Imaging of activated macrophages in experimental osteoarthritis using folate-targeted animal single-photonemission computed tomographycomputed tomography.код для вставкиСкачать
ARTHRITIS & RHEUMATISM Vol. 63, No. 7, July 2011, pp 1898–1907 DOI 10.1002/art.30363 © 2011, American College of Rheumatology Imaging of Activated Macrophages in Experimental Osteoarthritis Using Folate-Targeted Animal Single-Photon–Emission Computed Tomography/Computed Tomography Tom M. Piscaer,1 Cristina Müller,2 Thomas L. Mindt,3 Erik Lubberts,1 Jan A. N. Verhaar,1 Eric P. Krenning,1 Roger Schibli,2 Marjon De Jong,1 and Harrie Weinans1 Objective. Evaluation of macrophage activation may provide essential information about the etiology and progression rate of osteoarthritis (OA). Activated macrophages abundantly express folate receptor ␤ (FR␤), which can be targeted using radioactive-labeled folic acid. This study was undertaken to investigate whether macrophage activation can be monitored in small animal models of OA using a folate radiotracer and to determine whether macrophage activation differs in different models of OA with different OA progression. Methods. Two rat models of OA were used: the mono-iodoacetate (MIA) model, which is a fastprogressing biochemically induced model, and the anterior cruciate ligament transection (ACLT) model, which induces OA at a slower pace. Images were obtained using high-resolution small animal singlephoton–emissioncomputedtomography/computedtomography. The specificity of the technique was tested by eradicating macrophages using clodronate-laden liposomes and blockade of FR␤ by cold folic acid. Results. The MIA model had high initial macrophage activation, with a peak after 2 weeks which disappeared after 8 weeks. The ACLT model showed less activation but was still active 12 weeks after induction. The technique allowed monitoring of the disease process over time, in which late stages of the disease showed less macrophage activation than early stages, especially in the fast-progressing MIA model of OA. Conclusion. Our findings indicate that macrophage activation in experimental OA can clearly be demonstrated and monitored by the folate radiotracer. The high resolution, high sensitivity, and high specificity of the technique allow clear localization of macrophage activity in a disease model that is not known for abundant macrophage involvement. Macrophages might play a role in the development and progression of osteoarthritis (OA). Once activated, macrophages can produce large quantities of growth factors, enzymes, and proinflammatory cytokines. It has previously been shown that macrophages enhance osteophyte growth in animal models of OA (1,2). Furthermore, it is possible that macrophages play a role in the maintenance and progression of the disease by their contribution to synovial fibrosis, cartilage catabolism, and maintenance of an inflammatory state in the joint (3,4) by production of growth factors (5,6), enzymes (7), and cytokines, respectively (8,9). Currently, there are no validated measures of disease activity for OA. Anatomic imaging techniques, such as radiography and magnetic resonance imaging (MRI), measure structural changes that are a result of the disease process and do not visualize the activity of the disease process itself. A number of studies have demonstrated a possible association between synovitis and OA inflammation and the formation of these structural changes (10). A recently introduced technique allows monitoring of activated macrophages in vivo (11,12). Supported by the Dutch Arthritis Association (LLP 11), the Netherlands Ministry of Economic Affairs (Smart Mix Program), and the Netherlands Ministry of Education, Culture, and Science. 1 Tom M. Piscaer, MD, Erik Lubberts, PhD, Jan A. N. Verhaar, MD, PhD, Eric P. Krenning, MD, PhD, Marjon De Jong, PhD, Harrie Weinans, PhD: Erasmus University Medical Center, Rotterdam, The Netherlands; 2Cristina Müller, PhD, Roger Schibli, PhD: Center for Radiopharmaceutical Sciences ETH-PSI-USZ, and Paul Scherrer Institute, Villigen-PSI, Switzerland; 3Thomas L. Mindt, PhD: University Hospital Basel, Basel, Switzerland. Address correspondence to Tom M. Piscaer, MD, Erasmus University Medical Center, Dr. Molenwaterplein 50, Box 1738, Rotterdam 3000 DR, The Netherlands. E-mail: email@example.com. Submitted for publication September 9, 2009; accepted in revised form March 15, 2011. 1898 IMAGING OF MACROPHAGES IN EXPERIMENTAL OA USING SPECT/CT It is only after activation that macrophages express the functional form of folate receptor ␤ (FR␤) (11,13). The vitamin folic acid binds to FR␤ with high affinity (Kd ⬍ 10⫺9M). The presence of the functional form of FR␤ on macrophages is specific for macrophages in an activated state (expression of CD80, CD86, and Ly-6C/G), which are known to secrete tumor necrosis factor ␣ and reactive oxygen species (14). The function of folate receptors on activated macrophages is still unknown. Most cells use a folate carrier, such as the reduced folate carrier and the proton-coupled folate transporter, for folate uptake. On the other hand, the high-affinity folate receptor, which internalizes folates and folate conjugates via endocytosis, is expressed on only a few cell types, including malignant tissue, with the highest frequency on ovarian carcinoma cells (15). In normal tissue and organs, the renal brush border membrane of the proximal tubule cells, the choroid plexus, and the placenta are the only sites that express the folate receptor (15). By labeling a folic acid analog with a diagnostic radionuclide, the presence of the folate receptor can be traced with very high sensitivity using positron emission tomography (PET) or single-photon–emission computed tomography (SPECT). This new molecular imaging technique has already been applied in animal and human studies for the detection of ovarian carcinoma and the specific detection of activated macrophages in (experimental) rheumatoid arthritis (RA) (16–19). The introduction of high-resolution multipinhole SPECT systems provided submillimeter resolution and up to 50 times increased sensitivity compared to clinical SPECT and PET systems. This allows assessment and quantification of arthritic activity in small animals such as mice and rats (20). When SPECT is used in combination with microfocal computed tomography (micro-CT) as an anatomic reference, the radioactive signal can be located in the body in 3 dimensions. Although it has been shown that activated macrophages play a role in experimental OA, this disease is not considered to be immune mediated (characterized by inflammation and the presence of an abundance of activated macrophages), as is the case in RA. It will therefore be challenging to image a relatively small number of activated macrophages in OA, especially in small animal models of OA, since very high sensitivity and resolution are required to produce high-quality images with proper visualization. The aim of this study was to test the feasibility of imaging activated macrophages in experimental OA in vivo by targeting the functional form of FR␤ using a 1899 radiolabeled folic acid conjugate in combination with animal SPECT/CT. Furthermore, macrophage involvement in experimental OA was studied using 2 different rat models of OA to detect potential differences and exclude model-specific macrophage activation. The 2 models were the mono-iodoacetate (MIA) model, which is a fast-progressing biochemically induced model of OA, and the anterior cruciate ligament transection (ACLT) model, which is a more slowly progressing, surgically induced biomechanical model of OA. MATERIALS AND METHODS Radiosynthesis. The synthesis of diethylenetriaminepentaacetic acid (DTPA)–folate has been described previously (21). Radiolabeling was performed according to the following procedure. A stock solution of DTPA-folate (10⫺3 moles/liter) was mixed with phosphate buffered saline (PBS; pH 6.5) and incubated with 111InCl3 (Mallinckrodt-Tyco), resulting in a specific activity of 16 MBq/g of DTPA-folate. The reaction mixture was incubated for 1 hour at room temperature. Quality control was carried out by high-performance liquid chromatography (Waters HPLC System, with a tunable absorbance detector and a UniSpec MCA ␥-detector [Canberra] and a Symmetry C-18 reverse-phase column [5 m, 25 cm ⫻ 4.6 mm; Waters]). Elution was performed with aqueous 0.05M triethylammonium phosphate buffer (pH 2.25) and methanol with a linear gradient to 80% methanol over 30 minutes and a flow rate of 1.3 ml/minute. After quality control, an excess of 0.1 l/MBq of free DTPA was added for complexation of potentially free 111In. The radiofolate was diluted and administered to the animals in PBS (pH 6.5). Scanning procedure. Twenty hours prior to scanning, rats were injected with 60 MBq of 111In-DTPA-folate (15 MBq in mice), diluted with sterile saline into a volume of 600 l, into the tail vein. During injection, the animals were under isoflurane anesthesia. All animal SPECT/CT scans were performed with a 4-head multiplexing multipinhole dedicated small animal SPECT/CT camera (NanoSPECT/CT; Bioscan). Each detector head was fitted with a tungsten-based collimator of nine 1.5-mm diameter pinholes, with a field of view of 350 mm in width and 28 mm in length. The NanoSPECT/CT is fitted with a fast (200 m) helical micro-CT system for anatomic reference of the radioactivity. Before scanning, the animals were placed in a dedicated, temperature-controlled animal bed, and the hind limbs were stretched and fixed in a custommade Perspex frame to reduce noise from radioactivity accumulation in the kidneys and the intestinal tract in the field of view. Energy peaks were set at 170 keV and 240 keV (⫾10%). All animals were scanned from the midtibia to the pelvis. Acquisition time was 67 minutes per scan for the rats (60 minutes for SPECT and 7 minutes for CT) and 35 minutes for the mice (30 minutes for SPECT and 4 minutes for CT, with 1 minute for the transition between SPECT and CT.) Animal SPECT/CT data reconstruction and analysis. After scanning, SPECT data were reconstructed iteratively with HiSPECT software (Scivis) at an isotropic voxel size of 600 m. SPECT and CT data were automatically coregistered, 1900 PISCAER ET AL Figure 1. Timelines for the different experiments, indicating the time points at which experimental osteoarthritis (OA) or rheumatoid arthritis (RA) was induced (by anterior cruciate ligament transection [ACLT] or mono-iodoacetate [MIA] injection for OA or by antigen-induced arthritis for RA), the time points at which folate-enhanced imaging (single-photon–emission computed tomography/computed tomography [SPECT/CT]) was performed, and the species and number of animals used. FR ⫽ folate receptor. since both modalities used the same axis of rotation. All of the acquired images were analyzed using InVivoScope processing software (Bioscan). The fused images were reoriented to achieve the same orientation for all images. A region of interest (ROI) was determined for quantification of the radioactivity. The ROI was placed using only the micro-CT scan, such that the researcher was not influenced by the radioactive signal. The ROI was a cylindrical shape around the knee area; the epiphysis of the distal femur was used as the proximal limit, and the epiphysis of the proximal tibia was used as the distal limit. The ROI contained a volume of 1 cm3 for the rats and 0.2 cm3 for the mice. Animals and experimental animal models. All procedures were approved by the animal ethics committee of the Erasmus Medical Center (EUR 1456). The rats (used in the MIA, ACLT, naive, and macrophage depletion groups) were all male Wistar rats (obtained from Harlan) that were 14 weeks of age at the beginning of the experiment. The mice (used in the antigen-induced arthritis [AIA] group) were 6-week-old C57BL/6 mice (obtained from Harlan). Because a normal rodent diet is highly enriched with folic acid, the animals were fed a folate-deficient rodent diet starting 7 days prior to the experiments. We chose to test 2 commonly used models of OA, namely, the MIA model and the ACLT model. Figure 1 shows the planning scheme for the experiments. MIA-induced OA in rats. The MIA model is a biochemically induced OA model that initiates direct, fast-progressing cartilage degeneration and is used primarily for studying OA-related pain (22). MIA inhibits GAPDH activity, and thereby diminishes chondrocyte metabolism, resulting in chon- drocyte death and lesion formation (22). Within 6 weeks, the model develops to complete OA, characterized by cartilage cracks, cell death, bone erosion, and partial denudation. One milligram of MIA (Sigma-Aldrich) was dissolved in 50 l of sterile saline. We injected the MIA solution into one rat knee joint (n ⫽ 5 rats); the contralateral knee joint was used as a control condition and was injected with an equal amount of sterile saline. The knees of the rats were shaved and disinfected with iodine. The injections were performed under sterile conditions using a 27-gauge needle, while the rats were under isoflurane anesthesia. These animals were imaged 1, 2, 4, and 8 weeks after intraarticular injection of MIA (Figure 1). Ratios of radioactivity within each scan were determined using the saline-injected knee as a reference (experimental OA side versus control side). The results were analyzed using a paired t-test, and P values less than 0.05 were considered significant. ACLT model of OA in rats. The ACLT model is a surgically induced, biomechanical model of OA, which progresses more slowly than the MIA model. It is characterized by mild superficial fibrillation of the cartilage in the first weeks to more pronounced fibrillation throughout the whole cartilage surface and proteoglycan loss after 10 weeks (23). After shaving the knee joint (n ⫽ 5 rats), the skin was disinfected with iodine and an incision was made through the patellar tendon in the longitudinal direction to obtain access to the knee joint. The patellar tendon was spread using 2 surgical hooks. Once the ACL was visible, it was transected using a von Greaffe knife (Fine Science Tools) with the knee in a semiflexed position of ⬃30°. A positive anterior drawer test ensured complete transection of the ligament. Sham operations IMAGING OF MACROPHAGES IN EXPERIMENTAL OA USING SPECT/CT were performed in the contralateral knee joint and consisted of opening the joint space without transection of the ACL. The procedures were all performed under sterile conditions. This procedure was modified from the protocol described by Stoop et al (23). These animals were imaged at 2, 4, 9, and 12 weeks after transection of the ACL (Figure 1). The time points for scanning in this group differed from those in the MIA group, since experimental OA is known to progress much more slowly in the ACLT model. Ratios of radioactivity within each scan were determined, using the sham-operated knee as a reference (experimental OA side versus control side). The results were analyzed using a paired t-test, and P values less than 0.05 were considered significant. AIA model of RA (positive control) in mice. To ensure that the radioactivity uptake indicated joint macrophage activity, an animal model in which activated macrophages are known to be involved abundantly, namely, the AIA model of RA (24), was used as a positive control condition. Methylated bovine serum albumin (mBSA; 8 mg/ml) (Sigma) was emulsified in an equal volume of Freund’s complete adjuvant (CFA) containing 1 mg/ml heat-killed Mycobacterium tuberculosis (strain H37Ra; Difco). On day ⫺8, mice (n ⫽ 2) were immunized by intradermal injection of 100 l mBSA/CFA emulsion into the base of the tail. Thereafter, on day 0, arthritis was induced by unilaterally injecting mice knee joints with 60 g of mBSA in 0.9% NaCl in 6 l. The contralateral side was injected with an equal amount of sterile saline and was used as a control condition. These animals were imaged 10 days after intraarticular injection of mBSA (Figure 1). The mean amount of radioactivity uptake in the mouse knees with AIA was compared to the uptake in the contralateral saline-injected mouse knees within each scan. Student’s paired t-test was used for analysis. Determination of the systemic effect of experimental OA on the contralateral control knee in naive animals. To test the possible systemic effect of experimental OA of one knee on macrophage activation in the contralateral knee, rats (n ⫽ 4) that did not receive any treatment in either knee (naive) were assessed. These animals were imaged once at the age of 14 weeks. The mean radioactivity uptake in the knees of these rats was compared to the mean radioactivity uptake in the saline-injected knees of the animals in the MIA group 2 weeks after injection and in the sham-operated knees of the ACLT animals 4 weeks after operation (2 weeks later than MIA animals to rule out any effect of sham operation). Additionally, the radioactivity of the knees of the naive animals was compared to that in MIA-injected rat knees 2 weeks after MIA injection. Student’s unpaired t-tests were used for statistical analysis. Macrophage depletion to determine macrophage specificity. To test the specificity of the folate tracer for macrophage targeting, clodronate-laden liposomes were used for macrophage depletion. Clodronate-laden liposomes are selectively ingested by phagocytosing cells. After uptake, the clodronate is released, and subsequently, the cell undergoes apoptosis (25). Seven days after injection of clodronate-laden liposomes, optimal depletion of macrophages is established (26). The animal model used was similar to the MIA model described above. These animals (n ⫽ 4) were injected intraarticularly with 50 l of clodronate-laden liposomes 1 week after injection of MIA. The animals were imaged 2 weeks after 1901 Figure 2. Frontal, sagittal, and coronal images of a rat with MIAinduced OA obtained using microfocal computed tomography (microCT), microfocal single-photon–emission computed tomography (micro-SPECT), and micro-SPECT/CT. Micro-CT shows bone (high attenuation), muscle (intermediate attenuation), fat (low attenuation), and skin/fur (intermediate attenuation), as indicated by the arrows. Boxed and encircled areas are the regions of interest. Micro-SPECT images show radioactivity accumulation. Micro-SPECT/CT images show radioactivity accumulation and its anatomic localization. The contralateral untreated control knee served as control. See Figure 1 for other definitions. Color figure can be viewed in the online issue, which is available at http://onlinelibrary.wiley.com/journal/10.1002/ (ISSN)1529-0131. MIA injection and 1 week after injection of clodronate-laden liposomes (Figure 1). The mean radioactivity uptake levels in the MIA-injected/macrophage-depleted rat knees were compared to the contralateral saline-injected/non–macrophagedepleted control rat knees within each scan. Student’s paired t-test was used for statistical analysis. Folate receptor blocking to determine receptor specificity. In order to determine whether the compound 111In-DTPAfolate was receptor specific, we performed one blocking experiment. Two animals from the MIA group were assessed 3 weeks after intraarticular injection of MIA for folate receptor– specific radioactivity accumulation after blockade (Figure 1). Fifteen minutes prior to intravenous injection of 111In-DTPAfolate, the animals were injected with a 500-fold excess of free folic acid (Sigma-Aldrich) to obtain folate receptor blocking on folate receptors in both the experimental OA knees and the contralateral control knees. Because folate receptor blocking by intravenous injection of folic acid affects the MIA-injected knee as well as the contralateral control knee in the same 1902 PISCAER ET AL Figure 3. Representative small animal SPECT/CT images of macrophage activity at different time points in 3-dimensional (3-D) projections (frontal view) and 2-dimensional (2-D) transections (sagittal view). Left, Images of a representative rat injected with MIA. Right, Images of a representative rat subjected to ACLT. One fixed threshold was used for all images. The contralateral untreated control knee served as control. See Figure 1 for other definitions. animal, the amount of radioactivity in each scanned knee (MIA and saline injected) after blocking was compared to the amount of radioactivity uptake 2 weeks and 4 weeks after induction of OA in the same knee and not to the contralateral knee at the same time point. RESULTS In vivo folate receptor–targeted imaging. The high-resolution SPECT/CT scans allowed accurate localization of the distribution of radioactivity within each hind limb. Activity was localized in the joint areas (especially the affected OA knees) and the fatty tissue: inguinal fat, popliteal fat, and anterior and posterior subdermal fat (intermediate attenuation). Most likely, this activation was located in the lymph nodes, since the draining lymph nodes are embedded in this fatty tissue. (Figure 2 shows an example of a scan obtained 4 weeks after MIA injection.) Radioactivity was also located in the gastrointestinal tract (results not shown) (15). Areas of folate receptor–specific radioactivity accumulation, such as the kidneys, choroid plexus, or salivary glands, were not imaged. All of the animals had a mean activity level between 250 and 850 kBq per voxel in the selected ROI (the knee). Imaging of activated macrophages in the MIA and ACLT models of OA. In the MIA-treated rats, the MIA-injected knees showed a mean increase in radioactivity accumulation 1, 2, and 4 weeks after injection compared to the contralateral control knees (Figure 3, left panels and Figure 4A). At the first scan, the mean ⫾ SD difference in radioactivity in the MIA-injected knees of the rats was 24 ⫾ 12% compared to the contralateral saline-injected knees (P ⬍ 0.005). Two weeks after injection, the mean difference increased to 47 ⫾ 17% (P ⬍ 0.001), and 4 weeks after injection, the difference returned to a mean ⫾ SD of 24 ⫾ 13% (P ⬍ 0.01). Eight weeks after the injection of MIA (i.e., at the last scan), there was no longer any difference noted between the MIA-injected and the contralateral saline-injected sides (0 ⫾ 17%; P ⬎ 0.74) (Figure 4A). The first scan of the ACLT animals was performed 2 weeks after induction of ACLT to overcome the bias of macrophage activation due to wound healing of the ACLT procedure. At all time points measured, IMAGING OF MACROPHAGES IN EXPERIMENTAL OA USING SPECT/CT 1903 Figure 4. Percentage differences in mean radioactivity uptake in OA knees versus contralateral control knees at the indicated time points in rats injected with MIA (A) and rats subjected to ACLT (B). Data are presented as box plots, where the boxes represent the 95% confidence interval, the lines within the boxes represent the median, and the lines outside the boxes represent the minimum and maximum values. ⴱ ⫽ P ⬍ 0.05. See Figure 1 for definitions. the uptake of radioactivity in the affected knees was higher than that in the sham-operated knees (Figure 3, right panels, and Figure 4B). At 2 weeks, the uptake of activity in the affected knees was 37 ⫾ 34% higher than that in the unaffected sides (P ⬍ 0.05). The level of activation remained high until the end of the experiment, with a mean ⫾ SD difference of 31 ⫾ 11% at 4 weeks (P ⬍ 0.05), 34 ⫾ 12% at 9 weeks (P ⬍ 0.005), and Figure 5. Small animal single-photon–emission computed tomography/computed tomography images of the antigen-induced arthritis model of rheumatoid arthritis (RA). Three-dimensional (3-D) projections (frontal view) of the contralateral control paw and methylated bovine serum albumin (mBSA)–injected paw and 2-dimensional (2-D) transections (sagittal and coronal views) of the affected hind limb of a representative mouse are shown. The images show uptake of radioactivity in the knees as well as in the lymph drainage of the knees (arrowheads) and the popliteal lymph node (arrows). Color figure can be viewed in the online issue, which is available at http://onlinelibrary.wiley.com/journal/10.1002/(ISSN)1529-0131. 1904 PISCAER ET AL Figure 6. A, Mean radioactivity accumulation in the knees of untreated (naive) animals, saline-injected knees of animals in the MIA group 2 weeks after injection, sham-operated knees of animals in the ACLT group 4 weeks after transection of the ACL, and MIA-injected knees of animals in the MIA group 2 weeks after injection. P values were determined by Student’s unpaired t-test. B, Percentage difference in mean radioactivity uptake in the macrophage-depleted MIA-injected knees of rats compared to their saline-injected non–macrophage-depleted knees. C, Percentage difference in mean radioactivity uptake in MIA-injected and control rat knees with folate receptor blocking (performed 3 weeks after MIA injection) compared to the same MIA-injected and control rat knees without folate receptor blocking scanned 2 weeks and 4 weeks after MIA injection (1 week before and 1 week after blocking). Data are presented as box plots, where the boxes represent the 95% confidence interval, the lines within the boxes represent the median, and the lines outside the boxes represent the minimum and maximum values. ⴱ ⫽ P ⬍ 0.05. See Figure 1 for definitions. 24 ⫾ 10% at 12 weeks (P ⬍ 0.01) (Figure 4B). Besides joint (synovial) radioactivity uptake, a clear pattern of radioactivity uptake was seen at sites outside the knee region, probably the lymphoid system, in most animals in both the MIA and the ACLT groups. Imaging of mice in the AIA model of RA (positive control) group. Ten days after induction of arthritis, the mice with AIA had a mean ⫾ SD of 330 ⫾ 81% higher radioactivity uptake on the affected side than on the contralateral control side. As expected, this was far higher than the lesions in the rats with OA (ACLT and MIA groups). The mice with AIA also showed a clear activation pattern from the knee joint up to the inguinal regions through the fat, as well as clearly activated spots in the popliteal fat. This might indicate macrophage activation in the whole lymph draining passage of the joint, including the popliteal lymph node as well as the subdermal lymph draining passage to the inguinal lymph nodes (Figure 5). Systemic effects of experimental OA in naive animals. We scanned 4 animals that were not affected by experimental OA to assess the normal radioactivity uptake pattern in the hind limbs. These hind limbs showed a base level of radioactivity uptake in both knee joints, as well as in regions that were probably the draining lymph node areas. The radioactivity pattern resembled the pattern of radioactivity seen in the contralateral control (saline-injected or sham-operated) knees of all animals with experimental OA (Figure 6A). There was no difference detected between the mean radioactivity level of this control condition and the contralateral saline-injected knees of the MIA-injected rats 2 weeks after induction of OA (P ⬎ 0.35) or the contralateral sham-operated knees of the animals with ACLT 4 weeks after transection (P ⬎ 0.47) (Figure 6). As expected, there was a difference detected between the radioactivity uptake in the MIA-injected knees 2 weeks after injection and the radioactivity uptake in the knees of the naive animals (P ⬍ 0.025). Macrophage specificity of DTPA-folate determined by clodronate-induced depletion of macrophages. To test cell-specific targeting of 111In-DTPA-folate, macrophages were depleted from the OA knee joint of rats by injection of clodronate-laden liposomes. The IMAGING OF MACROPHAGES IN EXPERIMENTAL OA USING SPECT/CT liposomes were injected 1 week after MIA injection. The contralateral side was injected with sterile saline and was used as a control condition. The signal in the macrophage-depleted knees of rats 2 weeks after MIA injection decreased and was a mean ⫾ SD of 67 ⫾ 18% less than in the contralateral, saline-injected, non– macrophage-depleted sides (P ⬍ 0.009) (Figure 6B). At this time point after MIA injection, rats showed a 47% higher uptake in MIA-injected knees than in contralateral control knees. This confirms the macrophagespecific targeting of 111In-DTPA-folate. Folate receptor specificity of DTPA-folate determined by folate receptor blocking. An excess amount of folic acid injected intravenously 15 minutes prior to injection of the 111In-DTPA-folate almost completely blocked the radioactivity uptake seen in the knees of the animals with MIA-induced OA, confirming the folate receptor specificity of the compound. The mean decrease in uptake of radioactivity under blocking conditions was 93% compared to the mean radioactivity uptake in the same knees 1 week prior to blocking as well as 1 week after blocking (Figure 6C). DISCUSSION The findings of this study indicate that macrophage activation in experimental OA can be monitored using dedicated animal SPECT/CT with a folate radiotracer for targeting the folate receptor that is expressed on activated, but not resting, macrophages. This technique allows the localization of macrophage activity in a disease model that is not known for abundant macrophage involvement. As expected, we found increased macrophage activity around the knee area in the animals. Extraarticular accumulation of radioactivity was also visible, mostly in the subdermal and inguinal fat regions, but because of the inconsistency of this signal and the small number of animals, we cannot draw conclusions based on these findings. In the MIA model, the increased activation of macrophages was reduced in all animals at the latest time point in the experiment. The MIA model is known to progress quickly, even in the first weeks (27,28). The ACLT model is known to progress more slowly (23). The fast progression of the MIA model in the first weeks matches the relatively high uptake of radioactivity, and the progressive state of OA at 6–8 weeks might be the reason that fewer macromolecular complexes end up in the synovial fluid, diminishing the macrophage response and corresponding uptake of radiofolate tracer. Care 1905 was taken to exclude model-specific influences regarding macrophage activation by sham operation on and injection into the contralateral hind limb. However, we cannot exclude the possibility that naive triggers, such as pain, direct activation by MIA, and wound healing, resulted in macrophage activation. Although these animal models of OA cannot be compared with human OA directly, these results are consistent with previous findings in human OA, in which increased mononuclear cell infiltration and overexpression of mediators of inflammation are much more pronounced in early than in late disease (29,30). The positive control condition (RA model) clearly proved that it is possible to trace inflammation using a folate-based imaging agent, as previously shown (11,31). The presence of activated macrophages in the popliteal lymph node and other efferent lymph nodes has never been imaged before using folate-based imaging. The RA model also showed that the activity in the popliteal and inguinal fat region is most likely an enhancement of the lymphoid system, since the whole draining pathway could clearly be visualized (Figure 5). The relatively small increase in macrophage activation in the OA knees of the animals compared to the contralateral knees and the knees of naive animals suggests that the amount of activation in OA models is only slightly enhanced compared to the normal state, even in the faster progressing MIA models. However, we cannot exclude the possibility that species-dependent differences exist between mice and rats. The cell specificity of folate-based imaging has been thoroughly investigated in multiple previous studies, using flow cytometry, confocal microscopy, and in vivo macrophage elimination using clodronate-laden liposomes as determinants for cell phenotyping (11,14, 32–34). In the present study, an almost total absence of radioactivity in the knee joint after elimination of macrophages using clodronate-laden liposomes was observed. This indicates that macrophages are indeed the target of the radioactive folate. Furthermore, images obtained after receptor blockade with free folic acid showed that highly receptor-specific targeting of the folate receptor by 111 In-DTPA-folate occurred. Several 99m Tcradiofolates have already been developed and studied, and the radiofolate EC20 has been used in clinical trials (35). Other options for imaging macrophages in vivo include the use of ultrasmall paramagnetic iron oxide particles in MRI (36) or CD11b/CD14 and anti-CD14⫹ antibody–labeled tracers (37). However, these tech- 1906 PISCAER ET AL niques are by far not as sensitive (MRI) and specific (CD14⫹) for activated macrophage targeting. OA is a heterogeneous disease. Its etiology and progression involve biomechanics, hormones, and a diverse spectrum of cytokines, enzymes, and growth factors that can all affect joint tissue. Cartilage disruption leads to leakage of large molecular complexes, such as collagens and proteoglycans, into the synovial fluid. Chondrocytes generate a repair response characterized by increased syntheses of inferior types of collagen (e.g., type I and type IX) and proteoglycans. In OA, catabolic processes appear to increase faster in relation to anabolic processes in the cartilage. This results in the presence of various macromolecules in the joint fluid that will end up in the synovial tissue (38). Furthermore, it is clear that multiple signaling molecules produced either by chondrocytes or by neighboring fibroblasts or macrophages can also cause an activity response, as was recently shown by Blom and colleagues (39). Although we investigated the possibility of imaging activated macrophages in experimental OA through FR␤ and demonstrated its presence in experimental OA, this might also create the opportunity to test the effect of specific folate antagonists such as BCG 495 in OA (13). This study clearly demonstrates that in vivo imaging and quantification of macrophage activity by radiolabeled folate conjugate in small animal models of OA is feasible. The technique allows monitoring of the disease process over time, in which late-stage disease showed less macrophage activation than early stages. Although the technique was tested on small animal models of OA, it can easily be translated to humans with OA, thereby possibly providing a new imaging biomarker for the activity and potential progression of the disease. ACKNOWLEDGMENT We thank Dr. N. van Rooijen (Department of Molecular Cell Biology, Vrije Universiteit, VUMC, Amsterdam, The Netherlands) for kindly providing the clodronate-laden liposomes. AUTHOR CONTRIBUTIONS All authors were involved in drafting the article or revising it critically for important intellectual content, and all authors approved the final version to be published. Dr. Piscaer had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. Study conception and design. Piscaer, Müller, Mindt, Verhaar, Krenning, Schibli, De Jong, Weinans. Acquisition of data. Piscaer, Müller, Lubberts, Verhaar, Schibli. Analysis and interpretation of data. Piscaer, Müller, Verhaar, Krenning, Schibli, Weinans. REFERENCES 1. Blom AB, van Lent PL, Holthuysen AE, van der Kraan PM, Roth J, van Rooijen N, et al. Synovial lining macrophages mediate osteophyte formation during experimental osteoarthritis. Osteoarthritis Cartilage 2004;12:627–35. 2. Van Lent PL, Blom AB, van der Kraan P, Holthuysen AE, Vitters E, van Rooijen N, et al. Crucial role of synovial lining macrophages in the promotion of transforming growth factor ␤–mediated osteophyte formation. Arthritis Rheum 2004;50: 103–11. 3. Bondeson J, Wainwright SD, Lauder S, Amos N, Hughes CE. The role of synovial macrophages and macrophage-produced cytokines in driving aggrecanases, matrix metalloproteinases, and other destructive and inflammatory responses in osteoarthritis. Arthritis Res Ther 2006;8:R187. 4. Lauder SN, Carty SM, Carpenter CE, Hill RJ, Talamas F, Bondeson J, et al. Interleukin-1␤ induced activation of nuclear factor-B can be inhibited by novel pharmacological agents in osteoarthritis. Rheumatology (Oxford) 2007;46:752–8. 5. Blaney Davidson EN, Vitters EL, Mooren FM, Oliver N, van den Berg WB, van der Kraan PM. Connective tissue growth factor/ CCN2 overexpression in mouse synovial lining results in transient fibrosis and cartilage damage. Arthritis Rheum 2006;54:1653–61. 6. Bakker AC, van de Loo FA, van Beuningen HM, Sime P, van Lent PL, van der Kraan PM, et al. Overexpression of active TGF-␤1 in the murine knee joint: evidence for synovial-layer-dependent chondro-osteophyte formation. Osteoarthritis Cartilage 2001;9: 128–36. 7. Blom AB, van Lent PL, Libregts S, Holthuysen AE, van der Kraan PM, van Rooijen N, et al. Crucial role of macrophages in matrix metalloproteinase–mediated cartilage destruction during experimental osteoarthritis: involvement of matrix metalloproteinase 3. Arthritis Rheum 2007;56:147–57. 8. Haynes MK, Hume EL, Smith JB. Phenotypic characterization of inflammatory cells from osteoarthritic synovium and synovial fluids. Clin Immunol 2002;105:315–25. 9. Sakkas LI, Johanson NA, Scanzello CR, Platsoucas CD. Interleukin-12 is expressed by infiltrating macrophages and synovial lining cells in rheumatoid arthritis and osteoarthritis. Cell Immunol 1998;188:105–10. 10. Pelletier JP, Martel-Pelletier J, Abramson SB. Osteoarthritis, an inflammatory disease: potential implication for the selection of new therapeutic targets [review]. Arthritis Rheum 2001;44: 1237–47. 11. Turk MJ, Breur GJ, Widmer WR, Paulos CM, Xu LC, Grote LA, et al. Folate-targeted imaging of activated macrophages in rats with adjuvant-induced arthritis. Arthritis Rheum 2002;46:1947–55. 12. Low PS, Henne WA, Doorneweerd DD. Discovery and development of folic-acid-based receptor targeting for imaging and therapy of cancer and inflammatory diseases. Acc Chem Res 2008;41: 120–9. 13. Van der Heijden JW, Oerlemans R, Dijkmans BA, Qi H, van der Laken CJ, Lems WF, et al. Folate receptor ␤ as a potential delivery route for novel folate antagonists to macrophages in the synovial tissue of rheumatoid arthritis patients. Arthritis Rheum 2009;60:12–21. 14. Xia W, Hilgenbrink AR, Matteson EL, Lockwood MB, Cheng JX, Low PS. A functional folate receptor is induced during macrophage activation and can be used to target drugs to activated macrophages. Blood 2009;113:438–46. 15. Muller C, Forrer F, Schibli R, Krenning EP, de Jong M. SPECT study of folate receptor-positive malignant and normal tissues in mice using a novel 99mTc-radiofolate. J Nucl Med 2008;49:310–7. 16. Fisher RE, Siegel BA, Edell SL, Oyesiku NM, Morgenstern DE, Messmann RA, et al. Exploratory study of 99mTc-EC20 imaging IMAGING OF MACROPHAGES IN EXPERIMENTAL OA USING SPECT/CT 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. for identifying patients with folate receptor-positive solid tumors. J Nucl Med 2008;49:899–906. Ke CY, Mathias CJ, Green MA. Folate-receptor-targeted radionuclide imaging agents. Adv Drug Deliv Rev 2004;56:1143–60. Sega EI, Low PS. Tumor detection using folate receptor-targeted imaging agents. Cancer Metastasis Rev 2008;27:655–64. Siegel BA, Dehdashti F, Mutch DG, Podoloff DA, Wendt R, Sutton GP, et al. Evaluation of 111In-DTPA-folate as a receptortargeted diagnostic agent for ovarian cancer: initial clinical results. J Nucl Med 2003;44:700–7. Ostendorf B, Scherer A, Wirrwar A, Hoppin JW, Lackas C, Schramm NU, et al. High-resolution multipinhole singlephoton–emission computed tomography in experimental and human arthritis. Arthritis Rheum 2006;54:1096–104. Mindt TL, Muller C, Stuker F, Salazar JF, Hohn H, Mueggler T, et al. A “click chemistry” approach to the efficient synthesis of multiple imaging probes derived from a single precursor. Bioconjug Chem 2009;20:1940–9. Pomonis JD, Boulet JM, Gottshall SL, Phillips S, Sellers R, Bunton T, et al. Development and pharmacological characterization of a rat model of osteoarthritis pain. Pain 2005;114:339–46. Stoop R, Buma P, van der Kraan PM, Hollander AP, Billinghurst RC, Meijers TH, et al. Type II collagen degradation in articular cartilage fibrillation after anterior cruciate ligament transection in rats. Osteoarthritis Cartilage 2001;9:308–15. Santos L, Hall P, Metz C, Bucala R, Morand EF. Role of macrophage migration inhibitory factor (MIF) in murine antigeninduced arthritis: interaction with glucocorticoids. Clin Exp Immunol 2001;123:309–14. Van Rooijen N, Sanders A. Liposome mediated depletion of macrophages: mechanism of action, preparation of liposomes and applications. J Immunol Methods 1994;174:83–93. Van Lent PL, van den Bersselaar L, van den Hoek AE, van de Ende M, Dijkstra CD, van Rooijen N, et al. Reversible depletion of synovial lining cells after intra-articular treatment with liposome-encapsulated dichloromethylene diphosphonate. Rheumatol Int 1993;13:21–30. Guzman RE, Evans MG, Bove S, Morenko B, Kilgore K. Monoiodoacetate-induced histologic changes in subchondral bone and articular cartilage of rat femorotibial joints: an animal model of osteoarthritis. Toxicol Pathol 2003;31:619–24. Piscaer TM, Waarsing JH, Kops N, Pavljasevic P, Verhaar JA, van Osch GJ, et al. In vivo imaging of cartilage degeneration using microCT-arthrography. Osteoarthritis Cartilage 2008;16:1011–7. 1907 29. Benito MJ, Veale DJ, FitzGerald O, van den Berg WB, Bresnihan B. Synovial tissue inflammation in early and late osteoarthritis. Ann Rheum Dis 2005;64:1263–7. 30. Farahat MN, Yanni G, Poston R, Panayi GS. Cytokine expression in synovial membranes of patients with rheumatoid arthritis and osteoarthritis. Ann Rheum Dis 1993;52:870–5. 31. Paulos CM, Turk MJ, Breur GJ, Low PS. Folate receptormediated targeting of therapeutic and imaging agents to activated macrophages in rheumatoid arthritis. Adv Drug Deliv Rev 2004; 56:1205–17. 32. Leamon CP, Jackman AL. Exploitation of the folate receptor in the management of cancer and inflammatory disease. Vitam Horm 2008;79:203–33. 33. Nagayoshi R, Nagai T, Matsushita K, Sato K, Sunahara N, Matsuda T, et al. Effectiveness of anti–folate receptor ␤ antibody conjugated with truncated Pseudomonas exotoxin in the targeting of rheumatoid arthritis synovial macrophages. Arthritis Rheum 2005;52:2666–75. 34. Nakashima-Matsushita N, Homma T, Yu S, Matsuda T, Sunahara N, Nakamura T, et al. Selective expression of folate receptor ␤ and its possible role in methotrexate transport in synovial macrophages from patients with rheumatoid arthritis. Arthritis Rheum 1999;42: 1609–16. 35. Matteson EL, Lowe VJ, Prendergast FG, Crowson CS, Moder KG, Morgenstern DE, et al. Assessment of disease activity in rheumatoid arthritis using a novel folate targeted radiopharmaceutical Folatescan. Clin Exp Rheumatol 2009;27:253–9. 36. Thorek DL, Chen AK, Czupryna J, Tsourkas A. Superparamagnetic iron oxide nanoparticle probes for molecular imaging. Ann Biomed Eng 2006;34:23–38. 37. Goupille P, Chevalier X, Valat JP, Garaud P, Perin F, Le Pape A. Macrophage targeting with 99mTc-labelled J001 for scintigraphic assessment of experimental osteoarthritis in the rabbit. Br J Rheumatol 1997;36:758–62. 38. Akira S, Takeda K, Kaisho T. Toll-like receptors: critical proteins linking innate and acquired immunity. Nat Immunol 2001;2: 675–80. 39. Blom AB, Brockbank SM, van Lent PL, van Beuningen HM, Geurts J, Takahashi N, et al. Involvement of the Wnt signaling pathway in experimental and human osteoarthritis: prominent role of Wnt-induced signaling protein 1. Arthritis Rheum 2009;60: 501–12.