Cyclodextrin/Adamantane-Mediated Targeting of Inoculated Bacteria in Mice.

Cyclodextrin (CD)-based host-guest interactions with adamantane (Ad) have demonstrated use for functionalizing living cells in vitro. The next step in this supramolecular functionalization approach is to explore the concept to deliver chemical cargo to living cells in vivo, e.g., inoculated bacteria, in order to study their dissemination. We validated this concept in two rodent Staphylococcus aureus models. Bacteria (1 × 108 viable S. aureus) were inoculated by (1) intramuscular injection or (2) intrasplenic injection followed by dissemination throughout the liver. The bacteria were prefunctionalized with 99mTc-UBI29-41-Ad2 (primary vector), which allowed us to both determine the bacterial load and create an in vivo target for the secondary host-vector (24 h post-inoculation). The secondary vector, i.e., chemical cargo delivery system, made use of a 111In-Cy50.5CD9PIBMA39 polymer that was administered intravenously. Bacteria-specific cargo delivery as a result of vector complexation was evaluated by dual-isotope SPECT imaging and biodistribution studies (111In), and by fluorescence (Cy5); these evaluations were performed 4 h post-injection of the secondary vector. Mice inoculated with nonfunctionalized S. aureus and mice without an infection served as controls. Dual-isotope SPECT imaging demonstrated that 111In-Cy50.5CD9PIBMA39 colocalized with 99mTc-UBI29-41-Ad2-labeled bacteria in both muscle and liver. In inoculated muscle, a 2-fold higher uptake level (3.2 ± 1.0%ID/g) was noted compared to inoculation with nonfunctionalized bacteria (1.9 ± 0.4%ID/g), and a 16-fold higher uptake level compared to noninfected muscle (0.2 ± 0.1%ID/g). The hepatic accumulation of the host-vector was nearly 10-fold higher (27.1 ± 11.1%ID/g) compared to the noninfected control (2.7 ± 0.3%ID/g; p < 0.05). Fluorescence imaging of the secondary vector corroborated SPECT-imaging and biodistribution findings. We have demonstrated that supramolecular host-guest complexation can be harnessed to achieve an in vivo cargo delivery strategy, using two different bacterial models in soft tissue and liver. This proof-of-principle study paves a path toward developing innovative drug delivery concepts via cell functionalization techniques.

Introduction

Cyclodextrin (CD)-based host–guest supramolecular interactions with adamantane (Ad) have recently received considerable attention for the potential they hold for advancing precision medicine.[1,2] In the past, CDs’ ability to bind lipophilic drugs has led to CD mostly being used to improve solubility and biodistribution.[3] More recently, CD-based host–guest chemistry has been used to create a variety of nanoparticle and surface functionalizations.[4,5] Unique to such systems is the specificity of well-defined (multivalent) host–guest interaction, which allows a wide range of functionalities, like nanoparticles or polymers, to be introduced on various surfaces.[6−8] This chemical concept has even been used to bind bacteria to functionalized surfaces.[9]

A further step has been taken by applying CD-based host–guest chemistry in vitro on surfaces of macroaggregated albumin microspheres (MAA; diameter 10–90 μm).[10−12] Hepatic transplantation of Ad-functionalized MAA microspheres helped drive local accumulation of a β-CD- and Cy5-functionalized poly(isobutylene-alt-maleic-anhydride, PIBMA) polymer, in essence, targeted chemical cargo delivery. The PIBMA backbone of the secondary vector, first of all, serves as a backbone that connects multiple CD-host molecules, a feature that helps improve the affinity for the Ad-guest functionalized surfaces by exploiting multivalent interactions.[6] This same backbone can also serve as a carrier for various cargo molecules, e.g., fluorescent dyes and radioisotopes for imaging purposes. The same host–guest chemistry has also proven to be effectively introduced in vitro onto the surface of mammalian cancer cells (20–30 μm diameter), macrophages (21 μm diameter), and cardiac stem cells (4–6 μm in diameter).[6,13,14] For all these cell types, host–guest chemistry created a chemical means of influencing the interaction between living cells and their environment—converting living cells into pseudochemical scaffolds suitable for supramolecular chemistry in a biological environment. Based on the above, the next step in further developing the technology would be validating whether these pseudochemical scaffolds could be applied to deliver chemical cargos in vivo. Such in vivo cargo delivery concepts create new opportunities for more efficient delivery of, e.g., vaccines and immunomodulatory drugs, a concept we have already explored in earlier in vitro work in the area of host–guest functionalizations.[14]

As an extension of our earlier in vitro work on host–guest functionalization, we hypothesized that we could translate the same concept into an in vivo cargo delivery system. To demonstrate the proof-of-principle, we made use of S. aureus inoculation models where functionalized bacteria would serve as a target for the delivery of cargo. For this purpose, S. aureus bacteria functionalized with Ad (using the bacteria-specific UBI29–41 vector as previously described[15]) were inoculated into thigh muscle or liver to establish infection (Scheme ). Subsequently, this infection site, characterized by the presence of the Ad-functionalizations (the guest), then drove the local accumulation of an intravenously administered cargo, 111In-Cy50.5CD9PIBMA39 (the host).

Scheme 1

Experimental Setup

(A) S. aureus bacteria were Ad-functionalized using a 99mTc-UBI29-41Ad2 vector (guest) in vitro. (B) Ad-functionalized bacteria were inoculated into either thigh muscle (shown) or liver (not shown), followed by intravenous injection of a 111mIn-Cy50.5CD9PIBMA39 cargo (host). In vivo complexation of this host cargo with Ad-functionalized bacteria could then be assess by dual-isotope SPECT imaging.

Results

In Vitro Host–Guest Interactions

In line with the wealth of literature that describes bacteria-specific labeling of bacterial with UBI29–41,[16,17]in vitro scintillation studies revealed that the guest-vector 99mTc-UBI29–41-Ad2 gave a labeling efficiency with S. aureus of 86.6 ± 7.0% after 1 h of incubation, and showed that the tracer remained stably associated for at least 24 h at 37 °C (Figure S4B). Confocal microscopy confirmed binding of the host-vector (cargo) Cy50.5CD9PIBMA39 to UBI29–41-Ad2-labeled S. aureus in vitro, with the distribution of fluorescent signals throughout the bacteria (Figure B); additionally, the bacterial functionalization was also shown here to remain stable for at least 24 h (Figure A). Localization of the host-vector Cy50.5CD9PIBMA39 was overwhelmingly on the bacterial surface, as evinced by the absence of overlap between Cy5 and Hoechst emissions (Figure A).

Figure 1

In vitro host–guest complexation in S. aureus. (A) Time-dependent analysis of the presence of Cy50.5CD9PIBMA39 complexed onto UBI29–41-Ad2-functionalized S. aureus over 24 h of observation. (B) Consistent complexation of host-vector Cy50.5CD9PIBMA39 onto UBI29–41-Ad2-functionalized bacteria after 1 h of labeling.

In Vivo Host–Guest Interactions

To establish whether host–guest interactions allow for chemical cargo delivery, we subjected animal models inoculated with 99mTc-UBI29–41-Ad2-functionalized S. aureus (the guest) to an i.v. injection of 111In-Cy50.5CD9PIBMA39 (the host) (Scheme ). In the muscle inoculation model, SPECT 99mTc imaging at 24 h post-injection confirmed focal radioactivity of the 99mTc-UBI29–41-Ad2-functionalized S. aureus guest in the leg, though of relatively weak signal intensity after 4 physical half-lives of 99mTc (Figure A; left panel). At this same location, a distinct accumulation of the host 111In-Cy50.5CD9PIBMA39 could be observed (Figure A). Calculations based on %ID/g data (Figure B, Table S1) showed that host accumulation in infected muscle was 16-fold (p < 0.005) higher compared to that in contralateral uninfected muscle (3.2 ± 1.0%ID/g vs 0.2 ± 0.2%ID/g). Accumulation of 111In-Cy50.5CD9PIBMA39 in thigh muscles of mice without an infection yielded similar uptake as the contralateral muscle, namely, 0.4 ± 0.3%ID/g. Accumulation of 111In-Cy50.5CD9PIBMA39 in infections with UBI29–41-Ad2-functionalized S. aureus was 2-fold higher compared to infections with nonfunctionalized S. aureus (3.2 ± 1.0%ID/g vs 1.9 ± 0.4%ID/g; p < 0.05; 2-fold), indicating that the contribution of the functionalization step is essential in host–guest complexation in vivo. Fluorescence imaging confirmed these findings, with a strong Cy5 signal (host-vector) in UBI29–41-Ad2-functionalized S. aureus-infected muscle compared to noninfected contralateral thigh muscle (Figure C).

Figure 2

Cargo accumulation in bacteria-inoculated muscle. (A) In vivo SPECT imaging of 99mTc-UBI29–41-Ad2-labeled S. aureus 44 h after inoculation in the thigh muscle (left panel; purple-to-pink color coding) and imaging of host-vector 111In-Cy50.5CD9PIBMA39 (rainbow color coding). Clear colocalization of 99mTc and 111In was seen for mice colonized with Ad-functionalized S. aureus, whereas colocalization of such signals was less or altogether absent in nonfunctionalized bacteria or uninfected muscle, respectively. Organs are marked as (1) heart/lungs, (2) liver, (3) intestines, and (4) inoculation site (thigh muscle). (B) Biodistribution studies of 111In-Cy50.5CD9PIBMA39 in mice inoculated with 99mTc-UBI29–41-Ad2-functionalized S. aureus (blue bars), nonfunctionalized S. aureus (orange bars), or no infection (gray bars). The data (expressed as the mean ± SD ratios of the %ID/g) showed comparable activity in blood and major tissues for the 3 groups, whereas uptake between infected muscle and controls differed significantly (n = 6 for each group). * p< 0.05. (C) Ex vivo fluorescence imaging of muscle tissue corroborated the these findings, with the infected muscle indicated with a red arrow. The scale bar indicates the intensity of fluorescence expressed as photons/s/cm2.

In the hepatic inoculation model, SPECT imaging (Figure A; left panel) of 99mTc-UBI29–41-Ad2-functionalized S. aureus (guest) identified radioactivity at various sites in the liver. This observation is in line with previous reports using 99mTc-Ad-MAA.[11,15] Unfortunately, the short half-life of 99mTc meant that the signal intensity in the 140 keV detection window after about 4 physical half-lives was weak (see Figure A). Imaging in the 240 keV window of 111In revealed accumulation in the spleen and the liver, which was not observed in the control mice (no functionalization; Figure A), suggesting bacterial dissemination. Biodistribution data indicated that these accumulations were significantly (p < 0.001) higher for mice with infected livers compared to control mice (27.1 ± 11.1%ID/g vs 7.6 ± 2.3%ID/g) and intestines (19.5 ± 8.7%ID/g vs 2.6 ± 0.5%ID/g) (Figure B, Table S1). Liver–muscle ratios revealed a nearly 6-fold increase for infected mice (20.0 ± 7.8%ID/g vs 3.4 ± 0.8%ID; p < 0.001). These data were further corroborated by fluorescence imaging of the liver, showing a higher Cy5 signal in infected (i) livers than in uninfected (ii) ones (Figure C).

Figure 3

Cargo accumulation in the bacteria-inoculated liver. (A) SPECT imaging of 99mTc-UBI29–41-Ad2-functionalized S. aureus showed activity in the liver and spleen (left panel; purple-to-pink color coding). Imaging of 111In-Cy50.5CD9PIBMA39 (rainbow color coding) showed clear colocalization in the liver (middle panel), whereas mice without infection did not show any liver colocalization (right panel). Organs are marked as (1) heart/lungs, (2) inoculation site (spleen), (3) liver, and (4) intestines. (B) Biodistribution studies of 111In-Cy50.5CD9PIBMA39 showed significantly higher uptake in the infected liver and intestines (blue bars) compared to mice with a noninfected liver (orange bars). Data are expressed as the mean ± SD ratios of the %ID/g in the liver and blood measured at 4 h p.i. of the host-vector (n = 6 for each group). * p< 0.05. (C) Ex vivo fluorescence imaging of 111In-Cy50.5CD9PIBMA39 in infected (top panel - i) and noninfected liver (bottom panel - ii). The scale bar indicates the intensity of fluorescence expressed as photons/s/cm2.

To rule out dissociation of the guest-vector 99mTc-UBI29–41-Ad2 from the bacteria, the biodistribution of unbound 99mTc-UBI29–41-Ad2 was separately assessed. Gauged by the 99mTc signal (Table S3), 99mTc-UBI29–41-Ad2 was cleared via the renal pathway. Thus, as we did not observe 99mTc-activity in the bladder (Figures A, 3A), it appeared that throughout the experiment, 99mTc-UBI29–41-Ad2 remained attached to bacteria. This result was in good agreement with an earlier study of ours.[15]

Discussion

In this study, we addressed the feasibility of harnessing cyclodextrin/adamantane (CD/Ad) host–guest chemistry to achieve targeted accumulation of a chemical cargo. To this end, we functionalized living S. aureus bacteria with Ad, inoculated the bacteria into either muscle or liver, and then intravenously administered a chemical cargo to target these Ad-functionalized bacteria. We found that for both muscle and liver models, the presence of Ad-functionalized bacteria (the guest) was able to drive accumulation of the injected chemical cargo (the host). Hence, this study demonstrates a crucial proof-of-principle of the utility offered by supramolecular host–guest chemistry in advancing precision medicine.

While our earlier efforts to translate CD-Ad-based supramolecular host–guest chemistry into in vivo applications made use of chemical microspheres designed for hepatic embolization,[10,11] the work presented here uses actual living cells. Furthermore, we show here that the approach has value outside the liver, in this case, being able to target a focal point in the muscle. While the presence of nonfunctionalized bacteria by themselves led to a slight accumulation of the chemical cargo, the accumulation was significantly improved upon by the presence of the Ad functionalization. These results are in line with previous work done in our group, which has shown that bacterial surfaces appear to possess structures capable of serving as a weaker “guest” (analogously to Ad) for binding host-vectors like the CD-polymers used in this study.[14] Despite this property, bacteria nevertheless represented the most intriguing next step to take in developing this technology, because this technology could allow for long-term tracking of bacterial infections in animal models and provide opportunities for exploring targeted treatment modalities which may be grafted onto the CD polymer.

To expand upon this study’s findings, future studies will need to further tweak the presented chemical cargo delivery system to increase its clinical value. For instance, while the Ad-functionalization of bacteria in this study took place in vitro prior to inoculation into mice to more reliably assess the concept’s feasibility, clinical applications will likely depend on an initial in vivo Ad-functionalization. To this end, Ad-vectors must be designed to functionalize the target of interest specifically. For targeting bacteria, UBI29–41 potentially represents one such vector. In earlier studies, we reported a 10-fold higher binding to bacteria over host cells in vitro, which increased to 200-fold in vivo.[18] Since then, several other studies have demonstrated that this preferential binding to bacteria is maintained when a range of functionalizations is conjugated to UBI29–41.[16] Other targeting vectors that instead target overexpressed receptors on tumor cells can represent another attractive option for specific in vivo Ad-functionalization and would widen the application of the cargo delivery concept beyond infectious diseases.[6] An additional consideration for future clinical applications will be to investigate how compatible the presented chemical cargo delivery system is with various therapeutic agents. Such investigations have recently been initiated in our laboratory, and we hope that the findings presented here will also spur others to investigate supramolecular chemistry as a chemical instrument for advancing precision medicine.

Conclusion

In conclusion, we have here shown that host–guest chemistry is well-suited for in vivo cargo delivery to bacterial inoculations. Further development and refinement of the concept could pave the path toward therapeutic studies and innovative new treatment paradigms.

Experimental Procedures

General

All chemicals were obtained from commercial sources and used without further purification. Solvents were obtained from Actu-All Chemicals (Oss, The Netherlands) in HPLC grade and used without further purification. The reactions were monitored by thin-layer chromatography (TLC) and mass spectrometry using a Bruker microflex LRF MALDI-TOF. HPLC was performed on a Waters (Etten-Leur, The Netherlands) HPLC system using a 1525EF pump and a 2489 UV/vis detector. For preparative HPLC, a Dr Maisch GmbH (Ammerbuch, Germany) Reprosil-Pur 120 C18-AQ 10 μm (250 × 20 mm) column was used (12 mL/min). For analytical HPLC, a Dr Maisch GmbH Reprosil-Pur C18-AQ 5 μm (250 × 4.6 mm) column was used, applying a gradient of 0.1% TFA in H2O/CH3CN 95:5 to 0.1% TFA in H2O/CH3CN 5:95 in 40 min (1 mL/min).

Synthesis

UBI29–41-Ad2

Synthesis of adamantane-functionalized antimicrobial peptide ubiquicidin, UBI29–41-Ad2, was carried out as follows (Figure S1). The UBI29–41 peptide (H-TGRAKRRMQYNRR-NH2) was synthesized by Fmoc solid-phase peptide synthesis (SPPS). Four molar equivalents of Fmoc-Lys(Fmoc)–OH, PyBOP, and 1-hydroxybenzotriazole, and 15 mol equiv of DiPEA were dissolved in anhydrous DMF (3 mL), and this mixture was added to the UBI29–41 peptide on solid polymer support bearing a free amine on the N-terminal. After 2 h of shaking at room temperature, the solid was washed using DMF, followed by DCM. A bromophenol blue test (BPB) was used to confirm the full conversion. After standard Fmoc deprotection using 20% piperidine in DMF and reswelling of the polymer beads using DCM 4 mol equiv of Fmoc-Gly-OH, PyBOP, 1-hydroxybenzotriazole, and 15 equiv of DiPEA were dissolved in anhydrous DMF (3 mL) and added to the polymer beads. The suspension was shaken at room temperature for 18 h, and then washing, Fmoc deprotection, and reswelling were performed. Again, BPB was used to show full conversion. Hereafter, Fmoc protection and subsequent washing steps were performed, followed by coupling of 1-adamantane-carbonyl chloride (4 equiv) using DiPEA (15 equiv) and HOBt (4 equiv). Following confirmation by BPB and washing steps, the final product and protective groups were cleaved using TFA/TIPS/H2O 38:1:1 and purified by preparative HPLC. After pooling the relevant fractions and lyophilization, a white solid was obtained; MALDI-TOF [M + H]+ calculated: 2258.7; found 2285.2. An analytical HPLC chromatogram is shown in Figure S2.

Cy50.5CD9PIBMA39

Synthesis and characterization of Cy5 functionalized β-cyclodextrin-poly(isobutylene-alt-maleic-anhydride) (Cy50.5CD9PIBMA39, ∼18.7 kDa, diameter ∼11.7 nm) was carried out as previously described.[6]

Radiolabeling and Stability Testing

Radiolabeling of Guest Vector UBI29–41-Ad2 with 99mTc

Radiolabeling of UBI29–41-Ad2 with 99mTc (99mTc-UBI29–41-Ad2) and stability of the 99mTc-chelation were carried out as follows: to 5 μL of UBI29–41-Cy5-Ad2 (797 μM/mL H2O), 4 μL of SnCl2·2H2O (0.44 mg/mL saline, Technescan PYP, Mallinckrodt Medical B.V. Petten, The Netherlands), 4 μL of NaOH (0.1 M), and 100 μL of a freshly eluted 99mTc–Na-pertechnetate solution (1000 MBq/mL, Mallinckrodt Medical B.V.) were added, and the mixture was gently stirred in a shaking water bath for 1 h at 37 °C.[15,18] The schematic structure of 99mTc-UBI29–41-Ad2 is depicted in Figure S3. The stability of the radiolabeling was determined at 37 °C at various intervals up to 24 h in 1 mL of PBS or fetal calf serum (20% vol/vol, FCS, Life Technologies Inc. CA). The release of radioactivity was assessed by instant thin-layer chromatography (ITLC) on 1 × 7 cm ITLC-SG paper strips (Agilent Technologies, USA) using PBS as the mobile phase (Figure S4A).

Radiolabeling of Cy50.5CD9PIBMA39 with 111In

Labeling of Cy50.5CD9PIBMA39 and stability testing of host-vector 111In-Cy50.5CD9PIBMA39 was carried out as previously described.[10,11]

Labeling of S. aureus with 99mTc-UBI29–41-Ad2

To allow imaging of the administered bacteria in a whole mouse model, S. aureus (ATCC 25922) was labeled with the primary guest-vector 99mTc-UBI29–41-Ad2 as was previously used for labeling with the hybrid tracer 99mTc-UBI29–41-Cy5 15. In short, 100 μL of 99mTc-UBI29–41-Ad2 (20–100 MBq) was added to 1 mL of S. aureus (1 × 108 CFU/mL in 25 mM Na-NH4-acetate buffer pH 5), and the mixture was gently stirred at room temperature. The efficiency of the rate of bacterial functionalization was measured at different intervals within a 24 h time frame. Before determining the tracer uptake at each interval, 0.1 mL of the bacteria in the labeling suspension was washed two times with phosphate-buffered saline (PBS) and centrifugation steps (4 min × 3500 rpm). The amount of radioactivity associated with bacteria was determined at various intervals. The efficiency of the bacterial functionalization was expressed as the % of the total added amount of radioactivity: (([MBq]added/[MBq]associated with bacteria) × 100).[15]

Mixtures of 0.1 mL containing 37 nmol UBI29–41-Ad2 labeled to 1 × 108 colony forming units (CFU) of S. aureus were incubated with 0.1 mL 111In-Cy50.5CD9PIBMA39 (10 μg/mL, 1 MBq) and 0.8 mL of PBS. The mixtures were incubated for 1 h in a shaking water bath at 37 °C. Cells were washed after spinning twice for 5 min at 1500 × g and resuspending the pellet with PBS. The decay-corrected radioactivity associated with the pellet containing the bacteria and the supernatant was measured in a dose calibrator (VDC101, Veenstra Instruments, Joure, The Netherlands). Following correction for background activity, the host–guest interaction was expressed as the percentage of the total amount of radioactivity binding to the bacterial pellet. As a specificity control, nonspecific binding of 111In-Cy50.5CD9PIBMA39 to the same number of bacteria without UBI29–41-Ad2 labeling was evaluated.

Confocal Microscopy

Confocal microscopy was used to confirm the in vitro accumulation of Cy50.5CD9PIBMA39 on UBI29–41-Ad2 labeled bacteria. 0.2 mL of 25 mM Na-NH4-acetate buffer pH 5 solution containing 1 × 108 CFUs S. aureus, 10 μM UBI29–41-Ad2, and 10 μM Hoechst 33342 was incubated for 30 min in a dark shaking water bath at 37 °C. Samples were washed twice in PBS as described above, after which they were incubated in 0.2 mL of a 1 μM solution of Cy50.5CD9PIBMA39 for 60 min in a dark shaking water bath at 37 °C. Samples were washed thrice in PBS as described above, and finally resuspended in 100 μL PBS. Ten microliters of labeled bacteria were pipetted onto culture dishes with glass inserts (ø35 mm glass-bottom dishes No. 15, poly(d-lysine) coated, γ-irradiated, MatTek Corporation). Images were acquired at intervals of 1, 4, and 24 h in a single field of view using a Leica SP8 WLL confocal microscope (λex 633 nm, λem 650–700 nm) under 100× magnification using Leica Application Suite Software Suite 4.8. Pictures were loaded in ImageJ software (ImageJ 1.44p, NIH, USA) to draw a profile line over a single representative stained bacterium in the Cy5 and the Hoechst spectrum to estimate the distribution of both dyes in the bacterial membrane and cytoplasm.

In Vivo Imaging Experiments

Animals

In vivo studies were performed using 2–4-month-old Swiss mice (20–35 g, Crl:OF1 strain, Charles River Laboratories, USA). All animal studies were approved by the Institutional Animal Ethics Committee (DEC permit 12160) of the Leiden University Medical Center (LUMC). Mice were kept under specific pathogen-free conditions in the animal housing facility of the LUMC. Food and water were provided ad libitum. Two models were used: a thigh muscle and a hepatic embolization model.

For Muscle Inoculation

Mice were injected in the right thigh muscle with 0.1 mL of 99mTc-UBI29–41-Ad2-labeled S. aureus (n = 6). After 24 h, 0.1 mL secondary vector (host), 111In-Cy50.5CD9PIBMA39 (10 μg/mL, approximately 15 MBq), was i.v. administered. As controls for the muscle model, 6 mice were either infected with nonfunctionalized S. aureus to demonstrate nonspecific uptake of the secondary vector or not infected at all, to study the general biodistribution of 111In-Cy50.5CD9PIBMA39.

For Liver Inoculation

A liver embolization setup was performed according to previously described procedures.[10,11] In brief, animals were anaesthetized by intraperitoneal injection of a mixture containing Hypnorm (Vetapharma, Leeds, United Kingdom), dormicum (Roche, Basel, Switzerland), and water (1:1:2). After shaving and cleaning with ethanol (70%), the abdominal cavity was incised for 0.5 cm, and the spleen was exposed outside the peritoneum. After that, 1 × 108 CFU 99mTc-UBI29–41-Ad2-labeled S. aureus in a volume of 0.1 mL (5 MBq) was slowly injected into the spleen of mice (n = 6) using a Myjector U-100 insulin syringe (29G × 1/2 in. 0.33 × 12 mm, Terumo Europe, Leuven, Belgium). After 5 s, the needle was removed, and the spleen was repositioned in the peritoneal cavity. 2–4 stitches sutured the incision, and the animals were placed under a heating lamp to maintain until recovery. After 24 h, 0.1 mL 111In-Cy50.5CD9PIBMA39 (10 μg/mL, approximately 15 MBq) was administered intravenously (i.v.). Here, mice with noninfected livers were included as a control study (n = 6) as we studied the nonspecific uptake to bacteria in vivo and the general biodistribution of 111In-Cy50.5CD9PIBMA39 already in the muscle inoculation model.

Imaging Studies and Biodistribution Assays

SPECT imaging of mice either intramuscularly or hepatically inoculated by 99mTc-UBI29–41-Ad2-labeled S. aureus was performed[15,19] using a U-SPECT-2 scanner and continuous 1–2% isoflurane anesthesia, with imaging of 111In-Cy50.5CD9PIBMA39 in mice with non-UBI29–41 functionalized S. aureus and without bacterial infection as a control. Image acquisition took 20 min, and images were reconstructed according to the optimized settings for 99mTc (140 keV) and 111In (170 and 240 keV). Mice were imaged 4 h after the administration of 111In-Cy50.5CD9PIBMA39 using SPECT. After imaging, mice were euthanized by an intraperitoneal injection of 0.25 mL Euthasol (ASTfarma, Oudewater, The Netherlands). Tissues were excised for use in quantitative biodistribution studies. Various organs and tissues were weighed and counted for their radioactive content using a gamma counter (2470 automatic gamma counter, Perkin–Elmer). Counts per minute were converted into decay-corrected MBq at the time point of injection (t = 0), and the percentage of the injected dose per gram of tissue (%ID/g) was calculated as follows: ((([MBq]tissue/[MBq]injected) × 100)/g tissue). From biodistribution data, infection-to-background ratios and the % of excreted radioactivity were determined. To calculate the amount of excreted radioactivity, the entire mouse was weighed and counted for radioactivity in a dose-calibrator. At the time of dissection, and after the removal of various tissues, the radioactivity in the mouse was recalculated. Excreted radioactivity (in urine and feces) was then calculated as (radioactivity of injected dose) – (radioactivity of excised tissues + radioactivity carcass). To determine the biodistribution of free 99mTc-UBI29–41-Ad2 (not associated with S. aureus), three noninfected mice were injected and imaged as described above.

Ex vivo fluorescence imaging of excised infected and noninfected tissues of both models was carried out using the IVIS Spectrum imaging system (Caliper Life Science, Hopkinton, MA).[19] This helped verify the localization of the hybrid polymer via optical imaging.

Statistical Analysis

All data are presented as mean value (±SD) of 3–6 independent measurements. Student’s two-tailed independent sample t test was used to performed statistical analysis for differences between groups in the animal studies. Significance was assigned for p-values <0.05. All analyses and calculations were performed using Microsoft Office Excel 2010 and GraphPad Prism ver. 5.01 for Windows (GraphPad Software, San Diego, CA, USA).