Introduction
The last decade has witnessed a fast-growing interest in manipulating matter by strong coupling to the so-called vacuum electromagnetic field (Garcia-Vidal et al., Reference Garcia-Vidal, Ciuti and Ebbesen2021; Nagarajan et al., Reference Nagarajan, Thomas and Ebbesen2021). For instance, strong coupling between electronic transitions of molecules and optical modes has shown great potential for modifying photochemistry (Hutchison et al., Reference Hutchison, Schwartz, Genet, Devaux and Ebbesen2012; Zeng et al., Reference Zeng, Perez-Sanchez, Eckdahl, Liu, Chang, Weiss, Kalow, Yuen-Zhou and Stern2023), energy transport (Zhong et al., Reference Zhong, Chervy, Wang, George, Thomas, Hutchison, Devaux, Genet and Ebbesen2016, Reference Zhong, Chervy, Zhang, Thomas, George, Genet, Hutchison and Ebbesen2017; Sandik et al., Reference Sandik, Feist, Garcia-Vidal and Schwartz2025) and charge carrier conductivity (Orgiu et al., Reference Orgiu, George, Hutchison, Devaux, Dayen, Doudin, Stellacci, Genet, Schachenmayer, Genes, Pupillo, Samori and Ebbesen2015; Nagarajan et al., Reference Nagarajan, George, Thomas, Devaux, Chervy, Azzini, Joseph, Jouaiti, Hosseini, Kumar, Genet, Bartolo, Ciuti and Ebbesen2020). On the other hand, strong coupling between vibrational transitions of molecules and the vacuum field, known as vibrational strong coupling (VSC), has also shown great potential for modifying ground-state processes, such as chemical reactivity (Thomas et al., Reference Thomas, George, Shalabney, Dryzhakov, Varma, Moran, Chervy, Zhong, Devaux, Genet, Hutchison and Ebbesen2016; Thomas et al., Reference Thomas, Lethuillier-Karl, Nagarajan, Vergauwe, George, Chervy, Shalabney, Devaux, Genet, Moran and Ebbesen2019; Pang et al., Reference Pang, Thomas, Nagarajan, Vergauwe, Joseph, Patrahau, Wang, Genet and Ebbesen2020; Nagarajan et al., Reference Nagarajan, Thomas and Ebbesen2021; Ahn et al., Reference Ahn, Triana, Recabal, Herrera and Simpkins2023; Patrahau et al., Reference Patrahau, Piejko, Mayer, Antheaume, Sangchai, Ragazzon, Jayachandran, Devaux, Genet, Moran and Ebbesen2024), supramolecular assemblies (Hirai et al., Reference Hirai, Ishikawa, Chervy, Hutchison and Uji2021; Joseph et al., Reference Joseph, Kushida, Smarsly, Ihiawakrim, Thomas, Paravicini-Bagliani, Nagarajan, Vergauwe, Devaux, Ersen, Bunz and Ebbesen2021; Sandeep et al., Reference Sandeep, Joseph, Gautier, Nagarajan, Sujith, Thomas and Ebbesen2022; Joseph et al., Reference Joseph, de Waal, Jansen, van der Tol, Vantomme and Meijer2024) and electrochemistry (Fukushima et al., Reference Fukushima, Yoshimitsu and Murakoshi2022, Reference Fukushima, Yoshimitsu and Murakoshi2023). The strong coupling conditions are usually achieved by collective coupling of a large number (N) of molecules with an optical mode of a Fabry–Perot cavity, which gives rise to two hybrid polaritonic states (upper and lower polaritonic states, UP and LP, respectively) and N − 1 dark collective states (DS). It is important to note that strong coupling occurs even in the dark due to the involvement of the electromagnetic fluctuations of the cavity (that is, the vacuum field). VSC can be achieved either by coupling directly to the cavity a vibration of the target molecule or indirectly by the so-called cooperative coupling where solvent and solute vibrations are coupled simultaneously to the same cavity mode (Lather et al., Reference Lather, Bhatt, Thomas, Ebbesen and George2019; Schütz et al., Reference Schütz, Schachenmayer, Hagenmüller, Brennen, Volz, Sandoghdar, Ebbesen, Genes and Pupillo2020).
In biology, VSC has also been harnessed to modify biochemical processes such as enzymatic activity in aqueous solution (Vergauwe et al., Reference Vergauwe, Thomas, Nagarajan, Shalabney, George, Chervy, Seidel, Devaux, Torbeev and Ebbesen2019; Lather and George, Reference Lather and George2021; Gao et al., Reference Gao, Guo, Si, Wang, Zhang and Yang2023; Gu et al., Reference Gu, Si, Li, Gao, Wang and Zhang2023). The ubiquity of water as the natural medium for biological activities makes it the optimal solvent for exploring biochemical processes. Furthermore, the VSC of water is typically in the ultra-strong regime (where the energy separation of UP and LP is larger than 10% of the vibrational band) (Fukushima et al., Reference Fukushima, Yoshimitsu and Murakoshi2021; Kadyan et al., Reference Kadyan, Suresh, Johns and George2024) and can be even observed in micro-sized water droplets without photonic structures (Canales et al., Reference Canales, Kotov, Kucukoz and Shegai2024). Thus, water is an excellent solvent to explore VSC-induced effects in biochemical and biophysical processes.
Double-stranded deoxyribonucleic acid (ds-DNA) is a key supramolecular structure in biological systems, comprising two sequence-complementary DNA strands held together in a double helix by non-covalent interactions, e.g., hydrogen bonding between nucleotide base pairs. The non-covalent interactions provide ds-DNA high thermodynamic stability in biological environments, which is crucial for their functionalities allowing DNAs to interact with a range of biological molecules (Norden and Kurucsev, Reference Norden and Kurucsev1994; Jensen et al., Reference Jensen, Orum, Nielsen and Norden1997; Vologodskii and Frank-Kamenetskii, Reference Vologodskii and Frank-Kamenetskii2018). The stability of ds-DNA helixes depends among other things on the composition and sequence of the individual DNA strands, which in turn determine the strength of several non-covalent interactions, such as hydrogen bonding, dipole–dipole, dipole-induced dipole and London dispersion forces (Devoe and Tinoco, Reference Devoe and Tinoco1962; Yakovchuk et al., Reference Yakovchuk, Protozanova and Frank-Kamenetskii2006; Vologodskii and Frank-Kamenetskii, Reference Vologodskii and Frank-Kamenetskii2018; Feng et al., Reference Feng, Sosa, Martensson, Jiang, Tong, Dorfman, Takahashi, Lincoln, Bustamante, Westerlund and Norden2019; Nordén, Reference Nordén2019). The stability of ds-DNA also depends on its solvation in solution, thus sensitive to the latter’s composition, salt concentration and pH (Vologodskii and Frank-Kamenetskii, Reference Vologodskii and Frank-Kamenetskii2018). The thermodynamic stability of ds-DNA can be inferred from their melting behaviour, that is, the dissociation process of ds-DNA into single-stranded (ss) DNA. As VSC has been shown to influence non-covalent interactions in chemical and biological systems (Joseph et al., Reference Joseph, Kushida, Smarsly, Ihiawakrim, Thomas, Paravicini-Bagliani, Nagarajan, Vergauwe, Devaux, Ersen, Bunz and Ebbesen2021; Sandeep et al., Reference Sandeep, Joseph, Gautier, Nagarajan, Sujith, Thomas and Ebbesen2022; Joseph et al., Reference Joseph, de Waal, Jansen, van der Tol, Vantomme and Meijer2024; Patrahau et al., Reference Patrahau, Piejko, Mayer, Antheaume, Sangchai, Ragazzon, Jayachandran, Devaux, Genet, Moran and Ebbesen2024), we sought to explore how VSC influences the stability of ds-DNA. Notably, it was recently reported that VSC affects the hybridisation process of DNA, with the thermal stability of the resulting DNA was analysed outside the optical cavity after VSC (Hou et al., Reference Hou, Gao, Zhong, Li, Zhu, Wang, Zhao and Zhang2024).
Probing ds-DNA melting inside a Fabry–Perot cavity (an optical cavity consisting of two parallel mirrors) using common spectroscopy techniques remains a significant challenge due to several issues, including the limited optical pathlengths of the infrared (IR) cavities (~10 μm), the low transmission in the UV–visible region due to the metallic mirrors (less than 10%) and the low sample concentrations usually used for biological samples (1–10 μM). Therefore, employing fluorescence spectroscopy represents a promising approach, thanks to its high sensitivity. In addition, fluorescence spectroscopy based on Förster resonance energy transfer (FRET) is widely used to study biological processes (Ha, Reference Ha2001; Quan et al., Reference Quan, Yi, Yang, He, Huang and Wang2020), including the denaturation of nucleic acids (Johansson et al., Reference Johansson, Fidder, Dick and Cook2002; Marras et al., Reference Marras, Kramer and Tyagi2002).
In this work, we used our newly developed microfluid cavities (Patrahau et al., Reference Patrahau, Piejko, Mayer, Antheaume, Sangchai, Ragazzon, Jayachandran, Devaux, Genet, Moran and Ebbesen2024) as a platform to probe the effects of VSC on the melting behaviour of ds-DNA under equilibrium conditions. These microfluid cavities made in fused silica substrates support multiple cavity modes in the IR region and allow for sufficient light outcoupling in the UV–visible region for luminescence measurements. Additionally, they offer several advantages, such as fixed and homogeneous pathlength everywhere in the cavity with good thermal stability against temperature variations. A set of Fabry–Perot cavities with different pathlengths allows us to choose the right one for coupling the vibrational modes of water and thereby study systematically the effects induced by VSC.
Methods
Sample preparation
The DNA oligonucleotides used in this work were purchased from Integrated DNA Technologies (IDT) with HPLC purification. The sequences used for all experiments were as follows:
Strand 1: 5′-FAM-ACTCGCACCTAGT-3′.
Strand 2: 5′-ACTAGGTGCGAGT-BHQ1-3′.
where FAM and BHQ1 on strand 1 and strand 2 are, respectively, a fluorescein emitting moiety and a so-called black hole quencher. FAM was chosen as a fluorescence reporter for its high extinction coefficient (~76,900 M−1·cm−1) and its near-unity fluorescence quantum yield (~0.93) (Sjöback et al., Reference Sjöback, Nygren and Kubista1995). BHQ-1 is a widely used quencher for FAM.
Oligonucleotides were rehydrated in Milli-Q water to a final concentration of ~500 μM. Concentrations were confirmed by UV absorbance at 260 nm using the extinction coefficients provided by IDT (141,560 M−1 cm−1 for strand 1 and 138,700 M−1 cm−1 for strand 2). A stock solution of ds-DNA was prepared by mixing equimolar amounts of strand 1 and strand 2 solution in a 10 mM sodium phosphate buffer (pH 7.0) in a final concentration of 10 μM. The mixture was heated between 65 and 75 °C, that is, above the melting temperature estimated for the ds-DNA system (Tm ≈ 50 °C), for 15 min using a heating block (Thermo Fisher Scientific, Inc.). The solution was then allowed to slowly cool down to room temperature to ensure the assembly of ds-DNA. Samples were either used immediately or stored at −4 °C.
To prepare 1 μM ds-DNA solution samples, the 10 μM ds-DNA stock solution was further diluted 10 times in 10 mM sodium phosphate buffer (pH 7.0). For ds-DNA samples in D2O/H2O (90%/10%) mixed solvent, the stock solution was diluted with D2O.
Fabry–Perot microfluid cavities
The microfluid cavities were fabricated by LioniX International and consist of fused silica substrates coated with 10 nm gold (Au) mirrors. An additional SiOx layer (100 nm) was coated to protect the Au mirrors and to prevent any potential fluorescence quenching by physical contact with the gold. Reference structures without Au mirrors were also prepared. Schematic illustrations of the microfluid cavities and the reference structures are shown in Supplementary Figure S1. More detailed descriptions about the microfluid cavities can be found in previously published work (Patrahau et al., Reference Patrahau, Piejko, Mayer, Antheaume, Sangchai, Ragazzon, Jayachandran, Devaux, Genet, Moran and Ebbesen2024).
Melting curve measurements
ds-DNA solutions were gently injected into microfluid cavities with a syringe equipped with a flat needle. The cavity was sealed at each end with round glass platelets (Φ = 3 mm) and thin tape. Glue or wax as sealant was avoided not to contaminate the samples. The ds-DNA melting curve measurements inside microfluid cavities were performed in a home-built high-sensitivity fluorescence setup (Supplementary Figure S2). To perform temperature-dependent measurements, the cavities were fixed onto a commercial temperature-controlled sample holder (CD 250, Quantum Northwest, Inc., equipped with a temperature controller TC1, Quantum Northwest, Inc.). An electronically controlled optical beam shutter (SH1 and SC10, Thorlabs Inc.) was employed to minimize photobleaching and only allow laser illumination during the fluorescence measurements. Optical filters (Semrock, Inc.) optimized for the FAM fluorophore were used for collecting the signal and attenuating the excitation beam. Note that the shape of the FAM fluorescence spectrum (Figure 1b) is modified by the use of a dichroic beam-splitter with an edge at 506 nm and 540/50 nm bandpass for collection. The cutoff close to 570 nm is due to this bandpass filter.
Figure 1. Principle and protocol for probing melting behaviour of ds-DNA inside the microfluidic cavity. (a) Schematic illustration of the principle of measurement based on the FRET process (yellow cavity field, green melting curve as a function of temperature). (b) Measured fluorescence spectra at different temperatures of ds-DNA at 1 μM in the reference structure (R0). (c) Extracted melting curves from four independent measurements in R0, with their corresponding first derivatives shown in (d).
To ensure thermal equilibrium at each temperature, cavities were held at the set temperature for 5 min before recording the emission spectrum, except when testing temperature ramping effects. By plotting the maximum emission intensity of the collected spectra as a function of temperature (Figure 1b), melting curves, as shown in Figure 1c, were obtained. Measurements were repeated several times for each experimental condition to ensure reproducibility. It should be noted that the FAM emission intensity is by itself insensitive to the VSC of water. It does show a small temperature variation of about 10% (Supplementary Figure S3).
Results
To begin, we established a general experimental protocol by measuring the melting curves of ds-DNA inside the reference structure R0. In the latter, the microfluidic channel is without mirrors and the two substrates are spaced by 12 μm. R0 thus acts as a control in our experiments. Figure 1a demonstrates the principle of our measurements of melting curves based on the FRET mechanism. The fluorescein (FAM) and the black hole quencher (BHQ-1) (described in the Methods section) form a FRET pair. When the two fluorophores are in proximity, as in the case of the ds-DNA involved in this study, the FAM fluorescence is quenched by BHQ-1. At high temperatures, ds-DNA disassembles into two ss-DNA oligonucleotides, and the fluorescence of FAM is restored.
A 1 μM ds-DNA solution was injected into the reference structure R0, which was fixed onto a temperature-controlled sample holder, then cooled down to 12 °C and held for an additional 1 h before measurements, thereby ensuring that all DNA was in a double-stranded state. Even then, some residual background emission can be detected with the high sensitivity of the setup. Therefore, this background spectrum collected at 12 °C was subtracted from all the spectra collected at higher temperatures. Figure 1b shows an example of fluorescence spectra measured in R0 where the intensity increases monotonously with temperature. The emission intensities were further averaged over a large spectral range (from 510 to 550 nm), yielding the melting curves shown in Figure 1c. The intensity reaches a plateau at ca. 55 °C, suggesting that ds-DNA is dissociated into ss-DNA above this temperature. As shown in Figure 1c, the melting curves extracted from four independent measurements are superimposable, demonstrating a high reproducibility of our protocol. The melting temperature is defined as the temperature at which half of ds-DNA is dissociated into ss-DNA. To quantitively extract the melting temperature and its standard deviation, a first-derivative analysis was performed (Owczarzy, Reference Owczarzy2005). As shown in Figure 1d, the analysis yields a melting temperature of 46.5 ± 1.0 °C for the ds-DNA for the non-optical cavity conditions of reference R0.
We then performed similar measurements with cavity S1, which has the same depth as reference structure R0, that is, 12 μm. The presence of the two gold mirrors in cavity S1 gives rise to multiple well-defined cavity modes in the infrared (IR) region, as shown in Supplementary Figure S3. The free spectral range (FSR) of cavity S1 is measured to be 415 cm−1 and the full width of half-maximum (FWHM) of 64 cm−1, resulting in a cavity quality factor Q of 52. The eight-cavity mode at 3320 cm−1 indicated by the grey dashed line in Figure 2a has a good overlap with the broad OH stretching band of water at 3300 cm−1 measured by attenuated total reflection (ATR) (black line in Figure 2a). After injecting the solution, two polaritonic bands emerge around the spectral region of the OH stretching band (green line in Figure 2a), with the LP band at 2922 cm−1 and UP band at 3683 cm−1. The Rabi splitting is estimated to be ~762 cm−1. The Rabi splitting is much larger than either the FWHM of the OH stretching band (445 cm−1) or the FWHM of the cavity mode (64 cm−1), indicating that the system under investigation entered collective VSC.
Figure 2. Melting behaviour of ds-DNA at 1 μM inside microfluid cavities. (a) FTIR spectra of cavity S1 filled with ds-DNA solution, also shown in black line is the IR band of OH stretching mode of water. The cavity mode position is indicated in a grey-dashed line. (b) The melting curves of ds-DNA in cavity S1 and reference structure R0.
The average melting curve of ds-DNA inside cavity S1 is shown in Figure 2b, together with that obtained with the reference structure R0. As shown, no significant difference was observed between the melting curves of ds-DNA inside cavity S1 and that in R0. The first-derivate analysis gives a melting temperature of 47.0 ± 0.0 °C for the ds-DNAs inside cavity S1, which is within experimental error the same as inside the reference structure R0 (46.5 ± 1.0 °C).
To explore if cavity detuning influences the effects of VSC on the studied system, we further performed melting curve measurements (Supplementary Figure S4) and the first derivative analysis (Supplementary Figure S5) with other cavities of different FSRs and cavity profiles (cavity S2–S4, given in Table 1). However, no significant change of melting temperatures was observed across different cavity detuning, as given in Table 1, and illustrated in Supplementary Figure S6.
Table 1. Melting temperatures of ds-DNA under various experimental conditions and determined by first-derivative analysis
Besides the base composition and sequence of ds-DNA, the melting behaviour of ds-DNA is also known to depend on the DNA concentration and solution composition (Vologodskii and Frank-Kamenetskii, Reference Vologodskii and Frank-Kamenetskii2018). Furthermore, recent work demonstrated that the effects of VSC on supramolecular polymerisation of porphyrins are highly dependent on concentration and solvent (Joseph et al., Reference Joseph, de Waal, Jansen, van der Tol, Vantomme and Meijer2024). The modification of ionic conductivity in aqueous solution through VSC was also shown to be sensitive to the composition of the electrolytes (Fukushima et al., Reference Fukushima, Yoshimitsu and Murakoshi2023). Therefore, we further extended our measurements to a concentration 10 times higher (10 μM) and also explored the effect of D2O. As given in Table 1, the melting temperature increases at higher ds-DNA concentrations as expected (Breslauer, Reference Breslauer and Agrawal1994; Vologodskii and Frank-Kamenetskii, Reference Vologodskii and Frank-Kamenetskii2018). Still, no change was observed under VSC. The corresponding melting curves of ds-DNA at 10 μM inside the reference structure R0 and two representative cavities S1 and S4 are shown in Supplementary Figure S7.
We further prepared a 1 μM ds-DNA solution in D2O/H2O (90%/10%) mixed solvent. D2O has stronger hydrogen bonding than H2O, thus different solvent properties (Giubertoni et al., Reference Giubertoni, Bonn and Woutersen2023). The FTIR spectra of D2O/H2O (90%/10%) solution in reference structure R0 and cavity S1 are shown in Figure 3a. In reference structure R0, the vibrational bands of OH stretching and OD stretching were clearly observed at 3411 and around 2500 cm−1, respectively. In cavity S1, several polaritonic modes were observed due to the coupling between different cavity modes and solvent IR bands. The melting curves of ds-DNA in the mixed D2O/H2O solution shifted to a lower temperature compared with the melting curves measured in pure H2O solution (Figure 2b), as expected (Hou et al., Reference Hou, Gao, Zhong, Li, Zhu, Wang, Zhao and Zhang2024). Once again, no significant difference was observed under VSC (Figure 3b).
Figure 3. Melting behaviour of ds-DNA under other conditions. (a) FTIR spectrum of D2O/H2O (90%/10%) solution. (b) the melting curves of ds-DNA in reference structure R0 and cavity S1. (c) the melting curve after annealing in reference structure R0 and cavity S1. Also shown is the melting curve of ds-DNA without annealing in cavity S1.
During the measurement of the melting curves in the D2O/H2O solution, we noticed that the proton exchange process between H2O and D2O is very slow (hour timescale) (Printz and Vonhippel, Reference Printz and Vonhippel1965). As proposed in a theoretical study, the hydration spine present in the minor groove of ds-DNA is frozen in the helix and behaves as an integral part of the double helix, which increases the thermal stability of ds-DNA (Chen and Prohofsky, Reference Chen and Prohofsky1993). The slow proton exchange indeed influenced the melting behavior of fresh and aged samples (Supplementary Figure S8).
To ensure that such a slow feature did not affect other experiments, we repeated the study of the melting curves in H2O, first annealing the samples under VSC. The ds-DNA solution inside microfluid cavities was heated up to 75 °C and annealed for 1 h to fully dissociate the DNA into a single-stranded state and to fully exchange the proton between the water inside the minor grooves of ds-DNA and the solution. Then, the cavities were slowly cooled down to 12 °C and kept at that temperature for 1 h, during which time the ss-DNA re-hybridized into ds-DNA again, but under VSC. Again, this protocol did not introduce any difference in the melting temperatures of ds-DNA (Figure 3c) whether measured in the reference structure R0 or in the cavity S1.
Discussion
In summary, no significant effect from VSC on the melting curves ds-DNA was found for all tested conditions, which included cavity detuning, DNA concentration and solvent composition. One key difference between our work and the previous study (Hou et al., Reference Hou, Gao, Zhong, Li, Zhu, Wang, Zhao and Zhang2024) is that the melting behaviour was measured in equilibrium condition in our case, while in the previous work by Hou et al., it was measured in a continuous manner at a rapid ramp rate (2°C/min). Therefore, we repeated our experiments at a similar ramp rate and without equilibrating the DNA at each temperature, but again, no significant difference was observed (Supplementary Figure S9), which further rules out the possibility of a kinetic effect from VSC in our system.
In fact, our experiments are fundamentally different from the earlier work on DNA under VSC by Hou et al. (Reference Hou, Gao, Zhong, Li, Zhu, Wang, Zhao and Zhang2024). In that work, the ss-DNA solution was first injected into the optical cavity and the formation of ds-DNA was performed in situ at room temperature. The melting curves were then determined outside the cavity by fluorescence measurements and a decrease in Tm was found under VSC. This protocol assumes that the effects of VSC remain after extraction from the cavity. This would imply a stable structural change under VSC, which we do not detect with our DNA strand composition. In other words, it is possible that the difference in the results stems from the nucleotide composition differences in the two studies.
The insensitivity of the thermal stability of ds-DNA to VSC is in strong contrast to the supramolecular polymerisation of porphyrins, where even the relative stability between monomer and polymer can be reversibly tuned by VSC (Joseph et al., Reference Joseph, de Waal, Jansen, van der Tol, Vantomme and Meijer2024). The sensitivity of these porphyrin assemblies to VSC is probably due to the dynamic monomer addition and dissociation in solution and the related pathway complexity, which makes the system very sensitive to external stimuli (Mabesoone et al., Reference Mabesoone, Markvoort, Banno, Yamaguchi, Helmich, Naito, Yashima, Palmans and Meijer2018).
The fact that we detect no effect of VSC on the melting behaviour of DNA cannot be due to a lack of cooperative coupling between the stretching mode of H2O and the vibrational modes of DNA. The vibrational frequency of the OH stretching mode of H2O at around 3300 cm−1 is well-known to be very sensitive to the strength of hydrogen bonding (Brubach et al., Reference Brubach, Mermet, Filabozzi, Gerschel and Roy2005). For liquid water, there are at least three Gaussian components within the spectral profile of the OH stretching mode at 3295, 3460, and 3590 cm−1, which correspond to subpopulations of water clusters (Brubach et al., Reference Brubach, Mermet, Filabozzi, Gerschel and Roy2005). The vibrational frequency of the NH stretching mode in DNA pairs also lies between 3300 and 3600 cm−1 (Nir et al., Reference Nir, Plützer, Kleinermanns and de Vries2002), having a good overlap with the OH stretching mode of water. Therefore, the reason for the results must be other than an issue of cooperativity, reflecting the nature of the interaction forces at play in the DNA structure.
The thermal stability of ds-DNA is mainly determined by base-stacking interactions, which are partially from London dispersion forces and partially from hydrophobic interactions, and less by base pairing interactions, that is, hydrogen bonding (Herskovits, Reference Herskovits1962; Yakovchuk et al., Reference Yakovchuk, Protozanova and Frank-Kamenetskii2006; Vologodskii and Frank-Kamenetskii, Reference Vologodskii and Frank-Kamenetskii2018; Feng et al., Reference Feng, Sosa, Martensson, Jiang, Tong, Dorfman, Takahashi, Lincoln, Bustamante, Westerlund and Norden2019; Nordén, Reference Nordén2019). In our experiments, the water OH stretching mode is coupled to the vacuum field, where the hydrogen bonds between base pairs are also cooperatively coupled. The less important role of hydrogen bonding on the thermal stability of ds-DNA is consistent with the absence of a VSC effect. Furthermore, these hydrogen bonds are buried inside the double helix, which might also be the reason why VSC does not alter the thermal stability of ds-DNA. Therefore, our observation of the stability of ds-DNA to VSC can be considered as another new piece of evidence for the robustness of ds-DNA.
Open peer review
To view the open peer review materials for this article, please visit http://doi.org/10.1017/qrd.2025.5.
Supplementary material
The supplementary material for this article can be found at http://doi.org/10.1017/qrd.2025.5.
Data availability statement
The raw data are available upon reasonable request.
Author contribution
BN suggested the experiments. FM and CB designed the DNA sequence and prepared the corresponding samples. WT built the emission setup and carried out most of the VSC experiments. BP helped in the initial experiments. TWE supervised the project.
Financial support
We acknowledge the support from the International Center for Frontier Research in Chemistry (icFRC, Strasbourg), the Labex NIE Projects (ANR-11-LABX-0058 NIE), CSC (ANR-10-LABX-0026 CSC), USIAS (Grant No. ANR-10-IDEX-0002-02) within the Investissement d’Avenir program ANR-10-IDEX-0002-02, and the ERC (project no 788482 MOLUSC).
Open access
Comments
Dear Editor,
Please find uploaded our ms entitled “Probing DNA melting behaviour under vibrational strong coupling”. I have suggested two groups of referees one from the DNA physical chemistry community and one from the vibrational strong coupling community since this is a very pluridisciplinary topic. In regards to strong coupling referees, we do not want to have our ms refereed by scientists in the US as there are real conflicts of interest.
Thank you in advance for your consideration of our manuscript.
Best regards, Thomas Ebbesen