A Novel Near-Infrared Fluorescence Imaging Probe for in Vivo Neutrophil Tracking

Abstract

The development and validation of a multiscopic near-infrared fluorescence (NIRF) probe, cinnamoyl-F-(D)L-F-(D)L-F-PEG-cyanine7 (cFlFlF-PEG-Cy7), that targets formyl peptide receptor on neutrophils using a mice ear inflammation model is described. Acute inflammation was induced in mice by topical application of phorbol-12-myristate-13-acetate to left ears 24 hours before probe administration. Noninvasive NIRF imaging was longitudinally performed up to 24 hours following probe injection. The in vivo neutrophil-targeting specificity of the probe was characterized by a blocking study with preadministration of excess nonfluorescent peptide cFlFlF-PEG and by an imaging study with a scrambled peptide probe cLFFFL-PEG-Cy7. NIRF imaging of mice injected with cinnamoyl-L-F-F-F-L-PEG-cyanine7 (cFlFlF-PEG-Cy7) revealed that the fluorescence intensity for inflamed left ears was approximately fourfold higher than that of control right ears at 24 hours postinjection. In comparison, the ratios acquired with the scrambled probe and from the blocking study were 1.5- and 2-fold at 24 hours postinjection, respectively. Moreover, a microscopic immunohistologic study confirmed that the NIRF signal of cFlFlF-PEG-Cy7 was associated with activated neutrophils in the inflammatory tissue. With this probe, in vivo neutrophil chemotaxis could be correlatively imaged macroscopically in live animals and microscopically at tissue and cellular levels.

NEUTROPHILS are the most abundant type of white blood cells in mammals and are an essential part of the innate immune system, responding to a variety of stress signals. In a number of inflammatory disorders, such as bacterial infection, neutrophils are the first respondents activated and recruited to the site of injury, with subsequent release of reactive oxygen species and proteolytic enzymes from them to combat the infection. However, if the response from neutrophils is more than desired, the tissue damage associated with such disease conditions ensues.^1,2 An understanding of how neutrophils interact with a stressed tissue environment and how to manage their response is crucial for the development of effective intervention and therapies to attenuate inflammation. Noninvasive imaging technologies that enable visualization of neutrophil chemotaxis and activation in live animals are thus highly desirable.

Study of gross neutrophil recruitment and chemotaxis in patients and animals has been achieved by using nuclear imaging modalities with either ex vivo labeled leukocytes^3,4 or intravenously injected imaging probes that directly target formyl peptide receptor (FPR) expressed on neutrophils.^5–7 Nuclear imaging has the advantage of high sensitivity resulting from high tissue penetration of gamma photons, but its application in small-animal research settings is somehow impeded by the cost of the relevant instruments, laborious synthesis, and the limited shelf life of the nuclear probes. Moreover, the inherent low resolution of autoradiography prevents the study of neutrophil trafficking from minute details at tissue and cellular levels.

During the last decade, a novel modality, near-infrared fluorescence (NIRF) imaging of small animals, has gone through tremendous technological advances because of the merits of operational simplicity and low cost.^8–10 As a result, considerable efforts have been directed toward the development of new NIRF probes targeting in vivo molecular events and biochemical processes.^2,11 NIRF imaging has been widely used to monitor cell trafficking^12,13 macroscopically in live animals by labeling the cells with a variety of NIRF probes. In principle, the NIRF probe–labeled cells, along with their associated in vivo pathophysiologic process, can be studied with either fluorescence microscopy or an intravital fluorescence imaging device at the tissue or cellular level. Imaging at the cellular level can be performed before and after in vivo settings to delineate the function of the cells in various physiologic and pathophysiologic conditions. Eventually, correlation of whole-body imaging with high-resolution microscopic imaging should provide a better understanding of biochemical processes. Multiresolution data sets acquired from the exact same probe would make the correlation simpler and more straightforward.

To study neutrophil trafficking and function in vivo by NIRF modalities, labeling the cells with a NIRF fluorophore is required. The in vivo labeling procedure is preferred because the ex vivo separation and labeling method might result in undesired consequences, such as contamination and activation of neutrophils. We recently reported the synthesis of cinnamoyl-F-(D)L-F-(D)L-F-PEG-cyanine7 (cFlFlF-PEG-Cy7) for noninvasive NIRF imaging of neutrophils in small animals.¹⁴ In this article, validation of this probe in vitro and in vivo is presented and fluorescence images of live animals and microscopic tissue slices are correlated. The probe cFlFlF-PEG-Cy7 consists of three components, an FPR-binding peptide sequence (cFlFlF),¹⁵ a hydrophilic modifier moiety (PEG), and a fluorophore (Cy7), to attain desired pharmacologic and optical properties. The fluorescence spectrum (excitation/emission [E_x/E_m]: 745 nm/800 nm) of Cy7 provides better photon signal penetration through animal tissues because of the minimum tissue absorbance and autofluorescence in the near-infrared region (700–900 nm).¹⁰ The feasibility of in vivo neutrophil imaging was validated with a mouse model of acute ear inflammation. The in vivo neutrophil specificity of the probe was demonstrated by a blocking study and an imaging study with a scrambled peptide probe. The neutrophil-binding property of cFlFlF-PEG-Cy7 was also verified by fluorescence microscopy.

Materials and Methods

Chemicals used for synthesis were commercially available and of analytical grade and used without further purification. Fmoc-PAL-PS resin and Fmoc-amino acids were purchased from Applied Biosystems (Foster City, CA). t-Boc-PEG-NHS-3.4 kDa was purchased from Laysan Bio Inc (Arab, AL). Cy7 Mono NHS ester was purchased from GE Healthcare (Piscataway, NJ). Fetal bovine serum (FBS) was purchased from Invitrogen Corporation (Carlsbad, CA). High-performance liquid chromatography (HPLC) grade solvents were purchased from Fisher Scientific (Pittsburgh, PA). All other chemical reagents and solvents were purchased from Sigma-Aldrich (St. Louis, MO).

Semipreparative reversed-phase high-performance liquid chromatography (RP-HPLC) was performed on a Varian ProStar system (models: pumps, 210; column valve module, 500; fraction collector, 701) (Varian Instruments, Walnut Creek, CA) equipped with an ABI SpectroFlow 783 ultraviolet (UV) detector and an Apollo C18 reversed-phase column (5 μm, 250 × 10 mm) (Grace Davison Discovery Sciences, Deerfield, IL). For all chromatographic purification, the mobile phase gradient from 60% solvent

A, 0.1% trifluoroacetic acid (TFA) in water, and 40% solvent B, 0.1% TFA in 80% aqueous acetonitrile (CH₃CN), to 100% solvent B for 30 minutes at a flow rate of 3 mL/min. was used. In all purification experiments, the UV detector wavelength was set to 215 nm. Matrix-assisted laser desorption/ionization time-of-flight mass spectrometry (MALDI-TOF MS) analysis was performed on all peptide products at the W.M. Keck Biomedical Mass Spectrometry Laboratory at The University of Virginia, and the data were obtained on a Bruker Daltonics system (Billerica, MA).¹⁴

Mice

Male FVB mice, 3 months old (about 20 g/mouse), were purchased from the National Cancer Institute (Frederick, MD). All mice were housed in a controlled environment (12 hours light/12 hours dark photoperiod, 22 ° 1°C, 60 ° 10% relative humidity) and were provided with free access to autoclaved pellet food (type 7912, Harlan Laboratories, Madison, WI) and tap water. All in vivo studies were performed in compliance with The University of Virginia Animal Study Committee's requirements for the care and use of laboratory animals in research. Mice were anesthetized with isofluorane (2%) during probe injection and imaging.

Synthesis of cFlFlF-PEG-Cy7 6 and cLFFFL-PEG-Cy7 7

Peptide cFlFlF-PEG 4 and scrambled peptide cLFFFL-PEG 5 were synthesized using standard Fmoc chemistry following a previously reported procedure.^6,14 We scrambled the sequence of cFlFlF-PEG 4 and replaced (D)-Leu with (L)-Leu to form a scrambled peptide cLFFFL-PEG 5. Cy7 conjugation was carried out following a previously reported procedure¹⁴ (Figure 1). All intermediates and final products were purified by RP-HPLC to homogeneity, and the purity of the probes was > 95%. Structural characterization of peptides 4 and 5 and Cy7 conjugates 6 and 7 was performed by MALDI-TOF mass spectrometry.

Figure 1.

Synthesis of fluorescence imaging probes. Reagents and conditions: (i) and (ii) standard solid-phase Fmoc chemistry; (iii) 1, t-Boc-NH-PEG_3.4kD-NHS; 2, prechilled trifluoroacetic acid (95%); (iv) cyanine 7 Mono NHS ester, CH₃CN/sodium borate buffer, pH 8.5, 4°C to room temperature.

Fluorescence Spectra

Solutions of Cy7 conjugates (6 and 7, 20 μM each) were freshly prepared by dissolving samples in a mixture solvent (H₂O/dimethyl sulfoxide [DMSO], 97/3, v/v). Fluorescence emission spectra were recorded on a Fluorolog-2 spectrofluorometer (Jobin Yvon/Horiba, Edison, NJ). The slit width was 5 nm, and the integration time was 0.2 seconds for all measurements. The excitation wavelength used was 745 nm. The recorded fluorescence intensity was normalized and plotted versus the emission wavelength.¹⁴

Partition Coefficient Measurement

The measurement of partition coefficient (log P) was performed following a previously reported method.¹⁴ In brief, equal amounts (5 nmol) of Cy7 conjugates 6 and 7 were dissolved in an equal volume (600 μL) mixture of n-octanol and deionized water (preequilibrated overnight). The fluorescence intensities of 100 μL aliquots (in triplicate) from both layers were measured by a Xenogen IVIS Spectrum (Caliper Life Sciences, Hopkinton, MA). Log P was calculated using the equation log fluorescence intensity in octanol/fluorescence intensity in water, in which fluorescence intensity was corrected by a factor of solvent effect. The factor was calculated based on different fluorescence intensities of the same concentration of probes in n-octanol and water.

PMA-Induced Ear Inflammation

Topical application of phorbol-12-myristate-13-acetate (PMA) (5 μg in 20 μL of vehicle [DMSO and acetone, 1/9, v/v]) onto the left earlobe of the mice could induce acute dermatitis, manifested by local swelling, erythema, and infiltration of neutrophils.^16,17 The right earlobe served as the control and received only the vehicle. We used this model for neutrophil-specific fluorescence imaging at 24 hours after PMA application.

Measurement of Ear Thickness

The ear thickness was measured before and after PMA challenge on both ears. Measurements were made using an engineer's micrometer (Mitutoyo Corp., Kawasaki, Japan), with care being taken to measure only the outer two-thirds of the ear, avoiding skin folds located at the base. Measurements were taken in quadruplicate for each mouse (n = 6).¹⁷

PMA Dosage Test and Neutrophil Quantification

Various concentrations of PMA were applied to the left ears of the mice (n = 3 per PMA dose) to observe the effect of dose on the intensity of inflammation. The tested doses included 0.5, 1.5, 5, and 50 μg per 20 μL of vehicle. Twenty-four hours after PMA application, mice were sacrificed, and both ears were excised and fixed with formalin before paraffin embedding. The embedded ear tissues were sliced to 5 μm thick histologic sections that were specifically stained for neutrophils with a monoclonal rat-antimouse neutrophil IgG (MCA771G, Serotec, Raleigh, NC) as well as hematoxylin. Stained tissue slices were analyzed under a light microscope (Zeiss Primo Star, Carl Zeiss MicroImaging, LLC, Thornwood, NY) equipped with a camera (AxioCam ICc1, Carl Zeiss), and pictures were taken by AxioVision LE software (Carl Zeiss). Five random segments were selected on each slice, and three slices were examined for each ear. The neutrophil and total cell populations of PMA-challenged ears (left) were counted by ImageJ software (National Institutes of Health, Bethesda, MD). Neutrophils showed a dark brown color owing to staining with rat-antimouse neutrophil IgG, whereas all other cells exhibited a blue color derived from hematoxylin staining. The total cell count was calculated by summing the number of brown and blue cells. The percentage of neutrophil to total cell number was calculated (see supplementary material available online). The statistical analysis represents the whole ear. In addition, the percentage increase in left ear thickness was also calculated against various PMA doses.

Luminol-Bioluminescence Imaging of Myeloperoxidase

Myeloperoxidase (MPO) is the most abundant protein in azurophilic granules of neutrophils.^18,19 MPO activity has been assayed as an indicator of neutrophil migration and activation. An MPO-specific luminol-bioluminescence study was carried out to confirm the accumulation and activation of neutrophils in PMA-challenged ears.¹⁸ The mice were injected with 200 μL of luminol (5 mg/100 mL Dimethylformamide (DMF)) intraperitoneally 24 hours after PMA (5 μg in 20 μL of vehicle) application. Immediately after luminol injection, a 40-minute kinetic bioluminescence scan was performed with a Xenogen IVIS Spectrum processed on Live Image 3.1 software (Caliper Life Sciences) with an open filter and an exposure time of 60 seconds.

Effective Dose of Imaging Probe 6

Various doses of the probe cFlFlF-PEG-Cy7 6 were tested (1, 2, 5, and 10 nmol per 20 g mice) with PMA-challenged mice. It turned out that 1 nmol was not sufficient enough to provide fluorescence signal and 2, 5, and 10 nmol gave the same ratio of fluorescence signal of inflamed ear to control ear, which meant that 2 nmol was the minimal dose to observe reasonable signal contrast. As a result, 2 nmol was selected for all of the studies reported here.

In Vivo NIRF Imaging

To demonstrate the imaging feasibility and specificity of probe cFlFlF-PEG-Cy7 6, four groups of experiments (n = 5–6 per group) were performed, among which groups A, B, and C had PMA (5 μg per 20 μL of vehicle) applied to the left earlobes: (A) inflamed group: 2 nmol of probe 6 was injected intravenously; (B) blocked group: an excess of nonfluorescence peptide cFlFlF-PEG 4 (100 nmol, 50 equivalents) was intravenously injected 1 hour prior to probe 6 (2 nmol) injection; (C) scramble group: 2 nmol of scrambled peptide probe 7 was intravenously administered; (D) negative control group: mice without any PMA treatment were intravenously injected with probe 6 (2 nmol).

In vivo NIRF imaging was performed on a Xenogen IVIS Spectrum at various time points (30 minutes and 1, 2, 3, 6, and 24 hours postinjection) with a filter set of E_x/E_m = 745/800 nm. Identical illumination settings (autoexposure, medium binning, F/stop = 2) were used for acquiring all images. The fluorescence emission was normalized to photons per second per centimeter squared per steradian (p/s/cm²/sr). Images were acquired and analyzed on Live Image 3.1 and displayed in the same scale of fluorescence intensity. Measurement region of interest (ROI) was obtained from both PMA-challenged ears (left) and control ears (right). Background ROI was recorded as the fluorescence emission from the dark area of the image. Corrected fluorescence intensity was determined by subtracting background ROI from measurement ROI.

Organ Distribution

At the completion of in vivo imaging (24 hours postinjection), PMA-challenged mice that were injected with probe 6 (inflamed group) were euthanized (n = 4). Organs and tissues (left and right earlobes, lungs, heart, liver, kidney, spleen, stomach, small intestine, bone, and muscle) were excised, rinsed with saline, lined up on a black paper, and placed under the charge-coupled device (CCD) camera of the Xenogen IVIS Spectrum. Fluorescence images were taken using the identical illumination settings as in in vivo imaging. The normalized fluorescence intensities of each organ and tissue were measured and expressed as the average radiance (p/s/cm²/sr).

Immunohistochemical Staining and Fluorescence Microscopy

Immunohistochemical analysis was performed on harvested ear tissues 27 hours after PMA application on both inflamed and control earlobes (corresponding to the time point of 3 hours postinjection). After excision, ears were immediately transferred into 1 mL conical tubes and frozen into liquid nitrogen for 1 to 2 minutes and then stored at −80°C in a freezer overnight. The frozen tissues were kept frozen and dissected to adjacent histologic sections (3 μm thick) that were specifically stained for neutrophils with a monoclonal rat-antimouse neutrophil IgG (MCA771G, Serotec, Raleigh, NC).⁷ Stained ear tissue slices were observed under a light microscope (Alpha & Omega Microscope Service, Gaithersburg, MD) with a camera (DP71, Olympus Corporation, Japan), and pictures were taken and processed by MicroSuite Pathology software (Olympus). Furthermore, the fluorescence images were examined for the same batch of slices on a confocal fluorescence microscope (LSM 510-UV, Carl Zeiss, Germany) with a laser excitation filter of 633 nm and a long-pass emission filter of 650 nm, and pictures were taken and processed by LSM Image Browser software (Carl Zeiss).

Fluorescence Stability in Serum

Five nanomoles of probes 6 and 7 (in 0.5 mL phosphate-buffered saline buffer) were incubated separately with equal volume of FBS at 37°C.⁹ The purpose of the experiment was to test whether serum will affect the fluorescence emission property of this probe. Given that mice serum was not available, FBS was used instead. Aliquots of 50 μL were taken out at various time points (up to 24 hours) and transferred into a black, round-bottomed, 96-well plate (nontreated; Corning Incorporated, Corning, NY). The plate was placed under the CCD camera of the IVIS Spectrum, and fluorescence intensity was measured using the identical illumination settings as in in vivo imaging. In addition, a probe stability study was performed by HPLC from 0 to 24 hours postincubation (a detailed protocol is provided in the supplementary material online).

Statistics

Quantitative data are expressed as means ± SD. Means were compared using the Student t-test, and p values < .05 were considered statistically significant.

Results

Synthesis of Peptides

The synthesis of peptides was straightforward on the solid resin, following coupling sequences of c-F-l-F-l-F-K and c-L-F-F-F-L-K. PEG_3.4kDa was conjugated to ω-NH₂ of lysine (K) in aqueous solution. Cy7 conjugation was accomplished with moderate yields of 60 ± 1% (n > 3) (see Figure 1). The HPLC retention time of cFlFlF-PEG 4, cFlFlF-PEG-Cy7 6, cLFFFL-PEG 5, and cLFFFL-PEG-Cy7 7 was 18.0, 16.7, 17.9, and 16.6 minutes, respectively. The unconjugated peptides could be readily separated from Cy7-conjugated products by semipreparative HPLC. The chemical purities of Cy7 conjugates 6 and 7 were > 95% based on analytical HPLC. The molecular weights of peptides and Cy7 conjugates were characterized as follows: (1) cFlFlF-PEG 4 was previously reported⁷; (2) for cLFFFL-PEG 5, the average calculated mass was 4343, and major m/z peaks were observed at 4300 to 4477, with a center peak at 4345; (3) for cFlFlF-PEG-Cy7 6,¹⁴ the average calculated mass was 5007, and major m/z peaks were distributed at 4929 to 5150, with a center peak at 5000; (4) for cLFFFL-PEG-Cy7 7, the average calculated mass was 5007, and major m/z peaks were observed at 4864 to 5040, with a center peak at 5000.

Hydrophilicity of Probes

The log P values of probes 6 and 7 were calculated to be −1.13 ± 0.01 and −1.02 ± 0.10, indicating their similar hydrophilicity and blood solubility, which enabled the direct comparison of in vivo imaging between the two probes. Moreover, the log P value of fluorescence probe 6 was also comparable to that of our previously reported positron emission tomographic (PET) probe (cFLFLF-PEG-⁶⁴Cu, log P, −1.21),⁷ suggesting the major contribution of PEG moiety in adjusting the hydrophilicity of the whole molecule.

Fluorescence Spectra

Cy7 conjugates exhibited their maximum fluorescence emission wavelengths (λ_max) at 773 to 774 nm (Figure 2), whereas λ_max of Cy7 was at 771 nm in this solvent system.¹⁴ The conjugation of peptides and Cy7 did not significantly alter the fluorescence emission spectra of Cy7 dye. The peptide conjugates showed only 2 to 3 nm red shifts in their λ_max. Moreover, although λ_max of both Cy7 and Cy7 conjugates (6 and 7) was around 770 to 780 nm, a slightly different emission wavelength (800 nm) was selected for fluorescence imaging on the Xenogen IVIS Spectrum to take advantage of the existing filter pair of 745/800 nm (E_x/E_m) in the instrument.

Figure 2.

Representative fluorescence emission spectra of cFlFlF-PEG-Cy7 6 and cLFFFL-PEG-Cy7 7 in H₂O/DMSO (97/3, v/v) with the excitation wavelength at 745 nm and emission &#x03BN;_max observed at 773 to 774 nm.

In Vivo Model for Monitoring Inflammation

Ear redness could be visualized immediately after PMA (Figure 3A) application. Ear thickness was measured as a function of time until 96 hours after PMA application (Figure 3C). Inflamed ear thickness increased significantly from 0.23 ± 0.02 to 0.45 ± 0.03 mm within 6 hours and remained relatively at that level until 48 hours and began to drop down slowly. Therefore, the time frame of 24 to 48 hours after PMA application was chosen to inject fluorescence probes for monitoring neutrophil infiltration at inflamed areas and perform longitudinal NIRF imaging.

After 24 hours of PMA treatment, bioluminescence was locally emitted from PMA-treated ears (left ears), suggesting local generation and accumulation of MPO as a measurement of neutrophil accumulation and activation at the inflammation site (Figure 3B).

Figure 3.

In vivo model of ear inflammation. A, Chemical structure of PMA. B, Representative luminol-bioluminescence imaging of mice at 24 hours after PMA application to left ears to locate the inflammation in vivo. C, Inflamed ear (left) thickness of mice (n = 4) at various time points before and after PMA application at the constant dosage of 5 mg per 20 μL vehicle (DMSO:acetone = 1:9). D and E, Immunohistochemical staining of neutrophils (with rat-antimouse neutrophil IgG) and hematoxylin staining of both the left inflamed ear (D) and the right control ear (E) of PMA-challenged mice (×100 original magnification). F, Correlation of increase in ear thickness and accumulation of neutrophils in inflamed ears as plotted against various PMA dosages (0.5, 1.5, 5, and 50 μg per 20 μL vehicle (DMSO:acetone = 1:9).

Evaluation of the intensity of ear inflammation was sought by varying PMA doses applied to left ears. Microscopic images of both left and right ear sections were examined. All of the four doses chosen invoked inflammation, as suggested by the large amount of neutrophil infiltration and accumulation (dark brown cells) in the left PMA-challenged ear (Figure 3D), whereas only a few neutrophils were observed in the right control ear (primarily in blood vessels) (Figure 3E). This observation also confirmed that the inflammatory process was highly tissue selective (in this case, only the left ear) and triggered by PMA challenge in a site-specific manner in the whole body of the mouse. Quantification of PMA-induced inflammation was performed based on two factors: percentage of neutrophil accumulation to total cell amount and percentage of increase in left ear thickness after PMA application (see the supplementary material online). Except for the lowest PMA dose of 0.5 μg per 20 μL vehicle, all other doses did not show much of a difference in both factors (the intensity of inflammation in terms of neutrophil recruitment appears to remain the same); therefore, the dose of 5 μg per 20 μL vehicle was selected for further studies (Figure 3F).

In Vivo NIRF Imaging

Representative NIRF images of all four groups (inflamed, blocking, scrambled, and negative control) of mice were exhibited longitudinally (Figure 4). Fluorescence signal could be clearly visualized in the PMA-challenged ear as early as 30 minutes postinjection (image not shown). The inflamed group injected with probe cFlFlF-PEG-Cy7 6 exhibited the strongest fluorescence signal in the left ears among all groups at all time points. The absence of in vivo fluorescence signal from other organs was attributed to the low injection dose (2 nmol) of the probes and skin scattering.

Figure 4.

Representative in vivo near-infrared imaging of mice at 1, 2, 3, 6, and 24 hours following probe injection. A, Inflamed group: PMA-challenged mice injected with probe cFlFlF-PEG-Cy7 6 (2 nmol). B, Blocked group: PMA-challenged mice preinjected with excess of nonfluorescence peptide cFlFlF-PEG 4 (100 nmol) 1 hour before the injection of probe cFlFlF-PEG-Cy7 6 (2 nmol). C, Scrambled group: PMA-challenged mice injected with scrambled peptide probe cLFFFL-PEG-Cy7 7 (2 nmol). D, Negative control group: mice without PMA challenge injected with probe cFlFlF-PEG-Cy7 6 (2 nmol).

The fluorescence intensities of PMA-challenged ears (left) for groups A, B, and C (inflamed, blocked, scrambled) and those of the non–PMA-challenged group D (negative control) were plotted against various time points following probe injection as depicted in Figure 5. For the inflamed group, fluorescence intensity in the left ear reached the maximal level at 3 hours postinjection and showed a two-phase decay: in the first phase, fluorescence intensity decreased relatively quickly by 6 hours postinjection; in the second phase, a much slower decreasing trend was observed until 24 hours postinjection. The scrambled group demonstrated a similar trend of signal change: maximal signal intensity was also observed at 3 hours postinjection but in an “attenuated” pattern compared to that of the inflamed group. On the other hand, the blocking group did not show a similar trend: it exhibited a consistent drop in intensity from the beginning and never attained any maximal value. It is interesting to note that the fluorescence intensity of inflamed ears in the inflamed group was 2.5- and 3-fold greater at 3 hours postinjection and 3.2- and 4-fold greater at 24 hours postinjection compared to that of the scrambled and blocking groups, respectively. In addition, the fluorescence intensity of negative control mice as expected exhibited a low signal strength with a pattern very similar to either the blocked or the scrambled group, and signals from the left and right ears overlapped each other at all time points. It was also interesting to note that the fluorescence intensity of the right ears of both the scrambled and blocked groups was lower than that of the left ears, and this was presumably due to redistribution of blood because the amount of blood flow in inflamed areas is more than that in noninflamed ones.

Figure 5.

Analysis of fluorescence images. A, Three regions of interest (ROI) are depicted in a representative fluorescence image of the inflamed group. ROI 1 and ROI 2 were drawn around both earlobes, whereas ROI 3 was drawn in a dark area of the image as control. Fluorescence intensities of left and right ears were obtained by subtracting control ROI 3 from ROI 1 and ROI 2, respectively. B, Fluorescence intensity of left and right ears as a function of post-probe injection time. Eight lines represent four experimental groups (inflamed, blocked, scrambled, and negative control) and two lines (left and right ears) for each group. For all groups, average fluorescence intensities (n = 5–6) are plotted. Only for the inflamed group, standard deviation is shown owing to the significant difference (p < .0005) between signal intensities from the left and right ears.

Organ Uptake of Probe cFlFlF-PEG-Cy7 6

Ex vivo evaluation of probe 6 on dissected organs at 24 hours postinjection revealed that the fluorescence signal mainly existed in inflamed ear, liver, and kidney (Figure 6). The strong fluorescence intensity of the PMA-challenged ear indicated that the neutrophil binding of cFlFlF-PEG-Cy7 6 led to a very slow clearance of the probe from the inflammation site. The high uptake in liver suggested that this probe might be subjected to the hepatobiliary clearance pathway. This biodistribution pattern was comparable to our previous study on the PET tracer cFLFLF-PEG-⁶⁴Cu, which also demonstrated the highest uptake in liver and kidney 18 hours postinjection.⁷ It is of note that liver or kidney was not visible by whole-animal imaging.

Figure 6.

Organ biodistribution of probe cFlFlF-PEG-Cy7 6 at 24 hours postinjection. Fluorescence intensity measured from various organs and tissues represented in average radiance (p/s/cm²/sr) in the column. Upper right inset, representative fluorescence image of dissected organs and tissues.

Immunohistochemical Staining and Fluorescence Microscopy

Immunohistochemistry was carried out to assess the relative amount and distribution of neutrophils in tissue slices of PMA-challenged ears versus control ears at 3 hours postinjection (27 hours after PMA application) owing to the maximal fluorescence intensity of PMA-challenged ears at this time point. As can be seen in Figure 7A, PMA-challenged ears showed a large amount of neutrophil accumulation (A1, rat-antimouse neutrophil antibody–stained neutrophils in dark brown; A2, black-and-white version of A1 under a confocal microscope), and the fluorescence signal of probe 6 (A3, red color) correlated well with neutrophils (A4, fused image of A2 and A3). On the contrary, as indicated in Figure 7B, the control ear revealed very few neutrophils (B1) and little fluorescence signal was observed (B3 and B4). This microscopic result further confirmed the in vivo neutrophil specificity of probe 6.

Figure 7.

Neutrophil specificity of probe 6 at the tissue and cellular levels. Immunohistochemical staining of neutrophils (with rat-antimouse neutrophil IgG) and the corresponding microscopic fluorescence image of the left ears (A1-A4) and the right ears (B1-B4) of the inflamed group at 3 hours postinjection. Immunostained neutrophils appear dark brown (A1), whereas a few neutrophils are shown in the control ear (B1). Black-and-white images of the slices (A2, B2) could be fused with fluorescence images (A3, B3) to generate fused ones (A4, B4) that reveal the neutrophil-specific binding of probe 6 (×400 original magnification).

Fluorescence Stability in Serum

Figure 8 demonstrates that the fluorescence signal of probes 6 and 7 remained stable up to 24 hours in FBS, indicating that the fluorescence signal loss over time of in vivo imaging was attributed to the clearance and not to the decomposition of probes. The probe was stable (> 95%) at 24 hours in FBS as defined by HPLC analysis (see the supplementary material online).

Figure 8.

Fluorescence stability probes 6 and 7 in serum. The fluorescence intensity of both probes was plotted against the incubation time (5 minutes, 30 minutes, and 1, 2, 3, 4, 6, and 24 hours) with fetal bovine serum at 37°C (error bars are within the dots).

Discussion

To develop a NIRF imaging probe for in vivo imaging of inflammation and infection in live research animals, we designed and synthesized a series of Cy7-conjugated agents with various lengths of PEG moieties: cFlFlF-PEG_n-Cy7 (n = 0, 12, 76).¹⁴ The suitable water solubility of the PEG_3.4kDa-modified cFlFlF-PEG-Cy7 6 along with other favorable pharmacokinetic parameters renders its further investigation in vivo.

One of the major concerns in the development of an imaging agent for neutrophil to monitor inflammation has been the specificity of the probe, and most of the current research advancements are focused on this aspect.²⁰ In this report, we demonstrated the neutrophil specificity of fluorescence probe 6 by comparing in vivo NIRF imaging of the inflamed group with the blocked and scrambled groups as well as correlating immunohistochemical staining of neutrophils at the cellular level with a fluorescence microscopic study. In addition, evidence is provided by luminol-bioluminescence imaging of mice (24 hours after PMA application) by comparing left inflamed ears with right control ears, indicating the presence of functionally activated neutrophils in the inflamed area.

The neutrophil specificity of probe 6 was clearly demonstrated in the macroscopic in vivo imaging study (see Figure 4). The imaging was highly distinct and clearly shows that the left inflamed ear had greater fluorescence intensity than the right ear. The inflamed ear not only retained fluorescence over time but also showed an improved signal to noise ratio when compared to the control ear, along with mice without inflammation (negative control group). Although tissue distribution of the fluorescence probe after dissecting the organs indicated the presence of a significantly large amount of probe in liver and kidney, this characteristic did not interfere with live imaging. We attribute this loss of fluorescence signal to the scattering of deep tissue. The advantage of this technique is the ability to acquire good sensitivity for the fluorescence signal in inflamed areas compared to background, but the disadvantage is that only a surface organ or tissue can be monitored for inflammation, and additional tools may be needed for the detection of an inflammatory process in deeply located organs.

When probe 6 was injected in live mice with inflammation in the left ear, the maximal fluorescence signal was observed within 3 hours postinjection (climbing phase) and a subsequent decrease in the signal intensity between 3 and 6 hours postinjection (fast clearance phase), followed by a much slower loss of signal from 6 to 24 hours postinjection (slow clearance phase) of the left ear. The right ear had much less fluorescence signal (≈ 3.4-fold less) and showed continuous loss of intensity, indicating the amount of circulating neutrophils in the blood, its distribution volume and accumulation in inflamed areas, and clearance from noninflamed areas, thus improving the signal to noise ratio (resolution) of the probe. The observed pharmacokinetics of probe 6 in mice can be attributed to the combination of three factors: (1) the trafficking and recruitment of activated neutrophils bound to fluorescence probe 6 in inflamed areas from circulation along with free-flowing probe 6 in the blood reaching the inflamed site and targeting already recruited neutrophils; (2) the clearance of free probe 6 in blood from circulation; and (3) on-site leftover probe from dead neutrophils and its clearance. The climbing phase was assumed to be caused by factor 1 because inflammation manifests more blood flow. After maximal trafficking and trapping of probe 6 were reached in the inflammation site, blood clearance of the free fluorescence probe 6 dominated the fast clearance phase between 3 and 6 hours postinjection. The nuclear version of this probe, cFLFLF-PEG-⁶⁴Cu, was reported with blood half-life of 55 ± 8 minutes.⁷ The slow clearance phase of fluorescence signal (after 6 hours postinjection) was indicative of retention of neutrophil-bound probe in inflamed areas, which was presumably attributed to poor neutrophil clearance and probe degradation from dead neutrophils. This observation was further confirmed by the blocking group, in which the fluorescence signal was consistently decreasing because neutrophils were preoccupied with nonfluorescence peptide 4 and the blood clearance of floating fluorescent probe 6 dominated pharmacokinetics. For the scrambled group, similar pharmacokinetics was observed as the inflamed group for accumulation of signal in 3 hours postinjection, but in an “attenuated” pattern owing to the absence of neutrophil specificity of scrambled probe 7. The maximal peak at 3 hours postinjection was assumed to be caused by enhanced permeation (increased blood flow) to the inflammation site (see Figure 5B).

The immunohistochemical and fluorescence microscopic data provided evidence that fluorescence probe 6 was associated with neutrophils at the inflammation site, whereas few neutrophils and little fluorescence were observed in control tissue. It was also confirmed that microscopic study of neutrophil activation and trafficking with fluorescence probe 6 at tissue and cellular levels is feasible. The results from the studies demonstrated here suggest that the high potential for in vivo imaging of inflammation by recruiting activated neutrophils to the inflamed site is possible. However, quantifying the inflammation and monitoring it for therapeutic intervention will be the next highly challenging task.

Conclusion

We successfully performed a longitudinal NIRF imaging study of PMA-induced ear inflammation in live mice using fluorescence probe 6. With this imaging probe, macroscopic and microscopic visualization of neutrophilic activation and infiltration is enabled and will not only deepen our understanding on the inflammatory process but will also facilitate the preclinical screening of antiinflammatory agents.

Footnotes

Acknowledgments

We thank Jingyi Li for his help on image processing.

Financial disclosure of authors: This research was supported by the Nihon MediPhysics Fund (MD-RADL Pan NMP RDC-F001917E).

Financial disclosure of reviewers: None reported.

References

Stenfeldt

Karlsson

Wenneras

. The non-steroidal antiinflammatory drug piroxicam blocks ligand binding to the formyl peptide receptor but not the formyl peptide receptor like 1. Biochem Pharmacol 2007;74:1050–6, doi:10.1016/j.bcp.2007.06.049.

Panizzi

Nahrendorf

Wildgruber

. Oxazine conjugated nanoparticle detects in vivo hypochlorous acid and peroxynitrite generation. J Am Chem Soc 2009;131:15739–44, doi:10.1021/ja903922u.

McAfee

Thakur

. Survey of radioactive agents for in vitro labeling of phagocytic leukocytes. I. soluble agents. J Nucl Med 1976;17:480–7.

Thakur

Lavender

Arnot

. Indium-111-labeled autologous leukocytes in man. J Nucl Med 1977;18:1014–21.

Babich

Tompkins

Graham

. Localization of radiolabeled chemotactic peptide at focal sites of Escherichia coli infection in rabbits: evidence for a receptor-specific mechanism. J Nucl Med 1997;38:1316–22.

Zhang

Kundu

Fairchild

. Synthesis of novel neutrophil-specific imaging agents for positron emission tomography (PET) imaging. Bioorg Med Chem Lett 2007;17:6876–8, doi:10.1016/j.bmcl.2007.10.013.

Locke

Chordia

Zhang

. A novel neutrophil-specific PET imaging agent: cFLFLFK-PEG-64Cu. J Nucl Med 2009;50:790–7, doi:10.2967/jnumed.108.056127.

Wunder

Klohs

Dirnagl

. Non-invasive visualization of CNS inflammation with nuclear and optical imaging. Neuroscience 2009;158:1161–73, doi:10.1016/j.neuroscience.2008.10.005.

Cai

Chen

. Near-infrared fluorescence imaging of tumor integrin a_vb₃ expression with Cy7-labeled RGD multimers. Mol Imaging Biol 2006;8:226–36, doi:10.1007/s11307-006-0041-8.

10.

Frangioni

. In vivo near-infrared fluorescence imaging. Curr Opin Chem Biol 2003;7:626–34, doi:10.1016/j.cbpa.2003.08.007.

11.

Cheng

Xiong

. Near-infrared fluorescent RGD peptides for optical imaging of integrin a_vb₃ expression in living mice. Bioconjug Chem 2005;16:1433–41, doi:10.1021/bc0501698.

12.

Yamaoka

Kawasaki

. Establishment of in vivo fluorescence imaging in mouse models of malignant mesothelioma. Int J Oncol 37, 273–9.

13.

Yang

Shi

Tong

. Near IR heptamethine cyanine dye-mediated cancer imaging. Clin Cancer Res 2010;16:2833–44, doi:10.1158/1078-0432.CCR-10-0059.

14.

Xiao

Zhang

Liu

. Synthesis of the cyanine 7 labeled neutrophil-specific agents for noninvasive near infrared fluorescence imaging. Bioorg Med Chem Lett 2010;20:3515–7, doi:10.1016/j.bmcl.2010.04.136.

15.

Babich

Dong

Graham

. A novel high affinity peptide antagonist for infection imaging. J Nucl Med 1997;38:268P.

16.

Liu

Wang

. In vivo and in vitro antiinflammatory activities of neoandrographolide. Am J Chin Med 2007;35:317–28, doi:10.1142/S0192415X07004849.

17.

Cao

Cai

. PET imaging of acute and chronic inflammation in living mice. Eur J Nucl Med Mol Imaging 2007;34:1832–42, doi:10.1007/s00259-007-0451-0.

18.

Gross

Gammon

Moss

Rauch

. Bioluminescence imaging of myeloperoxidase activity in vivo. Nat Med 2009;15:455–61, doi:10.1038/nm.1886.

19.

Manel Querol Sans

JWC

Weissleder

Bogdanov

Jr . Myeloperoxidase activity imaging using ^{67Ga labeled substrate.} Mol Imaging Biol 2005;7:403–10, doi:10.1007/s11307-005-0020-5.

20.

Palestro

. Radionuclide imaging of infection: in search of the grail. J Nucl Med 2009;50:671–3, doi:10.2967/jnumed.108.058297.