Simultaneous multi-signal quantification for highly precise serodiagnosis utilizing a rationally constructed platform

Serodiagnosis with a single quantification method suffers from high false positive/negative rates. In this study, a three-channel platform with an accessional instrumented system was constructed for simultaneous electrochemical, luminescent, and photothermal quantification of H2S, a bio-indicator for acute pancreatitis (AP) diagnosis. Utilizing the specific reaction between platform and H2S, the three-channel platform showed high sensitivity and selectivity in the biological H2S concentration range. The three-channel platform was also feasible for identifying the difference in the plasma H2S concentrations of AP and normal mice. More importantly, the precision of AP serodiagnosis was significantly improved (>99.0%) using the three-signal method based on the three-channel platform and an optimized threshold, which was clearly higher than that of the single- or two-signal methods (79.5%–94.1%). Our study highlights the importance of constructing a multichannel platform for the simultaneous multi-signal quantification of bio-indicators, and provides rigorous ways to improve the precision of medical serodiagnosis.

The manuscript deals with a new strategy for H2S detection by the three-channel platform for the simultaneous three-signal quantification of H2S by modifying a glassy carbon electrode with a luminescent rare-earth-based nanothermometer and Cu-alginate (Cu-ALG) gel. The performance of this three-channel platform in quantifying H2S was studied in vitro and in vivo using an established AP mouse model. The study is interesting and informative in the field of H2S probes. However, it may not be acceptable for publication in nature communications. There are several matters that has to be addressed by the authors before considering to resubmit to another journal.
1. H2S detection by using this tri-channel platform is an irreversible process, so it does not have an advantage, how this system can be improved or modified to make it reversible? Reversible detection systems are more advantageous as compared to irreversible system. 2. Detection by using copper complexes have been known for ATP ADP AMP and PPi sensor. Therefore, authors need to do more experiment to find out the interference of ATP, ADP AMP and PPi. 3. For the selectivity studies of the tri-channel platform, the author also should test the response with other cellular signal molecules NO and CO. 4. How to confirm the H2S presence in solution as discussed on page 15 line 321-324, After the addition of Na2S in water. The Na2S with water change to SH-and NaOH. And it is not stable, but it is kept in a tri-channel platform for 1 hr. Did the authors confirm the H2S in test solution by using any other method? 5. The reference section has to be checked and modified according to the journal format. For example see reference 8, 10, 14, 16, and 19. 6. Title of the manuscript is not clear and it does not reflect the merit of the work, it has to be modified. 7. Title of the article in the manuscript file and SI file are different, it has to be corrected both should be same.
Reviewer #2 (Remarks to the Author): Manuscript ID: NCOMMS-19-19045 Journal: Nature Communications Title: Simultaneous Multi-signal Quantification of a Single Bio-indicator Improves the Precision of Serodiagnosis Authors: Yuxin Liu, Zheng Wei, Jing Zhou, Zhanfang Ma General Comments: Liu & Wey et al report the development of a three-channel H2S quantification platform constructed on a glassy carbon electrode using core-shell NaYbF4:Er@NaLuF4 nanoparticles. The system is designed and constructed to simultaneously measure the luminescence emission spectrum and the electrochemical signal of the probe with which the electrode modified by introducing an electrochemical workstation, visible and infrared luminescence emission spectroscopy. The Ln3+ doped nanoparticles synthesis and characterization are presented namely the size distribution. The photophysical emission of the prepared platform are studied in the presence of H2S. The use of dual irradiation wavelengths allow simultaneous excitation of the Ln3+-doped nanoparticles (by the 980 nm laser) and the photothermal effect due to the metallic electrode (by the 808 nm laser, particularly) and thus the authors combine the distinct readouts to improve the H2S quantification by combinatorial analysis of the three outputs. The novelty of this work is the incorporation of three outputs that resulted in an improved precision of acute pancreatitis serodiagnosis. In general, the paper is not reasonably well written however with too many abbreviations, repetition of the same ideas and unbalanced references to the relevant literature. The manuscript length is appropriate for this journal, although the presented figures are not adequate and deserve some effort looking forward to improving their clarity. One weak point of the work is the quantification of the temperature increase resulting (essentially) from the 808 nm radiation absorption readout ant that can be quantified using the visible emission of the Er3+ ion, however the strong points are the novelty in the combination of 3 signals for H2S quantification and acute pancreatitis serodiagnosis in animal models. In summary, the paper may be adequate for Nature Communications Journal if the authors can address the major concerns detailed next. In the present form this reviewer's recommendation is to accept it for publication after major revisions.
Particular Comments: 1. The text is very repetitive, and the same idea is repeated in every section thus the text flowing is very difficult. It is mandatory that the authors revise the text to eliminate the over repetition of the same ideas. 2. The overuse of abbreviations makes the text reading very difficult. In the perspective of a broad audience journal like Nature Communications this is not adequate because it results in an additional difficulty, especially for the non-experienced readers. Please consider reducing it to the minimum possible. 3. The references need to include some works of researchers in distinct latitudes. The list of references is biased towards Asiatic authors when the field presents important contributions from European and American groups. Please consider including them. 4. The major concern of this reviewer is the quantification of H2S towards the emission lines of Er3+ in the visible spectral range. These transitions are known to being used for luminescence thermometry and thus it is absolutely necessary to include the temperature raise resulting separately by the 980 nm, by the 808 and 980 nm irradiation (laser power densities are missing) and the cross effect between the temperature and the H2S quantification. 5. The luminescent thermometer performance parameters (thermal sensitivity and temperature uncertainty) need to be calculated and compared with the literature values. 6. The conclusions are quite qualitative so some numerical discussion is needed to give some strength to the main points focused in the work. solution of oxidants can oxidize this insoluble sulfide into soluble sulphate (Please see Response Figure 1A), which could decompose and eliminate CuS by washing-out (Please see Response Figure 1B). As a result, it would be applicable to eliminate CuS from electrode, which preliminarily allowed the reversion of the tri-channel platform.

Response Figure 1. Decomposition of produced CuS on H 2 S-treated Cu-ALG gel by
solution of oxidants. The absorbance at 808 nm A) and X-ray diffraction patterns B) of H 2 S-treated Cu-ALG gel treated with acidic H 2 O 2 solution (Cu-ALG + H 2 S + H 2 O 2 + H + ) and with HNO 3 solution (Cu-ALG + H 2 S + HNO 3 ). The H 2 S-treated Cu-ALG gel and un-treated Cu-ALG gel were used as positive or negative control. Data are represented as mean ± SD.
Though the elimination of CuS from electrode is likely to be achieved based on the pilot study shown above, it should be noted that, however, there are still some questions that have to be concerned for the reversion of the electrode due to the intrinsic limitation of range of biomedical applications, which was not only originated from our work but from most of previously researches and reported diagnostic systems (Gorbet, M. et Biotechnol. 2002, 20, 270-274). First of all, unlike those sensors and systems for only detection in vitro, the actual biological samples for diagnostic platform were biological fluids, which has complicated chemical composition. There were many macrobiomolecules, such as proteins and nucleic acids, in the biological fluids, which would adsorb on the platform during quantification process. To achieve the actual reversion of diagnostic platform, these macrobiomolecules must be removed from platform, which required sophisticated procedures and expensive reagents (proteinases or nucleases). Considering the low-cost and facile preparation procedure of our tri-channel platform, it was much more practical and easier to administrate new ones rather than reverse them. In addition, the platform for disease diagnosis should be clean and disposable to ensure the safety and reduce cross-infection. The re-use of reversed platform will result in the elevation of risk of infections, which was unacceptable for actual diagnosis applications. As a result of the above-mentioned questions, most of advanced platform for diagnosis, efficient and clinically applicable though they were, were irreversible (Hassibi, A. et Nat. Biotechnol. 2008, 36, 1373-1378. Therefore, though we believed that reversible diagnostic system may have advantage, it was still an insoluble issue in current design and construction of diagnostic system due to the imbalance between economic benefits and medical reliability. Meanwhile, we would like to iterate that the primary goal of this work is to propose a novel strategy to improve serodiagnosis precision by reducing false positive/negative rate via multi-signal quantification. To achieve this, a tri-channel platform was designed for H 2 S quantification by rationally optimizing probes and choosing signal types. By this multi-channel platform, the precision of AP diagnosis was significantly improved from 75.0-94.1% (single or bi-signal method) to 99% (tri-signal method). Instead of optimizing the sensor for H 2 S sensing alone, we attempted to propose a novel and convenient strategy to improve diagnostic precision.
We appreciated the reviewer for this suggestion and would further optimize our device in our future works. on-off assay for rapid and highly sensitive pyrophosphate and alkaline phosphatase detection. Sci. Rep. 2017, 7, 5849). In the presence of biophosphorus compounds, the Cu 2+ in acceptors complexed with negative phosphorus-contained groups, resulting in the reduction of FRET and hence luminescence recovery. However, in our cases, the luminescence of RENPs are not quenched by copper complex through FRET route due to the microscale distance between them, which indicated that the luminescence of RENPs won't recover in the presence of biophosphorus. Meanwhile, in the presence of H 2 S, CuS formed on gel in situ with strong near-infrared absorbing, which selectively quenched the red luminescence of RENPs through re-absorption route, which allowed the ratiometric luminescence quantification of H 2 S. Furthermore, unlike H 2 S, all studied biophosphorus (such as ATP, ADP, AMP and PPi) cannot react with gel and form products with both high resistance and photothermal conversion capacities. Therefore, typical biophosphorus won't affect the quantification of H 2 S. Response: We appreciate this valuable advice. To further illustrate the selectivity of the tri-channel platform to H 2 S, we further studied the response with other cellular signaling molecules, including NO and CO (Please see Response Figure 3 and Supplementary Figure 15). It was found that all three signals barely changed in response to either NO or CO, which could be explained by the mechanism of gel for H 2 S response: In the presence of H 2 S, CuS formed on gel in situ with capabilities of all high resistance, strong near-infrared absorbing and outstanding photothermal conversion, which thence led to the simultaneous change of all electrochemical, luminescence and photothermal signals. Unlike H 2 S, both NO and CO cannot react with gel and form products with the above three capacities. Therefore, cellular signal molecules like NO and CO won't possess influence on the quantification of H 2 S. Data are represented as mean ± SD. It was found that the C H2S of standard H 2 S solution only slightly fluctuated around 50 μM within 100 min, which suggested that the concentration of H 2 S standard solution was accurate if prepared freshly and used timely. It also ensured the reliability of C H2S within incubation time of 1 h. Data are represented as mean ± SD.
However, it was previously demonstrated that the H 2 S mainly existed as S 2and HSin aqueous biological system due to the neutral nature of serum (Sen, U. et al. Chem. Int. Ed. 2016, 128, 9511). Therefore, to simulate the existence of H 2 S in psychological conditions, the H 2 S standard solution (pH = 7.4) contains all dissolved H 2 S, ionic HSand ionic S 2-, but no NaOH. It was noteworthy that our tri-signal method can react with and quantify the total amount of all dissolved H 2 S, ionic HSand ionic S 2-, and also in high agreement with the result that obtained from standard methylene blue colorimetric protocol (Please see Response Figure  Corresponding C H2S was previously determined by standard colorimetric method and used for comparison. D) Data statistic of the above determined C H2S by various methods. The above results illustrated that the C H2S determined by all three methods were highly in accordance with that by standard method, which suggested that all three methods were applicable for accurate quantification of H 2 S. Data are represented as mean ± SD. Statistical significance was determined from one-way t tests. N.S. means not significant.
In addition, it should be noted that the incubation time of 60 min (1 h) was a combination of the time for reaction and the time for signal stabilization (Please see Response Figure 6). H 2 S (also including HSand S 2-) can react with ALG-Cu gel rapidly (< 2 min), which can be confirmed by the dynamic ultraviolet-visible-near infrared spectra (Please see Response Figure 7A and Supplementary Figure 12A). Therefore, the H 2 S in solution can be consumed in a short period, which means that the change of H 2 S concentration within long-term (> 100 min) may not possess significant influence on quantification result. Though the reaction period is short, the time for stabilizing signal varies due to different mechanism of signal change (Please see Response Figure  For luminescence signal, the grown CuS selectively absorbed the luminescence of RENPs in red spectral region but no SWIR spectral region, which allowed the change of luminescence ratio in these two wavelength range; 3) For photothermal signal, the grown CuS convert laser irradiation into heat by non-radiative relaxation, heating the RENPs and resulting in change of thermal-sensitive luminescence. Therefore, the incubation time of 60 min (1 h) was a comprehensive consideration to ensure the stabilization of all three signals instead of finishing reaction between gel and H 2 S.
In summary, according to results obtained by using standard methylene blue colorimetric protocol, H 2 S concentration of test solution will be reliable within the studied period of 60 min (1 h). Furthermore, though H 2 S ionized in biological aqueous system and existed in ionized state of HSand S 2-, it would not influence the quantification results as our tri-signal method was able to effectively quantify all kinds of H 2 S and in high accordance with standard colorimetric protocol. Response: We appreciate this valuable advice. In this work, we were proposing a strategy to reduce the false positive/negative rate of serodiagnosis and hence achieve higher precision by a multi-signal quantification method. For this purpose, a multi-channel platform was rationally constructed. To verify the practicability, acute pancreatitis was chosen as disease model and H 2 S was used as indicator for serodiagnosis. Therefore, based on the merit of this work, we changed the title from "Simultaneous Multi-signal Quantification of a Single Bio-indicator Improves the Precision of Serodiagnosis" to "Simultaneous Multi-signal Quantification for Highly Precise Serodiagnosis Utilizing Rationally Constructed Platform". We hope it can clearly demonstrate our novelty and main findings.

Response
Comment 7: Title of the article in the manuscript file and SI file are different, it has to be corrected both should be same.
Response: We appreciate this kind remind. The title of SI file has been corrected in accordance with that of manuscript file. We apologize for this unintentional mistake. Response: We appreciate this valuable advice. We tried our best to remove the repetitive sentences and ideas. For example, the following paragraphs were removed.
"To demonstrate the potential utility of the RENP-Cu-ALG-modified GCE as a platform for the simultaneous three-signal determination of H 2 S, the determination mechanism was studied" (Please see Line 16-18, Page 6 in the original manuscript).
"Therefore, as C H2S increased, the current signal and red UCL intensity decreased, whereas the photothermal temperature increased, corresponding to the CuS formed on the three-channel platform (Fig. 2D-F "To ascertain the difference in plasma C H2S between the AP and normal mice, the plasma C H2S was first determined with a standard colorimetric kit (Fig. 3E). The plasma C H2S in the AP mice was clearly higher than that in the normal mice, confirming that elevated plasma C H2S characterized the mice with AP. Therefore, the AP mouse model, with excessive plasma C H2S , was successfully established for further study" (Please see Line 12-17, Page 9 in the original manuscript) In addition, the following sentences were simplified.
"Therefore, the rational construction of an efficient multichannel platform is vital if we are to improve the precision of medical serodiagnostics. To develop a novel multichannel platform, a rationally selected quantification method with good applicability and associativity is required" (Please see Line 3-6, Page 3 in the original manuscript) were simplified to be "Therefore, the rational construction of an efficient multichannel platform by combining applicable quantification method is vital" (Please see Line 4-5, Page 3 in the revised manuscript).
"Therefore, by modifying the electrode with a rare-earth-based nanothermometer and an H 2 S-responsive electro-active material that can also act as a luminescence acceptor and a photothermal conversion agent, a multichannel platform can be established to simultaneously determine electrochemical, optical, and thermal signals for H 2 S quantification" (Please see Line 11-15, Page 4 in the original manuscript) was simplified to be "As a result of the above theoretical basis, a multichannel platform could be established for simultaneously multi-signal quantification" (Please see Line 11-12, Page 4 in the revised manuscript).
"To demonstrate the potential utility of the RENP-Cu-ALG-modified GCE as a platform for the simultaneous three-signal determination of H 2 S, the determination mechanism was studied" (Please see Line 16-18, Page 6 in the original manuscript) was simplified to be "After confirming successful construction, the determination mechanism of platform was then studied" (Please see Line 18-19, Page 6 in the revised manuscript).
"A detailed evaluation was conducted to assess the quantification performance of the as-developed three-channel platform. According to our previous findings, the biological C H2S range is from the nanomolar to the micromolar level" (Please see Line 3-5, Page 8 in the original manuscript) was simplified to be "Then, its quantification performance was tested in biological C H2S range, which is from nanomolar to micromolar level according to our previous findings" (Please see Line 11-12, Page 8 in the revised manuscript).
Therefore, 3 repetitive paragraphs have been removed and 4 sentences have been simplified. We hope the text was smooth to read after cutting down redundancy. Response: We appreciate this valuable advice. We deleted the unnecessary abbreviations and reduced them to the minimum possible. For example, most of the abbreviations of instruments and materials were deleted, including high resolution transmission electron microscopy, transmission electron microscopy, powder X-ray diffraction patterns, energy dispersive X-ray analysis, scanning electronic microscopy, electrochemical impedance spectroscopy, Fӧurier transform infrared spectrum, X-ray photoelectron spectroscopy, glassy carbon electrode, 1-octadecene, oleic acid and deionized water. Moreover, the abbreviations of those proper nouns that presented in manuscript less than 5 times were also removed, including ultraviolet-visible-near infrared, near-infrared, finite element method, limit of detection, haematoxylin and eosin, receiver operating characteristic curve and Vital River Institutional Animal Care and Use Committee. Therefore, more than 60 abbreviations have been removed and only those important retained in the manuscript. We hope it could be adequate for both professional and non-experienced readers of Nature Communications.

Comment 3: The references need to include some works of researchers in distinct latitudes. The list of references is biased towards Asiatic authors when the field presents important contributions from European and American groups. Please consider including them.
Response: We appreciate this kind remind. It was an unintentional mistake rather than bias that we cited more works from Asiatic authors than that from European and American groups. We apologize for our carelessness. To prevent this misunderstanding, we consulted a number of papers about this subject and added advanced pioneering works from European and American groups that initiatively studying luminescent nanoparticles or functionalized electrodes to quantify biomolecules and temperature. Therefore, 17 references have been added or replaced to keep the balance of reference. We hope it could prevent the misunderstanding.

Comment 4:
The major concern of this reviewer is the quantification of H 2 S towards the emission lines of Er 3+ in the visible spectral range. These transitions are known to being used for luminescence thermometry and thus it is absolutely necessary to include the temperature raise resulting separately by the 980 nm, by the 808 and 980 nm irradiation (laser power densities are missing) and the cross effect between the temperature and the H 2 S quantification.
Response: We appreciate this valuable advice. In our work, the mechanism of photothermal quantification of H 2 S can be described as the following processes: 1) Cu 2+ in gel reacted with H 2 S, forming typical photothermal conversion agent CuS; 2) Under 808 nm laser irradiation (1.5 W cm -2 ), CuS generated heat and increased surrounding temperature of RENPs; 3) Under 980 nm laser irradiation (0.75 W cm -2 ), RENPs emitted green upconversion luminescence (Please see Response Figure 8 and Figure 1H), which changed in response to temperature change (Please see Response Figure 9A and Supplementary Figure 8A). Therefore, H 2 S concentration can be correlated with the ratio of thermal-sensitive emission of Er 3+ , which allowed the luminescence-assisted temperature sensing and therefore photothermal quantification of H 2 S on electrode in situ (Please see Response Figure  To further understand the influence of laser irradiation on photothermal quantification and luminescence thermometry, temperature raise resulted by single 808 nm laser irradiation, by single 980 nm laser irradiation and by dual laser irradiations were studied and compared. It was found by heating curves that, under single 808 nm laser irradiation for 15 s, the H 2 S-treated gel had higher temperature raise (6.38 o C) than those without H 2 S treatment (1.04 o C), which indicated that the photothermal conversion capacities of gel were activated by H 2 S and hence reasonable to construct relationship between temperature and H 2 S concentration (Please see Response Figure 10 and Supplementary Figure 7). Furthermore, it was also noticed that the 980 nm laser cannot raise temperature efficiently (1.49 o C) under the studied power density (Please see Response Figure 11 and Supplementary Figure   10), which might be a combined result of following factors including decreasing molar extinction coefficient (ε = 5.64 × 10 9 mol -1 cm -1 at 980 nm and 6.71 × 10 9 mol -1 cm -1 at 808 nm) and photothermal conversion efficiency of H 2 S-treated gel at various spectral wavelengths (η = 7.2% at 980 nm and 18.1% at 808 nm), as well as the different laser power density used in this study (0.75 W cm -2 for 980 nm laser irradiation and 1.5 W cm -2 for 808 nm laser irradiation, Please see Response Figure 12 and Supplementary Figure 9). Due to these differences, the temperature raise by co-use of 808 nm and 980 nm laser was similar to that by single 808 nm laser irradiation, demonstrating that the 980 nm laser won't significantly influence the temperature rising (Please see Response Figure 13 and Supplementary Figure 11).
More importantly, the co-use of 808 nm and 980 nm laser irradiation can simultaneously activate efficient photothermal conversion process and strong thermal-sensitive luminescence, generating heat and heating NaYbF 4 :Er@NaLuF 4 , which was able to cause obvious change of luminescence ratio for photothermal quantification of H 2 S concentration.
Based on the above findings, it was rational to use 808 nm laser irradiation to activate photothermal conversion capacities of H 2 S-activated gel and use 980 nm laser irradiation to activate thermal-sensitive emission of NaYbF 4 :Er@NaLuF 4 for temperature monitoring, further allowing the indirectly photothermal quantification of H 2 S concentration by thermal-sensitive luminescence. where S is thermal sensitivity, R is I 525 /I 545 luminescence ratio and T is temperature in Kelvin unit. The thermal sensitivity of RENPs was calculated to be 0.0108 K -1 , which was comparable to most rare-earth-based luminescent nanothermometers (Please see Response Figure 14A and Supplementary Figure 3A, Response Table 1 and   Supplementary Table 1 Thermal sensitivity A) and temperature uncertainty of RENPs-based luminescent nanothermometer. The temperature uncertainty of RENPs was calculated from experimental data included in Response Figure 15 and Figure  where δT is temperature uncertainty, S is thermal sensitivity, ∆I is background noise of luminescence spectrum and I is luminescence intensity. The temperature uncertainty of RENPs was calculated to be 0.2559 K (Please see Response Figure 14B and Supplementary Figure 3B), which was less than 1 K and hence applicable to  Table 1). → 4 I 15/2 ), relatively high thermal sensitivity (0.0108 K -1 ) and low temperature uncertainty (0.2559 K). The Cu 2+ in the gel rapidly (< 2 min) and specifically reacted with H 2 S to form insoluble semiconducting CuS in situ, which increased not only the resistance of the gel but its visible/near-infrared absorbance."

Response
"It should be noted that either single-or two-signal methods based on this platform were suffer from relatively low serodiagnosis precision (79.5-94.1%, lower than 95%), while the three-signal method with unreasonable threshold (1/3 or 1) was also unfavourable for precise serodiagnosis (75.0% and 87.9%, lower than 95%). However, by using the three-signal method and an optimized threshold for the three-channel platform, the precision of AP serodiagnosis was significantly improved by 4.9%, which indicated that there was a 99% chance to identify AP cases correctly." Reviewers' comments: Reviewer #1 (Remarks to the Author): I am pleased that the authors considered the comments from the reviewers and revised the manuscript. The additional experiments and corrections made the manuscript in a better view. There are few mistakes that have to be corrected before recommending publication. Figure 2J caption is missing Figure 2E should be SWV curves Figure 1B and 1C caption need to be corrected.
Reviewer #2 (Remarks to the Author): attending to the comments of both the referees. After revision, the paper English was significantly improved, and references were carefully revised for better balance between Asiatic and non-Asiatic authors. The figures were modified to improving their clarity. The quantification of the temperature increase resulting (essentially) from the 808 nm radiation absorption is now properly described. In summary, the paper was significantly improved with respect to the original version, and, in this reviewer opinion, it is adequate for Nature Communications Journal if the authors address the major concerns of both referees. In the present form this reviewer's recommendation is to accept it for publication after minor revisions.
Particular Comments: 1. A particular point on the comments on the references about the relative thermal sensitivity and on the temperature uncertainty. The concept of relative thermal sensitivity was originally proposed by Wade et al. in the context of optical fiber temperature sensors and proposed as a figure of merit for comparing thermometers, irrespectively of their nature for the first time in a review work of 2012 from the group of prof. L. Carlos. 2. The major concern of this reviewer is the quantification of H2S using the emission lines of Er3+ in the visible spectral range (the same used for luminescence thermometry). The temperature raise resulting separately by the 980 nm and by the 808 nm irradiation (laser power densities are still missing). The temperature effect on the H2S quantification is not discussed and needs to be adressed. 3. The luminescent thermometer performance parameters (thermal sensitivity and temperature uncertainty) were calculated but were not compared with the literature values.
and luminescence quantification of H 2 S is orthogonal.
In order to evaluate the effect of temperature on luminescence quantification of H 2 S, we thence studied the relationship between luminescent ratio (Ln(I 654 /I 1550 )) and C H2S in the absence of 808 nm laser irradiation (Please see Response Figure 1A). It is noticed that, in the absence of heating, a linear relationship is also found between Ln(I 654 /I 1550 ) and C H2S while the function (Y = 1.5688 -0.3295 X) is similar to that in the presence of heating (Y = 1.5580 -0.3283 X, Please see Figure 2I in revised manuscript). This phenomenon can be a result of relatively ultralow thermal sensitivity of red emission in steady-state luminescence spectrum. We further discussed the effect of elevated absorbance on temperature sensing capacities of RENPs by green emissions (Please see Response Figure 1B). It is found that the Ln(I 525 /I 545 ) possessed linear relationship with C H2S in the presence of H 2 S while the function (Y = 2.0624 -0.6739 X) is similar to that in the absence of H 2 S (Y = 1.9672 -0.6411 X, Please see Figure 1D in revised manuscript). It can be explained by the mechanism of thermal-sensitive luminescence of Er 3+ : the increased temperature induced re-distributions in the populations of energy levels, further resulting in the change of luminescence ratio between 545 nm and 525 nm, which is ruled by Boltzmann distribution and is less relevant to absolute luminescence intensity.

Response
Therefore, though the elevated absorbance might result in some luminescence quenching, the luminescence ratio would not be affected significantly.
Based on the above findings, the photothermal and luminescence quantification method is orthogonal, which indicated that the above two methods, though both utilizing the emissions of Er 3+ in visible spectral range, will not affect the quantification each other's result.