Charge-based fingerprinting of unlabeled full-length proteins using an aerolysin nanopore

Charge-based fingerprinting of unlabeled full-length proteins using an aerolysin nanopore | bioRxiv /* */ /* */ <!-- <!-- /*! * yepnope1.5.4 * (c) WTFPL, GPLv2 */ (function(a,b,c){function d(a){return"[object Function]"==o.call(a)}function e(a){return"string"==typeof a}function f(){}function g(a){return!a||"loaded"==a||"complete"==a||"uninitialized"==a}function h(){var a=p.shift();q=1,a?a.t?m(function(){("c"==a.t?B.injectCss:B.injectJs)(a.s,0,a.a,a.x,a.e,1)},0):(a(),h()):q=0}function i(a,c,d,e,f,i,j){function k(b){if(!o&&g(l.readyState)&&(u.r=o=1,!q&&h(),l.onload=l.onreadystatechange=null,b)){"img"!=a&&m(function(){t.removeChild(l)},50);for(var d in y[c])y[c].hasOwnProperty(d)&&y[c][d].onload()}}var j=j||B.errorTimeout,l=b.createElement(a),o=0,r=0,u={t:d,s:c,e:f,a:i,x:j};1===y[c]&&(r=1,y[c]=[]),"object"==a?l.data=c:(l.src=c,l.type=a),l.width=l.height="0",l.onerror=l.onload=l.onreadystatechange=function(){k.call(this,r)},p.splice(e,0,u),"img"!=a&&(r||2===y[c]?(t.insertBefore(l,s?null:n),m(k,j)):y[c].push(l))}function j(a,b,c,d,f){return q=0,b=b||"j",e(a)?i("c"==b?v:u,a,b,this.i++,c,d,f):(p.splice(this.i++,0,a),1==p.length&&h()),this}function k(){var a=B;return a.loader={load:j,i:0},a}var l=b.documentElement,m=a.setTimeout,n=b.getElementsByTagName("script")[0],o={}.toString,p=[],q=0,r="MozAppearance"in l.style,s=r&&!!b.createRange().compareNode,t=s?l:n.parentNode,l=a.opera&&"[object Opera]"==o.call(a.opera),l=!!b.attachEvent&&!l,u=r?"object":l?"script":"img",v=l?"script":u,w=Array.isArray||function(a){return"[object Array]"==o.call(a)},x=[],y={},z={timeout:function(a,b){return b.length&&(a.timeout=b[0]),a}},A,B;B=function(a){function b(a){var a=a.split("!"),b=x.length,c=a.pop(),d=a.length,c={url:c,origUrl:c,prefixes:a},e,f,g;for(f=0;f<d;f++)g=a[f].split("="),(e=z[g.shift()])&&(c=e(c,g));for(f=0;f<b;f++)c=x[f](c);return c}function g(a,e,f,g,h){var i=b(a),j=i.autoCallback;i.url.split(".").pop().split("?").shift(),i.bypass||(e&&(e=d(e)?e:e[a]||e[g]||e[a.split("/").pop().split("?")[0]]),i.instead?i.instead(a,e,f,g,h):(y[i.url]?i.noexec=!0:y[i.url]=1,f.load(i.url,i.forceCSS||!i.forceJS&&"css"==i.url.split(".").pop().split("?").shift()?"c":c,i.noexec,i.attrs,i.timeout),(d(e)||d(j))&&f.load(function(){k(),e&&e(i.origUrl,h,g),j&&j(i.origUrl,h,g),y[i.url]=2})))}function h(a,b){function c(a,c){if(a){if(e(a))c||(j=function(){var a=[].slice.call(arguments);k.apply(this,a),l()}),g(a,j,b,0,h);else if(Object(a)===a)for(n in m=function(){var b=0,c;for(c in a)a.hasOwnProperty(c)&&b++;return b}(),a)a.hasOwnProperty(n)&&(!c&&!--m&&(d(j)?j=function(){var a=[].slice.call(arguments);k.apply(this,a),l()}:j[n]=function(a){return function(){var b=[].slice.call(arguments);a&&a.apply(this,b),l()}}(k[n])),g(a[n],j,b,n,h))}else!c&&l()}var h=!!a.test,i=a.load||a.both,j=a.callback||f,k=j,l=a.complete||f,m,n;c(h?a.yep:a.nope,!!i),i&&c(i)}var i,j,l=this.yepnope.loader;if(e(a))g(a,0,l,0);else if(w(a))for(i=0;i (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0];var j=d.createElement(s);var dl=l!='dataLayer'?'&l='+l:'';j.src='//www.googletagmanager.com/gtm.js?id='+i+dl;j.type='text/javascript';j.async=true;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-M677548'); Skip to main content Home About Submit ALERTS / RSS Search for this keyword Advanced Search New Results Charge-based fingerprinting of unlabeled full-length proteins using an aerolysin nanopore View ORCID Profile Verena Rukes , Evita Norkute , Georges Barnikol , Jingze Duan , Jiajie Gao , View ORCID Profile Chan Cao doi: https://doi.org/10.1101/2025.01.13.632743 Verena Rukes 1 Department of Inorganic and Analytical Chemistry, Chemistry and Biochemistry, University of Geneva ; 1211 Geneva, Switzerland 2 Institute of Bioengineering, School of Life Science, École Polytechnique Fédérale de Lausanne (EPFL) ; 1015 Lausanne, Switzerland Find this author on Google Scholar Find this author on PubMed Search for this author on this site ORCID record for Verena Rukes Evita Norkute 1 Department of Inorganic and Analytical Chemistry, Chemistry and Biochemistry, University of Geneva ; 1211 Geneva, Switzerland Find this author on Google Scholar Find this author on PubMed Search for this author on this site Georges Barnikol 3 Institute of Chemical Sciences and Engineering, School of Basic Sciences, EPFL ; 1015 Lausanne, Switzerland Find this author on Google Scholar Find this author on PubMed Search for this author on this site Jingze Duan 1 Department of Inorganic and Analytical Chemistry, Chemistry and Biochemistry, University of Geneva ; 1211 Geneva, Switzerland 4 Academy for Advanced Interdisciplinary Studies, Peking University ; 100871 Beijing, China Find this author on Google Scholar Find this author on PubMed Search for this author on this site Jiajie Gao 1 Department of Inorganic and Analytical Chemistry, Chemistry and Biochemistry, University of Geneva ; 1211 Geneva, Switzerland Find this author on Google Scholar Find this author on PubMed Search for this author on this site Chan Cao 1 Department of Inorganic and Analytical Chemistry, Chemistry and Biochemistry, University of Geneva ; 1211 Geneva, Switzerland Find this author on Google Scholar Find this author on PubMed Search for this author on this site ORCID record for Chan Cao For correspondence: chan.cao{at}unige.ch Abstract Full Text Info/History Metrics Supplementary material Preview PDF Abstract Proteins play essential roles in cellular processes and are involved in numerous diseases, driving the need for efficient proteoform identification. While nanopore technology was highly successful for DNA sequencing, this approach has not yet delivered its full potential in protein identification. Here, we demonstrate the capabilities of an aerolysin nanopore for direct identification of a range of unlabeled full-length proteins. Combining low pH and guanidinium chloride, we generate a strong electroosmotic flow that enables an efficient capture and translocation, resulting in distinct fingerprints from single-protein translocations. Using machine learning classifier, we achieve 80% accuracy in distinguishing seven related proteins with 38%–70% pairwise sequence identities. Differences in fingerprints largely reflect the distribution of positive charges in the protein sequences, providing a rational basis for the observed sensitivity. With further development, fingerprint-predictions could allow to infer de novo protein sequence information from single-molecule data, offering a powerful tool for proteomics. Introduction Proteins are the most abundant macromolecules in living systems and fulfill a plethora of functions such as providing cell structure, and signaling pathways, or driving metabolic processes 1 , 2 . They are highly diverse, accounting for millions of human proteoforms. Many of these are the product of post transcriptional alterations in the cell that dictate protein function and cannot be accessed through the ∼20,000 protein-encoding genes 3 – 5 . This is why tools that directly access the protein level have great significance e.g. in diagnostics. Currently, protein analysis is mostly based on mass spectrometry (MS), which despite much optimization 6 is still accompanied by drawbacks such as high costs, limited dynamic range, and detection limit as well as a limited reading length 7 – 9 . This has resulted in a rising interest in the development of single-molecule techniques to expand the repertoire of protein-analysis tools. Recent advancements have highlighted nanopores as promising tools for protein analysis, using voltage-driven ionic currents through a nanometer-scale pore 10 , 11 . As analytes interact with the pore, the ionic currents are modulated 12 , which is commercialized for DNA sequencing 13 . Notably, nanopores can measure single-molecule interactions and are compatible with mixture samples. To tailor the method to protein analysis, several approaches are being demonstrated, such as using DNA 14 – 16 or protein 17 , 18 motor enzymes to slowly thread peptides/proteins though nanopores. Another strategy is to fragment proteins and fingerprint the peptide spectra similar to bottom-up MS 19 – 21 . However, these require sample processing steps like chemical linkage or protein digest, that will reduce yield and introduce uncertainty to the analysis. A direct label-free approach can provide faster detection, specifically of low abundant species, which is desired in protein analysis. Conceptually, this can be achieved by directly threading the full-length proteins through the pore and identifying each protein based on prior knowledge of its current readout 22 . Important groundwork towards this goal has demonstrated that chemical denaturants can be used during nanopore measurements to weaken an analytes’ structure 23 , which is necessary to access sequence specific information. Since proteins are heterogeneously charged, it is not possible to rely solely on the electrophoretic force (EPF) as a driving force. This has been overcome by taking advantage of the Electroosmotic Flow (EOF), which can attract and transport analytes through nanopores independently of their charge. The EOF relies on the unequal transport of ions species of biological pores (illustrated in Fig. 1A ), which is referred to as ion selectivity and is caused by the charge of the pore lumen. Using the EOF, proteins and peptides have been successfully translocated through nanopores 24 – 26 . Selectivity can be induced and tuned through mutations 25 , pH 27 , or the electrolyte composition. For instance, guanidinium chloride (GdmCl) has been shown to universally induce EOF in nanopores 28 . Download figure Open in new tab Figure 1. Enhancing ion selectivity by combining pH and GdmCl. (A) Illustration of anion selectivity in aerolysin. (B) Permeability ratios based on reversal potential measurements with aerolysin K238A in KCl or GdmCl at pH 7.5 and pH 4.0. Equal permeability is indicated with a dashed line (P - /P + = 1). In each condition P - /P + was averaged over > 4 independent measurements and standard deviation s are indicated. MD results for GdmCl at pH 7.5 (C) and pH 4.0 (D) are shown together with an illustration of the pore structure. Residues carrying negative and positive charge are marked in red and blue respectively. Bar plots illustrate the duration of Gdm + binding to the pore surface normalized by the simulation time. Full-length protein identification was recently demonstrated using alpha hemolysin (α-HL) 22 or homologous pores (CytK) 25 . However, aerolysin is known for its high resolution when distinguishing analytes in free translocation 29 , 30 . Specifically, the aerolysin mutant K238A has been used for identification of site specific PTMs in peptides and direct decoding of data-storage polymers 31 , 32 . Due to its narrow cylindrical shape, aerolysin has been described as an ideal candidate for linear protein transport 33 . Additionally, single-file translocations of an unfolded protein has been reported with aerolysin 34 . However, to our knowledge, there has been no experimental and analytical investigation focused on establishing this nanopore for identification of full-length proteins. In this work, we demonstrate the optimization of EOF and unfolding conditions in aerolysin for efficient translocation of full-length proteins. In an optimized condition of 3 M GdmCl at pH 4.0, we measured 7 proteins from the same Turandot family of Drosophila Melanogaster to explore the sensitivity of the system in an extreme case. At an optimal potential of 170 mV, machine learning classifiers were able to discriminate between these 7 analytes with 80% accuracy. We further investigated the dependence of the current output on the sequence of the analyte. A simplified biophysical model along with the experiments of control analytes demonstrated that patterns in the current readout can be attributed to the distribution of positive charges along the protein analyte sequences. Current-sequence dependency has not previously been explored for freely translocating unfolded proteins. However, establishing such rational connection is the first step to develop an accurate model, which could significantly enhance this approach and potentially enable de novo protein identification. Results Enhancing EOF in the aerolysin system The central importance of the EOF for protein capture in nanopore systems is established. We thus characterized the impact of changes in the measurement condition to optimize the EOF in aerolysin K238A. While the structural integrity of the aerolysin pore is maintained up to 3 M GdmCl 35 , it is known that the EOF in nanopores is affected by both Gdm + 22 , 28 and changes in pH 27 . Based on reversal potential measurements we calculated the ratio of permeability ( P - /P + ) between anions and cations through the pore ( Fig. 1B and SI Fig. S1). This ratio is commonly used to determine the direction and relative magnitude of ion selectivity in biological nanopore systems 25 , 28 , 36 – 38 . Aerolysin K238A is slightly cation selective at neutral pH in KCl ( P - /P + = 0.73 ± 0.02). By changing the pH to 4.0 or changing the electrolyte to GdmCl, we observed a strong anion selectivity, resulting in P - /P + = 2.24 ± 0.10 and P - /P + = 2.31 ± 0.37, respectively. When combining both conditions (GdmCl at pH 4.0), an even larger selectivity is reached with P - /P + = 5.87 ± 1.11 in aerolysin K238A. Thus, we establish the combination of low pH and GdmCl ions as a fast and simple strategy to turn a slightly cation selective pore system into a strongly anion selective one. To investigate how or whether the binding of Gdm + is impacted by the change in pH, we have performed all atoms molecular dynamics (MD) simulations with aerolysin K238A in KCl and GdmCl both at pH 7.5 and pH 4.0. Interaction spots of the ions with the pore were defined, by considering ions that enter the pore and monitoring the time that each ion type spends in close proximity along the surface of the pore lumen, similar as previously reported 28 . We then normalize by the duration of the trajectory (see details in Methods). Gdm + showed specific interaction spots along the pore at both pH. At pH 7.5, Gdm + lingered at residues Asp216, Asp222, and (most prominently) Glu258, resulting in distinct peaks seen in Fig. 1C . At pH 4.0, the peaks at residues Asp216 and Glu258 largely disappear. While Asp222 remained a site of interaction, the residence of Gdm + at Glu254 increased ( Fig. 1D ). At either pH such peaks did not occur for Cl - in the GdmCl condition or for the ions in a KCl condition (SI Fig. S2). Overall, MD results – in agreement with experiments – showed that Gdm + ions can bind the pore lumen both at pH 7.5, and at pH 4.0. At both pH, the Gdm + interaction spots align well with negative charges in the lumen, showing the importance of electrostatic interactions for the binding. The combination of low pH and GdmCl results in more than double the P - /P + without noticeable negative interference, therefore providing a simple strategy to produce extreme ion selectivity and EOF. Optimizing the measurement condition for protein translocation We used Turandot A (TotA) from Drosophila Melanogaster to optimize the buffer condition for full length protein analysis with aerolysin K238A. At pH 7.5, TotA has a net charge of - 7.8 and is driven toward the pore by both the EPF and the EOF, while at pH 4.0 the protein’s net charge is + 8.0 and therefore the EOF is the only attracting force, opposing the EPF ( Fig. 2A ). The TotA protein has 115 amino acids (aa) and its native structure is composed of four α-helices 39 ( Fig. 2B ). Notably, changing the pH from 7.5 to 4.0 does not impact on the Circular Dichroism (CD) spectrum of TotA, indicating no change in its structure ( Fig. 2B ). Additionally, CD showed that TotA is not fully unfolded at 2 M GdmCl and largely unfolded at 3 M GdmCl. Therefore, by measuring the protein at 2 M GdmCl and 3 M GdmCl both at pH 7.5 and pH 4.0, we investigate the importance of residual structure, and the EPF-EOF interplay in aerolysin measurements of a full-length protein. TotA interactions with the aerolysin K238A pore were measured under a voltage range from 100 mV to 210 mV applied to the trans compartment. In all conditions and at all voltages, the analyte was added to the cis compartment, which caused deep current blockages with less than 15% mean residual current and dwell times between 3 and 30 ms ( Fig. 2C-E , SI Fig. S3). Download figure Open in new tab Figure 2. Optimizing conditions for protein translocations. (A) Illustration of the interplay of EOF and EPF in GdmCl at pH 7.5 (left) and pH 4.0 (right). (B) Results of CD measurements are averaged over 3 repeated measurements; the standard deviations are indicated. The TotA structure is shown (PDB: 8PBV). (C) 10 s of raw trace for each condition filtered to 1 kHz for illustration. (D) Example events filtered to 10 kHz for illustration. (E) Dwell time versus current scatterplots for individual events and each condition recorded at 120 mV. (F) Frequency of events calculated by fitting the interevent time of individual pores. Averaged over 8, 6, 5, and 4 recordings from left to right respectively weighted by the number of events in each recording, standard deviations are indicated. (G) The distribution of log dwell times for each condition at each voltage were fitted using a single exponentially modified gaussian distribution. The peak and the halfwidth of the fitted distributions are shown. We found that generating a strong EOF even against the EPF (pH 4.0) is the more efficient strategy to capture and translocate the protein analyte, compared to a combination of EPF and lower EOF (pH 7.5). As shown in Fig. 2F , at 100 mV shifting the pH from 7.5 to 4.0 increased capture at both 2 M and 3 M GdmCl, confirming that there is a stronger driving force toward the pore at low pH, even against the EPF. More strikingly, at pH 7.5, the dwell times did not decrease monotonously with increasing voltages ( Fig. 2G ). We observed a turnover voltage of 150 mV in 2 M GdmCl, which shifted to 170 mV at 3 M GdmCl. Therefore, we cannot conclude that TotA can translocated the pore at pH 7.5. However, despite being captured against the EPF, the analyte did translocate the engineered aerolysin pore at pH 4.0 in both 2 M and 3 M GdmCl ( Fig. 2G ). Interestingly, the residual structure of TotA in 2 M GdmCl did not hinder its translocation. At pH 4.0, the current pattern of the events as well as the shape of the event populations was very similar in 2 M and 3 M GdmCl, further indicating that the change in structure did not impact the process of protein translocation. Increasing the GdmCl concentration from 2 M to 3 M increased the current at both pH (SI Fig. S4). Since the EOF is a result of net ion translocation, it is expected to be stronger at higher current (given equal selectivity) – and thus larger at 3 M than at 2 M GdmCl. However, while dwell times were faster at pH 7.5 with increased salt concentration, they remained similar at pH 4.0. We also observed that the capture efficiency decreased with increasing GdmCl concentration. A likely cause is the more unfolded structure in 3 M GdmCl resulting in a larger radius of gyration and thereby lower diffusion coefficient in bulk. Following this evaluation, we chose a condition of 3 M GdmCl at pH 4.0 for further studies. Measurements of a protein family To further explore the interplay of full-length proteins with aerolysin in the optimized condition (in 3 M GdmCl, pH 4.0), we measured six more members of the Turandot family protein at 100 mV – 210 mV ( Fig. 3A-B ). These analytes have large structural similarities and pairwise sequence identities between 38% - 70% ( Fig. 3A , SI Fig. S5-6). The analytes produced event populations with 0-10% mean residual current at 120 mV and clear deviations in dwell time between the different proteins ( Fig. 3C , Example raw traces in SI Fig. S7-9). For all 7 proteins, the population dwell time decreased with increasing voltage, confirming their translocation ( Fig. 3D ). Download figure Open in new tab Figure 3. Measurements of related proteins. (A) Structural representation of each protein analyte (predicted using Alpha Fold 2 40 ). (B) Example events collected at 120 mV for corresponding proteins. (C) Dwell time versus current scatterplots for individual events recorded at 120 mV of (left to right) TotA (yellow), TotB (orange), TotC (light red), TotF (dark red), TotM (purple), TotX (blue), TotZ (turquoise). (D) Population dwell times were fitted with the Fokker Planck equation for each protein and plotted against applied voltages. Maximum and halfwidth of the distributions are shown. (E) Velocities of each protein against electric field fitted with a linear regression. (F) Pearson correlations between the population dwell time and the GRAVY (top) and the charge normalized by the number of amino acids of each protein (bottom) cross different voltages. The p-values of the correlations are indicated as error bars. (G) Illustration of the feature extraction and reduction based on the mean decrease in impurity (MDI) for machine learning classification. (H) Two-way classification of the Tot family proteins using data recorded at 120 mV. (I) Accuracy of seven-way classifications of the Tot family proteins using data at different voltages. For each voltage 10 rounds of data selection and model training were performed and the average with indicated standard deviation is shown. Translocation velocities ( Fig. 3E , proteins length divided by the fitted dwell time) depended roughly linearly on the electric field, which is in alignment with expectations (SI Note). However, velocities are not consistent between analytes, since the dwell times poorly correlate with the analyte length (SI Fig. S10). This suggests that sequence-specific effects dominate the kinetics during protein translocation. The grand average hydropathy (GRAVY, calculated based on the Kyte-Doolittle hydropathy scale 41 ) showed a weak positive correlation with the population dwell time across all voltages ( Fig. 3F ), while the charge per amino acid (aa) did not consistently correlate with the dwell times at all. Thus, while the velocity depended linearly on the electric field, the sequence of the analytes affected the absolute dwell times. Machine learning-based protein identification Nanopore protein identification ultimately has the potential to identify even low abundant molecules in undefined samples. This requires the development of advanced analytical methods capable of identifying proteins through their single-molecule interactions. Machine learning classifiers have demonstrated strong performance in tackling this challenge 17 , 22 , 42 . We chose to describe each single-molecule interaction in 88 statistical features ( Fig. 3G , SI Table 2). To allow maximal interpretability, we reduce the features by thresholding the mean decrease in impurity (MDI) in a random forest classification. This ensures that only features that are useful for identification were used further. From the MDI we observed that one of the most important features was the dwell time, showing that the protein-dependent variance in dwell time can ultimately be useful for identification. Additionally, the current and standard deviation in the first and last portions of the events were useful for identification (SI Fig. S11). We benchmark common classification algorithms and finally trained a voting classifier combining the top three models (see methods). An initial two-way classification demonstrated clear differentiation between signals from the related proteins with an accuracy of 90 ± 5% at 120 mV ( Fig. 3H ). Expanding to a seven-way classification of all Turandot proteins yielded an accuracy of 69 ± 1%, reflecting the increased challenge of distinguishing a larger set of closely related analytes. Interestingly, we found that tuning the applied voltage affected classification accuracies ( Fig. 3I ). Increasing the voltage to 170 mV improved accuracies to 80 ± 2%, suggesting that the higher voltage amplified the variance between the protein-pore interactions, resulting in more pronounced differences in the single molecule fingerprints. However, further increase in voltage (>170 mV) reduced accuracies, likely due to the shorter event times at those voltages. Interpreting nanopore current based on protein sequences To thoroughly analyze the current signature of each analyte, we randomly selected events and overlaid them in a single plot for detailed comparison ( Fig. 4A , SI Fig. S12). We found that the current signatures of each analyte are directional which is most evident in the patterns for TotF, TotM, and TotX. This suggests that the analytes translocated the pore from the same terminus, although no charged tag was added to facilitate unidirectionality. For TotA, TotB and TotC, currents tended to slightly decrease around the middle of the event, while for TotX an increase in current appeared toward the middle of the interaction. TotF and TotM show a clear downward trend in current and TotZ had the most stable current. By comparing the protein sequences (SI Fig. S6, S13), we found that these distinct features of the current patterns could be explained by the distribution of positive charges along the proteins ( Fig 4B ), while volume exclusion ( Fig 4C ) was less relevant for the patterns. Download figure Open in new tab Figure 4. Interpretation of current signals. (A) Time normalized overlay plots of 800 randomly chosen events for each Tot family protein recorded at 100 mV. Sum of positive charges (B) and relative residual volume (C) for a sliding window of 30 amino acid along the sequence of TotA (yellow), TotB (orange), TotC (light red), TotF (dark red), TotM (purple), TotX (blue), TotZ (turquoise). (D) Charge property along the unstructured loop of TotC, TotCneg, and TotCpos (left to right). (E) Average current of 800 randomly chosen events along normalized time recorded at 100 mV for TotC, TotCneg and TotCpos. (F) Population average over the mean currents of TotC (light red), TotCneg (brown), and TotCpos (green). Overlay plots of each protein are shown in SI Fig. S12. To interpret the current signatures, we simplified the system by assuming that the simultaneous effect of all residues inside the pore lumen results in one current level. We considered a sliding window of size l = 30 aa – approximately corresponding to the pore height – gliding along the analytes and monitored the sum of positive charge in each window ( Fig. 4B , SI Fig. S14). Strikingly, key features of the overlayed currents are recovered in these sliding windows of positive charges. The sliding windows represent N-terminal capture, since calculations from C-terminus capture showed lower agreement with experimental data. We further verified the effect of positive charges by replacing the unstructured loop of TotC with two control sequences ( Fig. 4C ): one with more negative charges (TotCneg), and one with more positive charges (TotCpos), their properties are shown in Fig. 4D . Indeed, when looking at the event patterns we found a higher current level in the signals for TotCpos, which was not present with TotC ( Fig. 4E ). Additionally, the population average for the current of TotCpos translocations (SI Fig. S15) was higher compared to TotC and TotCneg over all voltages ( Fig. 4F ). TotCneg behaved very similarly to TotC, suggesting that the positive charges produced a much more pronounced effect than negative ones in our system. This can be explained by the asymmetry of the ion transport in the GdmCl pH 4.0 system, where anions are more mobile than cations. Introducing higher positive charge inside the pore can thereby increase anion (Cl - ) fluxes and the total current. A similar effect was observed in the open pore current, which increased when the pore lumen was more positively charged at pH 4.0 compared to pH 7.5 (SI Fig. S4). As negative charged aa occupy the pore, they neutralize part of the asymmetry in ion transport and do not lead to a current increase. A typical model to explain nanopore sensing is volume exclusion 43 . In studies examining how aa properties influence current readouts – for instance when protein analyte motion through MspA or CsgG was controlled using a Helicase 14 , 44 or ClpX 17 – the volume of the amino acids within the reading site often correlated with the observed current blockages. To investigate volume exclusion in our system, we estimated the volume of the aerolysin pore lumen as 41.5 nm 3 and considered the relative residual volume during protein translocations using , where V30aa is the summed volume of aa 45 in each sliding window (see methods for details). As seen in Fig 4C , the relative residual volume is similar across our different analytes, and no correlation can be drawn between the volume and the event patterns. This could be because of the free translocation approach we investigated here. While a large part of the current blockage undoubtedly stems from volume exclusion, we expect the current to be influenced by an average over 30 aa and therefore the differences in volume exclusion are likely not substantial enough to significantly influence the observed signal patterns. Discussion As the field of nanopore protein analysis evolves, the technology is poised to become indispensable, offering speed, affordability, portability, and throughput compared to traditional methods. In this study, we investigated the capabilities of an aerolysin nanopore for label-free fingerprinting of full-length proteins. Our analysis underscores the critical role of the EOF as a driving force in protein measurements. While the long and narrow barrel of aerolysin provides a big surface that can yield large ion selectivity, it also provides more hydrodynamic resistance compared to other nanopores which reduces the resulting water flux 28 . This explains why in 2 M GdmCl at pH 7.5, TotA cannot translocate through aerolysin, whereas other proteins have been translocated through α-HL under the same condition 22 . Here, we overcome this challenge by combining GdmCl with a low pH condition which induces an even larger EOF. We observed that the residual structure of TotA at 2 M GdmCl did not impact its translocation or signal patterns. This is an important advantage, as even at 3 M GdmCl, some protein structures may remain. Generating reliable signals under such conditions is crucial for a broad applicability in protein identification. By evaluating related proteins, we observed that translocation kinetics were strongly influenced by interactions between sequence-specific properties of the protein and the pore. Aerolysin K238A demonstrated promising potential for protein identification when coupled with machine learning algorithms. By optimizing the applied voltage, we achieved 80% classification accuracy for seven proteins from the same family, which shared 38%–70% pairwise sequence identities. This highlights the potential of the nanopore system for high-accuracy, label-free protein identification. While a fingerprinting approach cannot directly provide the analytes sequence, its ability to operate without enzymatic processing or chemical labeling, combined with machine learning-based classification, positions it as a practical and efficient tool. Our findings suggest that the sequence-specific current patterns observed in aerolysin are predominantly influenced by the distribution of positive charges along the protein sequence. Positive charges of analyte proteins within our nanopore system appear to increase the measured current by elevating chloride ion fluxes. This model provides a rational explanation for the observed signal patterns and offers a foundation for future advancements. With continued development, like identifying additional factors that influence the current output and translocation kinetics, or engineer/design pore architecture to accommodate less amino acid and therefore higher resolution, it may become feasible to predict nanopore fingerprints from protein sequences or, conversely, infer de novo sequence information directly from single-molecule signals, shedding light on previously uncharted biology. Author contributions V.R. and C.C. conceived the project and designed the study. V.R. performed circular dichroism experiments. E.N. performed reversal potential measurements and analysis. V.R., E.N., G.B., and J.G. performed sensing experiments and signal processing. V.R. analyzed the data and performed machine learning classifications. J.D. performed molecular dynamics simulations and analysis. V.R. and C.C. interpreted the data. V.R., and C.C. wrote the manuscript with input from all authors. Acknowledgements We thank Dr. Samuel Rommelaere and Prof. Bruno Lemaitre for the generous gift of the proteins TotA, TotB, TotC, TotF, TotM, TotX, and TotZ. This work was supported in part using the resources and services of the Protein Production and Structure Core Facility at the School of Life Sciences of EPFL. We thank Dr. Kelvin Lau for support and scientific discussion concerning protein production. This work was supported by Swiss National Science Foundation (PR00P3_193090 to C.C). References 1. ↵ Urry , L. , Cain , M. , Wasserman , S. , Minorsky , P. & Reece , J. Campbell Biology . ( Pearson, New York, NY , 2016 ). 2. ↵ Timp , W. & Timp , G. Beyond mass spectrometry, the next step in proteomics . Sci. Adv . 6 , eaax8978 ( 2020 ). OpenUrl FREE Full Text 3. ↵ Conibear , A. C. Deciphering protein post-translational modifications using chemical biology tools . Nat Rev Chem 4 , 674 – 695 ( 2020 ). OpenUrl CrossRef PubMed 4. Timp , W. & Timp , G. Beyond mass spectrometry, the next step in proteomics . Sci. Adv . 6 , eaax8978 ( 2020 ). OpenUrl FREE Full Text 5. ↵ Smith , L. M. & Kelleher , N. L. Proteoform: a single term describing protein complexity . Nat Methods 10 , 186 – 187 ( 2013 ). OpenUrl CrossRef PubMed Web of Science 6. ↵ Kelly , R. T. Single-cell Proteomics: Progress and Prospects . Mol. Cell. Proteomics 19 , 1739 – 1748 ( 2020 ). OpenUrl Abstract / FREE Full Text 7. ↵ Nesvizhskii , A. I. & Aebersold , R. Interpretation of Shotgun Proteomic Data . Mol. Cell. Proteomics 4 , 1419 – 1440 ( 2005 ). OpenUrl Abstract / FREE Full Text 8. Chait , B. T. Mass Spectrometry: Bottom-Up or Top-Down? Science 314 , 65 – 66 ( 2006 ). OpenUrl Abstract / FREE Full Text 9. ↵ Hubbard , S. J. & Jones , A. R. Hughes , C. , Ma , B. & Lajoie , G. A. De Novo Sequencing Methods in Proteomics . in Proteome Bioinformatics (eds. Hubbard , S. J. & Jones , A. R. 105 – 121 ( Humana Press , Totowa, NJ , 2010 ). doi: 10.1007/978-1-60761-444-9_8 . OpenUrl CrossRef PubMed Web of Science 10. ↵ Ritmejeris , J. , Chen , X. & Dekker , C. Single-molecule protein sequencing with nanopores . Nat. Rev. Bioeng . 1 – 14 ( 2024 ) doi: 10.1038/s44222-024-00260-8 . OpenUrl CrossRef 11. ↵ Motone , K. & Nivala , J. Not if but when nanopore protein sequencing meets single-cell proteomics . Nat. Methods 20 , 336 – 338 ( 2023 ). OpenUrl CrossRef PubMed 12. ↵ Kasianowicz , J. J. , Brandin , E. , Branton , D. & Deamer , D. W. Characterization of individual polynucleotide molecules using a membrane channel . Proc. Natl. Acad. Sci . 93 , 13770 – 13773 ( 1996 ). OpenUrl Abstract / FREE Full Text 13. ↵ Wang , Y. , Zhao , Y. , Bollas , A. , Wang , Y. & Au , K. F. Nanopore sequencing technology, bioinformatics and applications . Nat Biotechnol 39 , 1348 – 1365 ( 2021 ). OpenUrl CrossRef PubMed 14. ↵ Brinkerhoff , H. , Kang , A. S. W. , Liu , J. , Aksimentiev , A. & Dekker , C. Multiple rereads of single proteins at single–amino acid resolution using nanopores . Science 374 , 1509 – 1513 ( 2021 ). OpenUrl CrossRef PubMed 15. Chen , Z. et al. Controlled movement of ssDNA conjugated peptide through Mycobacterium smegmatis porin A (MspA) nanopore by a helicase motor for peptide sequencing application . Chem. Sci . 12 , 15750 – 15756 ( 2021 ). OpenUrl CrossRef PubMed 16. ↵ Yan , S. et al. Single Molecule Ratcheting Motion of Peptides in a Mycobacterium smegmatis Porin A (MspA) Nanopore . Nano Lett . 21 , 6703 – 6710 ( 2021 ). OpenUrl CrossRef PubMed 17. ↵ Motone , K. et al. Multi-pass, single-molecule nanopore reading of long protein strands . Nature 633 , 662 – 669 ( 2024 ). OpenUrl CrossRef PubMed 18. ↵ Nivala , J. , Marks , D. B. & Akeson , M. Unfoldase-mediated protein translocation through an α-hemolysin nanopore . Nat Biotechnol 31 , 247 – 250 ( 2013 ). OpenUrl CrossRef PubMed 19. ↵ Versloot , R. C. A. , Straathof , S. A. P. , Stouwie , G. , Tadema , M. J. & Maglia , G. β-Barrel Nanopores with an Acidic–Aromatic Sensing Region Identify Proteinogenic Peptides at Low pH . ACS Nano ( 2022 ) doi: 10.1021/acsnano.1c11455 . OpenUrl CrossRef 20. Lucas , F. L. R. , Versloot , R. C. A. , Yakovlieva , L. , Walvoort , M. T. C. & Maglia , G. Protein identification by nanopore peptide profiling . Nat Commun 12 , 5795 ( 2021 ). OpenUrl CrossRef PubMed 21. ↵ Afshar Bakshloo , M. et al. Nanopore-Based Protein Identification . J Am Chem Soc 144 , 2716 – 2725 ( 2022 ). OpenUrl CrossRef PubMed 22. ↵ Yu , L. et al. Unidirectional single-file transport of full-length proteins through a nanopore . Nat. Biotechnol . 1 – 10 ( 2023 ) doi: 10.1038/s41587-022-01598-3 . OpenUrl CrossRef PubMed 23. ↵ Oukhaled , G. et al. Unfolding of Proteins and Long Transient Conformations Detected by Single Nanopore Recording . Phys Rev Lett 98 , 158101 ( 2007 ). OpenUrl CrossRef PubMed 24. ↵ 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 ). OpenUrl CrossRef PubMed Web of Science 25. ↵ Sauciuc , A. , Morozzo della Rocca , B. , Tadema , M. J. , Chinappi , M. & Maglia , G. Translocation of linearized full-length proteins through an engineered nanopore under opposing electrophoretic force . Nat. Biotechnol . 1 – 7 ( 2023 ) doi: 10.1038/s41587-023-01954-x . OpenUrl CrossRef 26. ↵ Bakshloo , M. A. et al. Polypeptide analysis for nanopore-based protein identification . Nano Res ( 2022 ) doi: 10.1007/s12274-022-4610-1 . OpenUrl CrossRef 27. ↵ Bonome , E. L. , Cecconi , F. & Chinappi , M. Electroosmotic flow through an α-hemolysin nanopore . Microfluid. Nanofluidics 21 , 96 ( 2017 ). OpenUrl CrossRef 28. ↵ Mehrafrooz , B. et al. Electro-osmotic Flow Generation via a Sticky Ion Action . ACS Nano 18 , 17521 – 17533 ( 2024 ). OpenUrl CrossRef PubMed 29. ↵ Ouldali , H. et al. Electrical recognition of the twenty proteinogenic amino acids using an aerolysin nanopore . Nat Biotechnol 38 , 176 – 181 ( 2020 ). OpenUrl CrossRef PubMed 30. ↵ Ge , Y. , Cui , M. , Zhang , Q. , Wang , Y. & Xi , D. Aerolysin nanopore-based identification of proteinogenic amino acids using a bipolar peptide probe . Nanoscale Adv . 4 , 3883 – 3891 ( 2022 ). OpenUrl CrossRef PubMed 31. ↵ Cao , C. et al. Deep Learning-Assisted Single-Molecule Detection of Protein Post-translational Modifications with a Biological Nanopore . ACS Nano 18 , 1504 – 1515 ( 2024 ). OpenUrl CrossRef PubMed 32. ↵ Cao , C. et al. Aerolysin nanopores decode digital information stored in tailored macromolecular analytes . Sci. Adv . 6 , eabc2661 ( 2021 ). OpenUrl 33. ↵ Sauciuc , A. et al. Blobs form during the single-file transport of proteins across nanopores . Proc. Natl. Acad. Sci . 121 , e2405018121 ( 2024 ). OpenUrl CrossRef PubMed 34. ↵ Pastoriza-Gallego , M. et al. Dynamics of Unfolded Protein Transport through an Aerolysin Pore . J Am Chem Soc 133 , 2923 – 2931 ( 2011 ). OpenUrl CrossRef PubMed 35. ↵ Lesieur , C. et al. Increased Stability upon Heptamerization of the Pore-forming Toxin Aerolysin * . J. Biol. Chem . 274 , 36722 – 36728 ( 1999 ). OpenUrl Abstract / FREE Full Text 36. ↵ Huang , G. et al. Electro-Osmotic Vortices Promote the Capture of Folded Proteins by PlyAB Nanopores . Nano Lett 20 , 3819 – 3827 ( 2020 ). OpenUrl CrossRef PubMed 37. Boukhet , M. et al. Probing driving forces in aerolysin and α-hemolysin biological nanopores: electrophoresis versus electroosmosis . Nanoscale 8 , 18352 – 18359 ( 2016 ). OpenUrl CrossRef PubMed 38. ↵ Bogard , A. et al. The Ionic Selectivity of Lysenin Channels in Open and Sub-Conducting States . Membranes 11 , 897 ( 2021 ). OpenUrl CrossRef PubMed 39. ↵ Rommelaere , S. et al. A humoral stress response protects Drosophila tissues from antimicrobial peptides . Curr. Biol . 34 , 1426 - 1437 .e6 ( 2024 ). OpenUrl CrossRef PubMed 40. ↵ Jumper , J. et al. Highly accurate protein structure prediction with AlphaFold . Nature 596 , 583 – 589 ( 2021 ). OpenUrl CrossRef PubMed 41. ↵ Kyte , J. & Doolittle , R. F. A simple method for displaying the hydropathic character of a protein . J. Mol. Biol . 157 , 105 – 132 ( 1982 ). OpenUrl CrossRef PubMed Web of Science 42. ↵ Im , J. , Lindsay , S. , Wang , X. & Zhang , P. Single Molecule Identification and Quantification of Glycosaminoglycans Using Solid-State Nanopores . ACS Nano 13 , 6308 – 6318 ( 2019 ). OpenUrl CrossRef PubMed 43. ↵ Wilson , J. , Sarthak , K. , Si , W. , Gao , L. & Aksimentiev , A. Rapid and Accurate Determination of Nanopore Ionic Current Using a Steric Exclusion Model . ACS Sens . 4 , 634 – 644 ( 2019 ). OpenUrl CrossRef PubMed 44. ↵ Liu , J. & Aksimentiev , A. Molecular Determinants of Current Blockade Produced by Peptide Transport Through a Nanopore . ACS Nanosci. Au 4 , 21 – 29 ( 2024 ). OpenUrl CrossRef PubMed 45. ↵ Perkins , S. J. Protein volumes and hydration effects . Eur. J. Biochem . 157 , 169 – 180 ( 1986 ). OpenUrl CrossRef PubMed Web of Science View the discussion thread. Back to top Previous Next Posted January 16, 2025. Download PDF Supplementary Material Email Thank you for your interest in spreading the word about bioRxiv. NOTE: Your email address is requested solely to identify you as the sender of this article. Your Email * Your Name * Send To * Enter multiple addresses on separate lines or separate them with commas. You are going to email the following Charge-based fingerprinting of unlabeled full-length proteins using an aerolysin nanopore Message Subject (Your Name) has forwarded a page to you from bioRxiv Message Body (Your Name) thought you would like to see this page from the bioRxiv website. Your Personal Message CAPTCHA This question is for testing whether or not you are a human visitor and to prevent automated spam submissions. Share Charge-based fingerprinting of unlabeled full-length proteins using an aerolysin nanopore Verena Rukes , Evita Norkute , Georges Barnikol , Jingze Duan , Jiajie Gao , Chan Cao bioRxiv 2025.01.13.632743; doi: https://doi.org/10.1101/2025.01.13.632743 Share This Article: Copy Citation Tools Charge-based fingerprinting of unlabeled full-length proteins using an aerolysin nanopore Verena Rukes , Evita Norkute , Georges Barnikol , Jingze Duan , Jiajie Gao , Chan Cao bioRxiv 2025.01.13.632743; doi: https://doi.org/10.1101/2025.01.13.632743 Citation Manager Formats BibTeX Bookends EasyBib EndNote (tagged) EndNote 8 (xml) Medlars Mendeley Papers RefWorks Tagged Ref Manager RIS Zotero Tweet Widget Facebook Like Google Plus One Subject Area Bioengineering Subject Areas All Articles Animal Behavior and Cognition (7622) Biochemistry (17648) Bioengineering (13870) Bioinformatics (41880) Biophysics (21423) Cancer Biology (18553) Cell Biology (25458) Clinical Trials (138) Developmental Biology (13364) Ecology (19866) Epidemiology (2067) Evolutionary Biology (24290) Genetics (15589) Genomics (22475) Immunology (17711) Microbiology (40326) Molecular Biology (17145) Neuroscience (88471) Paleontology (666) Pathology (2826) Pharmacology and Toxicology (4815) Physiology (7635) Plant Biology (15114) Scientific Communication and Education (2044) Synthetic Biology (4286) Systems Biology (9815) Zoology (2268)