To evaluate whether microbubble-enhanced ultrasound (US) treatment promotes the delivery of methotrexate (MTX) into synovial cells and the enhanced antiinflammatory effects of intraarticular MTX therapy in a rabbit arthritis model.
To evaluate whether microbubble-enhanced ultrasound (US) treatment promotes the delivery of methotrexate (MTX) into synovial cells and the enhanced antiinflammatory effects of intraarticular MTX therapy in a rabbit arthritis model.
Arthritis was induced in both knees of 53 rabbits by immunization with ovalbumin. MTX including a microbubble agent was then injected into the left and right knee joints, and the right knees were exposed to US (MTX+/US+ group), while the left knees were not (MTX+/US− group). The knee joints were evaluated histologically in 7 rabbits at 5 time points up to day 56. Quantitative gene expression of interleukin-1β (IL-1β) in synovial tissue was measured on days 7 and 28. Eight rabbits were used for the measurement of MTX concentration in synovial tissue 12 hours after treatment. To evaluate the effect of microbubble-enhanced US treatment in the absence of MTX, only the microbubble agent was injected into the left and right knee joints of 10 rabbits with or without US exposure, and these animals were evaluated histologically on days 7 and 28.
The MTX concentration in synovial tissue was significantly higher in the MTX+/US+ group than in the MTX+/US− group. Synovial inflammation was less prominent in the MTX+/US+ group compared with the MTX+/US− group, judging from the results of the histologic evaluation and the gene expression levels of IL-1β in synovial tissue. It also appeared that microbubble-enhanced US exposure itself did not affect inflammation.
Microbubble-enhanced US exposure promoted the uptake of MTX into synovial cells, which resulted in enhancement of the antiinflammatory effects of the intraarticular MTX injection. These results suggest that application of this technique may have clinical benefit.
Rheumatoid arthritis (RA) is a systemic inflammatory disorder characterized by pain, swelling, and destruction of the affected joints. The exact mechanism of RA pathogenesis is not well understood. Recently, remarkable progress in the area of anticytokine therapy has provided an alternative and successful approach for therapeutic intervention in RA. However, methotrexate (MTX) still plays a central role in the treatment of RA, although administration of this agent sometimes causes serious side effects, such as interstitial pneumonia, renal failure, and myelosuppression. Intraarticular injection of MTX is thought to be safe compared with the systemic administration of this agent, although the clinical effectiveness in controlling synovitis in RA patients is controversial. Most studies have documented insufficient antiinflammatory effects (1–3). Mechanisms of resistance to MTX are considered to consist of 3 parts: decreased transport, impaired polyglutamylation, and increased dihydrofolate reductase enzyme activity (4–6). It has been reported that there is a significant correlation between reduced levels of folate carrier protein (one of the MTX transporters) at diagnosis and the histologic responses to preoperative MTX chemotherapy for osteosarcoma (6). From these observations, it is possible to conclude that the efficacy of MTX is limited by its transport into cells. Thus, we hypothesized that poor delivery of MTX into RA synovial cells would lead to poor clinical efficacy of intraarticular MTX injection therapy.
To facilitate uptake of MTX into synovial cells, we chose an ultrasound (US) treatment technique (sonoporation) with enhancement by the use of an echo-contrast microbubble agent. Previous reports have indicated that US exposure increases transfection efficiency of gene constructs, due to increased cell membrane porosity and acoustic cavitation (7, 8), which is enhanced with the use of microbubble agents. This is one of the best techniques for in vivo work and clinical applications because it is simple and noninvasive. Additionally, there are no viral components, although the success rate for induction is lower than that found with viral technologies in vitro and in vivo, as previously reported (9–12). Our findings in the present study indicate that US irradiation treatment with a microbubble agent enhances the antiinflammatory effect of intraarticular MTX injection in an ovalbumin (OVA)–induced arthritis model in rabbits.
As an initial step, we determined the optimum concentration of echo-contrast microbubbles (Optison; Mallinckrodt, St. Louis, MO) for MTX (Iatron, Tokyo, Japan) induction into synovial cells. In vitro induction was performed according to a previously described procedure (10). Synovial cells were obtained during total knee arthroplasty from 2 RA patients, who had provided informed consent. Briefly, the tissue was minced and incubated with 0.25% collagenase (Roche, Indianapolis, IN) in phosphate buffered saline (PBS) for 2 hours at 37°C under continuous agitation. The cells were collected by centrifugation, resuspended, and cultured in Dulbecco's modified Eagle's medium (Invitrogen, Carlsbad, CA) containing 10% fetal calf serum and antibiotics (100 units/ml penicillin, 0.1 mg/ml streptomycin, 0.25 μg/ml amphotericin B; Invitrogen). Cultured cells were trypsinized, washed twice in PBS, and resuspended at 1 × 105/ml of PBS per well in a 48-well plate. Optison was added to the cell medium at concentrations of 0%, 5%, 10%, and 20%. Then, 10 μg of Texas Red–conjugated MTX (Molecular Probes, Leiden, The Netherlands) was added to the cell supernatant in each well.
The US probe and well plate were firmly fixed to a stand to avoid dislocation during US exposure. Immediately after fluorescent MTX and microbubbles were added to the well, US exposure was performed. The sonoporator (Sonitron 2000; Mallinckrodt) settings were as follows: frequency 1 MHz, duration 30 seconds, power 1.0 W/cm2, duty cycle 10%, and probe diameter 0.5 cm. The US probe was inserted directly into the cell suspension. A miniature stirrer was placed within the well and spun at 300 revolutions per minute to prevent cell adhesion to the plate. The cells were placed a row apart from each other to prevent interaction due to the transmission of US between the wells. After US exposure, cell viability was tested by counting the cells stained with trypan blue. The cell suspensions were harvested from the wells and attached to slides using a Cyto-Tek centrifuge (Sakura, Tokyo, Japan) at 1,500 rpm for 5 minutes. Texas Red–positive cells were detected by fluorescence microscopy. The average induction efficiency was calculated as the ratio of incorporated cells to all cells in 5 fields.
To confirm that MTX induction into synovial in vivo cells would be promoted by microbubble-enhanced US treatment, we designed in vivo experiments. Three NZW rabbits (Japan SLC, Hamamatsu, Japan) weighing 2.5, 2.7, and 2.8 kg were anesthetized by intramuscular injection of a mixture of ketamine (100 mg/ml [0.6 ml/kg body weight]; Sankyo, Tokyo, Japan) and xylazine (20 mg/ml [0.3 ml/kg body weight]; Bayer, Leverkusen, Germany). We injected 50 μg of Texas Red–conjugated MTX with 5% Optison in 2.5 ml saline, making sure to diffuse it into the left and right knee joints. Soon after injection, US exposure was applied in the medial, central, and lateral areas of the suprapatellar pouch of the right knees for 2 minutes per application. US treatment was not performed in the left knees. US was administered with a sonoporator using the following settings: frequency 1 MHz, duration 2 minutes, power 2.0 W/cm2, duty cycle 50%, and probe diameter 3 cm. Then, the rabbits were killed by excessive intravenous injection of anesthetic agents. Synovial tissue, with the associated muscle and tendon of the suprapatellar pouch of the joints, was obtained and dissected sagittally at the center. It was immediately chilled in liquid nitrogen. Sections of 7-μm thickness were cut in a cryostat, air-dried on slides, fixed in 4% paraformaldehyde, and stained with hematoxylin for counterstaining. The sections were examined by fluorescence microscopy.
Sixty-two NZW rabbits, each weighing ∼2.7 kg (2.5–3.1 kg) were anesthetized as described above. They then received intradermal injections of 4 mg OVA (Sigma, St. Louis, MO) in 0.5 ml Freund's complete adjuvant (Difco, Detroit, MI) and 0.5 ml PBS 3 times at 7-day intervals, as previously reported (13). Five days after the third injection, 1.5 mg OVA in 0.5 ml sterile saline was injected into the left and right knee joints of the rabbits. After confirming the establishment of arthritis 10 days after the injection, the experimental procedures were started. Fifty-three of the 62 rabbits (85%) showed signs of arthritis in both knees and were used for the experiments. In preliminary experiments, we confirmed that the inflammation was virtually identical histologically in both knees (data not shown).
We injected 0.1% MTX with 5% Optison (from the initial in vitro experiment described above) in 2.5 ml saline, ensuring diffusion into both knee joints of 43 rabbits that were anesthetized as described above. US was administered to the right knee in the same manner (MTX+/US+ group). US treatment was not performed on the left knees (MTX+/US− group). Seven rabbits each were killed on days 3, 7, 14, 28, and 56 for histologic examination. The remaining 8 rabbits were used for measurement of MTX in synovial tissue.
To determine whether microbubble-enhanced US itself affected inflammation of synovial tissue, we injected 5% Optison without MTX in 2.5 ml saline into the left and right knee joints of the other 10 rabbits. Then, US was administered to the right knees (MTX−/US+ group) in the same manner as described above, while the left knees were not exposed (MTX−/US− group). Five rabbits each were killed on days 7 and 28 for histologic evaluation.
Synovial tissue from the suprapatellar pouch of the joints was obtained at each time point. It was dissected sagittally at the center, washed, and fixed in 10% formaldehyde in PBS. It was then rinsed with deionized water and dehydrated in a graded ethanol series. Dehydrated tissue was embedded in paraffin, cut into 5-μm sections, mounted on glass slides precoated with poly-L-lysine, dried overnight at 50°C, and stained with hematoxylin and eosin (H&E).
Distal femurs were cut sagittally in the center of the patellar groove and fixed in 10% formaldehyde, decalcified in 10% EDTA, embedded in paraffin, cut into 5-μm sections, and stained with H&E and toluidine blue.
To evaluate the degree of synovial inflammation, we used modified scoring criteria as previously described by Sanchez-Pernaute and colleagues (14). These criteria (Table 1) consist of 5 categories: inflammatory cell infiltration, synovial lining layers, villus formation, vascularity, and cartilage damage. Synovial tissue was scored depending on the degree of inflammation (from 0 for normal tissue to 18 for most severe inflammation). Sections were examined blindly and scored independently by 3 of the authors (KT, TO, YN), without knowledge of the group being examined.
|Inflammation||0 = normal|
|1 = minimal inflammatory infiltration|
|2 = mild inflammatory infiltration|
|3 = moderate inflammatory infiltration|
|4 = marked infiltration with marked edema|
|5 = severe infiltration with edema|
|Synovial lining layers||0 = normal (1–2 cell layers)|
|1 = slightly hyperplasia (2–3 cell layers)|
|2 = moderate (3–5 cell layers)|
|3 = pronounced (5 or more cell layers)|
|Villus formation||0 = none|
|1 = minimal (1–2 villi)|
|2 = several (3 or more villi)|
|Vascularity||0 = normal (limited number of blood vessels)|
|1 = slightly hypervascular (focal occurrence of a small number of blood vessels)|
|2 = moderate (focal occurrence of a large number of blood vessels)|
|3 = pronounced (broadly distributed and large number of blood vessels)|
|Cartilage damage||0 = normal|
|1 = minimal (loss of toluidine blue staining only)|
|2 = mild (loss of toluidine blue staining and mild cartilage thinning)|
|3 = moderate (moderate diffuse or multifocal cartilage loss)|
|4 = marked (marked diffuse or multifocal cartilage loss)|
|5 = severe (diffuse or multifocal cartilage loss)|
To identify blood vessels accurately, anti–α-smooth muscle actin (α-SMA) immunostaining was performed using the mouse monoclonal anti–α-SMA antibody (1A4; Dako, Carpinteria, CA) and the Envision Plus HRP system (K4006; Dako). Formalin-fixed paraffinized sections of rabbit synovial tissue were baked, dewaxed, and rehydrated prior to a peroxidase block (0.1% [volume/volume] H2O2). The primary antibody and the horseradish peroxide–labeled polymer were used as per the Dako Envision kit, followed by staining with 3,3′-diaminobenzidine and counterstaining with hematoxylin before mounting. A negative control was prepared by omitting the primary antibody. A positive control was prepared on the vessels of the same section.
MTX and Optison in 2.5 ml saline were injected into the left and right knee joints of 8 rabbits. US was administered to the right knees in the same manner, while the left knees were not exposed to US. The concentration of MTX in the synovial tissue of 8 rabbits was measured by the enzyme immunoassay method with an MTX assay kit (Iatron) 12 hours after US irradiation. Briefly, the synovial tissue excised from the surface of inflamed synovium was homogenized in Tris buffer (pH 7.4; 3 ml/gm tissue), boiled for 5 minutes, and then centrifuged at 30,000 rpm for 30 minutes in a 4°C atmosphere. This was followed by the addition of 0.1 ml of the supernatant in 1 ml of reagent composed of dihydrofolate reductase (enzyme), NADPH (coenzyme), and 0.1 ml of dihydrofolate (substrate). The residual activity of the dihydrofolate reductase was assayed by absorptiometry at a wavelength of 340 nm.
Synovial tissue was obtained and messenger RNA (mRNA) expression was assessed quantitatively in 7- and 28-day samples of the MTX+/US+ and MTX+/US− groups. PCR primers and fluorogenic probes of interleukin-1β (IL-1β) (forward 5′-TTGCTGAGCCAGCCTCTCTT-3′, reverse 5′-GCTGGGTACCAAGGTTCTTTGA-3′, TaqMan 5′-TGCCATTCAGGCAAGGCCAGC-3′) were designed according to the published sequences (GenBank accession no. M_26295) using Primer Express software (Perkin-Elmer Applied Biosystems, Foster City, CA). They were obtained purified by high-performance liquid chromatography from Applied Biosystems. The fluorogenic probes contained a reporter dye (FAM) covalently linked at the 5′ end and a quencher dye (TAMRA) covalently attached at the 3′ end. Extension from the 3′ end was blocked by the attachment of a 3′-phosphate group.
As external controls for the target gene, plasmid recombinants containing the specific target sequence were generated, as well as 18S ribosomal RNA (rRNA; Perkin-Elmer Applied Biosystems). For this purpose, total RNA from individuals positive for the allele of interest was extracted and reverse transcribed as described above. Following reverse transcription and allele-specific PCR, amplicons were cloned using pCR 2.1 TOPO (Invitrogen). Recombinant plasmids were expressed in competent Escherichia coli (INVαF′; Invitrogen). Plasmid DNA was isolated using silica cartridges (QIAprep Spin Miniprep Kit; Qiagen, Hilden, Germany). Sequences of the cloned amplicons were verified using an automated sequencer (ABI PRISM 7700; Perkin-Elmer Applied Biosystems) with universal M13 primers. Concentrations of the recombinant plasmids were determined by optical density spectrometry. Serial dilutions from the resulting clones were used for standardization, as described in detail in the manufacturer's bulletin.
PCR was performed using 300 nM forward and reverse primers and 200 nM TaqMan probe (final concentration). Each PCR amplification was performed in triplicate wells using the following temperature and cycling profile: 50°C for 2 minutes and 95°C for 10 minutes, followed by 40 cycles of 95°C for 15 seconds and 58°C for 1 minute (15).
The relative expression of IL-1β alleles was determined with reference to the total amount of IL-1β mRNA after normalization against 18S rRNA as implemented in the ABI PRISM 7700 Sequence Detection System software. Results were considered only if the analysis of IL-1β showed all reactions to have the same amount of amplification as 18S rRNA. This procedure allowed for comparison of group-specific IL-1β expression as well as the total expression levels of 18S rRNA.
Results are expressed as the mean ± SD. The significance of the difference in histologic scores between the MTX+/US+ and MTX+/US− groups at the various posttreatment times (days 3, 7, 14, 28, and 56) was tested using two-way factorial analysis of variance (ANOVA), followed by Student's paired t-test for comparison.
The difference in histologic scores among the MTX+/US+ group and the other 3 control groups (MTX+/US−, MTX−/US+, and MTX−/US−) at 7 and 28 days was compared by Student's paired t-test. Student's paired t-test was also used to analyze the results of MTX concentration and real-time reverse transcriptase–PCR. P values less than 0.05 were considered significant.
In vitro MTX induction into synovial cells. To confirm the promotion of MTX induction into RA synovial cells by US irradiation and to determine the optimum concentration of Optison, fluorescence-conjugated MTX was administered into RA synovial cells by sonoporation in vitro. In the absence of US, MTX was administered into a few synovial cells. With US exposure, the administration rate was increased to almost 5% for MTX alone, and ∼31% with 5% Optison (Figure 1). Further increases in the concentration of this reagent elevated the induction rate to a maximum of 48%. For the studies described here, 5% Optison was used in each of the experiments.
In vivo MTX induction into synovial cells. To confirm the promotion of MTX induction in vivo into RA synovial cells by US treatment, fluorescence-conjugated MTX was administered into RA synovial cells by sonoporation in vivo. In the presence of US, fluorescence-positive cells were observed in several synovial lining layers, while only a weakly positive area was observed in the absence of US (Figure 1). The Texas Red–positive area was restricted to synovial cells, and was not observed in other tissue such as tendon or muscle in either group.
MTX concentration. The MTX concentrations in synovial tissue were measured in the MTX+/US+ and MTX+/US− groups (8 samples each) 12 hours after the injection of MTX and Optison. The mean ± SD weight of the tissue from right knees (the MTX+/US+ group) was 454 ± 37.0 mg, while that from left knees (the MTX+/US− group) was 467 ± 44.4 mg. There was no significant difference between these groups. The mean ± SD MTX concentration in the MTX+/US+ group was estimated at 7.575 ± 1.590 × 10−7 mg/dl, while that in the MTX+/US− group was 2.875 ± 0.7889 × 10−7 mg/dl (P < 0.01).
Histologic findings. In the MTX+/US+ group, moderate inflammation (inflammatory cell infiltration, thickened synovial lining layers, and hypervascularity) was observed after 3 days, but it became less severe over a period from 7 to 56 days. Evidence of inflammation decreased in the synovial lining layer and the vascular network. In the MTX+/US− group, moderate inflammation similar to that seen in the MTX+/US+ group was observed after 3 days, continued until 28 days, and became less severe after 56 days.
After 7 days (Figure 2), in the MTX+/US+ group, minimal inflammatory cell infiltration and slight hypervascularity (minimal inflammation) were observed. In the MTX+/US− group, marked inflammatory cell infiltration, proliferating synovial lining layers (as many as 5), and proliferating blood vessels (moderate inflammation) were observed.
After 28 days (Figure 3), in the MTX+/US+ group, moderately proliferating synovial lining layers and focal proliferation of a small number of blood vessels (minimal inflammation) were observed. In the MTX+/US− group, marked inflammatory cell infiltration, proliferating synovial lining layers (>5 cell layers), and general proliferation of blood vessels (moderate inflammation) were observed.
After 56 days, both groups showed moderately proliferating synovial lining layers and focal proliferation of a small number of blood vessels (minimal inflammation). There was no significant difference between the 2 groups.
Articular cartilage was examined histologically for signs of damage caused by US exposure and/or MTX injection. A very slight decrease in the intensity of metachromatic staining (histologic scoring 0 or 1) in the surface area was observed in some samples of the MTX+/US+ and MTX+/US− groups on days 7 and 28.
Histologic scores. The mean ± SD histologic scores at 3, 7, 14, 28, and 56 days after treatment were 9.00 ± 0.31, 5.42 ± 0.30, 4.71 ± 0.18, 4.57 ± 0.37, and 5.57 ± 0.30, respectively, in the MTX+/US+ group, and 8.42 ± 0.20, 8.57 ± 0.37, 8.43 ± 0.20, 8.29 ± 0.36, and 5.71 ± 0.29 in the MTX+/US− group. The scores in the MTX+/US+ group and the MTX+/US− group were significantly different by two-way factorial ANOVA (P < 0.05). In addition, when we compared them at each time point using Student's paired t-test, the scores in the MTX+/US+ group at 7, 14, and 28 days were significantly better than those in the MTX+/US− group (Figure 4).
To confirm that the antiinflammatory effect was not due to US exposure, we injected the right knees with Optison only and exposed them to US (MTX−/US+ group). Scores in the MTX+/US+, MTX+/US−, MTX−/US+, and MTX−/US− groups were compared on days 7 and 28, and a significant difference was observed between the MTX+/US+ group and the other 3 groups (P < 0.05) (Figure 5).
Findings of real-time PCR. Using real-time PCR, we examined the quantitative gene expression of IL-1β in synovial tissue obtained 7 and 28 days after US irradiation. The mean ± SD expression 7 and 28 days after treatment was 1.3 × 10−3 (1.5 × 10−4) and 3.3 × 10−3 (2.5 × 10−4), respectively, in the MTX+/US+ group, and 8.6 × 10−2 (9.0 × 10−3) and 8.4 × 10−2 (1.2 × 10−3) in the MTX+/US− group. There was a significant difference (P < 0.01) between the groups at each time point.
Our results show that, in a rabbit arthritis model, microbubble-enhanced US treatment promotes uptake of MTX into synovial cells, which results in acceleration of antiinflammatory effects following intraarticular MTX injection. The antiinflammatory effect was confirmed by histologic scoring and expression of IL-1β mRNA in synovial tissue. The increased uptake of MTX into synovial cells was confirmed histologically by analysis of MTX concentration in synovial tissue and induction of Texas Red–conjugated MTX in vitro and in vivo. To confirm that the antiinflammatory effect was not due to US irradiation treatment with or without Optison, we injected Optison only (without MTX), and administered US to the right knees in the rabbits. Because there was no antiinflammatory effect, we concluded that neither microbubble injection with US exposure nor MTX injection has an antiinflammatory effect. However, the combination of MTX, microbubble injection, and US exposure was very effective. To our knowledge, this is the first report to describe the use of US in conjunction with antiinflammatory joint therapy in vivo. US gained attention through its use in gene therapy and tissue engineering. It is safe, minimally invasive, and can accommodate different therapeutic applications.
An antiinflammatory effect of this procedure (significantly reduced histologic inflammation and IL-1β gene expression in the synovial tissue) was observed from 7 to 28 days, but not at 56 days. All of the inflamed synovium of the joint cannot be covered by the use of a US probe applied at the skin surface. Indeed, histologic evaluation of inflammation of synovial tissue in the femorotibial joint, where US was not administered, revealed no difference from that in the control group (data not shown). Another explanation is that 56 days (exactly 66 days after immunization) may be enough to decrease the inflammation of synovial tissue in AIA naturally.
One hypothesis regarding the mechanism for the efficacy of this procedure is that the bioeffects are consequences of inertial cavitation, violent oscillations, and the collapse of bubbles in the surrounding fluid. The Optison stock concentration is ∼6.5 × 108/ml, and 5% Optison (3.25 × 107/ml) was injected into the joint. The microbubbles consist of hollow albumin filled with octafluoropropane. They are collapsed by the cavitation produced by US, and are considered to be eliminated quickly by distribution or phagocytosis. Physical and chemical phenomena related to inertial cavitation include microstreaming, shock waves, microjets, extremely high localized temperatures, pressures inside the bubbles, and generation of free radicals (16). If US causes promotion of induction through increased membrane porosity, as described above, it is reasonable to expect that there would be some limitations for transducible agents in terms of molecular size, 3-dimensional structure, and chemical compositions. In this study, we chose MTX (MW 454.45) as a transducive agent. Because we showed that Texas Red–conjugated MTX (MW 1,257.49) was also incorporated into cells in an in vitro experiment, molecules of this size should be readily incorporated into cells with this method. Uptake into cells of some genes or other pharmacologic agents that have an antiinflammatory effect can be promoted by microbubble-enhanced US exposure.
In vivo US and Optison conditions were selected based on in vitro data, and it is clear that there are significant differences between in vitro and in vivo conditions. Since the power of US is reduced by the long distance between the joint and the skin in vivo, it may be necessary to elevate the US power to a higher setting, provided there is no resultant heat production on the skin. In this study, output power (1–2W), duration (30 seconds to 2 minutes), and duty cycle (10–50%) were elevated.
Electroporation is also an effective MTX transduction technique. It has been widely used for transduction of genes and pharmacologic agents in vivo and in vitro. It does not require viral vector construction or virus preparation. Many reports describe electroporation as a technique for promoting electrochemotherapy through the uptake of MTX into cancer cells (17–19). We also showed that electrochemotherapy was effective in digital chondrosarcoma (20). However, the electrical fields created also affect normal tissue beyond the target site. Furthermore, cell anomalies or tissue damage after electroporation have often been observed. Thus, it is not an appropriate technique for the treatment of joints in RA patients. The encapsulation of MTX into cells with the use of a viral vector, recombinant polyomavirus–like particle, has been described (21), but this is a complex procedure and may not be suitable for clinical application.
MTX injection and US treatment can be used together in the clinic because of their low-risk safety profile. This technique may reduce synovitis in human arthritis and take the place of surgical synovectomy. It has been reported that surgical synovectomy of RA joints may offer short-term symptomatic relief but no retardation of the bone destruction or the disease process (22). While the benefits of our procedure might not match those of surgical excision of inflamed synovium, it is much less invasive than surgery and it can be performed frequently to obtain symptomatic relief. Further experiments are needed to determine whether this procedure alters the course of the disease. Furthermore, this technique can be applied in other inflammatory diseases.
For broader application in the clinic, there are some additional considerations that must be addressed. First, the intraarticular injection of MTX, which at high doses is sometimes used as an immunosuppressive agent, may cause adverse effects. The risk of iatrogenic infection warrants close attention. But MTX is reported to suppress production of superoxide and nitric oxide, and not to affect glycosaminoglycan synthesis of chondrocytes in vivo and in vitro (23, 24).
Second, US may cause adverse effects. It has been reported that US increases matrix synthesis and that it does not affect the viability or proliferation of chondrocytes (25, 26). The procedure reported here caused no harm to joint tissue apart from the proliferation zone in the synovium. However, the thick skin of the human joint may restrict the penetration of the ultrasonic vibration into synovial cells, which may cause a decrease in the uptake of MTX. The Sonitron is not approved for human use, but there is other equipment with greater power that is approved for human use in Japan (not for sonoporation). Longer exposure or higher power may cause local heat production. In such a case, a lower percentage setting of the duty cycle will help to prevent adverse effects. The optimal conditions necessary for the effective administration of US in humans need further investigation.
Third, the echo-contrast microbubbles may cause local or systemic side effects. These may include drug-induced allergic shock or joint pain because of high osmotic pressure. Moreover, another concern is that the overdosage of microbubbles or overexposure to ultrasonic waves may cause irreversible tissue injury. A minimum concentration of microbubbles would be desirable for clinical use. Optison has been used in the US and other countries for cardiac applications only, but it is not allowed for use in humans in some countries, such as Japan. Alternative clinical echo-contrast agents are now available in Japan. Levovist is one of these agents, although its gene transduction efficacy is reported to be inferior to that of Optison (27). Both are classified as second-generation microbubble agents. Optison consists of micrometer-sized (mean diameter 2–5 μm), denatured hollow albumin microspheres with a shell thickness of ∼15 nm. The microbubbles are filled with octafluoropropane. Levovist is a galactose-based, air-filled microbubble agent, 99% of which is smaller than 7 μm. Further studies are planned to compare the performance of these agents in vivo.
This method can be used to introduce not only MTX but also other agents, such as steroids and some genes, into synovial cells. The procedure can be used not only for intraarticular injection but also for a systemic approach with intravenous or oral administration of the therapeutic drug. Studies are under way to examine these approaches. This technique enables a very targeted application of reagents to the diseased tissue, thus enabling healthy tissue to be spared from treatment.