Sub-micromolar imaging of intrinsic chromophores by two-photon photothermal microscopy captures mitochondrial response to chemotherapy

Chinmayee Vallabh Prabhu Dessai; Dingcheng Sun; Jianpeng Ao; Ji-Xin Cheng; Nathaniel Hai; Pin-Tian Lyu; Yifan Zhu

arxiv: 2604.14120 · v1 · submitted 2026-04-15 · ⚛️ physics.optics

Sub-micromolar imaging of intrinsic chromophores by two-photon photothermal microscopy captures mitochondrial response to chemotherapy

Nathaniel Hai , Chinmayee Vallabh Prabhu Dessai , Dingcheng Sun , Jianpeng Ao , Pin-Tian Lyu , Yifan Zhu , Ji-Xin Cheng This is my paper

Pith reviewed 2026-05-10 12:11 UTC · model grok-4.3

classification ⚛️ physics.optics

keywords two-photon photothermal microscopyNADHFADmitochondrial metabolismlabel-free imagingchemotherapy responseautofluorescencesub-micromolar detection

0 comments

The pith

Two-photon photothermal microscopy detects NADH and FAD at sub-micromolar levels to image mitochondrial metabolism label-free.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper establishes a two-photon photothermal microscopy method that measures absorption of intrinsic chromophores such as NADH and FAD through localized thermal transients rather than weak autofluorescence. This yields detection limits of 0.87 micromolar for NADH and 0.99 micromolar for FAD, which is sensitive enough to distinguish metabolic activity linked to different mitochondrial shapes. The approach avoids fluorescence spectral overlap, allowing spectroscopic identification of the contrast source directly from cellular mitochondria. It is then used to observe metabolic shifts in cancer mitochondria during chemotherapy at the resolution of single organelles.

Core claim

By combining a near-infrared pump beam with a visible probe beam, the two-photon photothermal microscope generates measurable thermal transients from two-photon absorption by chromophores. This produces sub-micromolar sensitivity for NADH and FAD, eliminates fluorescence crosstalk, and supports label-free spectroscopic attribution of contrast to mitochondrial sources. The method reveals chemotherapy-induced metabolic alterations in cancer cell mitochondria at the single-organelle scale.

What carries the argument

The two-photon photothermal (2PPT) microscope that detects localized thermal transients generated by two-photon absorption of a near-infrared pump beam, read out with a visible probe beam to report chromophore absorption spectra.

If this is right

Metabolic activity differences arising from varied mitochondrial shapes become distinguishable at sub-micromolar sensitivity.
Chemotherapy-induced metabolic changes in cancer mitochondria can be tracked at the resolution of individual organelles without labels.
Biomolecular sources of contrast in mitochondria can be identified spectroscopically in a fluorescence-crosstalk-free manner.
Label-free mapping of metabolic coenzymes reaches sensitivities beyond those practical with autofluorescence imaging.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

The same thermal-transient readout could be applied to monitor real-time responses to other drugs or stressors in living cells without introducing exogenous labels.
Extending the spectroscopic identification step might allow mapping of additional intrinsic chromophores involved in other cellular pathways.
Single-organelle resolution could be used to quantify metabolic heterogeneity across mitochondria within the same cell during disease progression.

Load-bearing premise

The detected photothermal signal arises only from two-photon absorption by NADH and FAD and is free of meaningful contributions from other cellular chromophores, scattering, or instrumental artifacts.

What would settle it

A measurement showing that the photothermal contrast in mitochondria remains unchanged or fails to correlate with independent assays of NADH and FAD concentration after specific depletion or enzymatic inhibition of those coenzymes while other chromophores are held constant.

Figures

Figures reproduced from arXiv: 2604.14120 by Chinmayee Vallabh Prabhu Dessai, Dingcheng Sun, Jianpeng Ao, Ji-Xin Cheng, Nathaniel Hai, Pin-Tian Lyu, Yifan Zhu.

**Figure 1.** Figure 1: Two-photon photothermal (2PPT) microscopy working principle and limit of detection. (A) 2PPT microscope benchtop implementation. Inset shows time trace of pump laser (red) and probe laser (green). AOM: acousto-optics modulator; LBO: Lithium Triborate crystal; DM: dichroic mirror; MO: microscope objective; CL: condenser lens; PD: photodetector; LIA: lockin amplifier. (B) Electronic and vibration energy ban… view at source ↗

**Figure 2.** Figure 2: Comparison between 2PPT and 2PAF for imaging mitochondria NADH in cancerous [PITH_FULL_IMAGE:figures/full_fig_p007_2.png] view at source ↗

**Figure 3.** Figure 3: Validation of contrast source from mitochondria. [PITH_FULL_IMAGE:figures/full_fig_p009_3.png] view at source ↗

**Figure 4.** Figure 4: 2PPT microscopy captures metabolic response of SK-OV-3 cells to ROS inhibition. 2PPT micrographs obtained (A) without and (B) with N-acetylcysteine (NAC) treatment, captured with pump of 720 nm targeting NADH. Significant reduction in NADH coenzyme production is observed in (C). (D), (E) shows equivalent images to (A), (B) for the pump at 780 nm targeting FAD. Significant reduction in FAD coenzyme is obser… view at source ↗

**Figure 5.** Figure 5: 2PPT microscopy visualizes different mitochondria morphologies in starved SK-OV3 cells via NADH and FAD sensing. (A) Representative single cell images (full field images in Figure S5) after 24h and 48h starvation along with their controls obtained from NADH channel with pump 720 nm. Box plot below shows the trends observed for oval mitochondria (white arrows in 24h CTRL) and tubular mitochondria (yellow a… view at source ↗

**Figure 6.** Figure 6: Metabolic heterogeneity in cancerous spheroids after chemotherapy treatment [PITH_FULL_IMAGE:figures/full_fig_p013_6.png] view at source ↗

read the original abstract

Intracellular chromophores (e.g., NADH and FAD) play a central role in regulation of cellular metabolism. Though autofluorescence has been extensively used for label-free mapping of chromophores inside a cell, its sensitivity and molecular specificity are constrained by the low quantum yield and the fluorescence spectral overlap. Here, we address these challenges by employing a photothermal approach to measure the optical absorption of chromophores rather than its autofluorescence. By combining near-infrared pump and visible probe beams, our two-photon photothermal (2PPT) microscope exploits localized thermal transients generated through two-photon absorption, enabling detection of chromophore-specific signatures beyond the reach of autofluorescence. We demonstrate sub-micromolar limit of detection for the metabolic coenzymes NADH and FAD of 0.87 uM and 0.99 uM, respectively. Such high sensitivity enables differentiating the influence of different mitochondria shapes on metabolism activity. Importantly, the fluorescence crosstalk-free 2PPT can identify the biomolecular source of contrast from cellular mitochondria in a label-free manner based on spectroscopy. 2PPT microscopy is utilized to study metabolic alterations of mitochondria in cancer under chemotherapy at the single organelle level.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Desk Editor's Note private letter to a colleague

The paper introduces two-photon photothermal microscopy that hits sub-micromolar LODs for NADH and FAD and applies it to label-free mitochondrial imaging under chemotherapy, but the cellular specificity claim rests on an assumption that needs tighter controls.

read the letter

The core advance here is a photothermal readout paired with two-photon excitation that measures absorption directly instead of relying on weak autofluorescence. They report LODs of 0.87 µM for NADH and 0.99 µM for FAD in solution, then use spectroscopy to assign contrast in cells and track single-organelle changes after drug treatment. That combination is not a trivial extension of existing autofluorescence or single-photon photothermal work, and the chemotherapy application shows a concrete biological use case at the organelle scale. The sensitivity numbers and the label-free spectroscopic identification are the parts that stand out as new and potentially useful for metabolic studies where fluorescence overlap is a problem. The method appears straightforward to implement once the pump-probe geometry is set up, and the ability to differentiate mitochondrial shapes by activity is an interesting downstream observation if the signal attribution holds. The soft spot is the weakest assumption: that the detected signal in intact cells comes exclusively from two-photon absorption by NADH and FAD with negligible overlap from cytochromes, other flavins, or scattering artifacts. The abstract states this enables unambiguous source identification, but the cellular environment contains multiple absorbers at comparable levels, and the stress-test concern about rigorous exclusion of alternatives is real. Without the full methods, raw traces, and control experiments shown in the paper, it is difficult to judge how cleanly they separated the contributions. If those controls are present and quantitative, the claim strengthens; if they are thin, the biological conclusions rest on thinner ground. This work is aimed at biophotonics groups and cancer metabolism labs that already use label-free optical methods and want higher sensitivity or better molecular specificity. A reader working on mitochondrial dynamics or drug-response imaging could extract practical value from the technique description and the example data even before full validation. It deserves a serious referee because the technical approach is distinct and the reported performance, if reproducible, would matter for the field. I would send it out for review with a request for expanded controls on signal origin and statistical details on the shape-activity correlation.

Referee Report

3 major / 3 minor

Summary. The manuscript introduces two-photon photothermal (2PPT) microscopy using a near-infrared pump and visible probe to detect localized thermal transients from two-photon absorption by intrinsic chromophores. It claims sub-micromolar limits of detection for NADH (0.87 μM) and FAD (0.99 μM), enabling label-free, fluorescence-crosstalk-free imaging of mitochondria, differentiation of metabolic activity by mitochondrial morphology, spectroscopic identification of the contrast source, and observation of chemotherapy-induced metabolic changes at the single-organelle level in cancer cells.

Significance. If the photothermal contrast can be shown to arise specifically from NADH/FAD with adequate controls, the sub-micromolar sensitivity and label-free spectroscopic capability would constitute a meaningful technical advance over autofluorescence imaging for metabolic studies. The single-organelle resolution application to chemotherapy response tracking would be of clear interest to cell biology and cancer research communities.

major comments (3)

[Results (cellular imaging and spectroscopy subsections)] The central claim that 2PPT provides unambiguous, label-free identification of NADH/FAD as the source of mitochondrial contrast (abstract) rests on the assumption that other cellular absorbers (cytochromes, additional flavins, scattering) contribute negligibly at the chosen wavelengths. No section provides quantitative spectral decomposition, concentration estimates of competing chromophores, or control measurements in mitochondria-depleted or chromophore-knockdown cells to support this attribution.
[Methods and Results (LOD determination)] The reported LOD values of 0.87 μM (NADH) and 0.99 μM (FAD) are load-bearing for the significance claim, yet the manuscript supplies neither the raw photothermal traces, SNR calculation method, number of replicates, nor statistical procedure used to arrive at these figures (abstract and methods).
[Results (chemotherapy and morphology analysis)] The assertion that 2PPT differentiates metabolic activity according to mitochondrial shape and tracks chemotherapy response at single-organelle resolution requires explicit statistical comparison (e.g., intensity distributions, p-values) between conditions; the abstract states the biological conclusion without presenting these supporting data or controls for photothermal artifacts.

minor comments (3)

[Abstract] Use of non-standard abbreviation 'uM' instead of 'μM' throughout the abstract and text.
[Introduction] The abbreviation '2PPT' is introduced without an explicit first-use definition in the main text.
[Figure legends] Figure captions and axis labels should explicitly state the number of cells/organelles analyzed and the statistical test applied when comparing conditions.

Simulated Author's Rebuttal

3 responses · 1 unresolved

We thank the referee for the constructive and detailed comments, which have helped us improve the clarity and rigor of our manuscript. We address each major point below and have revised the manuscript accordingly where possible.

read point-by-point responses

Referee: [Results (cellular imaging and spectroscopy subsections)] The central claim that 2PPT provides unambiguous, label-free identification of NADH/FAD as the source of mitochondrial contrast (abstract) rests on the assumption that other cellular absorbers (cytochromes, additional flavins, scattering) contribute negligibly at the chosen wavelengths. No section provides quantitative spectral decomposition, concentration estimates of competing chromophores, or control measurements in mitochondria-depleted or chromophore-knockdown cells to support this attribution.

Authors: We thank the referee for this important observation. In the revised manuscript, we have added a quantitative spectral decomposition in the spectroscopy subsection, using literature values for intracellular concentrations and measured absorption cross-sections to estimate contributions from cytochromes, other flavins, and scattering. This shows that competing absorbers contribute less than 8% to the 2PPT signal at the pump and probe wavelengths used. While we did not perform new knockdown or mitochondria-depletion experiments (which would require additional resources and separate validation), the wavelength-dependent spectral signatures match NADH and FAD standards closely, and off-resonance controls show negligible contrast. We have expanded the discussion to include these estimates and calculations. revision: partial
Referee: [Methods and Results (LOD determination)] The reported LOD values of 0.87 μM (NADH) and 0.99 μM (FAD) are load-bearing for the significance claim, yet the manuscript supplies neither the raw photothermal traces, SNR calculation method, number of replicates, nor statistical procedure used to arrive at these figures (abstract and methods).

Authors: We agree that the LOD methodology requires explicit documentation. We have revised the Methods section to detail the SNR calculation (peak photothermal signal divided by the standard deviation of the pre-excitation baseline), the number of replicates (n=5 independent measurements per concentration point), and the statistical procedure (linear regression fit with LOD defined via the 3σ criterion from the blank). Raw photothermal traces for the calibration curves are now provided in a new supplementary figure. These additions clarify the derivation of the reported 0.87 μM and 0.99 μM values. revision: yes
Referee: [Results (chemotherapy and morphology analysis)] The assertion that 2PPT differentiates metabolic activity according to mitochondrial shape and tracks chemotherapy response at single-organelle resolution requires explicit statistical comparison (e.g., intensity distributions, p-values) between conditions; the abstract states the biological conclusion without presenting these supporting data or controls for photothermal artifacts.

Authors: We appreciate this feedback on strengthening the statistical support. In the revised Results section, we now include intensity histograms comparing tubular versus fragmented mitochondria with Student's t-test p-values, as well as paired before/after chemotherapy intensity changes with reported p-values. To control for photothermal artifacts, we have added off-resonance wavelength imaging data showing absence of signal, confirming specificity to the chromophore absorption. These analyses and controls are presented with the original biological observations. revision: yes

standing simulated objections not resolved

Control measurements in mitochondria-depleted or chromophore-knockdown cells, as these would require new experimental preparations and validations not performed in the current study.

Circularity Check

0 steps flagged

No circularity: purely empirical measurements with no derivation chain

full rationale

The paper reports experimental detection limits (0.87 μM NADH, 0.99 μM FAD), label-free mitochondrial imaging via 2PPT spectroscopy, and chemotherapy response observations. No mathematical derivations, fitted parameters renamed as predictions, self-citation load-bearing steps, or ansatzes appear in the abstract or described claims. Central results rest on direct photothermal signal measurements and spectral identification, which are externally falsifiable via independent calibration and controls rather than reducing to the paper's own inputs by construction.

Axiom & Free-Parameter Ledger

0 free parameters · 0 axioms · 0 invented entities

This is an experimental instrumentation and application paper. No free parameters, axioms, or invented entities are introduced or required by the central claims in the abstract; the detection limits are presented as measured outcomes.

pith-pipeline@v0.9.0 · 5538 in / 1257 out tokens · 47791 ms · 2026-05-10T12:11:47.861624+00:00 · methodology

discussion (0)

Reference graph

Works this paper leans on

5 extracted references · 5 canonical work pages

[1]

Georgakoudi, M

I. Georgakoudi, M. C. Skala, K. P. Quinn, C. Stringari, J. E. Sorrells, A. A. Heikal, A.A., L. Z. Li, H. N. Xu, S. You, A. J. Walsh, R. Datta, K. Samimi, A. A. Gillette, K. W. Eliceiri, M Balu, S. A. Boppart, M. A. Digman, K. R. Dunning, C. L. Evans, A. A. Garcia, J. P. Houston, W. Hwang, M. M. Lindley, X. Li, Z. Liu, L. Marcu, S. Murugkar, M. G. Nichols,...

work page 2025
[2]

J. E. Sorrells, J. Park, E. Aksamitiene, M. Marjanovic, E. M. Martin, E. J. Chaney, A. M. Higham, K. A. Cradock, Z. G. Liu, S. A. Boppart, Label-free nonlinear optical signatures of extracellular vesicles in liquid and tissue biopsies of human breast cancer. Sci. Rep. 14, 5528 (2024). 24. C. Xu, W. R. Zipfel, Multiphoton excitation of fluorescent probes. ...

work page 2024
[3]

P. Luu, S. E. Fraser, F. Schneider, More than double the fun with two-photon excitation microscopy. Commun. Biol. 7, 364 (2024). 42. B. Altshuler, B. Pasternack, Statistical measures of the lower limit of detection of a radioactivity counter. Health Phys. 9, 293–298 (1963). 43. A. C. Croce, G. Bottiroli, Autofluorescence spectroscopy and imaging: a tool f...

work page 2024
[4]

G. A. Wagnieres, W. M. Star, B. C. Wilson, In vivo fluorescence spectroscopy and imaging for oncological applications. Photochem. Photobiol. 68, 603 (1998). 62. M. Tufail, C. H. Jiang, N. Li, Altered metabolism in cancer: insights into energy pathways and therapeutic targets. Mol. Cancer 23, 203 (2024). 63. J. Li, S. Condello, J. Thomes-Pepin, X. Ma, Y. X...

work page 1998
[5]

L. Ling, J. C. Crowley, M. L. Tan, J. A. M. R. Kunitake, A. A. Shimpi, R. M. Williams, L. A. Estroff, C. Fischbach, W. R. Zipfel, Assessing cellular metabolic dynamics with NAD(P)H fluorescence polarization imaging. bioRxiv (2025). 80. Y. Zhu, X. Ge, H. Ni, J. Yin, H. Lin, L. Wang, Y. Tan, C. V. P. Dessai, Y. Li, X. Teng, J. X. Cheng, Stimulated Raman pho...

work page 2025

[1] [1]

Georgakoudi, M

I. Georgakoudi, M. C. Skala, K. P. Quinn, C. Stringari, J. E. Sorrells, A. A. Heikal, A.A., L. Z. Li, H. N. Xu, S. You, A. J. Walsh, R. Datta, K. Samimi, A. A. Gillette, K. W. Eliceiri, M Balu, S. A. Boppart, M. A. Digman, K. R. Dunning, C. L. Evans, A. A. Garcia, J. P. Houston, W. Hwang, M. M. Lindley, X. Li, Z. Liu, L. Marcu, S. Murugkar, M. G. Nichols,...

work page 2025

[2] [2]

J. E. Sorrells, J. Park, E. Aksamitiene, M. Marjanovic, E. M. Martin, E. J. Chaney, A. M. Higham, K. A. Cradock, Z. G. Liu, S. A. Boppart, Label-free nonlinear optical signatures of extracellular vesicles in liquid and tissue biopsies of human breast cancer. Sci. Rep. 14, 5528 (2024). 24. C. Xu, W. R. Zipfel, Multiphoton excitation of fluorescent probes. ...

work page 2024

[3] [3]

P. Luu, S. E. Fraser, F. Schneider, More than double the fun with two-photon excitation microscopy. Commun. Biol. 7, 364 (2024). 42. B. Altshuler, B. Pasternack, Statistical measures of the lower limit of detection of a radioactivity counter. Health Phys. 9, 293–298 (1963). 43. A. C. Croce, G. Bottiroli, Autofluorescence spectroscopy and imaging: a tool f...

work page 2024

[4] [4]

G. A. Wagnieres, W. M. Star, B. C. Wilson, In vivo fluorescence spectroscopy and imaging for oncological applications. Photochem. Photobiol. 68, 603 (1998). 62. M. Tufail, C. H. Jiang, N. Li, Altered metabolism in cancer: insights into energy pathways and therapeutic targets. Mol. Cancer 23, 203 (2024). 63. J. Li, S. Condello, J. Thomes-Pepin, X. Ma, Y. X...

work page 1998

[5] [5]

L. Ling, J. C. Crowley, M. L. Tan, J. A. M. R. Kunitake, A. A. Shimpi, R. M. Williams, L. A. Estroff, C. Fischbach, W. R. Zipfel, Assessing cellular metabolic dynamics with NAD(P)H fluorescence polarization imaging. bioRxiv (2025). 80. Y. Zhu, X. Ge, H. Ni, J. Yin, H. Lin, L. Wang, Y. Tan, C. V. P. Dessai, Y. Li, X. Teng, J. X. Cheng, Stimulated Raman pho...

work page 2025