{"id":"770a31ec-973f-41aa-b5e4-d8d88cb5e31c","arxiv_id":"2412.06364","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"high","formal_verification":"none","parameter_count":1,"one_line_summary":"THz irradiation at 0.10 and 0.29 THz non-thermally increases lipid lateral diffusion in HeLa cell membranes below 37 °C, suggesting a phase transition, but 0.48 THz shows no effect.","lead":"This paper measures how terahertz waves affect the movement of fat molecules in living cell membranes. It reports that 0.10 and 0.29 THz radiation increase membrane diffusion below body temperature, while 0.48 THz does not.","discovery_kind":"new_method","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Thermocouple calibration may underestimate membrane heating under THz, and Table 1's ΔTs values are internally inconsistent with power and absorption, so the non-thermal diffusion increase is not yet established.","rationale":"The reader's weakest-assumption identification matches my own: the non-thermal interpretation is the paper's central claim, and the only evidence separating thermal from non-thermal is the thermocouple-based temperature correction. The internal inconsistency of Table 1's ΔTs values is a warning sign that the calibration is not trustworthy, and the physical placement of the thermocouple on the bottom side of a 0.15-mm film plausibly leads to underestimation of the membrane temperature. I also note that the 0.48 THz null result is confounded by its lower power density, but that issue affects the frequency-specific secondary claim rather than the primary positive effect at 0.10 and 0.29 THz. The direct membrane-temperature measurement proposed above would settle the main concern: if the temperature correction survives, the diffusion increase would be a genuine non-thermal effect, though the phase-transition interpretation would still require a direct membrane-order readout. The paper has merits — a novel THz-FRAP setup, a clear measurement protocol, and honest reporting of raw diffusion data — but the calibration control is currently inadequate for the strength of the claim. This keeps the verdict at CONDITIONAL rather than ACCEPT or REJECT, and my read does not change the reader's conditional verdict.","tokens_in":7290,"tokens_out":7305,"duration_ms":78792,"concrete_test":"Repeat the temperature calibration using a temperature-sensitive fluorescent dye loaded into HeLa cell membranes (e.g., C-Laurdan or ERthermoAC) and image the dye during 0.10, 0.29, and 0.48 THz irradiation at the Table 1 powers, without the thermocouple present in the beam. Compare the dye-derived membrane temperature to the reported ΔTs. If the dye-reported ΔT at 0.10 or 0.29 THz exceeds the thermocouple ΔT by more than about 1 °C at the relevant incubator temperatures, the diffusion increase may be thermal and the central non-thermal claim would not survive. As a secondary check, remeasure 0.48 THz at 65–89 mW/cm² to test whether the null result is only a power artifact.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central non-thermal claim rests on the temperature correction in Methods, 'Calibration of cell temperature.' The 13-µm K-type thermocouple is placed at the bottom of the film-bottom dish, whereas the THz beam is absorbed primarily by the adherent cell layer on the upper surface of the 0.15-mm film; the intervening plastic film has low thermal conductivity, so the thermocouple can read cooler than the actual membrane temperature. If ΔTs is underestimated, the THz data are plotted at too-low Ts values, and the apparent increase in D relative to control at equal Ts is exactly what a purely thermal effect would produce. Table 1 reinforces this concern: 0.48 THz at 20 mW/cm² with α ≈ 75 cm⁻¹ is reported to give ΔTs = 2.4–2.8 °C, nearly the same as 0.10 THz at 89 mW/cm² with α ≈ 37 cm⁻¹ (ΔTs = 2.5–3.0 °C), although a simple absorbed-power estimate predicts a roughly 0.5× smaller ΔT for 0.48 THz. This inconsistency suggests the calibration is not faithfully reporting membrane temperature, possibly because the thermocouple itself absorbs THz or because the thermal pathway to the sensor differs from that of the cells. Since no independent membrane-temperature measurement is provided, the observed diffusion increase could be thermal at the membrane even after correcting the incubator temperature.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The manuscript reports THz-FRAP measurements of lateral diffusion coefficients D of HeLa cell membranes under continuous-wave irradiation at 0.10, 0.29, and 0.48 THz, with power densities from 20 to 89 mW/cm2. After correcting the incubator temperature using a thermocouple-based calibration of the THz-induced temperature rise, the authors find that 0.10 and 0.29 THz irradiation increases D at sample temperatures below about 37 °C compared with no-THz controls, while 0.48 THz does not. They interpret this as a non-thermal, THz-induced order-disorder phase transition of membrane lipids mediated by hydration-water dynamics. The central claim is that the diffusion increase is genuinely non-thermal once the heating contribution is subtracted.","tokens_in":7607,"tokens_out":6796,"duration_ms":71883,"significance":"The paper presents a directly measured membrane-dynamics observable under controlled THz exposure and reports a clear, visually apparent difference in diffusion between 0.10/0.29 THz and control conditions. If the non-thermal interpretation survives scrutiny, the result is significant for THz bioeffect safety standards and for proposed THz-based cell manipulation. The study is not circular: the FRAP recovery time constant is measured from the data, and no equation is constructed to force the reported conclusion. The main significance therefore rests on the reliability of the temperature correction and on the statistical treatment of the repeated FRAP measurements.","major_comments":[{"comment":"The non-thermal interpretation rests entirely on the calibrated sample temperature Ts. The 13-µm thermocouple is placed at the bottom of the dish, while the THz beam is absorbed mainly by the cell layer on the upper side of the 0.15-mm film; the thermal resistance of the film means the sensor can read lower than the actual membrane temperature. If ΔTs is underestimated, the THz data are compared at too-low Ts values, and a purely thermal elevation of D would appear as a non-thermal effect. Table 1 reinforces this concern: 0.48 THz at 20 mW/cm2 with α ≈ 75 cm⁻¹ is reported to give ΔTs = 2.4–2.8 °C, essentially the same as 0.10 THz at 89 mW/cm2 with α ≈ 37 cm⁻¹ (ΔTs = 2.5–3.0 °C), although a simple absorbed-power estimate gives roughly half the temperature rise for 0.48 THz. This inconsistency suggests that the calibration does not faithfully report membrane temperature. I request an independent membrane-temperature measurement (for example, a temperature-sensitive fluorescent dye or a thermocouple at the upper film surface) or a quantitative thermal model, and a discussion of how the reported ΔTs values are consistent with the stated power densities and absorption coefficients.","section":"Methods, 'Calibration of cell temperature'; Table 1"},{"comment":"The statistical significance of the central comparison is not established as reported. The Methods state that approximately 20 points were measured without THz and 20 with THz on each sample, alternating ON/OFF, with more than four samples per condition. If all individual FRAP measurements are pooled as independent observations for the t-test in Fig. 3, the analysis suffers from pseudoreplication, making the p-values anti-conservative. The paper should report the number of cells or independent samples per condition, average the repeated measurements by sample before hypothesis testing, or use a mixed-effects model that accounts for repeated measures.","section":"Methods, 'FRAP measurement'; Fig. 3"},{"comment":"The title and abstract state that THz irradiation induces an order-disorder phase transition of the membrane lipids, but the measured observable is only the lateral diffusion coefficient. An increase in D at temperatures below the growth temperature is consistent with an order-disorder transition, but it is also consistent with other mechanisms such as altered hydration, changed cytoskeletal coupling, or membrane remodeling. No direct order parameter (for example, Laurdan generalized polarization, 2H NMR order parameter, or calorimetry) is measured under THz irradiation. I recommend either softening the phase-transition conclusion or adding a direct order-parameter measurement; as written, the interpretive step from increased D to 'order-disorder phase transition' is an unsupported assumption.","section":"Discussion; Fig. 4"}],"minor_comments":[{"comment":"Lengths that should be micrometers are written as millimeters: the ROI diameter is approximately 6 mm, the mode-field diameter is 4.6 mm, the cell thickness is 10–30 mm, the image scale is 0.075 mm/pixel, the THz penetration depths are 117/73/58 mm, and D is reported in mm2/s. These values are physically implausible as printed, and they affect the numerical value of the central observable D through the formula D = ω2/4τ. Please correct all units to µm and report D in µm2/s.","section":"Throughout the manuscript"},{"comment":"The text says the fluorescence intensity was 'fitted to an exponential decay function, and the decay rate τ was calculated,' but FRAP recovery is an increase in fluorescence, not a decay. Please clarify the fitted recovery model (for example, F(t) = A(1 − exp(−t/τ)) with a mobile fraction) so that the relationship between the fitted time constant and the reported D is unambiguous.","section":"Methods, 'FRAP measurement'"},{"comment":"The figures would benefit from explicit scale bars with correct units and from axis labels with units on the intensity-recovery plot; the current captions leave the reader to infer the length and time scales from the text.","section":"Fig. 1 and Fig. 5"},{"comment":"The reported ΔTs ranges (for example, 2.5–3.0 °C) have no associated uncertainty, number of replicate measurements, or description of how the range was estimated; adding these details would help the reader assess the calibration's reliability.","section":"Table 1"}],"recommendation":"major_revision","confidential_remarks":null},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"The genuinely new thing here is the THz-FRAP combination: living HeLa cells exposed to cw THz at three frequencies, with FRAP readout of lateral diffusion. That is a reasonable way to ask whether sub-THz photons affect membranes beyond heating. The alternating ON/OFF design and multiple samples are decent, and the frequency-dependent pattern (0.10 and 0.29 THz show an effect, 0.48 THz does not) is interesting if it holds.\n\nThe big problem is the temperature calibration. The 13-µm thermocouple sits at the bottom of the dish, while the cells adhere to the upper side of the 0.15-mm film. The THz beam is absorbed mainly by cells and water, and the film is in the thermal path. If the thermocouple reads cooler than the membrane, the correction undercorrects and the non-thermal claim evaporates. Table 1 makes this concrete: 0.48 THz at 20 mW/cm² gives ΔTs ≈ 2.4–2.8 °C, nearly the same as 0.10 THz at 89 mW/cm² (2.5–3.0 °C), even though the absorbed power at 0.48 THz should be roughly half. That inconsistency suggests the sensor is not faithfully reporting membrane temperature. Since the whole conclusion rests on that correction, this is load-bearing, not a footnote.\n\nSecond, the 0.48 THz null is not clean: it is also the lowest power (20 mW/cm² vs 89 and 65). So the frequency dependence could just be power dependence. The authors acknowledge this, but it still weakens the claim.\n\nThird, the manuscript uses mm everywhere it should use µm (ROI diameter ~6 mm, mode field diameter 4.6 mm, cell thickness 10–30 mm, D in mm²/s). That is obviously a units typo, but it makes the paper look careless and needs fixing.\n\nThe phase-transition title is stronger than the data. Diffusion constants going up at low temperature is consistent with increased fluidity, but calling it an order–disorder phase transition requires direct membrane-order evidence (e.g., Laurdan or 2H NMR). The bound-water mechanism is speculative, though Sugiyama et al. gives it plausibility.\n\nWho is this for? People working on THz bioeffects, membrane biophysics, and safety standards. They will find the method section useful and the question worth pursuing. It deserves a serious referee, but the referee should demand either a power-matched frequency comparison or an independent membrane-temperature measurement (e.g., a temperature-sensitive fluorophore) before the non-thermal claim is accepted. My recommendation: send it to peer review with the expectation of major revision. The core experimental idea is sound, the execution is mostly reasonable, but the central claim needs stronger calibration evidence.","headline":"New THz-FRAP combination with a plausible but not yet proven non-thermal membrane effect; the temperature calibration is the load-bearing uncertainty.","tokens_in":8107,"tokens_out":2009,"would_cite":true,"duration_ms":18817,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"THz irradiation increases cell-membrane diffusion without heating","keywords":["THz irradiation","non-thermal effects","cell membrane","order-disorder phase transition","fluorescence recovery after photobleaching","lateral diffusion","bound water","HeLa cells"],"falsifier":"Measure the cell-membrane temperature directly with a ratiometric fluorescent temperature probe during 0.10 THz irradiation: if the directly measured membrane temperature rise, combined with the control diffusion-versus-temperature curve, fully explains the diffusion increase, the non-thermal phase-transition claim is falsified. Conversely, if a direct lipid-order probe such as Laurdan generalized polarization drops under THz at constant measured temperature, the claim is supported.","tokens_in":1607,"feed_emoji":"🔬","tokens_out":2450,"duration_ms":66677,"temperature":0.7,"pith_summary":"This paper asks whether terahertz radiation can change living cell membranes through a non-thermal mechanism. Using fluorescence recovery after photobleaching (FRAP) under continuous THz irradiation, the authors measured lipid lateral diffusion in HeLa cell membranes at 0.10, 0.29, and 0.48 THz. After correcting for THz-induced warming with an ultrathin thermocouple, they report that 0.10 and 0.29 THz irradiation increases the diffusion constant at sample temperatures below the cell growth temperature, while 0.48 THz shows no significant effect. They interpret this as THz-driven excitation of bound water molecules that destabilizes the ordered lipid phase, effectively melting part of the membrane. If correct, this would mean safety standards for THz exposure and future cell-manipulation tools must account for non-thermal membrane effects.","feed_headline":"THz light loosens cell membranes without heating them","feed_subtitle":"At 0.10 and 0.29 THz, lipid diffusion in HeLa membranes rises below growth temperature—a non-thermal phase-transition hint.","key_machinery":"The central tool is THz-FRAP: fluorescence recovery after photobleaching performed while the sample is irradiated from below with a focused THz beam. A small spot of membrane fluorophores is bleached with a laser, and the recovery rate $\\tau$ of fluorescence yields the lateral diffusion constant $D = \\omega^2/(4\\tau)$, where $\\omega$ is the bleached-spot radius. The load-bearing companion is a temperature calibration in which an ultrathin (13 µm) thermocouple at the dish bottom measures the sample temperature during THz irradiation, letting the authors subtract thermal effects and attribute residual diffusion changes to non-thermal THz action. Mechanistically, the paper leans on the idea that sub-THz radiation excites the relaxation dynamics of hydration water around the lipid headgroups, mirroring earlier observations on protein hydration water.","core_discovery":"The paper claims that irradiating HeLa cells with continuous-wave THz light at 0.10 and 0.29 THz raises the lateral diffusion constant of membrane lipids when the sample temperature is below about 37 °C, even after the temperature increase caused by THz absorption is subtracted using a calibrated thermocouple. The effect is frequency-dependent: no significant diffusion increase is observed at 0.48 THz at the tested power density. The authors propose that sub-THz photons excite the dynamics of water molecules bound to the lipid bilayer, allowing water to penetrate hydrophilic regions and drive the membrane from an ordered (gel-like) state toward a disordered (fluid) state. This is presented as direct evidence for a non-thermal, water-mediated order-disorder phase transition in a living cell membrane.","pith_inferences":["A direct test the paper leaves implicit: the same 0.1–0.3 THz window should lower Laurdan generalized polarization in HeLa membranes at fixed temperature, because the proposed mechanism is a decrease in lipid order.","If bound-water relaxation drives the effect, it should be hydration-dependent; dehydrating the membrane surface should suppress the diffusion increase, while excess hydration should enhance it.","An independent optical measurement of membrane temperature during irradiation, for example with a ratiometric fluorescent probe, would settle whether the thermocouple calibration fully accounts for local warming.","The frequency window around 0.1–0.3 THz may correspond to a specific relaxation mode of membrane-bound water; if so, the effect should scale with absorbed power in that mode rather than with total power density."],"forward_implications":["THz exposure at 0.10 and 0.29 THz increases lipid lateral diffusion in HeLa membranes below the growth temperature, even after correcting for sample heating.","The absence of an effect at 0.48 THz indicates a frequency-dependent non-thermal response rather than a simple heating artifact.","Membrane properties that depend on lipid order, such as permeability, protein mobility, adhesion, and signaling, could be altered by sub-THz irradiation without a measurable bulk temperature rise.","Safety standards for THz radiation may need to include non-thermal membrane effects, not just tissue heating.","The same mechanism could become a contactless way to manipulate membrane fluidity in cultured cells or artificial lipid systems."],"supporting_citations":[{"why":"Supplies the precedent that sub-THz irradiation non-thermally accelerates hydration-water dynamics around biomolecules, the mechanism this paper extends to lipid membranes.","marker":"[20]"},{"why":"Establishes that lipid bilayers and living cells exhibit broad phase transitions a few degrees below the growth temperature, anchoring the interpretation of the diffusion increase as a phase transition.","marker":"[23]"},{"why":"Reports the phase transition of HeLa cell membranes measured by Laurdan fluorescence, providing the specific cell system's membrane behavior.","marker":"[24]"},{"why":"Links lipid order to the quantity and dynamics of water molecules at the glycerol backbone, supporting the proposed water-mediated mechanism.","marker":"[26]"},{"why":"Provides frequency-dependent water relaxation modes in hydrated membranes, used to explain why 0.48 THz is less effective than 0.10–0.29 THz.","marker":"[27]"},{"why":"Supplies the absorbance values of liquid water at 0.10, 0.29, and 0.48 THz used to estimate penetration depth into cells.","marker":"[28]"},{"why":"Supplies the FRAP method for measuring lateral diffusion of membrane molecules, the experimental core of the paper.","marker":"[21]"}],"fun_headline_variants":["THz waves flip cell membranes to fluid state without heat","Low-THz light triggers membrane phase shift in cells","Water-mediated membrane melting by THz irradiation","THz radiation non-thermally restructures cell membranes"],"cache_read_input_tokens":10240,"weakest_assumption_plain":"The thin thermocouple at the bottom of the dish is assumed to measure the true temperature of the cell membrane during THz exposure; if it under-reads the membrane temperature, the observed diffusion increase could be ordinary heating.","fun_headline_variants_meta":{"raw":{"variants":["THz waves flip cell membranes to fluid state without heat","Low-THz light triggers membrane phase shift in cells","Water-mediated membrane melting by THz irradiation","THz radiation non-thermally restructures cell membranes"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000529,"raw_usage":{"total_tokens":2531,"prompt_tokens":908,"completion_tokens":1623,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":524,"completion_tokens_details":{"reasoning_tokens":1559}},"tokens_in":524,"tokens_out":1623,"duration_ms":11584,"temperature":1.0,"reasoning_tokens":1559,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-11T19:44:10.475496+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Measure the cell-membrane temperature directly with a ratiometric fluorescent temperature probe during 0.10 THz irradiation: if the directly measured membrane temperature rise, combined with the control diffusion-versus-temperature curve, fully explains the diffusion increase, the non-thermal phase-transition claim is falsified. Conversely, if a direct lipid-order probe such as Laurdan generalized polarization drops under THz at constant measured temperature, the claim is supported.","supporting_citations":[{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Supplies the precedent that sub-THz irradiation non-thermally accelerates hydration-water dynamics around biomolecules, the mechanism this paper extends to lipid membranes."},{"cited_title":"Phase transitions in biological membranes","cited_arxiv_id":null,"evidence_quote":"Establishes that lipid bilayers and living cells exhibit broad phase transitions a few degrees below the growth temperature, anchoring the interpretation of the diffusion increase as a phase transition."},{"cited_title":"& Westerhausen, C","cited_arxiv_id":null,"evidence_quote":"Reports the phase transition of HeLa cell membranes measured by Laurdan fluorescence, providing the specific cell system's membrane behavior."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Links lipid order to the quantity and dynamics of water molecules at the glycerol backbone, supporting the proposed water-mediated mechanism."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Provides frequency-dependent water relaxation modes in hydrated membranes, used to explain why 0.48 THz is less effective than 0.10–0.29 THz."},{"cited_title":"& Ogawa, Y","cited_arxiv_id":null,"evidence_quote":"Supplies the absorbance values of liquid water at 0.10, 0.29, and 0.48 THz used to estimate penetration depth into cells."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Supplies the FRAP method for measuring lateral diffusion of membrane molecules, the experimental core of the paper."}],"review_version":1}