KIAA1199 as a potential diagnostic biomarker of rheumatoid arthritis related to angiogenesis

Introduction Our previous proteomic study on fibroblast-like synoviocytes (FLSs) derived from the synovial tissues found that the expression of KIAA1199 was higher in rheumatoid arthritis (RA) patients than in healthy controls. The aim of this study was to examine the biological function of KIAA1199 and evaluate its clinical diagnosis value in RA. Methods The over-expression of KIAA1199 was verified by quantitative real-time polymerase chain reaction (qPCR), Immunohistochemistry, Immunofluorescence and enzyme linked immunosorbent assay (ELISA) in inactive and active RA patients and healthy controls. The effect of KIAA1199 expression on FLSs proliferation, angiogenesis and related pathway were analyzed by MTT, cell migration, tube formation, chorioallantoic membrane (CAM) assay, qPCR and western-blotting after KIAA1199 knockdown and over-expression. Results The verification results show the up-regulation of KIAA1199 in RA patients at mRNA and protein level as compared to that in healthy controls. ELISA and receiver operator characteristic (ROC) analysis shows that KIAA1199 concentration in serum, synovial fluid and synovial tissues could be used as dependable biomarkers for the diagnosis of active RA, provided an area under roc curve (AUC) of 0.83, 0.92 and 0.92. Sensitivity and specificity, which were determined by cut-off points, reached 72% 84% and 80% in sensitivity and 80%, 93.3%, 93.3% in specificity, respectively. Moreover, KIAA1199 also enhance the proliferation and angiogenesis of synovial membrane, and KIAA1199/ PLXNB3/ SEMA5A/CTGF axis may be a newly found pathway enhancing cell proliferation and angiogenesis. Conclusion KIAA1199 may be a potential diagnostic biomarker of RA related to angiogenesis. Electronic supplementary material The online version of this article (doi:10.1186/s13075-015-0637-y) contains supplementary material, which is available to authorized users.


Introduction
Rheumatoid arthritis (RA) is a chronic inflammatory disease characterized by progressive joint damage. The pathogenesis of RA is complex and thought to be mediated by various mechanisms. Early events in RA disease progression are defined by hyperplasia of the synovial membrane, influx of leukocytes and inflammatory cells. Activated fibroblast-like synoviocytes (FLSs) in the lining layer of the synovial membrane are among the dominant cell types involved in pannus formation, and pannus is a key player in joint destruction [1,2]. Angiogenesis is now recognized as a key event in the formation and maintenance of the pannus in RA [3,4].
KIAA1199 gene is a member of the large transmembrane protein of the KIAA family with more than 1000 amino acids [5] discovered about 10 years ago. Human KIAA1199 gene is located on chromosome 15q25.1 segment, which encodes a 150 kDa protein originally described as an inner ear protein [6]. KIAA1199 was found to have a G8 domain [7] and two GG domains [8]. Although the basic function of KIAA1199 remains unknown, an inverse correlation between the expression level of KIAA1199 and disease stage/5-year survival rate suggests that KIAA1199 may be associated with cancer progression [9]. It was also demonstrated that KIAA1199 was over-expressed in excessively proliferated cancer tissues, including those from gastric cancer [9], breast cancer [10][11][12] and colon cancer [13][14][15][16][17][18]. In addition, our previous proteomic study on FLSs derived from the synovial membrane also found that KIAA1199 expression in RA patients was significantly higher than in healthy controls [19], but the biological function and mechanism of action of KIAA1199 in RA remain unknown.
The aim of the present study was to verify the overexpression of KIAA1199 mRNA and protein in the serum, synovial fluid and synovial tissues obtained from patients with active and inactive RA and healthy controls, explore the effect of KIAA1199 on FLSs proliferation and angiogenesis by MTT, cell migration, tube formation and chorioallantoic membrane (CAM) assay after KIAA1199 knockdown and over-expression.

Patients and primary culture of FLS cells
The serum was obtained from 44 RA patients, 15 osteoarthritis (OA) patients, 15 ankylosing spondylitis (AS) patients and 15 normal subjects. Knee synovial fluids and synovial tissues were from 44 RA patients undergoing synovectomy or joint replacement surgery and 15 normal subjects undergoing high-level amputations in Shanghai Changhai Hospital and Shanghai Guanghua Hospital (Shanghai, China). RA patients were further categorized as a group with active RA (n = 25) and a group with inactive RA (n = 19) depending on the elevation of disease activity score in 28 joints (DAS28) (inactive RA: DAS28 < 3.2; active RA: DAS28 > 3.2); DAS28 score correlates closely with clinical parameters of RA disease activity [20]. Patients fulfilled the 1987 American College of Rheumatology criteria for the diagnosis of RA [21]. The clinical data of the patients are shown in Table 1. Serum and synovial fluid were stored at −80°C immediately after centrifugation at 12,000 rpm. One part of synovial tissues was stored at −80°C, another part was isolated enzymatically according to the method previously described [19]. All FLSs of passages three to five were used for the experiment. This study was approved by Shanghai Changhai Hospital ethics committee (CHEC2013-194), with informed consent from all the participants concerned.

RNA preparation and quantitative real-time PCR analysis
Total RNA was extracted from the synovial tissues or FLS cells by Trizol reagent (Invitrogen, Carlsbad, CA, USA), precipitated with isopropanol and dissolved in DEPC-treated distilled water. The concentration of total RNA was determined by Eppendorf BioPhotometer Plus (Hamburg, Germany). Total RNA (2 μg) was then treated with RNase-free DNase (Invitrogen, Carlsbad, CA, USA) before the first-strand cDNA was generated using the random hexamer primer provided in the first-strand cDNA synthesis kit (MBI Fermantas, Vilnius, Lithuania). Specific amplification was performed using the primers of KIAA1199 genes (forward primer: 5′ TGC TGC CCG GGT ATT CAA AT 3' and reverse primer: 5′CGT CCA CTC CAC GTC TTG AA 3′), plexnB3 genes (forward primer: 5′ ACC CAG GTC AAG GAG AAG GT 3′ and reverse primer: 5′ GTC TTC GTC CGA TAG GGT CA 3′), sema5A genes (forward primer: 5′ GCT CCT TCC ACA AGA AGT GC 3′ and reverse primer: 5′ CAA GCT GCT TCC AAG AAT CC 3′), ctgf genes (forward primer: 5′ TGG AGT TCA AGT GCC CTG AC 3′ and reverse primer: 5′ GTA ATG GCA GGC ACA GGT CT 3′) and βactin (forward primer: 5′ATGG TGG GTA TGG GTC AGA AG 3′ and reverse primer: 5′TGG CTG GGG TGT TGA AGG TC 3′) used as an internal control for determining the cell number and metabolic status. Quantitative real-time PCR (ABI7300, Applied Bio-systems, Carlsbad, CA, USA) was done with SsoFast EvaGreen supermix PCR kit (Bio-Rad, Hercules, CA, USA). A total of 40 cycles Values are expressed as mean ± SD unless stated otherwise. NA, not applicable.
of PCR was performed for 15 s at 95°C, and 60 s at 60°C. The relative expression of each target gene compared with β-actin was calculated using the 2 -ΔΔCt . All reactions were conducted in triplicate.

Immunohistochemical analysis
Synovial tissues were fixed overnight in 4% paraformaldehyde, embedded in paraffin, and sectioned in 5-to 8-μm intervals. In brief, sections on slides were de-paraffinized, re-hydrated, antigens unmasked by incubating in target retrieval solution at 95°C for 30

Enzyme-linked immunosorbent assay for human KIAA1199
KIAA1199-specific antigens were detected in serum, synovial fluid and synovial tissues extraction samples from patients with active RA, inactive RA, OA and AS, and healthy subjects. The homogenate was centrifuged in a micro-centrifuge for 5 minutes at 3,000 g, and 100 μl diluted supernatant (1:50 with incubation buffer) was incubated at room temperature in a microtiter plate coated with KIAA1199 monoclonal antibody (sc-164775, Santa Cruz, CA, USA). After incubation, washing and addition of a detection antibody coupled to horseradish peroxidase, the substrate was added and incubated, followed by addition of a stop solution. The absorption rate was determined at an optical density of 450 nm. All reactions were conducted in triplicate.

Western blotting analysis
An equal amount of protein extract (50 μg per lane) was separated by SDS-PAGE of 10% polyacrylamide, and electrotransferred onto polyvinylidene fluoride (PVDF) membranes. The membranes were incubated for 1 h with a blocking solution containing 0.1% Triton X-100, 5% nonfat milk in PBS. The KIAA1199 monoclonal antibody (Abnova, H00057214-M01), PLXNB3 (Santa Cruz, sc-46240), SEMA5A (sc-67953, Santa Cruz, CA, USA), CTGF (R&D Systems, AF660, Minneapolis, MN, USA) were then added. The membranes were incubated overnight at 4°C, and then with the appropriate horseradish peroxidase-conjugated secondary antibody at room temperature for 1 h. The filter was then incubated with the substrate and exposed to radiographic film. All reactions were conducted in triplicate.

Cell proliferation assay
Cell proliferative activities were examined using FLSs.

Cell migration and endothelial tube formation assays
Transwell motility assays were performed by 6.5 mm Transwell® with 8.0-μm pore polycarbonate membrane filters (Corning Corp, Corning, NY, USA). Human umbilical vein endothelial cells (HUVEC) transfected with scrambled siRNA, KIAA1199 siRNA duplex (100 nM), pcDNA3.2DEST and pcDNA3.2DEST-KIAA1199 (100 nM) were cultured to 85% to 95% confluence, subjected to serum starvation for 24 h before cell migration assay, dissociated by incubation with trypsin-EDTA, washed twice with PBS, and counted using a hemocytometer. Then 600 μl medium was added to each lower chamber of the 24-well transwell for 24 h at 37°C. Inserts (the upper chambers) were then placed in the wells. Cells (1 × 10 5 in 300 μl serum-free media) were added to each upper chamber and incubated for 6 h. Nonmigrating cells were removed with a cotton swab. Cells that migrated to the lower phase of the upper chamber were then fixed in methanol for 30 minutes and stained with crystal violet for 3 minutes at room temperature. Excess stain was removed with distilled water, and the chambers were air-dried. Pictures were taken under the microscope and the cell number was quantified by software Image-Pro.
The assay was performed in triplicate. The tube formation assay was performed as follows: 24well plates were pre-coated with Matrigel and incubated at 37°C to promote gelling. HUVEC were added to each well transfected with scrambled siRNA, KIAA1199 siRNA duplex (100 nM), pcDNA3.2DEST and pcDNA3.2DEST-KIAA1199 (100 nM). After 6 h incubation, the plates were fixed with 4% paraformaldehyde and a blinded observer assessed the morphology of the tubes. Tube-like structures were quantified by counting the number of intersections between branches of the endothelial cell networks in the whole field. The assay was performed in triplicate.

Chorioallantoic membrane assay
Angiogenic activity of KIAA1199 was assayed on CAM as described by Takigawa et al. [22]. Embryonic CAM were treated on day 7 with scrambled siRNA, KIAA1199 siRNA duplex (100 nM), pcDNA3.2DEST and pcDNA3.2DEST-KIAA1199 (100 nM) absorbed on sterile Whatman GB/B glass fiber filter disks (6 mm in diameter; Reeve-Angel, Clifton, NJ, USA). The disks prepared with 20 μl of factor were placed upside down on windows that had been made in the eggshells on day 7 of incubation. The embryos were examined 3 days later under a stereomicroscope. ImageJ 2.43 s was used to calculate the vascular area and CAM area.

Statistical analysis
Data analysis involved estimation of the mean and SD using SPSS 17.0. The Shapiro-Wilk method and histograms were used to test whether the data were normally distributed. The Levene method was used to test homogeneity of variance. Two sets of data that met the normal distribution and homogeneity of variance were analyzed by independent samples t-test. The Kruskal-Wallis and Mann-Whitney non-parametric tests were used to compare interassay differences in data that did not meet the normal distribution or the homogeneity of variance. Inspection level P-values <0.05 were considered statistically significant.

Elevated expression of KIAA1199 in RA patients
Our previous comparative proteomic study results showed that KIAA1199 protein was 5.19 times over-expressed in RA FLSs [19]. The results of automated 2D-Nano-LC-ESI-MS/MS identification are shown in Figure 1A and Additional file 1. In this study quantitative real-time PCR was performed to confirm whether there was a real difference between RA and normal synovial tissues in the expression of KIAA1199. Compared to healthy subjects, the KIAA1199 mRNA expression was 2.77-fold and 6.62-fold up-regulated in inactive RA and active RA patients than in healthy subjects, respectively ( Figure 1B), which was consistent with the change in the previous proteomic study. The immunohistochemical analysis shows the expression of KIAA1199 was significantly higher in the vascular endothelium of synovial membrane in active RA and inactive RA than in normal synovial membrane ( Figure 1C). Immunofluorescence microscopy shows that KIAA1199 and CTGF proteins were co-expressed in FLS cells, and they were either not expressed or only weakly expressed in the nucleus, and were strongly expressed in the cytoplasm and membrane of FLS cells ( Figure 1D).

Correlation between KIAA1199 and disease activity
Knowing that DAS28 score is an indicator for measuring clinical disease activity, RA patients were divided into an active or inactive group according to DAS28 score (inactive RA: DAS28 < 3.2; active RA: DAS28 > 3.2). The abundance levels of KIAA1199 in the serum, synovial fluid and synovial tissues of patients with inactive and active RA and healthy subjects were determined by ELISA. KIAA1199 protein was expressed in human serum, synovial fluid and synovial tissue, while the expression of KIAA1199 protein was increased in the RA group compared to the healthy group (P <0.05). There were no significant differences in KIAA1199 in serum among patients with OA or AS, or healthy controls, and the concentration of KIAA1199 in RA was different from that of OA and AS. Therefore the concentration of KIAA1199 can be used as a biomarker of RA. The levels of KIAA1199 in serum, synovial fluid and synovial tissue were significantly elevated in patients with active RA as compared to those with inactive RA. Interestingly, the number of samples that were KIAA1199-positive was significantly greater in the active RA than that in the inactive RA group (18 versus 11 in the serum, 21 versus 9 in the synovial fluid, and 20 versus 9 in the synovial tissues).These data suggest that the level of KIAA1199 in serum, synovial fluid and synovial tissues was correlated with the disease activity and was associated with the process of RA (Figure 2).  Experiments were performed at least in triplicates; bars represents mean ± SD; *P <0.05 compared with normal. (C) Immunohistochemical staining of KIAA1199 was performed on synovial tissues from patients with inactive or active RA and normal subjects. The KIAA1199 protein was over-expressed in vascular endothelium in active and inactive RA synovial membrane compared to normal synovial membrane. (D) Immunofluorescence microscopy of KIAA1199 and CTGF was performed in FLS cells from RA patients. Result shows KIAA1199 and CTGF were co-expressed in FLS cells, were either not expressed or only very weakly expressed in the nucleus, and were strongly expressed in the cytoplasm and membrane of FLS cells.

The effect of KIAA1199 on FLS cells proliferation
The proliferation of synovial membrane is now recognized as a key event in RA progression. To determine the effects of KIAA1199 expression on FLS cell proliferation, the MTT assay was performed on FLS cells treated with scrambled siRNA, KIAA1199 siRNA duplex, pcDNA3.2D-EST and pcDNA3.2DEST-KIAA1199 for 24 h, 48 h, 72 h and 96 h at 37°C. The MTT results showed that the viability of FLSs treated with KIAA1199 siRNA decreased significantly after KIAA1199 knock-down and increased after KIAA1199 over-expression; significant differences were observed between every time point examined, indicating a linear change in proliferation with treatment time, P <0.05. These data support the assumption that KIAA1199 may be an inducer of FLSs proliferation (Figure 3).

The effect of KIAA1199 on angiogenesis
Angiogenesis is now recognized as a key event in the formation and maintenance of the pannus in RA. To further evaluate the effects of KIAA1199 expression on cell angiogenesis, the transwell and tube formation assays were performed on HUVEC cells treated with scrambled siRNA, KIAA1199 siRNA duplex, pcDNA3.2DEST and pcDNA3.2DEST-KIAA1199 for 6 h at 37°C. The transwell assay result shows KIAA1199 knock-down reduced the migration ability and KIAA1199 over-expression increased the migration ability of HUVEC cells, P <0.05 ( Figure 4A). The tube formation experiment shows a similar result to the transwell assay, P <0.05 ( Figure 4B). The above in vitro findings allowed us to investigate the angiogenic activity of KIAA1199 in vivo. First, we carried out a CAM assay. After 3 days treatment of scrambled siRNA, KIAA1199 siRNA duplex, pcDNA3.2DEST and pcDNA3.2DEST-KIAA1199 on CAM of chick embryos, the formation of small vessels decreased in KIAA1199 knock-down and increased in the KIAA1199 over-expression experiment, P <0.05, suggesting that KIAA1199 is a requirement for angiogenesis ( Figure 4C).

Effect of the KIAA1199/PLXNB3/ SEMA5A/CTGF axis on angiogenesis
To clarify the pathway of KIAA1199 regulation of angiogenesis in FLS cells in RA, hypothetical-pathway-related proteins were determined by qPCR and western blotting in triplicate. We observed an decrease of 77.0%, 70.3%, 51.7%, and 64.9% in RNA levels and a decrease of 76.5%, 54.6%, 47.0%, and 30.7% in protein levels of KIAA1199, PLXNB3, SEMA5A and CTGF after KIAA1199 knockdown, and an increase of 182.6%, 135.5%, 110.3%, and 132.2% in RNA levels and an increase of 77.5%, 40.7%, 75.4%, and 112.0% in protein levels of KIAA1199, PLXNB3, SEMA5A and CTGF after KIAA1199 overexpression, respectively. This is to say, the expression difference of PLXNB3, SEMA5A and CTGF were proportional to the expression of KIAA1199 in mRNA and protein levels ( Figure 5A and B).

Discussion
A hallmark of RA is the pseudo-tumoral expansion of FLSs, which induces the pannus formation and erodes the cartilage and bone. Many observations [23] have prompted speculation that joint FLSs in RA evolve genetically to form a locally invasive and metaplastic tissue. So cytokineenhanced FLS proliferation in RA would be a potential biomarker and has therefore been proposed as a therapeutic target by knocking down its expression.    Our previous comparative proteomic study showed that KIAA1199 was 5.19 times over-expressed in RA FLS cells as identified by automated 2D-Nano-LC-ESI-MS/MS [19]. This finding is supported by Yoshida's result showing that the level of KIAA1199 expression in non-inflamed synovial tissues was lower than in that of rheumatoid synovial tissues as shown by real time PCR and immunoblotting (n = 3) [24]. However, this has not been confirmed in a large set of samples, neither with regard to RNA or protein levels, nor with regard to gene function. In this study, we evaluated KIAA1199 expression by quantitative RT-PCR and ELISA in 44 patients with RA (19 with inactive and 25 with active RA) and 15 healthy controls. The data were analyzed to determine the clinical significance of KIAA1199 levels in RA. It was found that the expression of KIAA1199 mRNA and protein were higher in the synovial tissues of RA patients than healthy subjects; ELISA also showed that the levels of KIAA1199 protein expression were higher in the serum, synovial fluid and synovial tissue from RA patients than healthy subjects, suggesting that the overexpression of KIAA1199 plays an important role in the pathogenesis of RA. These data not only indicate that the level of KIAA1199 expression is higher in RA but suggest positive correlation between KIAA1199 expression and the DAS28 score, an RA index correlated closely with clinical parameters of RA disease activity. ROC curve analysis indicated high diagnostic value of KIAA1199 in active RA. These data also imply that KIAA1199 may be a potential diagnostic biomarker of RA, but its role in RA procession remains elusive.
Yoshida et al. carried out further research on the relationship between KIAA1199 and hyaluronan metabolism, a very important polysaccharide in synovial fluid for minimizing friction between the bones. They found that the cleavage of N-terminal 30 amino acids occurs in functionally matured KIAA1199, resulting in altered intracellular trafficking of the molecule and loss of cellular hyaluronic acid (HA) depolymerization. This suggests that the N-terminal portion of KIAA1199 functions as a cleavable signal sequence required for proper KIAA1199 translocation and KIAA1199-mediated HA depolymerization. Notably, the secreted mature form of KIAA1199 showed no HA degrading activity, together supporting the idea that KIAA1199mediated HA depolymerization occurred through rapid vesicle endocytosis. These results show that KIAA1199 protein promoted the degradation of HA, which is normal in physiological tissues and fast in inflammatory and neoplastic diseases. In osteoarthritis or RA synovial fibroblasts, the enhancement of HA metabolism is associated with increased expression of KIAA1199 [19,[24][25][26]. These data indicate that KIAA1199 plays a key role in HA catabolism as a unique hyaluronic cadherin in the dermal and OA synovial tissues.
The over-expression of KIAA1199 has been identified in many proliferative tissues, including synovial tissues in RA. Several reports [27] support the importance of the Wnt pathway activation in FLS proliferation. Wnt signaling is now well-recognized as a critical pathway in the regulation of growth and development. Birkenkamp-Demtroder et al. [17] report that KIAA1199 expression is markedly reduced by inactivation of the β-catenin/Tcell factor transcription complex, the pivotal mediator of Wnt signaling. Thus, they identified KIAA1199 as a novel target of the Wnt signaling, the regulation of KIAA1199 by Wnt signaling was observed as a proteinprotein interaction.
Immunohistochemical analysis in this study showed the expression of KIAA1199 was significantly higher in the vascular endothelium. Jami also revealed the involvement of KIAA1199 in breast cancer growth, motility and invasiveness by functional proteomic analysis [28]. Together these data suggest that KIAA1199 not only correlated with the proliferation of FLSs but may also be related to angiogenesis. The enhancing effects of KIAA1199 on angiogenesis were verified successfully in this study by transwell, tube formation and CAM assays after KIAA1199 knock-down and over-expression. CTGF is another overexpressed protein in RA patients identified in our previous proteomic study, which promotes the proliferation and migration of HUVEC [19]. Furthermore, the expression of CTGF was decreased or increased as KIAA1199 was knocked down or over-expressed. The question remains as to which pathway regulates the expression of CTGF by KIAA1199. Immunofluorescence microscopy of KIAA1199 and CTGF was performed in FLS cells from RA patients. The result shows that KIAA1199 and CTGF were coexpressed in FLS cells, and they were either not expressed or only very weakly expressed in the nucleus, and were strongly expressed in the cytoplasm and membrane of FLS cells. This experimental study gave us much inspiration. Nakayama found out that PLXNB3 is the only protein that can interact with KIAA1199, as identified by a yeast twohybrid system [29]. PLXNB3 belongs to the plexin family [30,31] and is the receptor of SEMA5A, which is a membrane protein belonging to semaphorin (arm board protein) gene family with seven repeated platelet-response protein-1 domains [32]. Furthermore, there is a plateletresponse protein-1 binding domain in CTGF [33], reminding us of the interaction between SEMA5A and CTGF by bioinformatics analysis. The binding of PLXNB3 and SEMA5A is involved in combination axon guidance [34], cell invasive growth [35], angiogenesis [36] and the process of cell migration [37]. Interestingly the synovial membrane also showed the characteristics of invasive growth and angiogenesis in the progression of RA. In our experiment the expression of PLXNB3, SEMA5A and CTGF are in accordance with KIAA1199 knock-down and over-expression. So we boldly assume that the KIAA1199/ PLXNB3/SEMA5A/CTGF axis may accelerate the proliferation of FLS cells and activate the downstream angiogenic signaling pathways, leading to the formation of the pannus, and erode the cartilage and bone in the progression of RA.

Conclusion
One of the interesting findings in this study is the more abundant expression of KIAA1199 in RA, and the KIAA1199/PLXNB3/SEMA5A/CTGF axis may be a newly found pathway enhancing cell proliferation and angiogenesis. In addition, KIAA1199 seems to have clinical correlation with the DAS28 and disease activity. In summary, KIAA1199 may be a potential biomarker and therapeutic target in RA.