Syndecan-1 (CD138) immunohistochemical expression patterns in lupus nephritis; reflections on different clinicopathological parameters


 Introduction: Syndecan 1 (SCD1) is a lectin expressed at the surface of renal tubular epithelial cells and plasma cells. In epithelial cells, cell surface syndecan1 is cleaved by inflammation-induced proteases (eg, ADAMTS, MMP), since loss of cell surface syndecan1 is associated with higher susceptibility to cell damage. Objectives: To explore a potential additional value of SCD1 immunohistochemical expression in lupus nephritis specimens of different ISN/RPS classes and NIH activity and chronicity indices. Patients and Methods: This retrospective study included 50 renal biopsy specimens diagnosed as lupus nephritis at the pathology laboratory, and electron microscopy (EM) laboratory of Ain-Shams University specialized hospitals. Data were collected from records as personal data, medical history and laboratory results including serum creatinine and proteinuria. Immunohistochemical expression of syndecan-1 was evaluated in renal tubular epithelial cells (TECs) followed by correlation with different clinicopathological parameters. Results: Fifty renal biopsy specimens with lupus nephritis including 14 cases of class II, 4 cases of class III, 20 cases of class IV and 12 cases of class V were re-evaluated. The mean serum creatinine was 1.57 ± 0.67 mg/dL. Nine cases (18%) were negative for proteinuria, while 41 cases (82%) were presented with proteinuria with a mean of 1.5 ± 0.9 g/24 h. There was no statistically significant difference in the percentage of SCD-1 expression with different lupus classes. Serum creatinine and albumin showed a statistically significantly different correlation with semiquantitative score of SCD-1 expression. The highest value of creatinine detected with score 1 of SCD-1 expression (P=0.038) and the highest value of urinary albumin was recorded with score 1 of SCD-1 expression. Accordingly, the lowest mean of urinary albumin recorded in SCD-1 score 3 (P<0.001). There was a weekly negative association between loss of SCD-1 expression and increased NIH activity and chronicity indices. Conclusion: Loss of syndecan immunohistochemical expression in renal TECs in lupus nephritis is highly associated with proteinuria and elevated serum creatinine and can be used as a predictive marker for disease severity and progression.


Introduction
Lupus nephritis (LN) represents a severe manifestation of systemic lupus erythematosus (SLE) initiated by deposition of anti-double stranded DNA (anti-dsDNA) autoantibodies in glomerular basement membranes (1)(2)(3). In the last few years, it was hypothesized that lupus nephritis has local renal immune mechanisms rather than a systemic activity (4,5).
Several studies have been conducted for early detection of lupus nephritis activity before the progression to end-stage renal disease (ESRD). These studies reported a strong association between low glomerular filtration rate (GFR), high chronicity indices and poor renal outcome (6)(7)(8)(9)(10)(11). Tubulointerstitial lesions in the form of interstitial inflammation and fibrosis, and tubular atrophy were known as one of the prognostic markers in lupus nephritis (12). Several underlying mechanisms for tubulointerstitial lesions include proteinuria, immune complex deposition in the interstitium, and rupture of the Bowman's capsule and induction of pro-inflammatory molecules at the surface of renal tubular cells (13)(14)(15).
Previous studies showed that changes in heparan sulfate proteoglycans (HSPGs) affect renal inflammatory responses as observed in experimental renal IRI (ischemia reperfusion injury) and primary human kidney disease (16,17). HSPGs were proved to be involved in angiogenesis, wound healing, tumor progression and inflammation through binding and presenting growth factors, chemokines and cytokines, as well as facilitating inflammatory cells adhesion by binding to the leukocyte adhesion molecule L-selectin (16,(18)(19)(20)(21)(22).
Syndecan-1 (SCD1) is a transmembrane HSPG involved in epithelial differentiation and re-epithelialization during wound healing and inflammation (18,(23)(24)(25). There are four members of syndecans. SCD1, known as CD138, is expressed on the surface of plasma, endothelial and epithelial cells (18). During tissue injury, the SCD1 shedding is accelerated through the release of different proteases and growth factors, thus it can be detected in inflammatory fluids. Epithelial cell surface SCD1 is cleaved by inflammation-induced proteases (ADAMTS and MMP) (26).
Few studies have assessed the role of serum SCD1 in lupus nephritis. Serum syndecan-1 was found to be significantly high in active lupus nephritis patients and correlated with disease activity. It was also significantly related to the presence of musculoskeletal manifestations and serositis which give a promising role in prediction of disease activity (31,32). Moreover, loss of cell surface SCD1 is associated with higher susceptibility to cell damage, in gastroenterological and renal disorders, and it has been recently identified as a marker of renal tubular stress (33)(34)(35)(36).
To date, SCD1 expression in lupus nephritis has scarcely been studied. Even in these few studies, the focus was directed on its expression in plasma cells regardless of its assessment in renal tubular epithelial cells (TECs). In these studies, SCD1 stained plasma cells were not seen in glomeruli even in proliferative glomerulonephritis and they were mostly seen as scattered single cells in the interstitium of the cortex and outer medulla. Additionally, the average medullary plasma cell SCD1 expression was found to be associated with higher disease activity and chronicity as well as all lupus nephritis classes except class II (37). Accordingly, the study by Pamfil et al (36) showed the link between renal TEC damage (indicated by lost SCD1 expression) and poorer renal outcome in LN, which was assessed by lower GFR (36). Therefore, we sought to explore the patterns of SCD1 TEC expression in lupus nephritis patients and test potential reflections of such patterns on the prognostic indices of lupus nephritis. The role played by SCD1 in inflammation/re-epithelialization sequel has given us insights to test its expression in TEC in lupus nephritis patients to find if there is a link with different clinicopathologic parameters.

Objectives
The aim of the present study is to evaluate SCD1 immunohistochemical expression in renal TECs in lupus nephritis in correlation with various clinicopathological parameters.

Study population
This retrospective study included 50 renal biopsy specimens diagnosed as lupus nephritis at the pathology laboratory, and electron microscopy (EM) laboratory of Ain-Shams University specialized hospitals. Data were collected from records as personal data, medical history and laboratory results including serum creatinine and proteinuria. Cases with adequate paraffin tissue blocks for immu nostaining were enrolled in the study. All renal biopsy specimens were graded according to the 2003 international society of nephrology/renal pathology society (ISN/RPS) classification system (38) and national institutes of health (NIH) activity and chronicity indices (39,40). The renal specimens were re-evaluated by light mi croscopy (LM) and immunohistochemical staining using (syndecan-1).

Light microscopy
Formalin-fixed, paraffin-embedded tissues from all cases were retrieved from the archive of the pathology laboratory. Slides were re-evaluated by two pathologists using LM (Olympus BX31) according to the histopathological guidelines for the evaluation and scoring of lupus nephritis (38)(39)(40). The pathologists were blinded to the clinical parameters of the patients.

Immunohistochemical staining method
Tissue sections were deparaffinized, rehydrated, and the endogenous peroxidase activity was quenched by 10-minute incubation in 3% hydrogen peroxide in methanol. After antigen retrieval and protein block, the primary antibody (mouse monoclonal anti-human syndecan-1 antibody; sc-390791, Santa Cruz) was supplied with dilution (1:50). The secondary antibody (supersensi tive immunodetection system (Biogenex, catalog No: AD000-SL) was then applied followed by peroxidase labeled streptavidin. Slides were incubated for 10 minutes with substrate chromogen (DAB) mixture. Finally, the slides were counterstained and mounted with Canada balsam. Any staining detected in membranes of TECs is considered positive.

Evaluation of immunostaining
Two scoring methods for SCD-1 expression in renal tubular epithelium were used as follows: (a) Quantitative score; by evaluation of the percentage of SCD-1 positive TECs to the whole area of renal cortical tubules in the examined specimen (typically 5-10 visual fields per biopsy) (41). (b) Semiquantitative score; -Score 0; cortical renal tubules are negative for SCD-1 -Score 1; positive membranous SCD-1 in <30% of renal cortical tubules -Score 2; positive membranous SCD-1 in 30-60% of renal cortical tubules -Score 3; positive membranous SCD-1 in >60% of renal cortical tubules The intensity of immunostaining was graded into weak, moderate, strong.

Ethical issues
This investigation was in accordance with the Declaration of Helsinki. This investigation was conducted retrospectively on paraffin embedded blocks of kidney biopsies to test SCD-1 immunohistochemical expression in renal TECs. The study was approved by the ethical board committee of Faculty of Medicine, Ain Shams University.

Statistical methods
Data was revised for its completeness and consistency. Double data entry on SPSS program version 16 was done. Quantitative data were summarized by mean along with standard deviation while qualitative data were summarized by frequencies and percentages. Analysis of variance, chisquare test and Spearman's correlation coefficient test were used for the analysis of significant differences or associations.

Results
Fifty renal biopsy specimens with lupus nephritis were reevalu ated including (14 cases of class II, 4 cases of class III, 20 cases of class IV and 12 cases of class V). Forty-two cases (84%) were females and eight (16%) were males, with mean age of 27.96 ± 12.9 years.
Regarding comparison of different variables based on ISN/ RPS class; no statistically significant difference in percentage of SCD-1 expression and the lupus classes was detected. We found a significantly higher mean age in class III (P = 0.011). Moreover, serum creatinine showed a significant difference between groups, with the highest value in class IV, followed by class III (P < 0.001). Accordingly, proteinuria showed a significant difference between groups, with the highest value in class V, and the lowest mean in class II (P < 0.001; Table 1).
Serum creatinine and albumin showed a statistically significantly different correlation with semiquantitative score of SCD-1 expression based on percentage of stained cells, with the highest value of creatinine which was detected with score 1 of SCD-1 expression (P = 0.038). Additionally, the highest value of urinary albumin was detected with score 1 of SCD-1 expression ( Figures  1A and 1B), while the lowest mean of urinary albumin recorded in SCD-1 score 3 (P < 0.001; Table 2; Figures  2A and 2B).
Regarding staining intensity, a statistically significant positive correlation between moderate intensity of SCD-1 expression and both serum creatinine (P = 0.043), and urinary albumin (P < 0.001) was found (Table 3; Figure 3).
Correlation between SCD-1 expression and NIH activity and chronicity indices was not significant. However a negative association between SCD-1 percentage and SCD-1 score and also intensity with increased NIH activity and chronicity indexes was noted, nonetheless, such association didn't mount to statistical significance (Table 4).

Discussion
SCD-1 is one of the four transmembrane HSPGs of the syndecan's family. All family members contain a transmembrane core protein to which glycosaminoglycan side chains are extracellularly attached (42). In adults, SCD-1 is mainly expressed by hepatocytes (43), epithelial cells (44) and plasma cells (45). The role of SCD-1 in re-epithelialization during wound healing and its role in restoration of TEC damage after ischemia reperfusion injury was established (18,23-25,41).
In the present work, evaluation of tissue SCD-1 immunohistochemical expression in renal TEC in lupus nephritis patients and its correlation with various clinicopathological parameters was our main concern.
We found a statistically significant inverse relationship between the loss of SCD-1 expression and increased proteinuria and serum creatinine. Our results were concordant with the study by Salam et al (32) who investigated the role of serum SCD-1 in lupus nephritis and revealed a significantly higher SCD-1 in patients with active lupus nephritis compared to those with extrarenal flare and inactive disease (P < 0.001 and P < 0.001 respectively). They found a highly statistically significant correlation between serum SCD-1 and 24h urinary proteinuria, decreased serum complement as well as SLEDAI (systemic lupus erythematosus disease activity  index) score indicating its potential role as a predictive marker for disease activity. Same results were detected by Kim et al (34) and Minowa et al too (46). Although there was no statistically significant correlation between loss of SCD-1 expression and increased NIH activity and chronicity indices in this study, however there was a weekly negative association between both parameters. These data may suggest that loss of SCD-1 expression might represent an early predictor of adverse clinical course and renal function deterioration before histopathological evidence manifestation.
In light of the previous data in the literature and the present study, we noted that both elevated serum SCD-1 and loss of tissue SCD-1 on TEC are related to progression of lupus nephritis. This inverse relationship between serum and tissue SCD-1 could be attributed to endothelial glycocalyx loss and/or shedding from the renal epithelium. During tissue injury, SCD-1 shedding is accelerated through the release of different proteases and growth factors, thus it can be detected in inflammatory fluids (26,44). Therefore, we recommend further studies investigating both serum and tissue SCD-1 expression  in TECs in patients with lupus nephritis to confirm this relationship and detect sensitivity of each procedure in prediction of disease activity. We also recommend investigating SCD-1 on desquamated/shed urinary TEC by flow cytometry in cases of lupus nephritis. Supporting this hypothesis; Celie et al (41) investigated the pattern of SCD-1 expression in renal TEC in murine ischemia reperfusion injury and human renal allograft transplantation and found increased epithelial SCD-1 in allografts was correlated with low-proteinuria and serum creatinine, less interstitial inflammation, less tubular atrophy, and prolonged allograft survival. They hypothesized that SCD-1 is involved in tubular repair and survival upon renal transplantation and proposed it as a novel marker of renal allograft survival.
Additionally, Pamfil et al (36) found that SCD-1 immunohistochemical expression was significantly higher in biopsies where no tubular damage was evidenced at histological evaluation in cases of lupus nephritis.
In our study, a direct relation between tubular SCD-1 expression and different ISN/ RPS classes could not  be detected. Classification of lupus nephritis classes varies according to site of immune complex deposition. Immune complexes deposit in the mesangium or the subendothelial and subepithelial spaces or in peritubular capillaries depending on the quality of the autoantibodies, the duration, and severity of lupus nephritis (47). This finding implies that immune complex formation in the mesangium causes class I and II lesions, subendothelial immune complex formation in class III and IV, and subepithelial immune complexes in class V (38). In the present work, we did not demonstrate SCD-1 expression in renal glomeruli in all classes, thus correlating a link between glomerular SCD-1 expression as a pathogenic factor determining the site of immune complex deposition couldn't be established. From the fore-mentioned results, the role of TEC-SCD-1 involvement in disease progression was highlighted while its role in disease initiation needs to be further investigated in larger studies.

Conclusion
In this study we revealed a clear statistically significant predictive role of TEC-SCD-1 immunohistochemical expression as an adverse predictive marker of disease progression and tubular damage in lupus nephritis patients regardless of their class. Accordingly, we propose SCD-1 immunohistochemical testing as a helpful adjunct tool to histopathological evaluation in improving insightful tailored management plans.

Limitations of the study
The missed relationship between loss of SCD1 expression and different ISN/RPS classes could be attributed to the limited number of cases. Hence further studies with larger sample size may reach a statistically significant value and prove the role of SCD1 in the process of disease initiation.