ORIGINAL RESEARCH published: 09 June 2022 doi: 10.3389/fcvm.2022.829490 Vitamin D Binding Protein and Renal Injury in Acute Decompensated Heart Failure Elisa Diaz-Riera 1,2 , Maisa García-Arguinzonis 1 , Laura López 3 , Xavier Garcia-Moll 3,4 , Lina Badimon 1,4,5 and Teresa Padró 1,4* 1 Cardiovascular-Program ICCC, Research Institute–Hospital Santa Creu i Sant Pau, IIB-Sant Pau, Barcelona, Spain, 2 Faculty of Medicine, University of Barcelona (UB), Barcelona, Spain, 3 Cardiology Department, Hospital Santa Creu i Sant Pau, Barcelona, Spain, 4 Centro de Investigación Biomédica en Red Cardiovascular (CIBERCV), Instituto de Salud Carlos III, Madrid, Spain, 5 Cardiovascular Research Chair, Universitat Autónoma de Barcelona (UAB), Barcelona, Spain Background: Renal function in acute decompensated heart faiulre (ADHF) is a strong predictor of disease evolution and poor outcome. Current biomarkers for early diagnostic of renal injury in the setting of ADHF are still controversial, and their association to early pathological changes needs to be established. By applying a proteomic approach, we aimed to identify early changes in the differential urine protein signature associated with development of renal injury in patients hospitalised due to ADHF. Edited by: Ernesto Martinez-Martinez, Materials and Methods: Patients (71 [64–77] years old) admitted at the emergency Universidad Complutense de Madrid, room with ADHF and hospitalised were investigated (N = 64). Samples (urine/serum) Spain were collected at hospital admission (day 0) and 72 h later (day 3). Differential Reviewed by: Silvia Spoto, serum proteome was analysed by two-dimensional electrophoresis and matrix-assisted Policlinico Universitario Campus laser desorption/ionisation-time of flight (MALDI-ToF/ToF). Validation studies were Bio-Medico, Italy Sonia Eiras, performed using ELISA. Health Research Institute of Santiago Results: Proteomic analysis depicted urinary vitamin D binding protein (uVDBP) as de Compostela (IDIS), Spain a two spots protein with increased intensity in ADHF and significant differences *Correspondence: Teresa Padró depending on the glomerular filtration rate (GFR). Urinary VDBP in patients with ADHF tpadro@santpau.cat at hospitalisation was > threefold higher than in healthy subjects, with the highest levels Specialty section: in those patients with ADHF already presenting renal dysfunction. At day 3, urine VDBP This article was submitted to levels in patients maintaining normal renal function dropped to normal values (P = 0.03 Heart Failure and Transplantation, vs. day 0). In contrast, urine VDBP levels remained elevated in the group developing renal a section of the journal Frontiers in Cardiovascular Medicine injury, with values twofold above the normal range (P < 0.05), while serum creatinine Received: 05 December 2021 and GF levels were within the physiological range in this group. Urinary VDBP in ADHF Accepted: 27 April 2022 positively correlated with markers of renal injury such as cystatin C and Kidney Injury Published: 09 June 2022 Molecule 1 (KIM-1). By ROC analysis, urinary VDBP, when added to cystatin C and Citation: Diaz-Riera E, KIM-1, improved the prediction of renal injury in patients with ADHF. García-Arguinzonis M, López L, Conclusion: We showed increased urine VDBP in patients with ADHF at hospital Garcia-Moll X, Badimon L and Padró T (2022) Vitamin D Binding admission and a differential uVDBP evolution pattern at early stage of renal dysfunction, Protein and Renal Injury in Acute before pathological worsening of GFR is evidenced. Decompensated Heart Failure. Front. Cardiovasc. Med. 9:829490. Keywords: renal dysfunction, vitamin D binding protein (GC), urine sample analysis, heart failure, proteomics, two doi: 10.3389/fcvm.2022.829490 dimension electrophoresis–MS/MS Frontiers in Cardiovascular Medicine | www.frontiersin.org 1 June 2022 | Volume 9 | Article 829490 Diaz-Riera et al. VDBP and Acute Heart Failure INTRODUCTION (25[OH]D3) and actively reabsorbed in the proximal tubule cells (22, 23). Renal insufficiency is present as comorbidity in nearly a Anomalous urinary loss of VDBP, due to impaired proximal quarter of patients hospitalised due to acute decompensated tubular reabsorption, has been reported in the setting of heart failure (ADHF) (1, 2), and the evolution of renal renal damage in preclinical studies and metabolic disease- function has prognostic relevance in these patients (2). Indeed, related nephropathies in humans (24, 25). In this hypothesis- ADHF is frequently complicated by acute kidney injury generating study, we analysed urinary VDBP levels in association during hospitalisation (2, 3), which is associated with adverse with renal function in patients with ADHF during the outcomes and longer hospital stays (2, 4, 5). In this respect, a first 72 h hospitalisation after an acute event to investigate recent meta-analysis indicated that all-cause mortality increased the value of VDBP to provide accurate pathophysiologic significantly in patients with ADHF compared with those information on kidney function deterioration in ADHF before without acute kidney injury (5). The complex relationship creatinine levels rise. between heart failure and renal dysfunction may result in a combination of hormonal, hemodynamic, and inflammatory factors (6) and aggressive decongestive treatment (7). In MATERIALS AND METHODS turn, kidneys are responsible for sodium homeostasis playing a key role to control intravascular and interstitial fluid Study Population and Study Design accumulation (congestion) in heart failure (8). Until now, This study refers to patients with ADHF that were hospitalised however, the mechanisms participating in the cardiorenal in Hospital de la Santa Creu i Sant Pau (HSCSP), in axis pathology need to be better understood, and the early Barcelona, between February 2017 and March 2020. Patients differential protein pattern associated to these processes remains hospitalised with ADHF were distributed in two different to be identified. groups depending on their kidney function at admission. Renal Deterioration of kidney function is frequently recognised by function was shown as MDRD-4 (ml/min/1.73 m2 ), a serum the rise of serum creatinine levels or by the decline of glomerular creatinine-based estimation obtained using clinical data of the filtration rate (GFR). However, creatinine levels can also be patients. Levels below 60 ml/min/1.73 m2 were considered influenced by several factors such as advanced age, sex, excessive pathological. Two groups were formed: I) patients with ADHF protein intake, muscle mass, medication, or intense exercise with renal dysfunction (RD) at hospital admission and II) (9). Nowadays, early changes in other molecular components patients with ADHF with normal renal function (NRF) at have been suggested including kidney injury molecule 1 (KIM- hospital admission. From this latter NRF group, two further 1), Cystatin C (CysC), or neutrophil gelatinase-associated subgroups were made based on the evolution of the renal lipocalin (NGAL) (10–13), but their usefulness for detecting early function during hospitalisation. Patients who developed renal pathological changes is controversial, and their value has not function deterioration to pathological levels (RI-Group; MDRD- been consistently proven (14, 15). Therefore, identification of 4 < 60 ml/min/1.73 m2 ) and those who maintained normal renal molecular components that accurately facilitate early detection function (noRI). A group of healthy subjects (HS, N = 13, 50 [37– of renal injury and may contribute to a better stratification of 63] years) was used to establish the normal range in urine of the patients with ADHF and enhance their management remains a identified proteins (Figure 1). medical challenge. The Ethics Committee of the Santa Creu i Sant Pau Over the last decade, urinary proteome has become a potential Hospital in Barcelona, Spain, approved this study, and it was source of novel disease molecular patterns and urinary proteomic performed according to principles of Helsinki’s Declaration. All analysis in heart failure has been used to identify the differential patients signed an informed consent prior being included in the expression profile associated to earlier diagnosis of the disease study. Patients under chemotherapy, pregnancy, or postdelivery (16–18) and to establish diagnostic classifiers in patients with ischemic heart syndrome in women and other causes of acute heart failure with specific disease conditions (19). episode (myocardial infarction, myocarditis, or toxic aetiology) In this study, we applied a mass spectrometry-based protein were excluded from the study. Medication was not considered as discovery approach (20, 21) aimed to identify urinary proteins exclusion criteria except those required in oncological treatment associated with renal dysfunction in hospitalised patients with (who were already excluded). ADHF. By proteomics, we have identified significant changes in vitamin D binding protein (VDBP), a member of the albumin Biological Samples superfamily of binding proteins that is produced in the liver, Urine and blood samples were collected at hospital admission filtered through the glomerulus as a complex with vitamin D3 and at day 3 of hospitalisation, when patients no longer showed signs of dehydration due to the acute event. Samples at hospital Abbreviations: ADHF, acute decompensated heart failure; CysC, cystatin C; GFR, admission were collected as soon as possible after hospitalisation. glomerular filtration rate; KIM-1, kidney injury molecule 1; LVEF, left ventricular Urine obtained at 72 h refers to the first urine in the morning ejection fraction; NGAL, neutrophil gelatinase-associated lipocalin; RD, renal and blood extraction. Blood extraction at day 3 was performed dysfunction patients art hospitalisation; noRI, patients who did not develop renal at the same time of urine collection. All samples were processed injury; NRF, normal renal function patients at hospitalisation; RI, patients who developed renal injury; VDBP, vitamin D binding protein; 2DE, two-dimension identically and within 30 min after obtention. Protease inhibitors electrophoresis. (cOmplete Mini, Roche, Basel, Switzerland) were added to urine Frontiers in Cardiovascular Medicine | www.frontiersin.org 2 June 2022 | Volume 9 | Article 829490 Diaz-Riera et al. VDBP and Acute Heart Failure samples at collection. Urine samples were centrifuged (1,200 g, for differences between gels was performed using the PDQuest 10 min, room temperature) to precipitate debris, aliquoted, and analysis software (Bio-Rad, San Diego, CA, United States). Each stored at –80◦ C until further analysis. The level of total protein spot was assigned a relative value (AU) that corresponds to the in urine was analysed with a Clima MC-15 analyser using the single spot volume compared with the volume of all spots in the specific Gernon kit (RAL S.A., Barcelona, Spain), as described gel, following background extraction and normalisation between by the providers. gels, as previously reported (28). This software specifically Serum was obtained after maintaining blood samples for analyses differences in protein patterns, in which a master gel is 30 min at 37◦ C followed by 30 min at 4◦ C. Thereafter, samples created wherein all gels are included and used to compare with were centrifuged at 1,800 g (30 min, 4◦ C), aliquoted, and stored each individual sample. at –80◦ C until analysed. Proteins were identified after in-gel tryptic digestion and Blood creatinine (Jaffe reaction), NT-proBNP extraction of peptides from the gel pieces by matrix-assisted laser (electroquimioluminiscence), urea (kinetic urease), and desorption/ionisation time-of-flight (MALDI-TOF) using an haemoglobin were analysed by standard laboratory methods as AutoFlex III Smart beam MALDI-TOF/TOF (BrukerDaltonics, part of the patients’ routine analyses. Glomerular filtrate was Billerica, MA, United States), as previously described (28). calculated using the MDRD-4 algorithm that includes patient’s Samples and calibrants were mixed 1:1 with alpha-cyano-4- plasma creatinine levels, age, sex, and race (26). Total protein hydroxycinnamic acid (HCCA) matrix (0.7 mg/ml) and were and creatinine in urine were analysed in a Clima MC-15 analyser applied to Anchor Chip plates (BrukerDaltonics, Billerica, using Gernon kits for creatinine and total protein (RAL S.A., MA, United States). Barcelona, Spain). Spectra were acquired with flexControl on the reflectron mode Urine samples for the HS reference group were collected in (mass range m/z 850–4,000; reflectron 1, 21.06 kV; reflectron the morning, and samples were processed as described for the 2, 9.77 kV; ion source 1 voltage, 19 kV; ion source 2, 16.5 kV; patients with ADHF. detection gain, 2.37×) with an average of 3,500 added shots at a frequency of 200 Hz. Samples processed with flexAnalysis (version 3.0, Bruker Daltonics, Billerica, MA, United States) Two-Dimensional Electrophoresis and considering a signal-to-noise ratio > 3, applying statistical Mass Spectrometry calibration, and eliminating background peaks. After processing, The analysis of differential protein patterns was performed by spectra were sent to the interface BioTools (version 3.2, Bruker a proteomic approach using two-dimensional electrophoresis Daltonics, Billerica, MA, United States), and MASCOT search (2DE) followed by mass-spectrometry for protein identification, on Swiss-Prot 57.15 database was carried out [taxonomy, as previously described (27). Homo sapiens; mass tolerance, 50–100; up to two trypsin Urine samples (4 ml) of patients with ADHF and healthy miss cleavages; global modification: carbamidomethyl (C); and controls were concentrated and desalted by centrifugation variable modification: oxidation (M)]. Identification was carried (3,220 g, 30 min, and 10◦ C) using 3 kDa cut-off filter devices out by peptide mass fingerprinting (PMF) where a MASCOT (Amicon Ultra-4, Millipore, Burlington, MA, United States) and scores > 56 and at least five matched peptides was accepted. 100 mM Tris-HCl, pH 7.6. A final volume of 1 ml was obtained Confirmation of identified protein was performed by peptide and depleted of albumin and IgGs using the ProteoExtract fragmentation working on the LIFT mode (MS/MS) (27, 28). Albumin/IgG Removal Kit (Calbiochem, San Diego, CA, United States), as reported by the providers. Thereafter, sample Western Blot buffer was exchanged to a urea-containing buffer (7 M urea, 2 M Urinary protein extracts were resolved by 1-DE under thiourea, 2% CHAPS), by centrifugation with the 3 kDa cut-off reducing conditions and electrotransferred to nitrocellulose filter devices (3,220 g, at room temperature) until a final volume membranes. Urinary VDBP detection was performed using of 400 µl was obtained. Protein concentration in urine extracts a rabbit monoclonal antibody (ab81307, 1:5,000 in 5% was measured with 2D-Quant Kit (GE Healthcare, Chicago, IL, NFDM-TBST, Abcam, Cambridge, United Kingdom). Band United States), and 100 µg total protein aliquots were prepared. detection was performed using chemiluminescent SuperSignal A 100 µg aliquot was used to perform the individual 2DE (FischerScientific, Waltham, MA, United States) and a molecular analytical gels and 300 µg for the preparative gels (pool from imager ChemiDoc XRS System, Universal Hood II (BioRad, San five representative samples), which refer to, approximately, 1 Diego, CA, United States). and 3 ml, respectively, of original urine samples. Each aliquot of the urea/thiourea/chaps urine extracts was applied to 17 cm Immunoassays dry strips (ReadyStrips IPG strips, pH 4–7 linear range; BioRad, Vitamin D binding protein was analysed in urine and San Diego, CA, United States), using the PROTEAN i12 IEF serum samples using Human Vitamin D BP Quantikine system (Bio-Rad, San Diego, CA, United States) for the first ELISA kit (DVDBP0B, R&D Systems, Minneapolis, MN, dimension, as previously described in our group (28, 29). The United States) with a sensitivity of 0.338 ng/ml and with second dimension was resolved in 12% SDS-PAGE. Gels were intraassay variability < 2.5% and variability <7% in interassay fixed for 2 h (40% ethanol and 10% acetic acid), developed with trials. Urinary Cystatin C (uCysC) levels were analysed using Flamingo (Bio-Rad, San Diego, CA, United States) for protein Human Cystatin C ELISA kit from Biovendor (RD191009100, fluorescent staining. Protein spot quantification and analysis Biovendor, Brno, Czech Republic), with a sensitivity of Frontiers in Cardiovascular Medicine | www.frontiersin.org 3 June 2022 | Volume 9 | Article 829490 Diaz-Riera et al. VDBP and Acute Heart Failure FIGURE 1 | Study design. Patients with ADHF were divided in two groups at hospital admission depending on their glomerular filtration rate. RD group included patients with glomerular filtration rate <60 ml/min/1.73 m2 , while NRF presented glomerular filtration rate above this cutoff. At day 3, patients with NRF were divided in two groups depending on their kidney function evolution. RI patients developed renal function deterioration defined by glomerular filtration rate <60 ml/min/1.73 m2 , while noRI patients maintained normal glomerular filtration rate. Urine and blood samples were obtained at hospital admission and at day 3 of hospitalisation. Protein extracts from urine were prepared, and samples were analysed by two-dimensional (2DE)-electrophoresis and mass spectrometry for identification of differential protein patterns. Validation studies were performed by ELISA. Sample size is given in Figure. 0.25 ng/ml, intraassay variability < 4%, and with interassay to check sample size in the validation quantitative analysis (i.e., variability < 11%. Urinary KIM-1 levels were analysed using ELISA method). Minimal required sample size was calculated Human KIM1 ELISA Kit (ab235081, Abcam, Cambridge, and validated using the JavaScript-based method for simple United Kingdom) with a sensitivity of 1.28 pg/ml and with power and sample size calculation when two independent groups intraassay variability <3% and interassay variability <7%. are compared, provided in http://www.stat.ubc.ca/~rollin/stats/ Each immunoassay was performed according to the providers’ ssize/n2.html (28). Specifically, based on the median value of protocol. Concentrations of the proteins in urine were expressed the ADHF group (at hospitalization) and the variability range per total protein content in the same sample to correct for of the total population, a N = 64 in the patient group results difference in total protein concentration. in a study power >0.75 (type I error = 0.05, two-sided test). Receiver operating characteristic (ROC) curve analyses were used Statistical Analysis to calculate the area under the curve for each variable along with its 95% CI and to determine the power to discriminate renal Data were expressed as median and interquartile range (IQR). function deterioration. The statistical analysis was performed N indicates the number of subjects tested. The normal using Stata v15 and SPSS v26. A P-value ≤ 0.05 was considered distribution was determined via the Kolmogorov-Smirnov test. statistically significant. Statistic differences between groups for non-normally distributed continuous variables were analysed by non-parametric tests, including Mann-Whitney or Kruskal-Wallis tests. Frequencies RESULTS of categorical variables were compared by Chi-square analyses. Correlations between variables were determined using single regression models and Spearman rank correlation. The Wilcoxon Clinical Characteristics of the Acute Signed Rank test was used to compare each patient evolution. Decompensated Heart Failure Patient Due to the exploratory character of this proteomic study, Population determination of the sample size was based on past experience Baseline demographic, clinical characteristics, and background with similar studies (28). Data of patients with ADHF at medication of the studied population are given in Table 1. hospital admission obtained from the proteomic studies served Briefly, this refers to 64 patients with ADHF with median age Frontiers in Cardiovascular Medicine | www.frontiersin.org 4 June 2022 | Volume 9 | Article 829490 Diaz-Riera et al. VDBP and Acute Heart Failure TABLE 1 | Patient baseline characteristics. All patients N = 64 NRF N = 34 RD N = 30 P-value Demographic characteristics; median [IQR] Female/male, N 20/44 9/25 11/19 0.146 Age, years 71.0 [64.0–77.0] 59.0 [58.0–75.0] 74.5 [69.0–78.0] 0.010 Weight, kg 72.6 [61.2–86.6] 70.0 [59.8–86.8] 75.8 [62.0–84.0] 0.591 Kidney function markers Creatinine, µmol/L 104.5 [78.0–146.3] 78.5 [67.8–96.8] 147.0 [120.8–195.5] <0.001 Glomerular filtration (MDRD-4) 61.3 [41.8–83.1] 82.3 [69.0–98.2] 41.1 [31.0–45.3] <0.001 Urea, mmol/L 10.1 [7.0–15.6] 7.1 [5.8–9.2] 16.5 [12.9–23.0] <0.001 Cardiac function markers NT-proBNP, µg/L 4.14 [2.5–8.9] 3.4 [1.9–6.4] 5.2 [3.0–14.7] 0.022 Left ventricular ejection fraction (LVEF),% 45.5 [33.0–57.5] 37.5 [33.0–57.0] 51.0 [33.0–58.0] 0.297 Preserved LVEF, N (%) 27 (42) 11 (32) 16 (53) Reduced LVEF, N (%) 27 (42) 18 (53) 9 (30) 0.158 Mid-range LVEF, N (%) 10 (16) 5 (15) 5 (17) Atrial fibrillation, N (%) 31 (48) 16 (47) 15 (50) >0.999 Cardiovascular disease, N (%) 20 (31) 10 (29) 10 (33) 0.791 Other biochemical markers Haemoglobin, g/L 122 [104–138] 129 [114–141] 113 [96–125] 0.008 Risk factors; N (%) Active smoking 10 (16) 8 (24) 2 (7) 0.088 Hypertension 47 (73) 21 (62) 26 (87) 0.046 Pulmonary hypertension 16 (25) 8 (24) 8 (27) 0.781 Diabetes mellitus type 2 28 (44) 11 (32) 17 (57) 0.079 Dyslipidaemia 44 (69) 20 (59) 24 (80) 0.108 Background medication; N (%) Diuretics 47 (73) 21 (59) 27 (90) 0.011 Statins 41 (64) 18 (53) 23 (77) 0.068 Anticoagulants 26 (41) 14 (41) 12 (40) >0.999 Antiplatelet agents 36 (56) 19 (56) 17 (57) >0.999 Beta-blockers 44 (69) 21 (62) 23 (77) 0.281 Antiarrhythmic agents 6 (9) 4 (12) 2 (7) 0.676 Antidiabetics 24 (38) 8 (21) 17 (57) 0.010 Insulin 11 (17) 2 (6) 9 (30) 0.018 Oral antidiabetic agents 21 (33) 8 (24) 13 (43) 0.114 ACE inhibitor/ARB 43 (67) 26 (76) 17 (57) 0.114 P-values compare NRF and RD patients, and it is calculated using the Fisher exact test, except for LVEF where χ2 was used. Diuretics: hydrochlorothiazide, furosemide, eplerenone, and spironolactone. Statins: atorvastatin, pravastatin, simvastatin, ezetimibe. Anticoagulants: warfarin, acenocoumarol, bemiparin, heparin, dabigatran, rivaroxaban, edoxaban, and apixaban. Antiplatelet agents: acetylsalicylic acid and clopidogrel. Beta-blockers: bisoprolol and carvedilol. Antiarrhythmic agents: amiodarone. Oral antidiabetic agents: metformin and repaglinide. Angiotensin-converting enzyme inhibitors (ACEIs) include: captopril, enalapril, and ramipril. Angiotensin receptor blockers (ARBs): losartan, olmesartan, and valsartan. Bold values emphasize a statistically significant result (p < 0.05). of 71[64–77] years and with 68% men, who were hospitalised creatinine, after day 5 of hospitalisation (Group RI; see at Hospital de la Santa Creu i Sant Pau (HSCSP), Barcelona. Figure 1). At the time of hospitalisation, 47% of patients with ADHF All patients with ADHF had plasma NT-proBNP values above presented pathological GFR levels (RD group; MDRD-4: the pathological cut-off (1.8 µg/L), being levels in the RD group 41.1 [30.7–45.3] ml/min/1.73 m2 , P < 0.001) while the other significantly higher than those in the NRF group (5.2 [3.0– 53% of ADHF (NRF group) presented GFR levels within the 14.7] µg/L vs. 3.4 [1.9–6.4] µg/L, P = 0.022). Forty-two percent of normal range with MDRD-4 > 60 ml/min/1.73 m2 (82.3 [69.0– patients with ADHF had reduced left ventricular ejection fraction 98.3] ml/min/1.73 m2 ). Notably, 38% of patients with ADHF (LVEF; 32 [21–35]%), and 42% had an LVEF above 50% (60 within the NRF group (normal kidney function at hospital [56–61]%) at admission. The remaining group of patients (16%) admission) showed renal function deterioration with GFR presented mid-range LVEF, with values between 40 and 49% values below the pathological cut-off (60 ml/min/1.73 m2 ), (46 [44–47]%). Mean hospitalisation time was 10.8 ± 5.3 days which referred to a mean increase of 45% of serum for all patients; however, a longer trend was observed in the Frontiers in Cardiovascular Medicine | www.frontiersin.org 5 June 2022 | Volume 9 | Article 829490 Diaz-Riera et al. VDBP and Acute Heart Failure subgroup of patients developing RI compared with those who range (12.6 [11.2–22.0] ng/mg total protein, P < 0.01) when maintained normal renal function (12 [9–17] days; 9 [7–11] days, quantified by specific ELISA (Figure 3A). More specifically, respectively, P = 0.075). the RD and NRF groups had median uVDBP levels (85.5 Regarding medication, 73% of our study group were already [17.7–171.7] ng/mg total protein and 38.8 [10.2–76.0] ng/mg under diuretic treatment before hospital admission. Upon total protein, respectively) sixfold (P = 0.001) and threefold hospitalisation due to ADHF, patients are given intravenous (P = 0.072), respectively, above the median value of the defined furosemide (dosage depending on their clinical condition, healthy range. In addition, patients with ADHF with pathological prior diuretic dosage and renal function) until stabilisation GFR levels at hospital admission (RD group) had higher VDBP for decongestion. Later, patients were administered furosemide loss in urine than those in the NRF group (P = 0.012) (Figure 3A). orally and complemented with other diuretics if needed, and the Urinary levels of VDBP (uVDBP) in patients with ADHF dose was dependent on the patients’ condition and response to at hospital admission (day 0) significantly correlated with NT- avoid increases in serum creatinine. proBNP (Rho = 0.337, P = 0.019) and GFR (Rho = –0.379, A representative subgroup of 17 patients with ADHF (76% P = 0.005) in patients with ADHF. In contrast, uVDBP did men, 72 [69–76] years old), with renal dysfunction (RD, N = 5), not associate with LVEF (Rho = 0.013, P = 0.924), and no and with normal renal function (NRF, N = 12) at hospitalisation differences were found in uVDBP levels between patients with was used for proteomic studies in an initial discovery phase. reduced LVEF (<40%) and preserved LVEF (>50%, P = 0.885). Demographic and clinical characteristics of this subgroup are Moreover, the uVDBP levels did not differ significantly between given in Supplementary Table 1. patients with and without atrial fibrillation (P = 0.337) or previous cardiovascular disease (P = 0.437). To notice, levels Urine Differential Proteomic Profile in of uVDBP were not affected by age (Rho = 0.106, P = 0.437) Patients With Acute Decompensated or sex (P = 0.563) nor by common comorbidities such as Heart Failure hypertension (75%, P = 0.0851) and dyslipidaemia (70%, The urinary proteome of patients with ADHF at hospital P = 0.212, Supplementary Table 3). admission was compared with that of a healthy control group Vitamin D binding protein has been previously studied within in the discovery phase. A total of 23 differential proteins were diabetic nephropathy. The 44% of the patients in the ADHF detected, with varying functions (cell signalling, cell metabolism, study population had diabetes, with a trend towards higher coagulation, acute phase response), as shown in Supplementary frequency of diabetes in the RD compared with that in the NRF Figure 1. Three differential proteins were associated to vitamin group (P = 0.079; Table 1). However, uVDBP levels did not metabolism and identified as transthyretin (TTR, Swiss-Prot significantly differ between diabetic and non-diabetic patients P02766, Mascot-score 61), retinol binding protein 4 (RBP4, (P = 0.635), neither in the NRF group (P = 0.572) nor in the Swiss-Prot P02753, Mascot-score 66), and VDBP (Swiss-Prot RD group (P = 0.594, see Supplementary Table 3). Furthermore, P21614, Mascot-score 72) (Supplementary Table 2). as shown in Supplementary Figure 4, patients with diabetes Vitamin D binding protein in urine was identified by 2DE- were homogenously distributed when the population was divided MS/MS as two spots with pI within 5.25–5.30 and a molecular in VDBP tertiles. A similar pattern was observed when the weight of 50.3 KDa (Figure 2A and Supplementary Figure 2). groups with normal and reduced renal function at hospital By Western blot analysis, VDBP was found as a single band in admission were separately considered (NRF group vs. RD group; urine samples of patients with ADHF (Figure 2B). Compared χ2 : P = 0.789 for differences in the pattern distribution of patients with healthy subjects, patients with ADHF at hospital admission with diabetes). Moreover, as summarised in Supplementary showed a twofold higher urinary levels (ADHF, N = 17, 1.71 Table 4, urinary VDBP levels did not depend on background [0.87–2.55] AU vs. 0.77 [0.66–0.98] AU of HS, N = 6, P = 0.042, medication, including statins, anticoagulants, and antiplatelets, Figure 2C). Within patients with ADHF at hospitalisation, the beta blockers, and ACE inhibitors (P > 0.05 between treated and RD group (N = 5, 2.98 [1.71–4.66] AU) presented the highest non-treated patients for all medications). values; these being twofold higher than those of the NRF group (N = 12, 1.58 [0.76–2.12] AU, P = 0.058) (Figure 2D). Vitamin D Binding Protein Levels at Day While no differences were observed between patients depending on presence of atrial fibrillation (P = 0.228), prior 3 of Hospitalisation cardiovascular disease (P = 0.428), hypertension (P = 0.113), In total, 38.2% of patients in the NRF group presented a dyslipidaemia (P = 0.833), or diabetes mellitus (DM2, P = 0.228). worsening of renal function to pathological values between day 5 hospitalisation and hospital discharge (RI group; N = 13), Increased Levels of Urinary Vitamin D whereas renal function was not significantly affected in the remaining 61.8% of the patients in the NRF group (noRI group; Binding Protein in Patients With Acute N = 21), with GFR > 60 ml/min/1.73 m2 (see Figure 1). Decompensated Heart Failure at At day 3, as shown in Figure 3B, patients who did not Hospital Admission develop renal injury (noRI group) during the hospitalisation In validation studies (N = 64), patients with ADHF presented period showed a significant drop of uVDBP to values within significantly higher VDBP levels in urine (48.2 [13.0– the HS range (noRI group: 11.1 [5.7–21.2] ng/mg total protein 100.8] ng/mg total protein) than those in the defined healthy at day 3; P = 0.030 in comparison with values at day 0). In Frontiers in Cardiovascular Medicine | www.frontiersin.org 6 June 2022 | Volume 9 | Article 829490 Diaz-Riera et al. VDBP and Acute Heart Failure FIGURE 2 | Proteomic analysis of urine samples. (A) Representative 2-DE image of human urine samples in a pI range of 4–7 and 12% SDS-PAGE gels. Vitamin D binding protein (VDBP) was identified as 2 spots (pI: 5.25–5.30; Mw: 50.3 KDa) in patients with ADHF (N = 17) and healthy subjects (N = 6). (B) Western blot analysis showed one single band for VDBP in urine samples. (C,D) Box plot diagrams [median (IQR)] of the intensity of both VDBP spots at hospital admission. In (C), comparison between patients with ADHF and heathy subjects (HS). (D) Differences in VDBP intensity between patients with ADHF with (N = 12) and without (N = 5) renal dysfunction (RD and NRF for GFR values below and above 60 ml/min/1.73 m2 , respectively). The Mann-Whitney test was used for comparison between groups in C and D. Statistical significance for P < 0.05. contrast, uVDBP levels remained high and did not significantly between urine and serum VDBP levels in patients with ADHF differ from values at admission (P = 0.530) in those patients who (Rho = 0.020; P = 0.879; Figure 4B). In addition, at day 3, no were going to develop renal injury (RI group) before hospital correlation was shown between GFR and serum VDBP levels discharge. It is to notice, however, that patients in the RI group (Rho = 0.060, P = 0.657) in ADHF, whereas GFR significantly still presented GFR levels within the normal physiological range correlated with VDBP levels in urine (Rho = –0.472, P = 0.001) (68.1 [61.5–77.8] ml/min/1.73 m2 ) at day 3 of hospitalisation. (Supplementary Figure 3). Patients in the RD group (GFR: 42.3 [31.6–52.4] ml/min/1.73 m2 ) maintained elevated levels of uVDBP (85.7 Urinary Vitamin D Binding Protein Levels [22.0–173.9] ng/mg total protein) at day 3 with a median Added to Cystatin C and KIM-1 Improved value > sixfold higher than in the noRI group (P = 0.007; Discrimination of Patients With Acute Figure 3B). The analysis of sera samples obtained at day 3 of Decompensated Heart Failure With hospitalisation did not show significant differences in VDBP Renal Deterioration During levels between noRI and RI groups (Figure 4A; P = 0.603). Hospitalisation Additionally, RD group did not present different levels Cystatin C and KIM-1 have been proposed to change during with noRI patients (P = 0.660, Figure 4A) nor with RI early renal injury (10, 11, 30). Cystatin C in urine of patients patients (P = 0.874, Figure 4A). No correlation was found with ADHF negatively correlated with GFR at admission Frontiers in Cardiovascular Medicine | www.frontiersin.org 7 June 2022 | Volume 9 | Article 829490 Diaz-Riera et al. VDBP and Acute Heart Failure FIGURE 3 | Urinary VDBP in patients with ADHF during hospitalisation: validation studies. Box plot diagrams [median (IQR)] refer to levels of urinary VDBP quantified by ELISA. VDBP in urine was normalised by the total protein content in each sample and expressed as ng VDBP/mg total protein. The Mann-Whitney test was performed for comparison between groups at: (A) hospital admission, (I) patients with ADHF (N = 64) vs. healthy subjects (HS, N = 13), (II) patients with ADHF with glomerular filtration rate (GFR) below (N = 34) and above 60 ml/min/1.73 m2 at hospital admission (N = 30); (B) 3 days of hospitalisation, (I) comparison between day 0 (D0) and day 3 (D3) for patients with ADHF maintaining normal renal function (noRI: N = 21) and presenting with renal function deterioration (RI: N = 13) after 3 days hospitalisation, and in patients with present renal dysfunction at hospitalisation (RD, N = 30); and (II) comparison between ADHF subgroups (Mann-Whitney test was performed after the Kruskal-Wallis test for multiple group comparison showed P-value < 0.05). (Rho = –0.363; P = 0.007), and values were twofold higher were observed between noRI and RI groups at 3 days of than those in the HS group (6.07 [2.34–11.04] vs. 3.90 hospitalisation (Figure 5B). [3.12–5.72] µg/mg total protein), although differences did Levels of uVDBP negatively correlated with GFR, and a not achieve statistical significance (P = 0.189). At day 3 significant positive correlation was found with uCysC and of hospitalisation, uCysC did not differ between ADHF uKIM-1 in patients with ADHF at day 3 of hospitalisation groups, although an increasing trend was observed in RD (Supplementary Table 5). Stronger positive correlations levels patients when compared with those maintaining normal were found when only patients with ADHF with GFR above renal function (noRI) during the hospitalisation period 60 ml/min/1.73 m2 were considered (noRI and RI groups; (P = 0.072) (Figure 5A). Similarly, the highest KIM-1 levels Figures 5C,D). in urine (uKIM-1) at day 3 were found in the RD group We tested whether each one of the proteins (uVDBP, with significant differences when compared with the RI uCysC, and uKIM-1) had the potential to discriminate renal and noRI groups (Figure 5B). However, no differences injury in patients with ADHF by the ROC curve analysis Frontiers in Cardiovascular Medicine | www.frontiersin.org 8 June 2022 | Volume 9 | Article 829490 Diaz-Riera et al. VDBP and Acute Heart Failure FIGURE 4 | Serum VDBP in patients with ADHF at day 3 of hospitalisation. (A) Box plot diagrams [median (IQR)] of VDBP levels quantified by ELISA in serum of patients with ADHF already presenting renal dysfunction at hospital admission (RD), developing renal function deterioration during hospitalisation (RI), and maintaining normal renal function (noRI). Groups’ comparisons were made by the Kruskal-Wallis and Mann-Whitney test. (B) Spearman correlation (Rho) between VDBP levels in serum and urine in patients with ADHF at day 3 of hospitalisation. (Supplementary Table 6A). None of the 3 proteins presented syndrome (6) and often referred to as type-1 cardiorenal a significant discriminating ability when taken individually, syndrome when having a primary and acute cardiac condition, with urinary levels of VDBP presenting the largest area under such as ADHF, that triggers acute kidney injury (4). In the curve (AUC = 0.579) compared with those of KIM- patients hospitalised with ADHF, occurrence of acute 1 (AUC = 0.487) and cystatin C (AUC = 0.476), although kidney injury during the treatment frequently affects key differences among them did not achieve statistical significance therapeutic decisions, resulting in incomplete decongestion of (Supplementary Table 6B). The combination of cystatin C the patients (5). and KIM-1 showed a greater area under the curve but There is a growing consensus about the poor effectiveness still not statistically significant (AUC = 0.604, P = 0.312). of serum creatinine measurement to detect initial stages However, the addition of VDBP levels to this combination of cardiorenal syndrome type 1 (32) and the urgent need significantly increased the discriminating value for renal function of more reliable biological variables for early detection of deterioration in patients with ADHF (AUC = 0.711, P = 0.018, kidney function deterioration in patients with ADHF (2, 4– Figure 5E), whereas the combination of VDBP with either 6) in order to better stratify their risk during the first days cystatin C or KIM-1 alone was not significant (P = 0.428 and of hospitalisation. NGAL (33), KIM-1 (34), and cystatin C P = 0.089, respectively). (35) had provided promising results at detecting severe renal Due to the impact of age in renal dysfunction, the damage in patients with ADHF and other cardiac syndromes; combination of aggregating VDBP, CysC, and KIM-1 for an however, negative studies (15, 32, 36–38) have also been early discrimination of patients with ADHF with subclinical reported. There is certainly controversy regarding their relative renal function deterioration during hospitalisation was further diagnostic power. reanalysed in an age-adjusted subgroup of patients (age; RI vs. Therefore, we performed an exploratory proteomic study noRI subgroups: 74 [69–77] vs. 69 [63–72] years, P = 0.304; aimed to identify urinary protein associated with early changes N = 13/group). As shown in Supplementary Figure 5, in this of renal function impairment in patients hospitalised with age-adjusted cohort, the discrimination power of combining diagnostic of ADHF. Urinary proteins have been extensively VDBP, CysC, and KIM-1 remained statistic significantly (AUC: studied in acute heart failure as they may be more sensitive 0.711 [0.536–0.885], P = 0.018). Adding age to the 3-variable to renal injury and structural damage than serum creatinine. cluster did not significantly modify the AUC to early discriminate Till present, however, there is insufficient evidence to translate patients with and without deterioration of kidney function during the usefulness of measuring these proteins as the clinical hospitalisation. diagnostic tool. Our hypothesis-generating study is based on an untargeted mass-spectrometry approach. We identified a differential pattern DISCUSSION of proteins involved in vitamin A and D homeostasis in prospectively collected spot-urine samples of patients with Heart and kidneys are closely interconnected through complex ADHF when arriving to the emergency room. More specifically, bidirectional mechanistic interactions in their underlying we have found that urinary VDBP levels are markedly elevated pathophysiology (31), which is collectively known as cardiorenal in patients with ADHF, being changes more evident in those Frontiers in Cardiovascular Medicine | www.frontiersin.org 9 June 2022 | Volume 9 | Article 829490 Diaz-Riera et al. VDBP and Acute Heart Failure FIGURE 5 | Markers of tubular damage (urinary CysC and KIM-1) in patients with ADHF at day 3 of hospitalisation. Box plot diagrams [median (IQR)] refer to (A) cystatin C (CysC) and (B) Kidney Injury Molecule 1 (KIM-1) in the subgroups of patients with ADHF as described in Figure 4A. (C,D) Spearman correlation (Rho) between urinary VDBP levels and CysC and KIM-1, respectively, in all patients with ADHF. (E) ROC curve analyses (C-Statistics) for the prediction of renal function deterioration presentation in patients with ADHF (N = 34). AUC indicates area under the curve. Frontiers in Cardiovascular Medicine | www.frontiersin.org 10 June 2022 | Volume 9 | Article 829490 Diaz-Riera et al. VDBP and Acute Heart Failure patients with reduced GFR, suggesting an association between changes in GFR (24, 25). Our results showed no association low GFR and high VDBP levels. Unfortunately, we cannot discern between urinary VDBP loss and serum levels of VDBP supporting if low GFR levels in the RD patients evidenced a chronic renal a tubular damage leading to diminished reabsorption and dysfunction or were a result of the acute decompensated event. consequently higher urinary excretion as the most possible Our findings are consistent with previous reports on urinary process accounting for the high VDBP levels in urine of patients VDBP in the setting of pathologies associated to nephropathies with ADHF. In this respect, Chaykovska et al. (24) also proposed in humans. In systemic lupus erythematosus, urinary VDBP the increase of VDBP levels could be associated with the severity was significantly elevated in patients with lupus nephritis and of renal damage. associated with disease severity (39). In addition, VDBP has Although the underlying mechanisms remain unclear, we been studied in diabetic nephropathy and has shown to correlate could speculate that elevated levels of VDBP in urine could with microalbuminuria, suggesting a diagnostic value in this regulate tubular cell function after endocytosis via megalin, since pathology (40). Interestingly, it was previously reported that megalin is involved in the active uptake of VDBP complexed to urinary VDBP was strongly and consistently elevated in rats 25-hydroxy vitamin D3 in the proximal tubules (45). A reduced with induced nephropathy very early in the course of the uptake of VDBP, as in kidney disease conditions, could lead disease (25). VDBP has also been associated with major contrast- to intrarenal losses of VDBP and vitamin D metabolites (45). induced nephropathy events 90 days after exposure to contrast Further studies are needed to evidence whether VDBP is a media in patients undergoing coronary angioplasty (24). To our relevant cardiorenal connector protein transmitting heart-to- knowledge, however, it was not shown so far that the changes in kidney signals. urinary VDBP levels are detected before renal injury is evidenced Other proteins in urine including KIM-1 and cystatin C in hospitalised patients with ADHF. have been related to renal failure and more specifically to It is to notice that urinary VDBP significantly correlated with tubular damage (14, 15). Supporting this finding, at day 3 of plasma NT-proBNP and the level of GFR, suggesting VDBP as a hospitalisation, KIM-1 levels were significantly elevated only protein associated with cardiorenal syndrome. This finding is in in those patients with a pathological GFR. Also, a trend to line with a previous study using NT-proBNP and cystatin C, as higher values was observed in the levels of uCysC although a marker of renal function, to identify the cardiorenal status in it did not achieve statistical significance. In contrast, patients patients with refractory heat failure on cardiac resynchronisation with ADHF with GFR > 60 ml/min/1.73 m2 at day 3 of therapy (41). The fact that no relationship between VDBP and hospitalisation had KIM-1 and cystatin C levels that did not ejection fraction was found might result from the heterogeneity differ from the normal healthy range regardless of the subsequent of our study population. Indeed, patients with heart failure evolution of the kidney function. Different pattern of VDBP with reduced and preserved ejection fraction are frequently and cystatin C and KIM-1 in urine of patients with ADHF considered as two different phenotype entities with differences at day 3 could be due to differences in the time response of in pathophysiological pathways deriving in the cardiorenal the cellular processes occurring at the proximal tubular cells syndrome (42, 43). after injury. Thus, KIM-1 is a protein expressed in proximal As suggested before, high urinary VDBP in patients tubular cells after injury, but barely found in normal kidney (46), with ADHF at hospital admission may partially reflect a whereas the presence of cystatin C and VDBP in urine result of temporary kidney function deterioration event due to impaired a deficient endocytic receptor-mediated uptake, catabolism, and renal perfusion as consequence of low cardiac output and degradation of this protein by the tubular cells (47, 48). Further systemic venous congestion (8, 44). However, the maintained studies are guaranteed to better understand whether differences urinary VDBP loss in patients with ADHF with GFR below in the endocytic process and/or cell catabolism account for the 60 ml/min/1.73 m2 after 3 days of hospitalisation suggest VDBP and cystatin C patterns in urine of patients with ADHF. that high VDBP levels in urine associate to a renal structural The positive correlation of urinary VDBP with cystatin C and damage in these patients, that might derive into cardiorenal KIM-1 and day 3 of hospitalisation strongly suggest that renal syndrome type 1. tubular insult accounts for acute kidney function deterioration Indeed, among patients with ADHF with normal renal in patients with ADHF. Differences with two previous studies function at hospital admission (NRF group), urinary VDBP suggesting a lack of association between worsening of renal remains high at day 3 in those patients presenting worsening of function and tubular injury in ADHF might be explained by the the renal function at a later stage of the hospitalisation period, fact that these studies just focused on the expression/secretion while patients who maintained normal renal function presented of proteins related to tubular cell injury such as KIM-1 and a decrease of urinary VDBP into the healthy range. Interestingly, NGAL (49, 50), without considering other molecules reflecting these changes at day 3 were observed before any apparent a dysfunctional cell phenotype. pathological change in plasma creatinine levels and the GFR, Subsequently, the ROC analysis was carried out to study that when occurred it was detected after 4–5 days hospitalisation. the value of VDBP compared with cystatin C and KIM-1 This finding suggests that urinary VDBP is sensitive to early in association with incident kidney function deterioration in pathological changes occurring in kidney and identifying those patients hospitalised with ADHF. In our study population, none patients with subclinical kidney function deterioration. of the three proteins showed enough discriminative power when High urinary VDBP levels in patients with ADHF may result considered separately, nor did it when considering the joint from renal tubular injury or reflect hemodynamic or functional detection of cystatin C and KIM-1. However, when VDBP was Frontiers in Cardiovascular Medicine | www.frontiersin.org 11 June 2022 | Volume 9 | Article 829490 Diaz-Riera et al. VDBP and Acute Heart Failure added the AUC achieved statistical significance, suggesting that DATA AVAILABILITY STATEMENT urine VDBP combined with urine values of cystatin C and KIM-1 might serve as new diagnostic tool to discriminate ADHF patients The original contributions presented in the study are included regarding their risk of developing renal function deterioration in the article/Supplementary Material, further inquiries can be during hospitalisation. directed to the corresponding author. Age is an important factor of kidney injury risk and development (51). In this respect, we want to highlight as a study limitation the age difference between ADHF patients ETHICS STATEMENT with and without renal dysfunction at admission. However, VDBP levels did not correlate with age in our study population, The studies involving human participants were reviewed and suggesting the independence of both variables. In addition, approved by the Ethical Committee for Clinical Research it is interesting to notice that the discriminating power of from Hospital Santa Cruz y San Pablo in March 2016 (ref VDBP, clustered with cystatin C and KIM-1 for kidney function number 16/022). The patients/participants provided their written deterioration in patients hospitalised with ADHF is maintained informed consent to participate in this study. when age-adjusted ADHF groups were compared. Thus, to the best of our knowledge, this is the first study reporting an association between urinary VDBP and renal worsening AUTHOR CONTRIBUTIONS in ADHF during hospitalisation. Still, the limited sample size and high heterogeneity of the study population in their XG-M, LB, and TP: study concept and design. LL and X-GM: characteristics, etiopathology, and disease background need to be patient inclusion, clinical data, and sample acquisition. ED-R, recognised. Therefore, further studies in larger groups are needed MG-A, and TP: methodology and formal analysis. ED-R, LB, and to confirm our findings and validate the impact of urinary VDBP TP: statistical analysis and result interpretation. ED-R, LB, and in ADHF and incident kidney disease. TP: writing–original draft preparation. ED-R, MG-A, LL, XG-M, Vitamin D binding protein is forming a complex with the 25- LB, and TP: manuscript review and editing. LB and TP: funding (OH) vitamin D3 when present in urine. Tubular reabsorption of acquisition. All authors have read and agreed to the submitted this complex is a critical process into the conversion of vitamin version of the manuscript. D into its active metabolite that is secreted into the circulation (52). Damage to proximal tubules in ADHF leading to a deficient reabsorption of VDBP-25-(OH) vitamin D3 complex could result FUNDING in diminished amounts of vitamin D active metabolite, also This study was supported by the “Fundació La Marató known as 1,25-dihydroxycholecalciferol and calcitriol. In this TV3” (Project number 20153110) to TP; the H2020-JTI-IMI2- respect, vitamin D deficiency has been previously associated to 2017-13-two-stage Project 821283 –TransBioLine to LB and several diseases concerning renal failure (53, 54), cardiovascular TP; the Spanish “Agencia Estatal de Investigación (AEI)”– health (55), certain types of cancer, type 2 diabetes, metabolic Institute of Health Carlos III, ISCIII [FIS PI19/01687 to TP; syndrome, inflammatory bowel disease, and several others (56, Centro de Investigación Biomedica en Red Cardiovascular 57). Vitamin D has been studied within chronic heart failure (58); (CIBERCV-CB16/11/00411) to LB; Spanish Ministry of Economy however, to our knowledge, there are no publications associating and Competitiveness of Science-PID2019-107160RB-I00 to LB; ADHF with vitamin D. and cofounded by FEDER “Una Manera de Hacer Europa.” In summary, VDBP has been consistently detected by Secretaria de Universitats i Recerca del Departament d’Empresa 2DE-MS/MS as two spots in urine of patients with ADHF i Coneixement de la Generalitat de Catalunya (2017 SGR 1480). at hospital admission in this exploratory study, wherein the We thank Fundación Jesús Serra and Fundación de Investigación patients with ADHF presented higher urinary levels than Cardiovascular (Barcelona) for their continuous support. ED-R healthy subjects. Additionally, patients with ADHF presenting is a predoctoral fellow funded by the “Fundació La Marató TV3” kidney dysfunction at hospitalisation showed the highest and the Cardiovascular Program-ICCC (IR-HSCSP). urinary loss of VDBP. A significant drop in urinary VDBP is observed within 3 days in patients with normal renal function during the hospitalisation period. On the contrary, ACKNOWLEDGMENTS those patients who developed renal function deterioration maintained high levels of VDBP loss in urine, while their We gratefully acknowledge the valuable help and support of GFR levels were within normal physiological range at that Montse Gómez-Pardo with the sample handling. time. When performing ROC analysis, the addition of urinary VDBP to renal injury markers such as uCysC and KIM-1 gave statistical power to anticipate renal damage. Therefore, SUPPLEMENTARY MATERIAL a daily follow-up of VDBP urinary secretion could help to determine whether patients’ kidneys are prone to failure The Supplementary Material for this article can be found online and prevent it using more appropriate and personalised at: https://www.frontiersin.org/articles/10.3389/fcvm.2022. therapeutic techniques. 829490/full#supplementary-material Frontiers in Cardiovascular Medicine | www.frontiersin.org 12 June 2022 | Volume 9 | Article 829490 Diaz-Riera et al. VDBP and Acute Heart Failure REFERENCES infarction. Eur Heart J. (2015) 36:965–75. doi: 10.1093/eurheartj/ehu 356 1. Grande D, Gioia MI, Terlizzese P, Iacoviello M. Heart failure and kidney 22. Eshbach ML, Weisz OA. Receptor-mediated endocytosis in the proximal disease. Adv Exp Med Biol. (2018) 1067:219–38. tubule. Annu Rev Physiol. (2017) 79:425–48. 2. Metra M, Nodari S, Parrinello G, Bordonali T, Bugatti S, Danesi R, et al. 23. Leheste JR, Rolinski B, Vorum H, Hilpert J, Nykjaer A, Jacobsen C, et al. Worsening renal function in patients hospitalised for acute heart failure: Megalin knockout mice as an animal model of low molecular weight clinical implications and prognostic significance. Eur J Heart Fail. (2008) proteinuria. Am J Pathol. (1999) 155:1361–70. doi: 10.1016/S0002-9440(10) 10:188–95. doi: 10.1016/j.ejheart.2008.01.011 65238-8 3. Aronson D. Cardiorenal syndrome in acute decompensated heart failure: a 24. Chaykovska L, Heunisch F, Von Einem G, Alter ML, Hocher CF, Tsuprykov new approach. Expert Rev Cardiovasc Ther. (2012) 10:177–89. O, et al. Urinary Vitamin D binding protein and KIM-1 are potent new 4. Ronco C, McCullough P, Anker SD, Anand I, Aspromonte N, Bagshaw SM, biomarkers of major adverse renal events in patients undergoing coronary et al. Cardio-renal syndromes: report from the consensus conference of the angiography. PLoS One. (2016) 11:e0145723. doi: 10.1371/journal.pone. acute dialysis quality initiative. Eur Heart J. (2010) 31:703–11. doi: 10.1093/ 0145723 eurheartj/ehp507 25. Mirković K, Doorenbos CRC, Dam WA, Lambers Heerspink HJ, Slagman 5. Damman K, Valente MAE, Voors AA, O’Connor CM, Van Veldhuisen DJ, MCJ, Nauta FL, et al. Urinary Vitamin D Binding protein: a potential novel Hillege HL. Renal impairment, worsening renal function, and outcome in marker of renal interstitial inflammation and fibrosis. PLoS One. (2013) patients with heart failure: an updated meta-analysis. Eur Heart J. (2014) 8:e55887. doi: 10.1371/journal.pone.0055887 35:455–69. doi: 10.1093/eurheartj/eht386 26. Tarwater K. Estimated glomerular filtration rate explained. Mo Med. (2011) 6. Ronco C, Haapio M, House AA, Anavekar N, Bellomo R. Cardiorenal 108:29–32. syndrome. J Am Coll Cardiol. (2008) 52:1527–39. 27. Cubedo J, Padró T, Garcia-Moll X, Pintó X, Cinca J, Badimon L. Proteomic 7. Palazzuoli A, Ruocco G. Heart–kidney interactions in cardiorenal syndrome signature of apolipoprotein J in the early phase of new-onset myocardial Type 1. Adv Chronic Kidney Dis. (2018) 25:408–17. doi: 10.1053/j.ackd.2018. infarction. J Proteome Res. (2011) 10:211–20. doi: 10.1021/pr100805h 08.013 28. Cubedo J, Padró T, Alonso R, Mata P, Badimon L. ApoL1 levels in high density 8. Verbrugge FH. Utility of urine biomarkers and electrolytes for the lipoprotein and cardiovascular event presentation in patients with familial management of heart failure. Curr Heart Fail Rep. (2019) 16:240–9. doi: 10. hypercholesterolemia. J Lipid Res. (2016) 57:1059–73. doi: 10.1194/jlr.P061598 1007/s11897-019-00444-z 29. García-Arguinzonis M, Padró T, Lugano R, Llorente-Cortes V, Badimon L. 9. Manguba AS, Vela Parada X, Coca SG, Lala A. Synthesizing markers of kidney Low-density lipoproteins induce heat shock protein 27 dephosphorylation, injury in acute decompensated heart failure: should we even keep looking? oligomerization, and subcellular relocalization in human vascular smooth Curr Heart Fail Rep. (2019) 16:257–73. doi: 10.1007/s11897-019-00448-9 muscle cells. Arterioscler Thromb Vasc Biol. (2010) 30:1212–9. doi: 10.1161/ 10. Sokolski M, Zymliński R, Biegus J, Siwołowski P, Nawrocka-Millward S, Todd ATVBAHA.109.198440 J, et al. Urinary levels of novel kidney biomarkers and risk of true worsening 30. Han X, Zhang S, Chen Z, Adhikari BK, Zhang Y, Zhang J, et al. Cardiac renal function and mortality in patients with acute heart failure. Eur J Heart biomarkers of heart failure in chronic kidney disease. Clin Chim Acta. (2020) Fail. (2017) 19:760–7. doi: 10.1002/ejhf.746 510:298–310. 11. Hu Y, Liu H, Du L, Wan J, Li X. Serum cystatin C predicts AKI and the 31. Kumar U, Garimella P, Wettersten N. Cardiorenal syndrome-pathophysiology. prognosis of patients in coronary care unit: a prospective, observational study. Cardiol Clin. (2019) 37:251–65. Kidney Blood Press Res. (2017) 42:961–73. doi: 10.1159/000485341 32. Atici A, Emet S, Cakmak R, Yuruyen G, Alibeyoglu A, Akarsu M, et al. Type 12. Katagiri D, Doi K, Honda K, Negishi K, Fujita T, Hisagi M, et al. Combination I cardiorenal syndrome in patients with acutely decompensated heart failure: of two urinary biomarkers predicts acute kidney injury after adult cardiac the importance of new renal biomarkers. Eur Rev Med Pharmacol Sci. (2018) surgery. Ann Thorac Surg. (2012) 93:577–83. doi: 10.1016/j.athoracsur.2011. 22:3534–43. doi: 10.26355/eurrev_201806_15180 10.048 33. Angeletti S, Fogolari M, Morolla D, Capone F, Costantino S, Spoto S, 13. Devarajan P. Neutrophil gelatinase-associated lipocalin: a promising et al. Role of neutrophil gelatinase-associated lipocalin in the diagnosis and biomarker for human acute kidney injury. Biomark Med. (2010) 4:265–80. early treatment of acute kidney injury in a case series of patients with doi: 10.2217/bmm.10.12 acute decompensated heart failure: a case series. Cardiol Res Pract. (2016) 14. Urbschat A, Gauer S, Paulus P, Reissig M, Weipert C, Ramos-Lopez E, et al. 2016:3708210. doi: 10.1155/2016/3708210 Serum and urinary NGAL but not KIM-1 raises in human postrenal AKI. Eur 34. Huang Y, Don-Wauchope AC. The clinical utility of kidney injury molecule J Clin Invest. (2014) 44:652–9. doi: 10.1111/eci.12283 1 in the prediction, diagnosis and prognosis of acute kidney injury: a 15. Breidthardt T, Sabti Z, Ziller R, Rassouli F, Twerenbold R, Kozhuharov N, systematic review. Inflamm Allergy Drug Targets. (2011) 10:260–71. doi: 10. et al. Diagnostic and prognostic value of cystatin C in acute heart failure. Clin 2174/187152811796117735 Biochem. (2017) 50:1007–13. doi: 10.1016/j.clinbiochem.2017.07.016 35. Zhang Z, Lu B, Sheng X, Jin N. Cystatin C in prediction of acute kidney injury: 16. Kuznetsova T, Mischak H, Mullen W, Staessen JA. Urinary proteome analysis a systemic review and meta-analysis. Am J Kidney Dis. (2011) 58:356–65. in hypertensive patients with left ventricular diastolic dysfunction. Eur Heart doi: 10.1053/j.ajkd.2011.02.389 J. (2012) 33:2342–50. doi: 10.1093/eurheartj/ehs185 36. Breidthardt T, Socrates T, Drexler B, Noveanu M, Heinisch C, Arenja N, et al. 17. Campbell RT, Jasilek A, Mischak H, Nkuipou-Kenfack E, Latosinska A, Welsh Plasma neutrophil gelatinase-associated lipocalin for the prediction of acute PI, et al. The novel urinary proteomic classifier HF1 has similar diagnostic and kidney injury in acute heart failure. Crit Care. (2012) 16:R2. doi: 10.1186/ prognostic utility to BNP in heart failure. ESC Hear Fail. (2020) 7:1595–604. cc10600 doi: 10.1002/ehf2.12708 37. Verbrugge FH, Dupont M, Shao Z, Shrestha K, Singh D, Finucan M, et al. 18. He T, Mischak M, Clark AL, Campbell RT, Delles C, Díez J, et al. Urinary Novel Urinary biomarkers in detecting acute kidney injury, persistent renal peptides in heart failure: a link to molecular pathophysiology. Eur J Heart Fail. impairment and all-cause mortality following decongestive therapy in acute (2021) 23:1875–87. doi: 10.1002/ejhf.2195 decompensated heart failure frederik. J Card Fail. (2013) 19:621–8. doi: 10. 19. Farmakis D, Koeck T, Mullen W, Parissis J, Gogas BD, Nikolaou M, et al. Urine 1016/j.cardfail.2013.07.004 proteome analysis in heart failure with reduced ejection fraction complicated 38. Legrand M, de Berardinis B, Gaggin HK, Magrini L, Belcher A, Zancla B, et al. by chronic kidney disease: feasibility, and clinical and pathogenetic correlates. Evidence of uncoupling between renal dysfunction and injury in cardiorenal Eur J Heart Fail. (2016) 18:822–9. doi: 10.1002/ejhf.544 syndrome: insights from the BIONICS study. PLoS One. (2014) 9:e112313. 20. Cubedo J, Padró T, Garcia-Moll X, Pintó X, Cinca J, Badimon L. Serum doi: 10.1371/journal.pone.0112313 proteome in acute myocardial infarction. Clin Investig Arterioscler. (2011) 39. Go D, Lee JY, Kang MJ, Lee EY, Lee EB, Yi EC, et al. Urinary vitamin D-binding 23:147–54. protein, a novel biomarker for lupus nephritis, predicts the development 21. Ramaiola I, Padró T, Peña E, Juan-Babot O, Cubedo J, Martin-Yuste V, et al. of proteinuric flare. Lupus. (2018) 27:1600–15. doi: 10.1177/096120331877 Changes in thrombus composition and profilin-1 release in acute myocardial 8774 Frontiers in Cardiovascular Medicine | www.frontiersin.org 13 June 2022 | Volume 9 | Article 829490 Diaz-Riera et al. VDBP and Acute Heart Failure 40. Bai X, Luo Q, Tan K, Guo L. Diagnostic value of VDBP and miR-155-5p in 25-(OH) vitamin D3. Cell. (1999) 96:507–15. doi: 10.1016/s0092-8674(00)80 diabetic nephropathy and the correlation with urinary microalbumin. Exp 655-8 Ther Med. (2020) 20:1. doi: 10.3892/etm.2020.9214 53. Teng M, Wolf M, Ofsthun MN, Lazarus JM, Hernán MA, Camargo CA, et al. 41. Truong QA, Szymonifka J, Januzzi JL, Contractor JH, Deaño RC, Chatterjee Activated injectable vitamin D and hemodialysis survival: a historical cohort NA, et al. Cardiorenal status using amino terminal pro brain natriuretic study. J Am Soc Nephrol. (2005) 16:1115–25. peptide and Cystatin C on cardiac resynchronization therapy outcomes: from 54. Wolf M, Shah A, Gutierrez O, Ankers E, Monroy M, Tamez H, et al. Vitamin the BIOCRT study Quynh. Hear Rhythm. (2019) 16:928–35. doi: 10.1016/j. D levels and early mortality among incident hemodialysis patients. Kidney Int. hrthm.2018.12.023 (2007) 72:1004–13. 42. Yamagishi T, Matsushita K, Minamishima T, Goda A, Sakata K, Satoh T, et al. 55. Anderson JL, May HT, Horne BD, Bair TL, Hall NL, Carlquist JF, et al. Comparison of risk factors for acute worsening renal function in heart failure Relation of vitamin D deficiency to cardiovascular risk factors, disease status, patients with and without preserved ejection fraction. Eur J Intern Med. (2015) and incident events in a general healthcare population. Am J Cardiol. (2010) 26:599–602. doi: 10.1016/j.ejim.2015.08.003 106:963–8. doi: 10.1016/j.amjcard.2010.05.027 43. Agrawal A, Naranjo M, Kanjanahattakij N, Rangaswami J, Gupta S. 56. Bouillon R, Carmeliet G. Vitamin D insufficiency: definition, diagnosis and Cardiorenal syndrome in heart failure with preserved ejection fraction—an management. Best Pract Res Clin Endocrinol Metab. (2018) 32:669–84. doi: under-recognized clinical entity. Heart Fail Rev. (2019) 24:421–37. doi: 10. 10.1016/j.beem.2018.09.014 1007/s10741-018-09768-9 57. Battistini C, Ballan R, Herkenhoff ME, Saad SMI, Sun J. Vitamin d modulates 44. Schefold JC, Filippatos G, Hasenfuss G, Anker SD, Von Haehling S. Heart intestinal microbiota in inflammatory bowel diseases. Int J Mol Sci. (2021) failure and kidney dysfunction: epidemiology, mechanisms and management. 22:1–22. doi: 10.3390/ijms22010362 Nat Rev Nephrol. (2016) 12:610–23. doi: 10.1038/nrneph.2016.113 58. Witte KK, Byrom R, Gierula J, Paton MF, Jamil HA, Lowry JE, et al. Effects of 45. Fowlkes JL, Bunn RC, Cockrell GE, Clark LM, Wahl EC, Lumpkin CK, vitamin D on cardiac function in patients with chronic HF: the VINDICATE et al. Dysregulation of the intrarenal vitamin D endocytic pathway in a study. J Am Coll Cardiol. (2016) 67:2593–603. doi: 10.1016/j.jacc.2016.03.508 nephropathy-prone mouse model of type 1 diabetes. Exp Diabetes Res. (2011) 2011:269378. doi: 10.1155/2011/269378 Conflict of Interest: LB received institutional research grants from AstraZeneca; 46. van Timmeren M, van den Heuvel M, Bailly V, Bakker S, van Goor H, consultancy fees from Sanofi, Pfizer, and Novartis; and speaker fees from Lilly, Stegeman C. Tubular kidney injury molecule-1 (KIM-1) in human renal Pfizer, and AstraZeneca. TP and LB are shareholders of the academic spin-off disease. J Pathol. (2007) 212:209–17. doi: 10.1002/path.2175 companies GlyCardial Diagnostics S.L. and Ivestatin Therapeutics S.L. All 47. Jensen D, Kierulf-Lassen C, Kristensen MLV, Nørregaard R, Weyer K, Nielsen unrelated to the present work. R, et al. Megalin dependent urinary cystatin C excretion in ischemic kidney injury in rats. PLoS One. (2017) 12:e0178796. doi: 10.1371/journal.pone. The remaining authors declare that the research was conducted in the absence of 0178796 any commercial or financial relationships that could be construed as a potential 48. Bennett MR, Pordal A, Haffner C, Pleasant L, Ma Q, Devarajan P. Urinary conflict of interest. vitamin D-binding protein as a biomarker of steroid-resistant nephrotic syndrome. Biomark Insights. (2016) 11:1–6. doi: 10.4137/BMI.S31633 Publisher’s Note: All claims expressed in this article are solely those of the authors 49. Dupont M, Shrestha K, Singh D, Awad A, Kovach C, Scarcipino M, et al. and do not necessarily represent those of their affiliated organizations, or those of Lack of significant renal tubular injury despite acute kidney injury in acute the publisher, the editors and the reviewers. Any product that may be evaluated in decompensated heart failure. Eur J Heart Fail. (2012) 14:597–604. doi: 10.1093/ this article, or claim that may be made by its manufacturer, is not guaranteed or eurjhf/hfs039 endorsed by the publisher. 50. Ahmad T, Jackson K, Rao VS, Tang WHW, Brisco-Bacik MA, Chen HH, et al. Worsening renal function in patients with acute heart failure undergoing Copyright © 2022 Diaz-Riera, García-Arguinzonis, López, Garcia-Moll, Badimon aggressive diuresis is not associated with tubular injury. Circulation. (2018) and Padró. This is an open-access article distributed under the terms of the Creative 137:2016–28. Commons Attribution License (CC BY). The use, distribution or reproduction in 51. Coca SG. Acute kidney injury in elderly persons. Am J Kidney Dis. (2010) other forums is permitted, provided the original author(s) and the copyright owner(s) 56:122–31. doi: 10.1053/j.ajkd.2009.12.034 are credited and that the original publication in this journal is cited, in accordance 52. Nykjaer A, Dragun D, Walther D, Vorum H, Jacobsen C, Herz J, et al. An with accepted academic practice. No use, distribution or reproduction is permitted endocytic pathway essential for renal uptake and activation of the steroid which does not comply with these terms. Frontiers in Cardiovascular Medicine | www.frontiersin.org 14 June 2022 | Volume 9 | Article 829490