Aberrant MHC class II expression in mouse joints leads to arthritis with extraarticular manifestations similar to rheumatoid arthritis -- Kanazawa et al. 103 (39): 14465 Data Supplement - HTML Page - index.htslp -- Proceedings of the National Academy of Sciences

Kanazawa et al. 10.1073/pnas.0606450103.

Supporting Information

Files in this Data Supplement:

Supporting Figure 6
Supporting Table 1
Supporting Table 2
Supporting Figure 7
Supporting Movie 1
Supporting Movie 2
Supporting Movie 3
Supporting Table 3
Supporting Text

Supporting Figure 6

Fig. 6. hCIITA expression vector, Southern blotting, and PCR analyses of the hCIITA transgene in the D1CC mouse. (A) pCII.CIITA plasmid construction: HA epitope-tagged human CIITA (h:hCIITA) cDNA and SV40 polyA (pA) signals were placed downstream of the rat collagen II promoter (CII pro) and upstream of its enhancer (CII en). Also, b-globin splice signals were placed between the CII promoter and h:hCIITA cDNA. Arrows above the 5' end of CIITA indicate primers used for amplification of genomic DNA from the D1CC mouse (PCR). The solid line above the 3' end of CIITA corresponds to the probe for Southern blotting of DNA from the D1CC mouse (SB). (B) Southern blotting of the hCIITA transgene. D1CC mice were generated by injecting the linearized pCII.CIITA plasmid into blastocysts and backcrossing F₁ transgenic mice with DBA/1 mice >15 times. Lanes 1 and 2 contain genomic DNA from D1CC and parental DBA/1 mice, respectively. Genomic DNA was cut with EcoRI. Radiolabeled probe (A, SB, followed by bar above CIITA) hybridizes with the endogenous mouse CIITA gene (wt, upper band) and the hCIITA transgene (tg, lower band). The size of the molecular weight marker is given to the right of the Southern blot. (C) PCR detects the hCIITA transgene in the D1CC mouse. Primers amplify a DNA fragment of 409 bp from the HA epitope tag and the body of the hCIITA cDNA (A, arrows surrounding PCR above CIITA). Lanes 1 and 2 contain PCR products from D1CC and DBA/1 mice, respectively. The size of the fragment is given to the left of the agarose gel. (D and E) hCIITA transcripts can be detected in articular cartilage of the D1CC mouse. In situ hybridization with the radiolabeled cDNA probe revealed hCIITA transcripts in the cartilage of D1CC (D, arrow) but not parental DBA/1 mice (E). Radioactive signals were detected by autoradiography. Arrow with CIITA points to grains on film generated by the probe. Bone (b), cartilage (c), and joint space (js) also are labeled. (F and G) Likewise, the expression of MHC class II determinants can be detected by immunostaining with an anti-IA^q specific monoclonal antibody on chondrocytes in the articular cartilage of D1CC (F, arrow) but not parental DBA/1 mice (G). Arrow with MHC class II points to chondrocytes expressing MHC class II determinants in the D1CC mouse. Bone (b), cartilage (c), and joint space (js) also are labeled.

Supporting Figure 7

Fig. 7. Mast cells infitrate the joints in loCII-D1CC mice. (A-D) Mast cells contain large granules (A, HE) that stain with toluidine blue (B) and are observed in joints of the loCII-D1CC mouse 15 weeks after the boost. Cells from the same sections also express MHC class II (C) and B7.1 (D) determinants. Arrows point to mast cells (mc) and cells expressing MHC class II and B7.1 determinants. (E) Higher numbers of mast cells migrate into joints of loCII-D1CC mice. In each histological slide of fore and hind limbs, five periarticular areas were chosen and numbers of mast cells were counted in each 0.1 mm². They were tabulated and are presented in bar graphs. Presented are numbers from all extremities (limbs) of loCII-D1CC (black bar), loCII-DBA/1 (gray bar), and WT mice as a control (open bar). Sixty areas were analyzed for each extremity from six independent mice. Error bars present SEM. The Student's t test was performed for each value between loCII-D1CC and loCII-DBA/1 mice (*, P < 0.05).

Supporting Movie 1

Movie 1. Joint space narrowing, erosions, and osteoporosis in the knee of the loCII-D1CC mouse 24 weeks after the boost. Movie data present CT scans of distal femur, the knee joint, the proximal tibia, and the fibula for a loCII-D1CC mouse.

Supporting Movie 2

Movie 2. Joint space narrowing, erosions, and osteoporosis in the knee of the loCII-D1CC mouse 24 weeks after the boost. Movie data present CT scans of distal femur, the knee joint, the proximal tibia, and the fibula for a loCII-DBA/1 mouse.

Supporting Movie 3

Movie 3. Joint space narrowing, erosions, and osteoporosis in the knee of the loCII-D1CC mouse 24 weeks after the boost. Movie data present CT scans of distal femur, the knee joint, the proximal tibia, and the fibula for a hiCII-DBA/1 mouse.

Table 1. Clinical score for inflammatory arthritis

Score	Criteria
0	No effects*
1	Redness and swelling (one or two joints)
2	Redness and moderate swelling (more than three joints)
3	Redness, severe swelling (entire paw), and functional impairment^†

*All joints are examined and evaluated, no differences with untreated mice are found.

^†Maximal score for all four extremities is 12.

Table 2. Onset and incidence of inflammatory arthritis

Induction	CFA	CFA/loCII*	CFA/hiCII^†
Boost	IFA	IFA/loCII	IFA/hiCII
DBA/1	0%	0%	57.1%^‡
D1CC	0%	88.9%^§	90.5%^¶

Onset, clinical score ≥ 7.

*loCII, low dose of bCII, refers to 10 g of bCII per mouse.

^†hiCII, high dose of bCII, refers to 200 g of bCII per mouse. ^‡Corresponds to 14 mice. ^§Corresponds to 27 mice. ^¶Corresponds to 21 mice.

Table 3. Serology

Reagents	Days*	loCII-DBA/1^†	loCII-D1CC	hiCII-DBA/1^‡
α-CII Ab.^§	Day 0	ND^¶	ND	ND
α-CII Ab.^§	7 wks	3.0 ± 0.8^¶	6.2 ± 1.6^¶	10.0 ± 2.5^¶
IgG2a"	Day 0	7.4 ± 1.2	9.0 ± 1.0	7.8 ± 1.4
IgG2a"	7-8 wks	20.1 ± 5.3	16.4 ± 3.6	19.9 ± 4.3
IgG1**^††	Day 0	7.8 ± 2.8	8.4 ± 1.4	5.9 ± 1.1
IgG1**^††	7-8 wks	18.4 ± 4.6	18.6 ± 2.0	14.0 ± 3.2
α-CCP Ab.^††	Day 0	ND	2.08 ± 1.8	ND
	3wks-3mos	35.8 ± 5.0	51.5 ± 11.4	39.6 ± 14.7
	6mos	12.5 ± 2.5	2.40 ± 0.90	ND

ND, not detectable.

*Days after first immunization.

^†Low-dose bCII, referred in Table 1. ^‡High-dose bCII, referred in Table 1. ^§Anti-CII antibody (units/ml, ´ 10⁴), corresponds to average of 12, 10, and 9 mice, respectively. ^¶Student's t test, significant difference among each value, P < 0.05.

"IgG2a (ng/ml, ´ 1/15,000), ^cIgG1 (ng/ml, ´ 1/60,000), in serum, corresponds to average of four mice, respectively.

**Anti-CCP antibody (units/ml), in serum, corresponds to average of four mice, respectively.

Supporting Materials and Methods

DBA/1, CII Promoter/Enhancer-Driven CIITA
(D1CC) Mice. To create the CIITA transgene (pCII.CIITA; Fig. 6A), the HA epitope-tagged human full-length CIITA (h:hCIITA) cDNA was placed downsteam of the rat collagen type II (CII) promoter and upstream of exon 1 of the CII gene, where the CII enhancer is located between exons 1 and 2. Linearized pCII.CIITA DNA then was injected into F₁ blastocysts from FVB/NJ and DBA/1 mice (Charles River Laboratories, Davis, CA). The resulting transgenic mouse was named the D1CC mouse. After germ-line transmission, D1CC mice were backcrossed with DBA/1 mice more than 10 times to obtain the pure DBA/1 background. WT mice (called DBA/1 in the text) were used as controls. To determine the gene dosage of the pCII.CIITA transgene in the D1CC mouse, we performed Southern blotting. The human CIITA fragment that has the highest similarity with the mouse sequence was made by using PCR primer pairs 5'-ATAGAATTCCGCTGAGTGAGAACAAGA (p1) and 5'-TATGAATTCGGCGTCCACATCGCCAGC (p2) and subcloned into the Bluescript KS^- plasmid vector. Approximately 10 copies of CII.CIITA transgene were detected in the D1CC mouse. Genotyping D1CC Mice.
To examine the genotype of the D1CC mouse, Shimadzu AmpDirect reagents (Shimazu Biotech, Kyoto, Japan) and protocols were used in PCR. The PCR product measured 409 bp and was observed only in the D1CC mouse. In brief, blood samples of <1 ml were mixed with 10 ml of AmpDirect/10 ml of Amp Addition 2/1 ml of dNTP/1 ml of each primer (PCR primers pairs 5'-GATGTTCCAGATTACGCTGCT (h:hCIITA) and 5'-CTTCATCTGGTCCTATGTGCT (h:hCIITA p)/0.5 ml of Blend Taq (Toyobo, Osaka, Japan) and diluted to the final volume of 50 ml with distilled water. PCR reactions were performed for 15 min at 80°C, 4.5 min at 94°C, leading to 35 cycles for the next three steps: 30 sec at 94°C, 1 min at 60°C, 1 min at 72°C, and an additional 7 min at 72°C. All samples were separated and analyzed on 1.5% agarose gels. For Southern blotting, we used the 429-bp fragment from the hCIITA gene that is highly conserved between humans and mice according to the protocol from the manufacturer (Amersham Pharmacia Biosciences, Piscataway, NJ). Quantification of Anti-Collagen Antibodies with Immunospot Assay By Using the Infrared Fluorescence Imaging System (IFIS).
PVDF membrane (Immobilon-P Transfer membrane, Millipore, Billerica, MA) was pretreated with methanol for 1 min and washed with transfer buffer (20 mM Tris·HCl, pH 7.4/150 mM glycine/20% Methanol/0.005% SDS) for 10 min and rinsed with PBS. Collagen samples at various concentrations were spotted on the PVDF membranes (100 ml for each well) by using a 96-well dot blot apparatus (Bio-Dot Microfiltration Apparatus; Bio-Rad, Hercules, CA). These membranes were washed with PBS three times, and each well was blocked with 200 ml of blocking buffer (buffer B, 0.03% casein/0.3% rabbit serum albumin/0.3% porcine serum albumin in PBS) for 1 h at room temperature. All wells were washed with PBS containing 1% Tween 20 (PBS/T) four times. Sera from CII injected (low dose and high dose) D1CC and DBA/1 mice were diluted with the antibody buffer (buffer A: 0.1% casein/PBS/0.01% SDS) at 1:2,000. Serum samples (100 ml) were applied and incubated for 1 h at room temperature. The PVDF membrane was detached from the apparatus and washed with buffer B for 30 min and with PBS/T three times for 20 min, respectively. The membrane was immersed into the secondary antibody solution, Alexa Fluor 680 goat anti-mouse IgG antibodies (H&L; Molecular Probes, Eugene, OR) were diluted 1:5,000 with buffer A (for a 9 ´ 12-cm membrane, 1 ml of 1:5,000 diluted secondary antibody solution gave the best result), and incubated in the dark for 1 h at room temperature. The antibody-treated membrane was washed with the following buffers: PBS/T three times for 5 min, PBS/T containing 0.01% of SDS for 5 min, and PBS for 15 min. The fluorescence signal that was emitted from each spot on the membrane (3.5 mm in diameter) was detected with the Odyssey Infrared Imaging System (LI-COR, Lincoln, NE). The Odyssey application software (LI-COR) was used to quantify the infrared fluorescence intensity (1). All measurements were performed in duplicate. In Situ
Hybridization. To generate sense and anti-sense ³⁵S-labeled RNA probes for the hCIITA cDNA, we subcloned the C-terminal 281-bp fragment from hCIITA cDNA into Bluescript KS^- plasmid vector. Decalcified bones were used for sectioning and in situ hybridization (2). The signals of h:hCIITA transcripts were visualized by autoradiography by using emulsion (NBT-2; Kodak, Rochester, NY) Measurements of anti-CCP and IgG1 and IgG2a Antibodies in Serum.
We performed ELISAs according to the manufacturer's protocols: IgG1 and IgG2a (Bethyl, Montgomery, TX) and CCP (Axis-Shield Diagnostics Limted, Dundee, Scotland). For the measurement of anti-CCP antibodies, we used alkaline phosphatase conjugated anti-mouse IgG antibody as the secondary antibody and MRL-lpr/lpr mice sera as a standard to define our original unit definition (3).

1. Ota S, Kanazawa S, Kobayashi M, Otsuka T, Okamoto T (2005) J. Immunol. Methods 299:189-198.

2. Ferguson C, Alpern E, Miclau T, Helms JA (1999) Mech Dev 87:57-66.

3. Lopez-Hoyos M, Marquina R, Tamayo E, Gonzalez-Rojas J, Izui S, Merino R, Merino J (2003) Arthritis Rheum 48:2353-2361.