{"id":"a74df7e0-d6f7-4d55-bf66-5b81a02c74d3","arxiv_id":"2501.04589","paper_version":1,"verdict":"CONDITIONAL","confidence":"LOW","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":3,"one_line_summary":"A single-protein electrical junction records catalytic turnovers of two redox enzymes as two-level conductance switching, with switching frequencies matching bulk turnover rates.","lead":"Researchers trapped single molecules of two enzymes in a tiny electrical gap and watched the current jump back and forth as the enzymes worked. The jump rate matched the enzymes' known reaction speeds, pointing to a label-free way to watch individual enzymes catalyze reactions in real time.","discovery_kind":"new_method","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Validation of switching-to-turnover assignment is compromised by an ill-defined background correction and overly broad kcat ranges; the Table 1 agreement is therefore non-discriminating.","rationale":"The reader's weakest assumption is that each conductance switch corresponds to a catalytic turnover. My stress-test agrees that this is the crux, but identifies a sharper, textually verifiable problem: the background-correction step used to rescue the Table 1 agreement is not well-defined. The text reports inactive switching as a percentage of traces (5-7%, Fig. 5) and then says these are 'subtracted' from the bulk frequencies (Eq. 1, which is events per minute). Without reporting an inactive-condition frequency in the same units, the corrected values are unreproducible. This is a concrete internal flaw, not merely a disagreement with the field. In addition, the cited kcat ranges are so broad that the post-correction values (1790 and 12871 min-1) falling inside them is a weak, low-information test. The proposed concrete checks would settle the concern: recomputing the corrected frequencies transparently with uncertainties, and, if possible, a substrate-titration experiment that tests the mechanistic assignment. Because the core idea is plausible and the issue is addressable with additional analysis or data, I do not recommend moving the verdict; the reader's CONDITIONAL assessment remains appropriate. My agreement is 'partial' because I sharpen the same weakest assumption with a specific dimensional/reproducibility critique rather than simply restating it.","tokens_in":15841,"tokens_out":5018,"duration_ms":52270,"concrete_test":"Reanalyze the raw current-time traces with the same Python/HMM pipeline to compute, for both active and inactive conditions, the switching-event rate per minute of total protein trapping time (Eq. 1). Report the raw active rate, the inactive rate, and the background-corrected rate with uncertainties (e.g., bootstrap across traces). If the inactive rate is not a true per-minute frequency, or if the corrected rate falls outside the cited kcat range by more than 20%, the claimed correlation is not established. As a complementary check, perform a substrate titration (e.g., 0.5x, 1x, 2x KM) and test whether the switching frequency follows Michaelis-Menten kinetics; this would directly test whether switches report catalytic turnover rather than binding or conformational noise.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The paper's central claim is that the measured conductance-switching frequency quantitatively reports single-enzyme catalytic turnover. The key evidence is the agreement in Table 1 between the 'bulk frequency' and literature kcat. This validation is insecure for two concrete reasons. First, the background correction is dimensionally ambiguous. Figure 5 reports 5-7% switching under inactive conditions as a percentage of traces, not as a rate (events per minute). The text states that 'subtracting them from the bulk frequencies' yields the corrected values (P450: 1790 min-1; GR: 12871 min-1), but a percentage of traces cannot be subtracted from a per-minute frequency without an explicit conversion. If an inactive-condition frequency was computed, it is not reported, so the corrected values cannot be reproduced or checked. Second, the cited kcat ranges are extremely broad: P450cam is given as 124-3960 min-1 and GR as 12600-17500 min-1, spanning roughly 30- and 1.4-fold ranges. Nearly any corrected value within an order of magnitude of the raw frequency would fall inside one of these ranges, so the agreement is weak evidence for the turnover assignment. The causal claim additionally requires that each two-level switch is caused by a transient cofactor oxidation during catalysis, but no direct single-molecule product detection or independent probe of the enzyme's redox state is provided to rule out substrate-binding fluctuations, conformational dynamics, or trapping artifacts as sources of the switching. Because the entire quantitative claim rests on this correlation, the load-bearing assumption is not yet secured.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The manuscript reports an electrochemical STM-based single-protein junction platform in which individual unmodified redox enzymes (cytochrome P450cam and glutathione reductase) are transiently trapped between a Au(111) substrate and an STM tip. Under electrocatalytic conditions, the authors observe two-level conductance fluctuations ('switching' events) in current-time traces, which they attribute to transient oxidation of the enzyme cofactor during each catalytic turnover. Using a Python-based classification algorithm and HMM/GMM analyses, they extract switching frequencies and compare them with literature kcat values: the 'bulk' frequencies after background correction are 1790 min^-1 for P450cam and 12871 min^-1 for GR, both within the cited literature ranges. The paper claims real-time, label-free electrical transduction of single-enzyme catalytic events, and further reports a 'single-enzyme' frequency roughly threefold higher than the bulk frequency, interpreted as evidence of dynamic disorder and catalytic heterogeneity.","tokens_in":16100,"tokens_out":2442,"duration_ms":25431,"significance":"If the switching-to-turnover assignment is valid, this would be a notable advance: label-free, real-time electrical detection of individual catalytic events in unmodified redox enzymes, with potential biosensing applications. The paper has genuine strengths: the active/inactive control design is appropriate; bulk electrocatalytic activity is verified by CV and by GC-MS and UV-visible product detection; two enzymes with different cofactors (heme vs FAD), chemistries, and rate ranges are compared; and a nontrivial automated classification pipeline (with both HMM and GMM) is used. The proposed mechanism (redox-state-gated sequential tunnelling) is physically plausible and consistent with prior work on redox protein junctions. However, the central validation rests on the agreement between measured switching frequencies and literature kcat values, and that agreement is currently not quantitative enough to be discriminating: the background correction is dimensionally ambiguous, the literature kcat ranges are very broad, and the reported frequencies carry no statistical uncertainties. These issues are fixable but must be addressed before the central claim can be accepted.","major_comments":[{"comment":"The background subtraction used to obtain the corrected bulk frequencies is dimensionally inconsistent as reported. Fig. 5 reports 5-7% switching under inactive conditions as a percentage of traces, but Eq. (1) defines bulk frequency as events per total protein trapping time (min^-1). A percentage of traces cannot be subtracted from a per-minute frequency without stating the conversion (e.g., the per-trace residence time and the event count per trace). The manuscript does not report an inactive-condition frequency, so the corrected values of 1790 and 12871 min^-1 in Table 1 cannot be reproduced or checked. Please provide the raw event counts, total residence times, and an explicit calculation showing how the 5-7% background translates into the subtracted frequency.","section":"Table 1, Eq. (1), Fig. 5"},{"comment":"The literature kcat ranges used for validation are too broad to be discriminating. For P450cam the cited range is 124-3960 min^-1, a factor of ~32, and the raw measured bulk frequency of 2573 min^-1 is already within this range before any background correction; for GR the range is 12600-17500 min^-1, and the corrected value of 12871 min^-1 sits at its lower boundary. With such wide windows, nearly any measured frequency within an order of magnitude would 'agree' with kcat. The claim of 'exquisite agreement' therefore requires either narrower benchmark values (ideally measured under the same solution conditions with the same enzyme preparation) or a statistical comparison that quantifies the expected spread of the measured frequencies.","section":"Table 1"},{"comment":"The reported bulk and single-enzyme frequencies are presented without error bars or confidence intervals, and the manuscript does not state the number of switching events, the total protein trapping time, or the number of traces contributing to each frequency. Because the central conclusion is a quantitative correlation between measured frequencies and kcat, the absence of uncertainty estimates makes it impossible to assess whether the observed 3-fold difference between bulk and single-enzyme frequencies is significant or whether the agreement with kcat is better than chance. Please report per-condition event counts, total times, and bootstrap or other uncertainty estimates for each frequency.","section":"Eqs. (1)-(2), Table 1"},{"comment":"The assignment of each two-level switching event to a catalytic turnover is not directly verified. The authors note that 5-7% of traces show switching under inactive conditions, indicating that switching is not exclusive to catalysis, and no single-molecule product detection or independent probe of the enzyme redox state is provided to rule out substrate-binding fluctuations, conformational dynamics, or trapping artifacts as the cause of the switching signal. The proposed mechanism in Fig. 6 (transient cofactor oxidation opening a sequential tunnelling channel) is plausible and consistent with earlier redox-protein junction work, but it is presented as the interpretation of the frequency correlation rather than being tested independently. A direct test, e.g., a substrate-concentration dependence of switching frequency following Michaelis-Menten behaviour, or a mutant/inhibitor control that abolishes catalytic activity while retaining redox switching, would substantially strengthen the turnover assignment.","section":"Results, 'Electrochemically Controlled Single-Enzyme Catalytic Junctions'; Fig. 5; Fig. 6"}],"minor_comments":[{"comment":"The GR conductance values are inconsistent between the main text and the caption: the text states G2 = 1.5 x 10^-4 G0 for GR, while the Fig. 4 caption gives 1.5 x 10^-5 G0. Please correct this discrepancy.","section":"Fig. 4 caption"},{"comment":"Equations (1) and (2) are referenced in the text but are not explicitly numbered in the displayed layout; please number them clearly and define all variables (e.g., whether 'total protein trapping time' includes silent traces in Eq. (1) and only switching traces in Eq. (2), as implied).","section":"Equations (1) and (2)"},{"comment":"The phrase 'exquisite agreement' (main text, Results section) is an overstatement given the broad kcat ranges and the lack of error bars; a more measured description of the comparison would be appropriate.","section":"Abstract and main text"},{"comment":"The y-axis label '% of traces displaying conductance switching events' is clear, but the text describing the 5-7% residual as 'enzymatic events that do not lead to the enzymatic chemical conversion' is speculative; the residual could equally arise from non-enzymatic junction instability, and this should be acknowledged.","section":"Fig. 5"}],"recommendation":"major_revision","confidential_remarks":"The paper addresses a high-impact question and the experimental platform is interesting, but the validation of the central claim is currently too weak. The authors should be asked to supply a properly dimensioned background correction, error statistics, and ideally a more discriminating test of the switching-to-turnover assignment (e.g., substrate dependence or an inactive mutant). If those can be provided, the paper would be a strong candidate for publication; in its present form the quantitative agreement with kcat is not convincing. I also note that the very broad literature kcat ranges cited for P450cam make the comparison nearly unfalsifiable, so the authors should either narrow the benchmark or present a different validation strategy."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"The paper reports label-free electrical detection of catalytic turnover in two redox enzymes, P450cam and GR, trapped in an EC-STM nanogap. The new concrete thing is the demonstration that unmodified redox enzymes show two-level conductance switching under catalytic conditions, with switching frequency roughly tracking the enzymes' kcat. That is not in the prior literature, which either used engineered proteins (azurin) or captured static catalytic states (formate dehydrogenase). The active/inactive control set is the right design, and the ~30% vs 5-7% switching fraction under active vs inactive conditions is a robust qualitative effect. The STM height vs potential data for P450 also support redox-gated tunneling.\n\nThe soft spots are quantitative. The central claim—each switch is a turnover—is validated only by comparing the measured bulk frequency to literature kcat values. That comparison is weak for two reasons. First, the kcat ranges cited are very broad (P450: 124-3960 min-1, about 30-fold), so almost any measured frequency in that order of magnitude would agree. Second, the background correction is dimensionally odd: they report 5-7% switching under inactive conditions as a percentage of traces, then state they 'subtract' that from the per-minute bulk frequency. You cannot subtract a percentage from a rate without defining how the percentage translates into events per minute. The corrected values in Table 1 (1790 and 12871 min-1) are therefore unreproducible from the text. The raw frequencies (2573 and 16292) are still consistent with the broad kcat ranges, so the main correlation survives, but the paper's emphasis on the corrected numbers is misplaced.\n\nThere are smaller issues: Table 1 has no error bars, and the GR conductance values differ by 10x between text (1.5e-4 G0) and figure caption (1.5e-5 G0). The supplementary material with raw traces and analysis code was not available to me, so I could not check the HMM/GMM event counting.\n\nThe mechanism proposed—transient cofactor oxidation opening a redox-mediated tunneling channel—is plausible and consistent with earlier azurin work, but it remains an interpretation. No direct single-molecule product detection supports the switch-to-turnover assignment. That said, the authors do acknowledge the possibility of non-productive dynamics and use the background to attempt a correction. The core idea is sound; the validation just needs tightening.\n\nWho is this for? Anyone working on single-molecule enzymology or protein electronics. It deserves peer review; a referee should ask for error bars, a clear definition of the background subtraction, and either raw data or a reproducible analysis pipeline. I would accept it with major revision.","headline":"A plausible new single-enzyme electrical readout with solid controls, but the quantitative link between switching and turnover needs tighter validation.","tokens_in":16700,"tokens_out":3350,"would_cite":true,"duration_ms":30426,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"Conductance switching in a single trapped enzyme tracks catalytic turnover in real time.","keywords":["single-enzyme catalysis","single-protein junction","conductance switching","redox enzyme","electrochemical scanning tunnelling microscopy","label-free biosensing","cytochrome P450cam","glutathione reductase"],"falsifier":"Record the timing of conductance switches in the same junction while detecting the reaction product (5-exo-hydroxycamphor for P450cam, GSH for GR) with a single-molecule-sensitive assay; the central claim would fail if switches occur with no product, if product is formed with no accompanying switch, or if the switching rate does not respond to substrate concentration or known inhibitors.","tokens_in":15648,"feed_emoji":"⚡","tokens_out":9323,"duration_ms":83491,"temperature":0.7,"pith_summary":"This paper reports a label-free, all-electrical way to watch individual enzyme molecules catalyse reactions in real time. The authors trap unmodified redox enzymes—cytochrome P450cam and glutathione reductase—in a nanoscale tunnelling junction under electrochemical control, and find that during catalysis the junction current switches between two conductance levels. They argue that each switch marks a catalytic cycle: a transient oxidation of the enzyme's cofactor momentarily opens a redox-mediated tunnelling channel, raising the conductance until the cofactor returns to its reduced state. Counting these switches over thousands of trapping events gives average frequencies that, after subtracting a background, fall inside the enzymes' reported bulk turnover ranges. If correct, this turns a single-protein electrical junction into a direct, label-free readout of single-enzyme activity and heterogeneity.","feed_headline":"Trapped enzymes report each catalytic cycle as a current jump","feed_subtitle":"Two enzymes show two-level conductance blips matching their bulk turnover rates, no labels needed.","key_machinery":"The central object is the electrochemically controlled single-protein tunnelling junction: an STM tip and Au(111) substrate separated by a 3.8-5.7 nm gap in aqueous buffer, with one enzyme transiently trapped between them. The carrying mechanism is the redox-gated sequential tunnelling channel: when the enzyme's cofactor (heme in P450cam, FAD in GR) is transiently oxidised during a catalytic cycle, its redox level comes into resonance with the electrode Fermi levels, opening an additional two-step electron-transfer channel that raises the junction conductance; returning to the reduced state closes it. The argument is carried by counting these conductance switches in long current-time traces, classifying blinks as silent or switching with an automated algorithm plus Gaussian mixture and hidden Markov models, and converting switch counts into bulk and single-enzyme frequencies via equations [1] and [2].","core_discovery":"The central claim is that the two-level conductance fluctuations observed in a single trapped redox enzyme junction are the electrical signature of individual catalytic turnovers. Under reducing potentials with substrate present (and mediator for GR), roughly 30% of protein trapping events show switching blinks, whereas inactive conditions yield only 5-7%. The switching frequencies extracted over long records are 2573 min$^{-1}$ for P450cam and 16292 min$^{-1}$ for GR; subtracting the inactive background gives 1790 and 12871 min$^{-1}$, both within the reported bulk $k_{\\mathrm{cat}}$ ranges of 124-3960 and 12600-17500 min$^{-1}$. The authors interpret this agreement as validation that the switching events transduce catalysis. They further report single-enzyme frequencies of 6908 and 42943 min$^{-1}$, about threefold higher than the bulk values, which they read as the signature of dynamic disorder and catalytic heterogeneity masked by ensemble averaging.","pith_inferences":["If the turnover assignment holds, varying substrate concentration while counting switches should recover single-molecule Michaelis-Menten behaviour, a testable extension the paper does not report.","The residual switching under inactive conditions suggests some switches may be redox or conformational fluctuations that do not produce product; direct single-molecule product detection in the junction would quantify that fraction.","The consistent ~3 ratio of single-enzyme to bulk frequency across two very different enzymes hints that dynamic disorder may scale with catalytic rate, a pattern worth testing across more enzyme families.","Because the trapping geometry is built from standard STM nanogap technology, the approach could in principle be ported to large arrays of nanogap electrodes for label-free enzyme screening."],"forward_implications":["Because the readout needs no fluorescent label, the method should extend to unmodified redox enzymes that are hard to label or prone to photobleaching.","Bulk catalytic turnover numbers are recoverable from purely electrical records, so the switching signal is a quantitative activity readout, not just an on-off indicator.","The roughly threefold gap between single-enzyme and bulk frequencies gives a direct measurement of catalytic heterogeneity and dynamic disorder.","The persistent 5-7% switching under inactive conditions implies that a background correction is needed before equating switches with successful turnovers.","The proposed mechanism predicts that any redox enzyme whose cofactor is transiently oxidised during turnover should show similar two-level conductance switching in such a junction."],"supporting_citations":[{"why":"predicted that an active trapped enzyme would show discrete conductance switching from individual turnover events, the hypothesis this paper tests.","marker":"[12]"},{"why":"single-molecule fluorescence study whose on-off blinking was ascribed to FAD redox toggling during turnover, supplying the analogy for reading turnover from switching.","marker":"[13]"},{"why":"demonstrated potential-tuned resonant tunnelling through redox molecules, the basis for redox-state-dependent conductance.","marker":"[9]"},{"why":"showed redox-gated tunnelling resonance in a single redox protein, supporting the cofactor-resonance mechanism invoked here.","marker":"[11]"},{"why":"reported direct electrical monitoring of a non-redox enzyme via tunnelling current noise, a precedent for activity-induced conductance fluctuations.","marker":"[5]"},{"why":"captured catalytic intermediate states of an enzyme as distinct protein conductance states in single-molecule junctions.","marker":"[6]"},{"why":"one of the bulk kinetic references for P450cam turnover that the measured switching frequency is compared against.","marker":"[22]"},{"why":"one of the bulk kinetic references for glutathione reductase turnover used to validate the extracted frequencies.","marker":"[25]"}],"fun_headline_variants":["Single enzymes flash electrical blips for each catalytic act","Enzyme in a trap turns each reaction into a current blip","Conductance twitches reveal single-enzyme catalysis in action","Redox enzyme blinks signal every catalytic turnover, label-free","Nanoscale junction reads single-enzyme turnovers as current jumps"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing premise is that each two-level conductance switch in an active junction is caused by a catalytic turnover—the transient oxidation of the enzyme's cofactor—and not by trapping artifacts, substrate binding, or redox fluctuations that do not lead to product formation.","fun_headline_variants_meta":{"raw":{"variants":["Single enzymes flash electrical blips for each catalytic act","Enzyme in a trap turns each reaction into a current blip","Conductance twitches reveal single-enzyme catalysis in action","Redox enzyme blinks signal every catalytic turnover, label-free","Nanoscale junction reads single-enzyme turnovers as current jumps"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000164,"raw_usage":{"total_tokens":1270,"prompt_tokens":993,"completion_tokens":277,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":609,"completion_tokens_details":{"reasoning_tokens":193}},"tokens_in":609,"tokens_out":277,"duration_ms":3518,"temperature":1.0,"reasoning_tokens":193,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-10T21:28:25.436113+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Record the timing of conductance switches in the same junction while detecting the reaction product (5-exo-hydroxycamphor for P450cam, GSH for GR) with a single-molecule-sensitive assay; the central claim would fail if switches occur with no product, if product is formed with no accompanying switch, or if the switching rate does not respond to substrate concentration or known inhibitors.","supporting_citations":[{"cited_title":"Electrochemical tunnelling sensors and their potential applications","cited_arxiv_id":null,"evidence_quote":"predicted that an active trapped enzyme would show discrete conductance switching from individual turnover events, the hypothesis this paper tests."},{"cited_title":"& X Xie, S","cited_arxiv_id":null,"evidence_quote":"single-molecule fluorescence study whose on-off blinking was ascribed to FAD redox toggling during turnover, supplying the analogy for reading turnover from switching."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"demonstrated potential-tuned resonant tunnelling through redox molecules, the basis for redox-state-dependent conductance."},{"cited_title":"& Ulstrup, J","cited_arxiv_id":null,"evidence_quote":"showed redox-gated tunnelling resonance in a single redox protein, supporting the cofactor-resonance mechanism invoked here."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"reported direct electrical monitoring of a non-redox enzyme via tunnelling current noise, a precedent for activity-induced conductance fluctuations."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"captured catalytic intermediate states of an enzyme as distinct protein conductance states in single-molecule junctions."},{"cited_title":"K., England, P","cited_arxiv_id":null,"evidence_quote":"one of the bulk kinetic references for P450cam turnover that the measured switching frequency is compared against."},{"cited_title":"& Brandt, K","cited_arxiv_id":null,"evidence_quote":"one of the bulk kinetic references for glutathione reductase turnover used to validate the extracted frequencies."}],"review_version":1}