Figure 1.
Schematic of human COMP showing shRNA target locations relative to mRNA structure and protein domains.
Solid gray blocks are used to indicate the 5′UTR, coding region and 3′UTR of COMP mRNA. The positions of COMP protein domains including the pentamer, epithelial growth factor (EGF)-like, type 3 calcium repeat and globular domains are shown below the protein structure. The position of the common D469del mutation in exon 17B is designated by arrow. Stem-loop hairpins indicate target positions of the shRNAs.
Figure 2.
shRNAs directed against human COMP effectively reduce steady-state levels of COMP mRNA and protein.
COS-7 cells were stably transfected with shRNAs directed against COMP, and individual clones identified and selected as described in Methods section. Cells were infected with an adenovirus expressing MT-COMP and collected 48 hours after infection. RNA and protein were purified for subsequent analysis. (A) Northern blot analysis of steady-state COMP mRNA in transfected cells. (B) Quantification of mRNA in a. Absence of a shRNA was set to 100% and all other lanes were compared to the no shRNA reference sample. snU6 mRNA was used to normalize protein loading. (C) Western blot analysis. (D) Quantification of protein in c in which the level of COMP protein from COS-7 cells expressing recombinant COMP protein in the absence of a shRNA was set to 100% and all other protein levels were compared to this reference. β-actin was used to control for protein loading. All experiments were replicated at least three times.
Figure 3.
Cells expressing MT-COMP with a COMP-directed shRNA 3B show reduction of intracellular COMP and CRT.
COS-7 cells with (I–L) and without (E–H) shRNA 3B were infected with adenoviruses expressing MT-COMP (E–L). Uninfected COS-7 cells with shRNA 3B serves as controls (A–D). Cells were grown on coverslips for 48 hours after adenoviral infection and immunostained with COMP and CRT antibodies as described in the Methods section. DAPI (blue) staining was used to localize all of the cells in the field. COMP (green in E) co-localizes (merge-yellow in H) with CRT (red in F).
Figure 4.
Long-term activity of shRNA COMP knockdown was maintained over 2 weeks.
COS-7 cells with shRNA 3B, 3A-A or 3A–C integrant were infected with an adenovirus expressing MT-COMP and protein lysate was collected at 2, 4, 7 and 14 days. (A) Western analysis of protein lysates of MT-COMP expression levels at 2, 4 and 7 days. After 14 days, appreciable amounts of COMP were not being produced (*). (B) Quantification of COMP protein levels from COS-7 cells expressing recombinant COMP in the absence of a shRNA was set to 100% (*a) and all other lanes were compared to this reference. Levels of β-actin were used to normalize protein loading.
Figure 5.
shRNA knockdown was maintained in the presence of high levels of COMP.
COS-7 cells with and without 3B shRNA integrant were infected with a DOX-inducible MT-COMP adenovirus. DOX dosage ranging from 1 ng/ml to 10 µg/ml was used. Protein lysate was collected 48 hours after adenovirus infection. (A) Western blot analysis of COMP expression in the absence of shRNA 3B. (B) Western blot analysis of COMP expression in the presence of shRNA 3B.
Figure 6.
shRNAs directed against human COMP reduces steady-state levels of endogenous COMP mRNA from primary chondrocytes.
A population of integrants for each shRNA was isolated using puromycin selection as described in the Methods section. Primary human chondrocytes, with and without an shRNA, were cultured in monolayer for one week and RNA was collected. COMP is expressed in primary human chondrocytes grown in monolayer culture. qRT-PCR was used to quantify the COMP mRNA levels and β-actin mRNA levels were used to normalize the RNA.
Figure 7.
Immunohistochemistry analysis of RCS cells expressing MT-COMP with and without a COMP-directed shRNA 3B.
RCS cells with (right panel) and without (left panel) shRNA 3B, grown on coverslips, were infected with adenovirus expressing MT-COMP-FLAG and GFP. Cells were fixed 4 days after adenoviral infection and immunostained with COMP, type IX collagen, MATN3 and type II collagen antibodies as described in the Methods section. GFP was used to identify infected cells. FLAG was used as a marker for recombinant COMP (red) and the rER was delineated with a CRT antibody (blue).
Table 1.
Antibodies used in studies.