Serum biomarkers discovery of steroid-induced femoral head osteonecrosis in Chinese female by comparative proteome CURRENT POSTED

Background Steroid-induced osteonecrosis of the femoral head (SONFH) is a disabling, aseptic and ischemic disease due to excessive glucocorticoids (GCs) usage. Patients with SONFH are commonly asymptomatic, which makes its early diagnosis is challenge, the pathological mechanisms of SONFH are not well-known, the purpose of the present study was to screen diagnostic biomarkers for SONFH. Methods The differential expression of serum proteins from SONFH, traumatic osteonecrosis of the femoral head (TONFH) patients and healthy volunteers (CK) in Chinese females was compared using iTRAQ, and potential diagnostic biomarkers were verified by western blotting. Results Gene Ontology, Kyoto Encyclopedia of Genes and Genomes (KEGG), Domain and Clusters of Orthologous Groups (COG) analyses revealed key groups of proteins, pathways and domains differentially regulated among SONFH, TONFH and healthy volunteers in Chinese female, the results showed that peptidase S1, fibrinogen, transferrin, lipid transport domains and hematopoietic cell lineage, fat digestion, absorption, peroxisome proliferator-activated receptor (PPAR) pathways were associated with the development of SONFH. Finally, C-reactive protein (CRP), serum amyloid A protein (SAA1), alpha-1-acid glycoprotein 1 (ORM1) and dopamine beta-hydroxylase were selected for verification of differential expression using western blotting. Conclusions Our data suggest that dysfunction of hematopoietic cell lineage, adhesion, fat digestion and absorption, PPAR pathways may be involved in the pathogenesis of SONFH, serum proteins SAA1, ORM1 could be used as new potential diagnostic biomarkers for SONFH. lineage, adhesion and several steroid pathways were up-regulated in SONFH, and the fat digestion and pathways with expression in TONFH group. also found lower expression level in SONFH, TONFH versus CK groups. (4) Several displayed

from blood moderately drawn from bodies and then stored at -80 ℃ until detection.

Protein isobaric labelling and sample cleanup
After trypsin digestion, peptides were desalted by Strata X SPE column and vacuum-dried. Peptides were reconstituted in 100 μL 100 mM TEAB and processed according to the manufacturer's protocol for 6-plex tandem mass tags (TMT) kit [

Data processing
The resulting MS/MS raw data were converted to mgf format profile with the software Proteome Expression-based and functional enrichment-based clustering for different protein groups was used to explore potential relationships between different protein groups at special protein function. Firstly, all the protein groups obtained after functional enrichment analysis along with their P values were collected. Secondly, those categories enriched were sorted in at least one of the protein groups with a P value <0.05. This filtered P value matrix was transformed by the function x = −log10 (P value).
Thirdly, z-transformed applies on x values for each functional category and z scores were clustered by one-way hierarchical clustering (Euclidean distance, average linkage clustering). Finally, the expression pattern of the protein was completed using the timeclust function in the R package 'TCseq', a heatmap was generated by the software Heml 1.0.3 [28]. We also present the STRING network of some known/novel protein-protein interactions as an evidence view by using the String 9.0

Protein identification and expression of different protein (EDPs) screening
A total of 950 proteins were identified in 5 replicates LC-MS/MS experiments at the high confidence of peptide selection (FDR=0.01), among which quantitative information was obtained for 529 proteins (Supplementary Table 2). Volcano diagram shows the EDPs of each comparison groups, with upregulated proteins marked by red points distributing over the top right of the graph and downregulated proteins marked by blue points distributing over the top left of the graph (Fig. 1). There were 17 proteins show a significant differential expression (fold change > 1.2 and p value < 0.05) between the SONFH and CK groups with 8 up-regulated and 9 down-regulated proteins (

Protein expression patterns and function clustering
To investigate the protein profile change under the two pathological situations compared to healthy volunteers, the protein Domain and KEGG pathway annotation were elaborated from InterProScan platform and KAAS online automatic annotation server (http://www.genome.jp/tools/kaas/). Then the confidence of function enrichment was evaluated by the fisher exact test, all quantified proteins were classified into 8 clusters based on the expression level (Fig. 2). Domain enrich-based clustering of each expression cluster shows that IgG enriched expressed highly in clusters 2, 4 and 6 in SONFH group, while the peptidase S1 and fibrinogen, transferrin presents present low expression enriched pathways. Interestingly, PPAR signal pathway was also find lower expression level in SONFH and TONFH groups.

PPI network analysis
STRING was then used to further analyze the functional correlation of differentially expressed proteins between each group. The Fig. 3A announced that alpha-1-acid glycoprotein 1 (P02763) and haptoglobin (P00738, HP) connected the complement pathway and the coagulation pathway that may take potential interactions. And the interaction network EDPs between SONFH and TONFH indicated that beta-2-glycoprotein 1 (P02749, APOH) was employed an important role between these two pathological states (Fig. 3B).

Western blot validations
Based on proteomic results, C-reactive protein, serum amyloid A protein, serum alpha-1-acid glycoprotein 1 and dopamine beta-hydroxylase were selectively examined by using western blot. As shown in Fig. 4, in SONFH versus TONFH and CK, significantly increased level of C-reactive protein, serum amyloid A protein, serum alpha-1-acid glycoprotein 1, which is consistent to iTRAQ (based on LC-MS/MS identification and quantification). Meanwhile, western blotting also verified the decreased expression of dopamine beta-hydroxylase (DBH) in TONFH versus SONFH and CK, respectively.
Western blotting was consistent with the results of the mass spectrometry analysis, all these data added confidence to the results obtained from iTRAQ.

Discussion
Currently, steroid-related SONFH is a frequently occurring disease that can lead to necrosis of bone tissue and arthritis of the hip joint in patients who receive high-dose corticosteroid therapy. Though lots of studies have been paid to clarify their pathogenesis, the pathogenesis of SONFH is still unclear, and very little proteomic research has been done to high-throughput investigate the simultaneous expression of serum proteins and diagnostic biomarkers in adult SONFH patients.
(2) In screening protein clusters, IgG enriched was up-regulated in SONFH group, the peptidase S1 and fibrinogen, transferrin presents down-regulated in SONFH group.
(3) Hematopoietic cell lineage, adhesion and several steroid derived signal pathways were upregulated in SONFH, and the fat digestion and absorption pathways with highly expression in TONFH group. PPAR signal pathway was also found lower expression level in SONFH, TONFH versus CK groups. (4) Several other proteins such as C-reactive protein, serum amyloid A protein, alpha-1-acid glycoprotein 1 and dopamine beta-hydroxylase displayed similar expression in SONFH, TONFH versus CK groups.
PPARγ are believed to be the master regulator of adipogenesis and also show well described anti-     Validation of 4 EDPs, protein expression level and unique peptide intensity of four proteins.