Biocompatibility study of different hyaluronan products for intra-articular treatment of knee osteoarthritis

SUPARTZ FX™ was purchased from Seikagaku Corporation (Tokyo, Japan). ORTHOVISC® was from Anika Therapeutics, Inc. (Bedford, MA). SYNVISC® was from Sanofi K.K. (Tokyo, Japan). Details of these products are given in Table 1. Water for Injection (WFI) and saline (Otsuka Normal Saline) was from Otsuka Pharmaceutical Factory, Inc. (Tokushima, Japan). The Glucatell® and Pyrochrome® plus Glucashield® kits were from Associates of Cape Cod, Inc. (Falmouth, MA), and used for (1 → 3)-β-D-glucan and endotoxin detection, respectively. The DC (detergent compatible) protein assay kit II was from Bio-Rad Laboratories, Inc. (Hercules, CA), and was used for protein detection in each HA product. The Pierce™ BCA Protein Assay kit was from Thermo Fisher Scientific (Waltham, MA), and was used for protein detection in synovial fluid in the biocompatibility study. Endo (1 → 3)-β-D-glucanase was prepared according to the method of Tanaka et al. [18].

Male New Zealand White rabbits (12 weeks old) were purchased from Oriental Yeast Co., Ltd. (Tokyo, Japan). The animals were housed under specific pathogen-free conditions at a room temperature between 20 °C and 26 °C, and humidity of approximately 50% on a 12-h/12-h light/dark cycle. The animals were quarantined and acclimatized for 1 week before dosing. The animal studies were reviewed and approved by the Institutional Animal Care and Use Committee of Seikagaku Corporation (approval number, 72–306) and performed under the animal husbandry/management system in an appropriate environment with animal protection/welfare in mind.

A biocompatibility study was conducted in accordance with our previous study [19]. Twenty-one rabbits were divided into four groups (n = 6 or 3) based on body weight and anesthetized with an intravenous injection of mixed anesthesia containing midazolam (Dormicum, Astellas Pharma Inc., Tokyo, Japan), xylazine (Selactar 2%, Bayer Yakuhin Ltd., Osaka, Japan), butorphanol (Vetorphale, Meiji Seika Pharma Co., Ltd., Tokyo, Japan), and saline (1:2:1:2), in a volume of 2 mL/animal. The animals received three consecutive weekly intra-articular injections of 0.25 mL of either saline, Avian-HA, Bio-HA, or Avian-CL-HA into both knee joints. All the rabbits were euthanized by exsanguination under deep anesthesia induced with the above-mentioned mixed anesthesia intravenous injection at 15 days after the first injection. Both knee joints were collected. Synovial lavage fluid was collected from the right knee joint by washing the joint cavity three times with 0.5 mL washing buffer (1 mM ethylenediaminetetraacetic acid in phosphate-buffered saline). The total number of cells in the synovial fluid was counted using a light microscope. For characterization of the cell types (monocytes, lymphocytes, and heterophils), the synovial fluid was smeared on a slide, stained with Diff-Quick (Sysmex Corporation, Hyogo, Japan), and then observed with a light microscope. The total protein content in synovial fluid was measured using the Pierce™ BCA Protein Assay kit. The synovium was collected from the left knee joint and was fixed in 10% neutral buffered formalin. Fixed samples were processed, embedded in paraffin, sectioned, and stained with hematoxylin and eosin (H&E) for histological examination. Blinded histological scoring was performed independently by two investigators. Inflammatory reaction with eosinophil infiltration was scored on a 4-point scale, from Grade 0 (no change) to Grade 3 (severe change). Synovial thickening was scored using criteria given in a previous report [19] (Table 2). The numbers of evaluated samples were n = 6 per each group except for the saline group, which was n = 3.

The protein level in each HA product was determined by the DC protein assay kit II according to the manufacturer’s instructions. Each HA product was diluted two-fold with purified water and incubated with the attached assay reagent for a minimum of 15 min at room temperature. The absorbance of the reaction mixture was measured in triplicate at 750 nm with a spectrophotometer (UV-1700; Shimadzu Corporation, Kyoto, Japan). The protein concentration range for the standard curve was 5–125 μg/mL. The sensitivity of the assay is defined as the lowest concentration used in the standard curve. To determine protein recovery, each HA product was spiked with a known amount of bovine serum albumin (BSA) and assayed, and the recovery rate was calculated.

All statistical analyses were performed using the Statistical Analysis System (SAS Institute Inc., Cary, NC). Parametric Tukey’s test was used for the protein contents and cell numbers in synovial lavage fluid and histological scores. The significance level was set at 5% (two-tailed). Data were represented as the mean and standard error of the mean (SE). For spike and recovery analysis, recovery rate (%) was calculated by dividing the measured value by the spiked value after subtracting any baseline signal from the measured sample.

The mean protein content of synovial fluid was 2.5 ± 0.2 mg/joint in the saline group. In the Avian-HA and Bio-HA groups it was higher, at 4.3 ± 0.1 and 4.6 ± 0.5 mg/joint, respectively, although the increase was not significant (Fig. 1). By contrast, the protein content in the Avian-CL-HA group was 7.9 ± 1.2 mg/joint, significantly higher than that in the saline, Avian-HA, and Bio-HA groups (parametric Tukey’s test, **P < 0.01 [vs. saline], †P = 0.01 [vs. Avian-HA], ¶P = 0.02 [vs. Bio-HA]).

In the Avian-HA and Bio-HA groups, three consecutive weekly injections did not induce a significant increase in any cell type in the synovial fluid (Fig. 2). In the Avian-CL-HA group, there was a significant increase in the average number of total cells (8.63 × 10⁵), monocytes (5.95 × 10⁵), and heterophils (1.22 × 10⁵), compared to the saline group.

The histological appearance of the synovium is shown in Fig. 3. One day after three weekly injections of Avian-HA and Bio-HA, no histological changes were detectable in the synovium of treated animals when compared with that of the saline group. A slight proliferation of the synovial cells was seen after injections of Avian-HA and Bio-HA as well as saline; however, eosinophil infiltration was not observed. After three injections of Avian-CL-HA, marked stratification of synovial lining cells, prominent cellular infiltration including eosinophils, and edema in the synovial tissues were observed in all animals. Furthermore, aggregations of inflammatory cells were observed in three of the six animals treated with Avian-CL-HA (black arrowhead). Histological scores are shown in Fig. 4.

To test whether our assays could efficiently recover protein from the HA products, we performed spike and recovery experiments in which a known amount of bovine serum albumin (BSA) was added to each HA product, and the protein content was determined thereafter. The recovery was close to 100%, ranging from 102 to 109%. As shown in Table 3, Avian-HA, Bio-HA, and Avian-CL-HA all had protein levels that were below the lower limit of quantification (< 10 μg/mL).

Spike and recovery experiments were performed to determine the accuracy of the method. Recovery rate varied between 76% (Bio-HA) and 151% (Avian-CL-HA). As shown in Table 3, all the HA products had endotoxin levels that were below the lower limit of quantification (< 0.05 EU/mL; except for Avian-HA, < 0.01 EU/mL).

Spike and recovery experiments were performed to determine the accuracy of the method. Recovery rates varied between 81% (Avian-HA) and 111% (Avian-CL-HA). As shown in Table 3, Avian-HA and Bio-HA had (1 → 3)-β-D-glucan levels that were below the lower limit of quantification (< 0.018 ng/mL and < 0.054 ng/mL, respectively). However, Avian-CL-HA contained 3.4 ng/mL (1 → 3)-β-D-glucan. To confirm the specificity of the test, Avian-CL-HA (100 μL) diluted 100-fold with water certified free of interfering glucan was mixed with 5 μL of 1.17 U/mL endo-(1 → 3)-β-D-glucanase and incubated at 37 °C for 1 h. As a result, 94.0% of the glucan signal was lost, confirming that the measured value for Avian-CL-HA was derived from (1 → 3)-β-D-glucan.