Label-free mid-infrared photothermal microscopy revisits intracellular thermal dynamics: what do fluorescent nanothermometers measure?

TL;DR

This study uses label-free mid-infrared photothermal microscopy to measure intracellular temperature dynamics, clarifying the nature of heat conduction in cells and comparing it with fluorescent nanothermometry results.

Contribution

It demonstrates that intracellular heat conduction is water-like and distinguishes between true temperature changes and slow signals in nanothermometry.

Findings

Intracellular thermal diffusivity is 93-94% that of water.

Label-free thermometry shows rapid temperature responses consistent with water-like conduction.

Fluorescent nanothermometers exhibit an additional slow variation not related to temperature.

Abstract

Fluorescent nanothermometry has revealed pronounced intracellular temperature heterogeneity, establishing the field of single-cell thermal biology. However, these observations have sparked a controversy known as the "10^5 gap issue", because heat conduction calculations in aqueous environments predict that such large temperature distributions cannot be sustained within cells. Here, we address this issue using label-free mid-infrared photothermal microscopy. This technique quantifies heat-induced temperature changes under local thermal equilibrium (LTE), in accordance with the conventional thermodynamic and statistical-mechanical definition of temperature, by detecting refractive index variations. From transient thermal decay measurements, we determined that intracellular thermal diffusivity corresponds to 93-94% that of water. This result indicates that intracellular heat conduction is essentially water-like and rules out the hypothesis that anomalously slow intracellular heat conduction underlies the 10^5 gap discrepancy. We then directly compared fluorescent nanothermometry with our label-free thermometry. Under seconds-long heating, the label-free method exhibited a rapid temperature response consistent with water-like heat conduction. In contrast, fluorescent nanothermometers showed not only a similarly fast response but also an additional slow variation that was absent in the label-free readout. This slow component cannot be explained by temperature changes defined under LTE, but instead likely reflects slower intracellular processes not governed by conductive heat transfer. These results suggest that the "10^5 gap issue" stems from comparing two fundamentally distinct physical quantities: the LTE-defined temperature and a slowly-varying, long-lived non-conductive signal detected by fluorescent nanothermometers.