Urinary Neuropilin-1: A Predictive Biomarker for Renal Outcome in Lupus Nephritis

At present, Lupus Nephritis (LN) is still awaiting a biomarker to better monitor disease activity, guide clinical treatment, and predict a patient’s long-term outcome. In the last decade, novel biomarkers have been identified to monitor the disease, but none have been incorporated into clinical practice. The transmembrane receptor neuropilin-1 (NRP-1) is highly expressed by mesangial cells and its genetic deletion results in proteinuric disease and glomerulosclerosis. NRP-1 is increased in kidney biopsies of LN. In this work we were interested in determining whether urinary NRP-1 levels could be a biomarker of clinical response in LN. Our results show that patients with active LN have increased levels of urinary NRP-1. When patients were divided according to clinical response, responders displayed higher urinary and tissue NRP-1 levels at the time of renal biopsy. Areas under the receiver operating characteristic curve, comparing baseline creatinine, proteinuria, urinary NRP-1, and VEGFA protein levels, showed NRP-1 to be an independent predictor for clinical response. In addition, in vitro studies suggest that NRP-1could promote renal recovery through endothelial proliferation and migration, mesangial migration and local T cell cytotoxicity. Based on these results, NRP-1 may be used as an early prognostic biomarker in LN.


Patients
Renal involvement was documented by either a total urinary protein level ≥0.5 g/day, an increment of serum creatinine levels of more than 0.5 mg/dL, or the presence of active sediment by microscopic examination; more details in supporting information. Renal flares at inclusion were categorized as nephritic (≥30% increase in serum creatinine or decrease in estimated glomerular filtration rate (eGFR) by 10% with an active urine sediment, irrespective of changes in proteinuria) or proteinuric (reproducible doubling of the Upro-Ucre ratio >0.9 mg/mg after complete response or reproducible doubling of the Upro-Ucre ratio to >1.8 mg/mg after partial response) [37]. The renal samples were classified according to the histological types of LN using the ISN/RPS 2003 classification [38]. Renal biopsies were examined by light and immunofluorescence microscopy and categorised according to the International Society of Nephrology/Renal Pathology Society Classification [ISN/RPS] [37] and rated for activity (AI) and chronicity (CI) [36].
All patients were treated with IV methyl-prednisolone (500 mgx3), followed by a tapering dose of oral prednisone along with at least a 24-month course of oral mycophenolate mofetil [39]. In the prospective study, urine and blood samples were also obtained on a three-monthly basis for 1 year. All patients also received supportive treatment with angiotensin-converting enzyme inhibitor or angiotensin-receptor blocker and statin-lowering therapy. In the prospective study, patients were followed up every three months for one year. At each time point, urine and blood samples were obtained.

Study Design
First, urinary NRP-1 levels were measured in 45 patients with active LN prior to renal biopsy (exploratory cohort 1) and from control groups (active non-renal SLE patients (n = 25), patients with non-lupus glomerular diseases (n = 25) and healthy controls (25)) (initial screening stage). Results were also analyzed according to clinical outcome after treatment (responders (n = 22) vs. non-responders (n = 23)). Second, results were further confirmed in the validation cohort (cohort 2) that included 25 patients with active LN, of whom 16 achieved complete response during follow-up and 9 were non-responders (confirmation stage). Finally, urinary NRP-1 and VEGF levels were further measured three-monthly for at least 12 months in a new prospective cohort of patients with active LN (cohort 3) (n = 39). The study was approved by the Vall d'Hebron Ethic Committee and written informed consent was obtained from all patients.

Quantitative Reverse Transcription-PCR
For the quantification of mRNAs, 200 ng total RNA was reverse transcribed using the High Capacity RNA-to-cDNA Kit (Applied Biosystems) with the thermal cycler program: 25 °C for 5 min, 42 °C for 45 min and 82 °C for 5 min.
As controls, we used disease-free kidney sections from tissue margins of total or subtotal nephrectomies obtained from patients undergoing surgery for renal malignancies.

Isolation of Primary T Cell Isolation from Healthy Donors
First, we isolated Peripheral Blood Mononuclear Cells (PBMCs) from total human blood by Ficoll gradient. To obtain a final volume of 5 × 10 7 PBMC, we obtained 4 tubes of 8 mL of total blood. After centrifugation (3000 rpm during 25 min at RT) we recovered the PBMCs. Second, we isolated Tcells by Dynabeads ® Untouched TM Human T cells protocol (Invitrogen) following the manufacturer's protocol. This protocol is intended for isolation of untouched human T cells from PBMC by depleting B cells, NK cells, monocytes, platelets, dendritic cells, granulocytes, and erythrocytes. Finally, T cells were characterized using APC Mouse Anti-Human CD4 (BD Pharmingen TM ), Anti-Human CD3 PE (Diaclone) and APC Mouse Anti-Human CD8 (BD Pharmingen TM ) by flow cytometry analyzer BD LSRFortessa™. The final step was the separation of CD3/CD4 and CD3/CD8 T cells by BD FACSAria II flow cytometer Cell Sorter (Biosciences).

Wound Healing Assay
To assess the potential angiogenesis of endothelial renal cells in different conditions (control and inhbited NRP-1) and stimulations, wound healing assays were performed. We used scratch assay in all the conditions. Firstly, 1 × 10 5 cells were plated in 24-well plate to obtain high confluence (70-80%). After that, the cell monolayer was scraped in a straight line to create a "scratch" with a p200 pipet tip. The derbis was removed and the edge of the scratch were smoothed by washing the cells with 1 mL of medium. To obtain the same field during the image acquisition, it is important to create markings to be used as references points close to the scratch. The cells were placed at 37 °C into the incubator and images were acquired at 0, 2, 4, 8, 12, and 24 h. To calculate migration, bright field microscopic images were analyzed using ImageJ software. The percentage of confluence index (CI) between the scratch was determined by the following equation: When the distance between wound edges starts to be minor, the %CI between them will be higher. If there is not migration or wound healing in the scratch, the %CI will be around to zero value.
At least, three independent scratch experiments were performed in each condition.

Immunofluorescence Assay
For Immunofluorescence, cells were plated on glasses into 24-well plates. They were fixed using PFA 4% solution during 20 min at room temperature. After washing them with PBS, they were treated with 0.1% triton during 10 min. To block them, a solution of PBS 5% BSA was used during 1 hour at room temperature. After that, primary antibody rabbit anti-VEGFR2 (1:200, Santa Cruz Biotechnology) and mouse anti-NRP1 (1:200, Santa Cruz Biotechnology) were incubated overnight at 4 °C. As secondary antibodies, Alexa 488 donkey anti-rabbit (Abcam, ab150061) and Alexa 647 goat anti-mouse (Abcam, ab150119) were incubated for 2.5 h at room temperature (dilution 1:100). Fluoromount-G with DAPI (Invitrogen, Thermo Fisher Scientific, Carlsbad, CA) and Olympus BX61 motorized upright microscope with fluorescence were used to visualized the staining cells.

Evaluation of Immunohistochemistry
Results were evaluated on blinded specimens by the Vall d'Hebrón pathologist unit under the supervision of the nephropathologist (Dr. Marta Vidal). The percentage of cells expressing the different probes was scored semiquantitatively as follows: 0 (no expression), 1 (11-20%), 2 (40-60%), or 3 (>80%). Staining intensity was scored semiquantitatively as 0 (no staining), 1 (weakly positive), 2 (moderately positive), or 3 (strongly positive). This scores were obtained after the evaluation of 5 pathologists. After that, the mean of them has been expressed in the figures as "Average score". Figure S1. Protein level of sNRP-1 in the different cohorts.    Table   Table S1. Clinical and histological variables at the time of the renal biopsy of the two cohorts and their combination. Values are means ± SE. BUN, blood urea nitrogen; eGFR, estimated glomerular filtration rate; anti-dsDNA, anti-double-stranded DNA; n.a., not applicable. a P-value refers to the comparison of the LN with healthy controls, the values in bracket referred to the comparison with active non-renal SLE and the values in square bracket referred to the comparison with other glomerular-disease cohort: Mann-Whitney U Test or Pearson χ2 test.