In Vivo High-resolution Ratiometric Fluorescence Imaging of

The Supporting Information is available free of charge on the ACS ... Materials and methods; figures providing additional spectra and characterization...
0 downloads 0 Views 1MB Size
Subscriber access provided by UNIV OF CAMBRIDGE

Communication

In Vivo High-resolution Ratiometric Fluorescence Imaging of Inflammation Using NIR-II Nanoprobes with 1550 nm Emission Shangfeng Wang, Lu Liu, Yong Fan, Ahmed Mohamed ElToni, Mansour Saleh Alhoshan, Dandan Li, and Fan Zhang Nano Lett., Just Accepted Manuscript • DOI: 10.1021/acs.nanolett.8b05148 • Publication Date (Web): 18 Mar 2019 Downloaded from http://pubs.acs.org on March 18, 2019

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Nano Letters

In Vivo High-resolution Ratiometric Fluorescence Imaging of Inflammation Using NIR-II Nanoprobes with 1550 nm Emission

Shangfeng Wang1, Lu Liu1, Yong Fan1, Ahmed Mohamed El-Toni2, Mansour Saleh Alhoshan3, Dandan li1 and Fan Zhang1*

1Department

of Chemistry, State Key Laboratory of Molecular Engineering of Polymers, Shanghai

Key Laboratory of Molecular Catalysis and Innovative Materials and iChem, Fudan University, Shanghai 200433, P. R. China 2King

Abdullah Institute for Nanotechnology, King Saud University, Riyadh 11451, Saudi Arabia

3Department

of Chemical Engineering, King Saud University, Riyadh 11421, Saudi Arabia

KEYWORDS: ratiometric fluorescence, NIR-II, bioimaging, biosensing, nanoprobe

ABSTRACT: Quantitatively imaging the spatiotemporal distribution of biological events in living organisms is essential to understand fundamental biological processes. Self-calibrating ratiometric fluorescent probes enable accurate and reliable imaging and sensing, but conventional probes using wavelength of 400-900 nm suffer from extremely low resolution for in vivo application due to the disastrous photon scattering and tissue autofluorescence background. Here, we develop a NIR-IIb (1500-1700 nm) emissive nanoprobe for high-

ACS Paragon Plus Environment

Nano Letters 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 30

resolution ratiometric fluorescence imaging in vivo. The obtained nanoprobe shows fast ratiometric response to hypochlorous acid (HOCl) with a detection limit down to 500 nM, through an absorption competition-induced emission (ACIE) bioimaging system between lanthanide-based downconversion nanoparticles and Cy7.5 fluorophores. Additionally, we demonstrate the superior spatial resolution of 1550 nm to a penetration depth of 3.5 mm in a scattering tissue phantom, which is 7.1-fold and 2.1-fold higher than that of 1064 nm and 1344 nm, respectively. With this nanoprobe, clear anatomical structures of lymphatic inflammation in ratiometric channel are observed with a precise resolution of ~477 μm. This study will motivate the further research on the development of NIR-II probes for high-resolution biosensing in vivo.

Fluorescent imaging and sensing have become indispensable technology for investigating biological systems because they provide dynamic information concerning the localization and quantity of molecules of interest.1,2 Traditional fluorescent imaging and sensing based on absoluteintensity of a single emission suffer from signal fluctuation caused by penetration depth, probes distribution, auto-fluorescence interference and instrumental parameters, thus can fail to accurately quantify the analyte concentration information.3-7 By introducing a self-calibrated reference signal, ratiometric fluorescence imaging provide more accurate and reliable quantitative information,8,9 which has been widely utilized for visualizing the spatiotemporal distribution of biological ions,10-12 pH,13,14 enzymes15,16 and signaling molecules17-21 in vitro or ex vivo. Nonetheless, for in vivo bioimaging and sensing, due to the light scattering and tissue autofluorescence, current ratiometric fluorescence imaging using probes operating in the short wavelength range of 400-900 nm still suffers from extremely low resolution. This will lead to the misinterpretation of the data, hindering the ACS Paragon Plus Environment

Page 3 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Nano Letters

accurate sensing and correct understanding of biological events in living organisms. In particular, the influence has never been more apparent when regarding the false ratiometric signals derived from scattering and autofluorence background as true signals. Fluorescent imaging in the second near-infrared window (NIR-II, 1000-1700 nm) is an emerging technology which offers deeper tissue penetration and unprecedented resolution compared with the traditional fluorescence imaging (400-900 nm), owing to the reduced light scattering and negligible background autofluorescence.22-29 As photon scattering in biological tissue scales with λ-α (α = 0.2-4 for different tissues) according to the Mie theory,30 wavelength beyond 1500 nm provides the lowest photon scattering in the entire NIR-II window.31 Additionally, laser-excited tissue autofluorescence also reaches undetectable levels beyond 1500 nm.32 These benefits have made the wavelength region of 1500-1700 nm to be defined as the transparent “NIR-IIb” window, motivating the development of a series of NIR-IIb emitters, including Er3+-doped lanthanide nanoparticles,33-38 carbon nanotubes,31 PbS39 and InAs40,41 quantum dots, for resolving fine-scale anatomical structures in vivo. However, in sharp contrast to the extensive exploration of these emitters for high-resolution fluorescence imaging in vivo, their sensing applications are extremely limited. One major obstacle is the lack of flexible strategies to regulate such long-wavelength fluorescence for responding to biological stimuli. This limitation along with the superior performance of NIR-IIb imaging prompts us to search a new strategy for high-resolution ratiometric fluorescence imaging and sensing in vivo. Due to the important roles of reactive oxygen species (ROS), such as H2O2, HOCl, HO•, O2•-and 1O

2,

in inflammation,42,43 we herein develop a novel ratiometric nanoprobe in response to HOCl for

high-resolution imaging of inflammation in vivo. This ratiometric sensing system is based on an absorption competition-induced emission (ACIE) process between Cy7.5 fluorophores and Er3+doped DCNPs (Figure 1A), therefore displaying fast and good linear ratiometric fluorescent response

ACS Paragon Plus Environment

Nano Letters 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 30

to HOCl at 1550 nm under dual-wavelength (808 nm, 980 nm) excitation with a detection limit down to 500 nM. Tissue phantom study demonstrates that ratiometric fluorescence imaging at 1550 nm has superior spatial resolution to a depth up to 3.5 mm, which is 7.1-fold and 2.1-fold higher than that at 1064 nm and 1344 nm, respectively. Using an LPS-induced lymphatic inflammation model, we successfully observe the advancing of inflammatory lymphatic drainage with clear vessel structures via ratiometric fluorescence imaging. The ACIE process is depicted in Figure 1A. Er3+-doped DCNP with multiexcitation bands (808 nm and 980 nm) is used as a NIR-IIb signal unit. Cy7.5 with a maximum absorption at 800 nm is chosen as the competition absorber because of its unique features including large molar extinction coefficient (6 orders of magnitude larger than that of Er3+ ion at 808 nm), no emission in the NIR-IIb region and proper absorption overlapping only with one of the multiexcitation bands (808 nm) of the DCNPs (Figure 1B). Thus, Cy7.5 fluorophores modified around the surface of DCNPs serve as photon filtration layer on 808-nm excitation energy, causing the notable quenching of 1550 nm emission of Er3+ at this excitation channel (recorded as F1550Em, 808Ex). By utilizing the 980-nm excited 1550 nm emission as a reference signal (recorded as F1550Em,

980Ex),

the nanoprobes exhibit a ratiometric

response for HOCl (ratio = F1550Em, 808Ex/F1550Em, 980Ex) because the degradation of Cy7.5 induced by HOCl could recover the 1550-nm signals under 808-nm excitation. The nanoprobes (termed as [email protected]) were prepared from a click conjugation of azidebearing phospholipids (DSPE-PEG-N3) coated DCNP (termed as DCNP@N3) and multiple dibenzocyclooctyne (DBCO)-bearing phospholipids (DSPE-PEG-DBCO) nanomicelles containing Cy7.5 fluorophores (termed as Cy7.5@DBCO) (Scheme S2). To obtain effective downconversion emission, hexagonal phase (β) NaErxY1-xF4 core nanocrystals with varying Er3+ dopant concentrations (5, 15, 50 mol%) were synthesized, followed by growing a thick (∼10 nm) NaYF4 epitaxial shell to

ACS Paragon Plus Environment

Page 5 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Nano Letters

minimize surface quenching. All the nanocrystals have a uniform core size of ~43 nm and shell thickness of ~10 nm with a discernible contrast for the core/shell structure (Figure 1C, 1D and S1). The high-resolution transmission electron microscopy (HRTEM) image shows that the nanoparticles are highly crystalline hexagonal phase without any significant impurity phases (Figure 1E), which is consistent with the result of X-ray diffraction (XRD) patterns (Figure S2). After coating with a monolayer of phospholipids (90 mol% DSPE-mPEG and 10 mol% DSPE-PEG-N3), the oleic acidcapped DCNPs were transferred from oil phase to aqueous solution with an average hydrodynamic diameter of ~59 nm and good monodispersity (Figure 1G). Among all three nanocrystals, 50 mol% Er3+-doped DCNPs have the strongest fluorescence (Figure S3). To conjugate Cy7.5 with DCNP@N3, the fluorophores were encapsulated into nanomicelles consisting of 90 mol% DSPE-mPEG and 10 mol% DSPE-PEG-DBCO at a dye loading capacity of ~2 wt%, where no aggregation-induced broadening and reducing of absorption was observed (Figure S4). Cy7.5@DBCO exhibits superior photostability to Cy7.5 monomer and linear variation of absorbance in a wide range of concentration (0-25 μM) (Figure S5 and S6), ensuring the accurate quantification of ratiometric signals. Subsequently, Cy7.5@DBCO with a monodispersed size of ~14 nm was conjugated to the surface of DCNP@N3 through efficient click chemistry, which was confirmed by the zero absorption of Cy7.5 in supernatant after ultracentrifugation. Dynamic light scattering measurements show the increase of hydrodynamic diameter from ~59 nm of DCNP@N3 to ~80 nm of [email protected], suggesting the successful preparation of ACIE nanoprobes (Figure 1G). As a proof of concept, the fluorescence quenching process induced by the absorption competition effect between DCNPs and Cy7.5 was first studied. Upon addition of Cy7.5 from 0 to 25 μM, NIRII emission at 1550 nm of DCNPs was gradually quenched under 808-nm excitation (Figure 2A), but without any disturbance to the same emission under 980-nm excitation (Figure 2B). A plot of

ACS Paragon Plus Environment

Nano Letters 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 6 of 30

fluorescence intensity ratio (F1550Em, 808Ex/F1550Em, 980Ex) shows that the quenching process displays exponential decay with a rapid decrease occurred at a concentration range of 0-10 μM (Figure 2C). Lifetime measurements show no difference between DCNP@N3 and [email protected], indicating there is no energy transfer interaction between DCNP and Cy7.5 (Figure S7). Therefore, the quenching process is based on an ACIE mechanism that Cy7.5 with strong absorption at 808 nm attenuates the excitation energy on DCNPs through a photon filtration effect. According to Lambert-Beer law, photon energy transmits through absorber layer (equivalent to the excitation energy on DCNPs) scales as 10-Kbc, where K, b and c represent as absorption coefficient, optical length and concentration of absorber, respectively. Since the fluorescence of DCNPs show the linear power dependence (Figure S8 and S9), to a fixed thickness of filtration layer, fluorescence under 808 nm excitation should have an exponential relationship with the concentration of Cy7.5, namely F1550Em, 808Ex ∝ 0-Kbc, which is confirmed by the nonlinear quenching process in Figure 2C. Similar phenomenon is observed for 5 and 15 mol% Er3+-doped DCNPs (Figure S10), further indicating the quenching process depends on the filtration effect of Cy7.5. Given the superior quenching efficiency, we speculate the quenching process is a collective behavior of all Cy7.5 on the travel path of excitation light. By varying the path length at a fixed DCNP/Cy7.5 ratio with 15 μM Cy7.5, reduced quenching is observed while the highest quenching efficiency of about 95% is achieved at a path length of 5 mm (Figure S11). The results demonstrate the complementary effect of controlling absorber concentration and path length on the fluorescence regulation, which makes the ACIE technique flexible enough for biological application. We also noticed that 50 mol% Er3+-doped DCNPs have the most sensitive response to Cy7.5 at a concentration range of 0-10 μM, which can be attributed to the most sensitive power dependence fluorescence property (Figure S8 and S12). As a result, [email protected] consisting of 50 mol% Er3+-doped DCNPs and 10 μM Cy7.5 was used for the following experiments.

ACS Paragon Plus Environment

Page 7 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Nano Letters

Next, the ratiometric response of [email protected] to HOCl was further characterized. It has been demonstrated that ROS like HOCl can bleach Cy7.5 through the oxidative cleavage of the central polymethine chain.44 As expected, the absorbance of Cy7.5 at 808 nm decreases by titrating [email protected] with HOCl from 0 μM to 30 μM (Figure S13), accompanying with the gradual recovery of 1550 nm fluorescence under 808-nm excitation (Figure 2D). The reaction of Cy7.5 to HOCl can be completed within 1 minute, allowing for real-time monitoring of the HOCl fluctuation in vivo (Figure S13C). By using the unchanged fluorescence under 980-nm excitation as a reference signal (Figure 2E), the fluorescence ratio increases linearly with the concentrations of HOCl ranging from 0 μM to 20 μM, giving the detection limit as low as 500 nM (Figure 2F). Additionally, the fluorescence response selectivity to various competing ROS shows that Cy7.5@DCNP has the most sensitive response to HOCl compared with H2O2, O2•-, NO and HO• (Figure 2G-I). Encouraged by the above results, we expected the 1550 nm-emissive nanoprobes could revolutionize ratiometric image resolution. Although greater fluorescence imaging performance has been predicted for progressively longer wavelengths in the NIR-II region, to the best of our knowledge, the effect of wavelength changes within the NIR-II region on in vivo ratiometric image resolution has not been examined. Therefore, we first performed a tissue phantom study in 1% Intralipid to investigate the wavelength dependence of ratiometric fluorescence imaging (Figure 3A). We chose three separated wavelengths (1064 nm, 1344 nm and 1550 nm) which belong to three optical sub-windows previously defined in the range of 1000-1700 nm according to their improved fluorescence imaging performance (Figure 3B). Capillary filled with Er3+-doped DCNP immersed into Intralipid was used for generating 1550 nm ratiometric fluorescence signal under 808/980-nm excitation. Similarly, another two ratiometric fluorescence signals of 1064 nm and 1344 nm were acquired from capillaries filled with Nd3+-doped DCNP under 808/860-nm excitation (Figure S14). ACS Paragon Plus Environment

Nano Letters 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 8 of 30

Owing to the reduced scattering of NIR-II photons, capillary profiles are discernible for all groups in the depth range of 0-3.5 mm in fluorescence channel (Figure 3C, S15). Nevertheless, improved signal-to-background ratio (SBR) and narrowing of feature width are observed at longer wavelength (Figure 3D, F) due to the inverse wavelength dependence of photon scattering in 1% Intralipid (∼λ2.95,

Figure 3B). However, ratiometric imaging widens the SBR and feature width difference between

different wavelengths, for example, only capillary in 1550 nm ratio channel shows sharp pseudocolor contrast to the background in a penetration depth of 3.5 mm. Specifically, the SBR difference has changed from 2.6 times (fluorescence channel) to 53.6 times (ratio channel) between 1550 nm and 1064 nm, and from 1.1 times (fluorescence channel) to 20.8 times (ratio channel) between 1550 nm and 1344 nm (Figure 3D, 3E). Similar behavior is also observed for the changes of feature width difference in two channels (Figure 3F, 3G). A reasonable explanation for the above phenomenon can be attributed to the internal normalization of ratiometric imaging (Scheme S3), where response signal is divided by reference signal in a ratiometric operation. Thus, the strong capillary signals and weak scattering signals in the two fluorescent channels are all normalized to similar ratio values, in analogy to the “leveling effect” in chemistry, resulting in the contrast reduction between signal and background as well as the broadening of feature width (Supplementary note 1). The results demonstrate the greater influence of photon scattering to the resolution of ratiometric imaging than fluorescence imaging. We further investigate the superiority of 1550 nm ratiometric fluorescence in pork tissue phantom imaging, where the light-tissue interactions such as inhomogeneous scattering and tissue autofluorescence, are much more complex than Intralipid. We compared the ratiometric imaging performance of 1064 nm and 1550 nm, because they show a great difference of scattering, absorption and autofluorescence in tissue.23 As shown in Figure 4, two pairs of capillaries filled with 1064 nm

ACS Paragon Plus Environment

Page 9 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Nano Letters

emissive Nd3+-doped DCNPs and 1550 nm emissive Er3+-doped DCNPs were parallel arranged in a spacing of 2.5 (Nd) and 1.9 (Er) mm, respectively. In both cases without tissue covered, capillary edges are clearly resolved from fluorescence (Figure 4A, F) and ratio channels (Figure 4B, G). Crosssectional line profiles in these channels show that the two fluorescent signals have almost the same cross-sectional distribution (Figure 4C, H), which give a nearly constant ratio value in ratio channel (Figure 4E, J) after the internal normalization, without broadening of the feature width observed. When capillaries are covered with a 2 mm pork slice, their profiles become blurry at both wavelengths, but being worse for 1064 nm. Especially, for 1064 nm fluorescent imaging, there is a losing of similarity and symmetry for the cross-sectional signal distributions in two excitation channels, with significant increase of scattering background between two capillaries (Figure 4D). This is caused by the background overlapping from inhomogeneous photon scattering and tissue autofluorescence. Therefore, we introduced a background removed method to improve the ratiometric image resolution (Figure 4D, I). Specifically, in sensing channels (here are 860Ex for Nd and 808Ex for Er), signals below 20% of the maximum signal intensity (Fmax) are defined as background (< 0.2Fmax) whilst ensuring the minimum loss of main feature information, and their values are further set to 0 to eliminate their contribution to the resolution of ratio images. Thus, even at a narrower capillary spacing, 1550 nm ratiometric imaging can resolves the two isolated capillaries with almost unaffected resolution (Figure 4J). By contrast, the capillary features are still merged in 1064 nm ratiometric imaging (Figure 4E). To demonstrate the ability of our nanoprobes for in vivo high-resolution ratiometric fluorescence detection, [email protected] was applied to the imaging of HOCl in an animal model of lymphatic inflammation induced by intradermal (i.d.) injection of lipopolysaccharide (LPS) (Figure 5). Lymphatic drainage in the hindlimb of mice consists of several anatomical microstructures including

ACS Paragon Plus Environment

Nano Letters 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 10 of 30

two lymph nodes and two connective lymph vessels, which is suitable for evaluating imaging resolution (Figure 5B).44 Mice were injected i.d. with LPS or saline for control at the rear paw of mice (n=3), and 4 h later followed by an additional i.d. injection of [email protected] for HOCl detection through a homebuilt NIR-II fluorescence imaging system (Figure 5A). Significantly, popliteal and sciatic lymph nodes are clearly identified from all the images acquired post-injection at two fluorescent channels (808-nm and 980-nm excitation) and their corresponding ratio channel (Figure 5C, D). After 10 minutes post-injection, the afferent and efferent lymphatic vessels connecting the injection site with the two lymph nodes are gradually distinguished. In contrast, we observe sustained increase of ratiometric signals as time extend from 0 to 30 min in all lymph structures. The ratiometric signals of popliteal lymph nodes in LPS-treated mice increase about 2.05-fold during this time, while there is no obvious change in the control group of saline-treated mice (Figure 5E). The results demonstrate the ability of [email protected] to differentiate inflammation and normal tissue in living mouse through NIR-II ratiometric fluorescence imaging. We further analyzed the resolution of fluorescence and ratiometric signals by plotting the crosssectional signal profiles of the afferent lymphatic vessels (Figure 5F, G). The results show that the SBR/Gaussian-fitted FWHM of fluorescent signal is ~8.9/~645 µm and ~38/~273 µm for LPS-treated and saline-treated group, respectively. Both the two groups show high contrast and µm-scale resolution at fluorescent imaging mode owing to the low scattering and low autofluorescence background at 1550 nm. In addition, the feature width of the same position at ratio channel shows approximately 2-fold broader, which is ~1114 and ~477 µm for LPS-treated and saline-treated group, respectively. This difference is consistent well with the result predicted from the capillary imaging model (Scheme S3), thus indicating there is no feature broadening in ratiometric imaging.

ACS Paragon Plus Environment

Page 11 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Nano Letters

Overall, this work pushed the ratiometric sensing window to the brand-new NIR-IIb region (15001700 nm), whereby we preliminarily realized mm- to μm-scale resolution in ratiometric imaging of lymphatic inflammation without significant feature broadening. This unprecedented resolution offer opportunities for quantitative visualization of biological events in vivo both in space- and time-scale. Although the idea that NIR-II light with reduced tissue scattering facilitates high imaging clarity has been widely accepted, we show that the resolution of ratiometric imaging will benefit more from the reduced scattering of long wavelength compared with fluorescent imaging. The underlying principle is that the ratiometric operation levels the true signal and all scattering backgrounds, resulting in the amplification of background signals in ratio channel. Furthermore, such “leveling effect” is even more prominent in deep biological tissue due to the additional interference from tissue autofluorescence and overlapping scattering background from densely emitting structures. Here, we show that a background removed solution by setting a threshold to eliminate the background interference can effectively improve resolution in 2 mm thick tissue. In this case, the inherent low background character of 1500-1700 nm permits a lower background threshold to be set so that more low-intensity details could be resolved. These results will then motivate the development of bright NIR-IIbemissive fluorophores and precise targeting technologies for clarifying feature structures in ratiometric imaging. Perhaps the chief obstacle to realize NIR-IIb ratiometric fluorescent imaging is the lack of available fluorophores with specific response at this long wavelength. Currently, only Er3+-doped lanthanide nanoparticles, carbon nanotubes, PbS and InAs quantum dots have NIR-IIb emission but they all lack flexible strategies to regulate fluorescence for sensing application. Here, the proposed ACIE technique provides a general solution to regulate the fluorescence of lanthanide activators in DCNPs by exploiting the absorption competition effect. Lanthanide ions have low absorption cross-section,

ACS Paragon Plus Environment

Nano Letters 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 12 of 30

and given that the absorptivity of organic dyes is normally orders of magnitude higher, the ACIE technique can therefore take full advantage of the efforts for years at short-wavelength dye probes, which will avoid the complex design of NIR-II dyes. In addition, owing to the unique set of distinguishable spectroscopic fingerprints of lanthanide ions,46,47 the ACIE technique may further expand the library of NIR-II nanoprobes for biosensing with high multiplexing capacity. Through this technique, our group has realized semi-quantitative monitoring of oral drug delivery by utilizing the sensing system composed of phthalocyanin dye (NPTAT) and Nd3+-doped DCNP.48 In this work, the difference in molar absorptivity of Cy7.5 versus Er3+ (106-fold) is three orders of magnitude greater than that of NPTAT versus Nd3+ (103-fold), which afford higher sensitivity to discriminate the pathological and physiological concentrations of HOCl. Although the accurately pathological concentrations of HOCl and various potentially interfering ROS species (H2O2, O2-, NO and ∙OH) are unclear to date due to the confounding factors such as in vivo cellular half-life of the ROS species, variability of tissue response to injury, and rate of extracellular ROS diffusion, we can roughly estimate their concentration ranges from reported references. For example, it is widely accepted that H2O2 with moderate half-life (10-3-10-5 s) shows the highest level among all ROS species in inflammation site,49 which has a concentration of about 50-300 μM.50,

51

And about

28%~72% of H2O2 was used for the formation of HOCl by stimulated neutrophils through the myeloperoxidase-H2O2-Cl- system.52 As HOCl has a relative long half-life (~1-~10 s),53 its concentration may be located at μM-scale range. Other ROS species such as O2- and ∙OH may have pM to nM concentrations due to their shorter half-lives (e.g., O2- ~ 10-6 s, ∙OH ~10-9-~10-15 s)53. Few reports about the concentration of NO was found, but it should be no more than that of H2O2. As a result, the selectivity results in Figure 2I ensure that this response is specific to HOCl in inflammation site. Furthermore, it has been reported that injection of nanoparticles has limited influence on the ACS Paragon Plus Environment

Page 13 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Nano Letters

drainage of the lymphatic fluid,54 while anesthesia and compression have more significant impacts.55 In this work, all mice were anaesthetized during the fluorescence imaging. The response time of nanoprobes is within 1 minute (Figure S13C). Therefore, the reasons for why ratiometric changes took ~30 min to reach maximum may involve multiply aspects, such as the slow lymphatic drainage under anesthesia, the diffusion-limited low reaction rate between nanoprobes and HOCl, and the HOCl homeostasis during ratiometric response. Notably, Figure 5E shows that the ratiometric signal increases about 1.69-fold over the first 5 minutes, which is significantly higher than the total increasement (0.36-fold) over the following 5-30 minutes. This indicates that the nanoprobe shows fast response to the cumulative HOCl that generated before injection and shows sustained response to the newly generated HOCl during the following 30 minutes. The Er3+-based ratiometric signals are relatively reliable within the depth range of 1-3.5 mm in Intralipid, showing a coefficient of variation lower than 11% (Figure S16). Although this value is higher compared with Nd3+-based ratiometric signals (< 6%) and does not ensure the accurate quantification in vivo, at least the influence of tissue depth to the observed 1550 nm ratiometric changes (2.05-fold enhancement) in Figure 5E can be ruled out. As a result, if taking the coefficient of variation of signal fluctuation in mimic tissue into account, an estimated concentration of HOCl can be derived as 7.0 ± 2.0 μM during the 30 minutes of inflammatory response (Supplementary note 2). This data is consistent with the estimated pathological concentration of HOCl as discussed above. Nevertheless, there remains room for improvement; for example, there would be a need to realize reversible imaging of HOCl due to the cellular redox homeostasis.56 This could be possible if using redox-activated probes56 as absorption competitors with colorimetric response at Er3+ absorption bands (655, 808 and 980 nm). In addition, a more accurate and high-resolution ratiometric imaging system may be realizable if future works focus on a Nd3+/ Er3+ coupled system (730, 808 and 860 nm ACS Paragon Plus Environment

Nano Letters 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 14 of 30

excitation, 1550 nm emission)34 due to the less attenuation difference between excitation wavelengths. In summary, we have developed a 1550 nm emissive ratiometric nanoprobe in response to HOCl and further demonstrated their utility for high-resolution (μm-scale) ratiometric sensing of lymphatic inflammation in vivo. The ratiometric system is based on an absorption competition-induced emission mechanism between Cy7.5 fluorophores and Er3+-doped lanthanide nanoparticles, thus integrating all advantages of the two components, such as the fast (within 1 min), sensitive (500 nM) and selective response of Cy7.5 to HOCl as well as the long-wavelength ratiometric fluorescence of Er3+ under 808-nm and 980-nm excitation. We believe this study will inspire further research on the development of NIR-II ratiometric fluorescent probes.

ACS Paragon Plus Environment

Page 15 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Nano Letters

ASSOCIATED CONTENT SUPPORTING INFORMATION. This Supporting Information is available free of charge via the Internet at http://pubs.acs.org. Materials and methods; figures providing additional spectra and characterization data. (PDF) AUTHOR INFORMATION Corresponding Author *[email protected] Notes The authors declare no competing financial interest. ACKNOWLEDGMENT The work was supported by the National Key R & D Program of China (2017YFA0207303), National Natural Science Foundation of China (NSFC, 21725502), Key Basic Research Program of Science and Technology Commission of Shanghai Municipality (17JC1400100). The authors extend their appreciation to the International Scientific Partnership Program ISPP at King Saud University for funding this research work through ISPP#0100.

REFERENCES (1) Vahrmeijer, A. L.; Hutteman, M.; van der Vorst, J. R.; van de Velde, C. J. H.; Frangioni, J. V. Nat. Rev. Clin. Oncol. 2013, 10, 507.

ACS Paragon Plus Environment

Nano Letters 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

(2) Kobayashi, H.; Ogawa, M.; Alford, R.; Choyke, P. L.; Urano, Y. Chem. Rev. 2009, 110, 2620. (3) Xu, G.; Yan, Q.; Lv, X.; Zhu, Y.; Xin, K.; Shi, B.; Wang, R.; Chen, J.; Gao, W.; Shi, P.; Fan, C.; Zhao, C.; Tian, H. Angew. Chem. Int. Ed. 2018, 57, 3626. (4) Jeong, S.; Song, J.; Lee, W.; Ryu, Y. M.; Jung, Y.; Kim, S.-Y.; Kim, K.; Hong, S. C.; Myung, S. J.; Kim, S. Nano Lett. 2017, 17, 1378. (5) Dang, X.; Gu, L.; Qi, J.; Correa, S.; Zhang, G.; Belcher, A. M.; Hammond, P. T. Proc. Natl. Acad. Sci. USA 2016, 113, 5179. (6) Shuhendler, A. J.; Pu, K.; Cui, L.; Uetrecht, J. P.; Rao, J. Nat. Biotech. 2014, 32, 373. (7) Welsher, K.; Sherlock, S. P.; Dai, H. Proc. Natl. Acad. Sci. USA 2011, 108, 8943. (8) Huang, X.; Song, J.; Yung, B. C.; Huang, X.; Xiong, Y.; Chen, X. Chem. Soc. Rev. 2018, 47, 2873. (9) Lee, M. H.; Kim, J. S.; Sessler, J. L. Chem. Soc. Rev. 2014, 44, 4185. (10)Zhou, L.; Zhang, X.; Wang, Q.; Lv, Y.; Mao, G.; Luo, A.; Wu, Y.; Wu, Y.; Zhang, J.; Tan, W. J. Am. Chem. Soc. 2014, 136, 9838. (11)Aron, A. T.; Loehr, M. O.; Bogena, J.; Chang, C. J. J. Am. Chem. Soc. 2016, 138, 14338. (12) Narayanaswamy, N.; Chakraborty, K.; Saminathan, A.; Zeichner, E.; Leung, K.; Devany, J.; Krishnan, Y. Nat. Meth. 2018, 16, 95.

ACS Paragon Plus Environment

Page 16 of 30

Page 17 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Nano Letters

(13)Richardson, D. S.; Gregor, C.; Winter, F. R.; Urban, N. T.; Sahl, S. J.; Willig, K. I.; Hell, S. W. Nat. Commun. 2017, 8, 577. (14)Wang, J.; MacEwan, S. R.; Chilkoti, A. Nano Lett. 2017, 17, 1226. (15) Makukhin, N.; Tretyachenko, V.; Moskovitz, J.; Míšek, J. Angew. Chem. Int. Ed. 2016, 55, 12727. (16)Ma, T.; Hou, Y.; Zeng, J.; Liu, C.; Zhang, P.; Jing, L.; Shangguan, D.; Gao, M. J. Am. Chem. Soc. 2017, 140, 211. (17)Pu, K.; Shuhendler, A. J.; Rao, J. Angew. Chem. Int. Ed. 2013, 52, 10325. (18)Bianchi-Smiraglia, A.; Rana, M. S.; Foley, C. E.; Paul, L. M.; Lipchick, B. C.; Moparthy, S.; Moparthy, K.; Fink, E. E.; Bagati, A.; Hurley, E.; Affronti, H. C.; Bakin, A. V.; Kandel, E. S.; Smiraglia, D. J.; Feltri, M. L.; Sousa, R.; Nikiforov, M. A. Nat. Meth. 2017, 14, 1003. (19)Umezawa, K.; Yoshida, M.; Kamiya, M.; Yamasoba, T.; Urano, Y. Nat. Chem. 2016, 9, 279. (20)Wu, L.; Wu, I. C.; DuFort, C. C.; Carlson, M. A.; Wu, X.; Chen, L.; Kuo, C. T.; Qin, Y.; Yu, J.; Hingorani, S. R.; Chiu, D. T. J. Am. Chem. Soc. 2017, 139, 6911. (21)Wang, N.; Yu, X.; Zhang, K.; Mirkin, C. A.; Li, J. J. Am. Chem. Soc. 2017, 139, 12354. (22)Smith, A. M.; Mancini, M. C.; Nie, S. Nat Nanotechnol 2009, 4, 710. (23)Hong, G.; Antaris, A. L.; Dai, H. Nat. Biomed. Eng. 2017, 1, 0010.

ACS Paragon Plus Environment

Nano Letters 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

(24)Shou, K.; Qu, C.; Sun, Y.; Chen, H.; Chen, S.; Zhang, L.; Xu, H.; Hong, X.; Yu, A.; Cheng, Z. Adv. Funct. Mater. 2017, 27, 1700995. (25)Fan, Y.; Wang, P.; Lu, Y.; Wang, R.; Zhou, L.; Zheng, X.; Li, X.; Piper, J. A.; Zhang, F. Nat. Nanotechnol. 2018, 13, 941. (26)Qi, J.; Sun, C.; Zebibula, A.; Zhang, H.; Kwok, R. T. K.; Zhao, X.; Xi, W.; Lam, J. W. Y.; Qian, J.; Tang, B. Z. Adv. Mater. 2018, 30, e1706856. (27)Kenry; Duan, Y.; Liu, B. Adv. Mater. 2018, 30, e1802394. (28)Zhu, S.; Hu, Z.; Tian, R.; Yung, B. C.; Yang, Q.; Zhao, S.; Kiesewetter, D. O.; Niu, G.; Sun, H.; Antaris, A. L.; Chen, X. Adv. Mater. 2018, 0, 1802546. (29)Tang, Y.; Li, Y.; Hu, X.; Zhao, H.; Ji, Y.; Chen, L.; Hu, W.; Zhang, W.; Li, X.; Lu, X.; Huang, W.; Fan, Q. Adv. Mater. 2018, 30, 1801140. (30)Bashkatov, A. N.; Genina, E. A.; Kochubey, V. I.; Tuchin, V. V. J. Phys. D: Appl. Phys. 2005, 38, 2543. (31)Diao, S.; Blackburn, J. L.; Hong, G.; Antaris, A. L.; Chang, J.; Wu, J. Z.; Zhang, B.; Cheng, K.; Kuo, C. J.; Dai, H. Angew. Chem. Int. Ed. 2015, 54, 14758. (32)Diao, S.; Hong, G.; Antaris, A. L.; Blackburn, J. L.; Cheng, K.; Cheng, Z.; Dai, H. Nano Res. 2015, 8, 3027. (33)Naczynski, D. J.; Tan, M. C.; Zevon, M.; Wall, B.; Kohl, J.; Kulesa, A.; Chen, S.; Roth, C. M.; Riman, R. E.; Moghe, P. V. Nat. Commun. 2013, 4, 2199.

ACS Paragon Plus Environment

Page 18 of 30

Page 19 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Nano Letters

(34)Wang, R.; Li, X.; Zhou, L.; Zhang, F. Angew. Chem. Int. Ed. 2014, 53, 12086. (35)Johnson, N. J. J.; He, S.; Diao, S.; Chan, E. M.; Dai, H.; Almutairi, A. J. Am. Chem. Soc. 2017, 139, 3275. (36)Zhong, Y.; Ma, Z.; Zhu, S.; Yue, J.; Zhang, M.; Antaris, A. L.; Yuan, J.; Cui, R.; Wan, H.; Zhou, Y.; Wang, W.; Huang, N. F.; Luo, J.; Hu, Z.; Dai, H. Nat. Commun. 2017, 8, 737. (37)Lei, X.; Li, R.; Tu, D.; Shang, X.; Liu, Y.; You, W.; Sun, C.; Zhang, F.; Chen, X. Chem. Sci. 2018, 9, 4682. (38)Liu, X.; Chen, Q.; Xie, X.; Huang, B.; Liang, L.; Han, S.; Yi, Z.; Wang, Y.; Li, Y.; Fan, D.; Huang, L. Angew. Chem. Int. Ed. 2017, 56, 7605. (39)Zhang, M.; Yue, J.; Cui, R.; Ma, Z.; Wan, H.; Wang, F.; Zhu, S.; Zhou, Y.; Kuang, Y.; Zhong, Y.; Pang, D.-W.; Dai, H. Proc. Natl. Acad. Sci. USA 2018, 115, 6590. (40) Franke, D.; Harris, D. K.; Chen, O.; Bruns, O. T.; Carr, J. A.; Wilson, M. W. B.; Bawendi, M. G. Nat. Commun. 2016, 7, 12749. (41)Bruns, O. T.; Bischof, T. S.; Harris, D. K.; Franke, D.; Shi, Y.; Riedemann, L.; Bartelt, A.; Jaworski, F. B.; Carr, J. A.; Rowlands, C. J.; Wilson, M. W. B.; Chen, O.; Wei, H.; Hwang, G. W.; Montana, D. M.; Coropceanu, I.; Achorn, O. B.; Kloepper, J.; Heeren, J.; So, P. T. C.; Fukumura, D.; Jensen, K. F.; Jain, R. K.; Bawendi, M. G. Nat. Biomed. Eng. 2017, 1, 0056. (42)Yang, J.; Zhang, X.; Yuan, P.; Yang, J.; Xu, Y.; Grutzendler, J.; Shao, Y.; Moore, A.; Ran, C. Proc. Natl. Acad. Sci. USA 2017, 114, 12384. (43)Strowig, T.; Henao-Mejia, J.; Elinav, E.; Flavell, R. Nature 2012, 481, 278. ACS Paragon Plus Environment

Nano Letters 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

(44)Oushiki, D.; Kojima, H.; Terai, T.; Arita, M.; Hanaoka, K.; Urano, Y.; Nagano, T. J. Am. Chem. Soc. 2010, 132, 2795. (45)Van den Broeck, W.; Derore, A.; Simoens, P. J. Immuno. Methods 2006, 312, 12. (46)Gai, S.; Li, C.; Yang, P.; Lin, J. Chem. Rev. 2013, 114, 2343. (47)Gorris, H. H.; Wolfbeis, O. S. Angew. Chem. Int. Ed. 2013, 52, 3584. (48)Wang, R.; Zhou, L.; Wang, W.; Li, X.; Zhang, F. Nat. Commun. 2017, 8, 14702. (49) Ermakova, Y. G.; Bilan, D. S.; Matlashov, M. E.; Mishina, N. M.; Markvicheva, K. N.; Subach, O. M.; Subach, F. V.; Bogeski, I.; Hoth, M.; Enikolopov, G.; Belousov, V. V. Nat. Commun. 2014, 5, 5222. (50) Chen, Q.; Liang, C.; Sun, X.; Chen, J.; Yang, Z.; Zhao, H.; Feng, L.; Liu, Z. Proc. Natl. Acad. Sci. USA 2017, 114, 5343. (51) Sen, C. K.; Roy, S. Biochim. Biophys. Acta 2008, 1780, 1348. (52) Klebanoff, S. J. J. Leukocyte Bio. 2005, 77, 598. (53) Parvez, S.; Long, M. J. C.; Poganik, J. R.; Aye, Y. Chem. Rev. 2018, 118, 8798. (54) Yang, Z.; Tian, R.; Wu, J.; Fan, Q.; Yung, B. C.; Niu, G.; Jacobson, O.; Wang, Z.; Liu, G.; Yu, G.; Huang, W.; Song, J.; Chen, X. ACS Nano 2017, 11, 4247. (55) Proulx, S. T.; Luciani, P.; Christiansen, A.; Karaman, S.; Blum, K. S.; Rinderknecht, M.; Leroux, J.-C.; Detmar, M. Biomaterials 2013, 34, 5128. (56)Lou, Z.; Li, P.; Han, K. Acc. Chem. Res. 2015, 48, 1358. ACS Paragon Plus Environment

Page 20 of 30

Page 21 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Nano Letters

Figure 1. (A, B) Schematic illustration showing the ratiometric response of [email protected] to HOCl based on an ACIE mechanism. Er3+-doped DCNP: NaErxY1-xF4@NaYF4 (x = 0.05, 0.15, 0.5). When excitation is performed in the absorption overlapping region (808 nm), fluorescence of DCNPs are quenched due to the energy filtration of excitation light by strong absorbing Cy7.5. The quenching process is reversed after the degradation of Cy7.5 by HOCl oxidation. In the absorption nonoverlapping region, fluorescence of DCNPs under 980-nm excitation are unaffected, thus providing ratiometric readout. (C) TEM, (D) HAADF-STEM, (E) HRTEM images of oleic acid-capped DCNP (NaEr0.5Y0.5F4@NaYF4) in cyclohexane. (F) TEM image of DCNP@N3 in water. (G) DLS measurements of Cy7.5@DBCO, DCNP@N3 and [email protected] in water.

ACS Paragon Plus Environment

Nano Letters 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Figure 2. (A, B) Fluorescence spectra of DCNP with varying concentrations of Cy7.5 (0-25 μM) under 808-nm and 980-nm excitation respectively. (C) Plot of fluorescence ratio changes as a function of Cy7.5 concentration. (D, E) Fluorescence spectra of [email protected] (10 μM Cy7.5) upon addition of 0-30 μM HOCl under 808-nm and 980-nm excitation respectively. (F) Plot of fluorescence ratio changes as a function of HOCl concentration. (G, H) Fluorescence spectra of [email protected] (10 μM Cy7.5) upon addition of various ROS (100 μM for H2O2, O2•-, NO, 50 μM •OH and 20 μM HOCl) under 808-nm and 980-nm excitation respectively. (I) Fluorescence ratio changes in the presence of various ROS. DCNP: NaY0.5Er0.5F4@NaYF4,

ACS Paragon Plus Environment

Page 22 of 30

Page 23 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Nano Letters

~2.4 mg/mL containing ~0.26 mg/mL of Er3+. The bars represent mean ± s.d. derived from n = 3 replicate measurements. p values were analyzed by Student’s two-sided t-test (**P < 0.01, n.s. represents no significant differences).

ACS Paragon Plus Environment

Nano Letters 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Figure 3. (A) A homebuilt NIR-II ratiometric fluorescence imaging setup for tissue phantom study. (B) Absorbance spectra of 1% Intralipid (red) and water (blue) measured with a 1 mm path length cuvette, Intralipid scattering (green) was acquired by subtracting Intralipid and water absorptions. Plot fitting shows photon scattering scales as λ-2.95. (C) NIR-II fluorescence images (1064Em, 1344Em and 1550Em) and corresponding ratiometric images (Ratio) of capillary immersed in 1% Intralipid with varied depth of 0 mm, 2 mm and 3.5 mm. Detailed images at 0, 1, 1.5, 2, 2.5, 3 and 3.5 mm are shown in Figure S15. Fluorescent signals of 1064Em and 1344Em were collected from Nd3+-doped DCNP under 808-nm and 860-nm excitation. Fluorescent signals of 1550Em was collected from Er3+-doped DCNP under 808-nm

ACS Paragon Plus Environment

Page 24 of 30

Page 25 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Nano Letters

and 980-nm excitation. (D, E) Normalized signal-to-background ratio (SBR, defined as (ROI1ROI2)/ROI1) of capillary images in fluorescent channel (D) and ratio channel (E) obtained for different wavelengths with respect to immersion depth in Intralipid. (F, G) Feature width of capillaries in fluorescent channel (F) and ratio channel (G) obtained for different wavelengths with respect to immersion depth in Intralipid. The bars represent mean ± s.d. derived from n = 3 replicate analysis results.

ACS Paragon Plus Environment

Nano Letters 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Figure 4. Comparison of ratiometric fluorescence imaging using 1064 and 1550 nm to resolve a pair of parallel arranged capillaries in pork tissue phantom study. Fluorescence images (A, F) and corresponding ratiometric fluorescence images (B, G) of capillaries with 0 mm/2 mm pock tissue covered, recorded in 1064 nm (A, B) and 1550 nm (F, G), respectively. Cross-sectional line profiles (white-dashed lines in images) of corresponding images recorded in 1064 nm (CE) and 1550 nm (H-J), respectively. The spacing is about 2.5 mm for the two capillaries in 1064 nm group, and 1.9 mm for 1550 nm group. A background removed method was introduced for processing ratiometric fluorescence images under 2 mm pock tissue covered. Signals with intensity below 20% of the maximum intensity in sensing channel (808Ex channel for 1550 nm,

ACS Paragon Plus Environment

Page 26 of 30

Page 27 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Nano Letters

860Ex channel for 1064 nm) were defined as background, and their values were further set to 0 to eliminate their contribution to the ratio images.

ACS Paragon Plus Environment

Nano Letters 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Figure 5. (A) Schematic illustration showing NIR-II ratiometric fluorescence imaging of LPSinduced lymphatic inflammation using nanoprobes. (B) The anatomical structure of lymphatic system in the hindlimb of mice, green arrow represents the lymphatic drainage from the paw to the sciatic lymph node. (C, D) In vivo NIR-II fluorescence images and corresponding ratiometric images of LPS-treated and saline-treated mouse lymphatic drainage at different times post-injection of [email protected]. (E) Ratiometric signals of popliteal lymph node (white dotted circle) obtained from LPS-treated and saline-treated groups over time. The bars represent mean ± s.d. derived from n = 3 biologically independent mice. p values were analyzed between LPS-treated and saline-treated mice at all time points starting from 5 min by Student’s two-

ACS Paragon Plus Environment

Page 28 of 30

Page 29 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Nano Letters

sided t-test (*P < 0.05). (F, G) Cross-sectional intensity (black)/ratio (blue) profile along the yellow-color bars in the images, Gaussian fit of the intensity curves are shown in red curves.

ACS Paragon Plus Environment

Nano Letters 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

TOC Graphic

ACS Paragon Plus Environment

Page 30 of 30