleotides of different lengths with a wild-type aerolysin nanopore. Nat. Nanotechnol. 11, 713–718 (2016).

69. 69.

Lairez, D., Pauthe, E. & Pelta, J. Refolding of a high molecular weight protein: salt effect on collapse. Biophys. J. 84, 3904–3916 (2003).

70. 70.

Nivala, J., Mulroney, L., Li, G., Schreiber, J. & Akeson, M. Discrimination among protein variants using an unfoldase-coupled nanopore. ACS Nano 8, 12365–12375 (2014).

71. 71.

Kennedy, E., Dong, Z., Tennant, C. & Timp, G. Reading the primary structure of a protein with 0.07 nm3 resolution using a subnanometre-diameter pore. Nat. Nanotechnol. 11, 968–976 (2016).

72. 72.

Xi, D., Li, Z., Liu, L., Ai, S. & Zhang, S. Ultrasensitive detection of cancer cells combining enzymatic signal amplification with an aerolysin nanopore. Anal. Chem. 90, 1029–1034 (2018).

73. 73.

Parker, M. W. et al. Structure of the Aeromonas toxin proaerolysin in its water-soluble and membrane-channel states. Nature 367, 292–295 (1994).

74. 74.

Degiacomi, M. T. et al. Molecular assembly of the aerolysin pore reveals a swirling membrane-insertion mechanism. Nat. Chem. Biol. 9, 623–629 (2013).

## Acknowledgements

This work was supported by grants from the Agence Nationale de la Recherche (ANR-17-CE09-0032-01 to F.P., M.P.-G., and A.O.; ANR-17-CE09-0044-02 to P.M., M.P.-G., A.O., and J.P.; ANR12-NANO-0012-03, to J.P. and A.O.), and from the Région Ile-de-France in the framework of DIM Nano-K (n° 094251, to A.O.). This work was also supported by the Direction Générale de l’Armement (the French Defense Procurement Agency, n° 2017 60 0042 to H.O. and A.O.) and by the Région Ile-de-France in the framework of DIM ResPore (n° 2017-05 to H.O. and A.O.). We thank Excilone company (France) which provided financial support to H.O. We thank F. Gisou van der Goot (Ecole Polytechnique Fédérale de Lausanne, Switzerland) which generously provided the pET22b-proAL plasmid containing the pro-aerolysin sequence. We thank Guillaume Lamour (Université d’Evry-Val-d’Essonne) for careful reading and for improving the use of English in the manuscript.

## Author information

J.P. and A.O. conceived the project. A.O. managed the project, designed the nanopore platform, and conceived the experiments. M.P.-G. produced the recombinant wild-type pro-aerolysin. F.P., H.O., and A.O. performed the experiments and analyzed the data. All authors participated to the interpretation of the data. F.P., H.O., and A.O. prepared the manuscript figures. F.P, H.O., P.M., J.P., and A.O. wrote the manuscript. All authors reviewed the manuscript.

### Competing interests

F.P., H.O., P.M., J.P., and A.O. are named as inventors in a patent application (Institut National de la Propriété Industrielle patent application number 16/01007, European patent application number EP17177951.5, Patent Cooperation Treaty application number 17/000129). M.P.-G. declares no competing interests.

Correspondence to Fabien Piguet or Juan Pelta or Abdelghani Oukhaled.

## Rights and permissions

### DOI

https://doi.org/10.1038/s41467-018-03418-2

• ### FraC nanopores with adjustable diameter identify the mass of opposite-charge peptides with 44 dalton resolution

• Gang Huang
• , Arnout Voet
•  & Giovanni Maglia

Nature Communications (2019)

• ### Thermostable virus portal proteins as reprogrammable adapters for solid-state nanopore sensors

• Benjamin Cressiot
• , Sandra J. Greive
• , Mehrnaz Mojtabavi
• , Alfred A. Antson
•  & Meni Wanunu

Nature Communications (2018)

• ### Mapping the sensing spots of aerolysin for single oligonucleotides analysis

• Chan Cao
• , Meng-Yin Li
• , Nuria Cirauqui
• , Ya-Qian Wang
• , Matteo Dal Peraro
• , He Tian
•  & Yi-Tao Long

Nature Communications (2018)