Peptide-based therapeutics hold great promise for the treatment of various diseases; however, their clinical application is often hindered by toxicity challenges. The accurate prediction of peptide toxicity is crucial for designing safe peptide-based therapeutics. While traditional experimental approaches are time-consuming and expensive, computational methods have emerged as viable alternatives, including similarity-based and machine learning (ML)-/deep learning (DL)-based methods. However, existing methods often struggle with robustness and generalizability. To address these challenges, we propose HyPepTox-Fuse, a novel framework that fuses protein language model (PLM)-based embeddings with conventional descriptors. HyPepTox-Fuse integrates ensemble PLM-based embeddings to achieve richer peptide representations by leveraging a cross-modal multi-head attention mechanism and Transformer architecture. A robust feature ranking and selection pipeline further refines conventional descriptors, thus enhancing prediction performance. Our framework outperforms state-of-the-art methods in cross-validation and independent evaluations, offering a scalable and reliable tool for peptide toxicity prediction. Moreover, we conducted a case study to validate the robustness and generalizability of HyPepTox-Fuse, highlighting its effectiveness in enhancing model performance. Furthermore, the HyPepTox-Fuse server is freely accessible at https://balalab-skku.org/HyPepTox-Fuse/ and the source code is publicly available at https://github.com/cbbl-skku-org/HyPepTox-Fuse/. The study thus presents an intuitive platform for predicting peptide toxicity and supports reproducibility through openly available datasets.

Objective:To investigate the value of soluble interleukin-6 (IL-6) receptor (sIL-6R) level in predicting ultrafiltration insufficiency in peritoneal dialysis (PD) patients.Methods:It was a prospective cohort study. The patients who received continuous ambulatory PD and regular follow-up between November 2016 and July 2018 in the PD Center of Renji Hospital, School of Medicine, Shanghai Jiao Tong University were enrolled. Enzyme-linked immunosorbent assay was used to determine dialysate sIL-6R and its appearance rate (AR) was calculated. Patients were divided into high sIL-6R AR group and low sIL-6R AR group according to median value of sIL-6R AR and prospectively followed up until death, PD cessation, or the end of the study (December 31, 2022). Multiple linear regression was used to analyze the related factors of sIL-6R AR. Kaplan-Meier method and log-rank test were used to compare the survival rate difference of ultrafiltration insufficiency between high sIL-6R AR group and low sIL-6R AR group. Multivariate Cox regression and multivariate competing risk models were used to assess the risk factors associated with occurrence of ultrafiltration insufficiency.Results:A total of 198 PD patients were enrolled, including 115 (58.1%) males, with age of (54.9±13.7) years old and PD duration of 22.5 (6.6, 65.0) months. The sIL-6R AR of the cohort was 2 094.7 (1 672.4, 2 920.9) pg/min. Compared with low sIL-6R AR（<2 094.7 pg/min）group, high sIL-6R AR（>2 094.7 pg/min）group had older age ( t=-3.269, P=0.001), higher body mass index ( t=-3.248, P=0.001), proportion of combined diabetes mellitus ( χ2=8.890, P=0.003), 24 h glucose exposure ( Z=-2.257, P=0.024), 24 h ultrafiltration capacity ( Z=-2.515, P=0.012), 4 h dialysate creatinine to serum creatinine ratio ( t=-2.609, P=0.010), mass transfer area coefficient of creatinine ( Z=-2.308, P=0.021), IL-6 AR ( Z=-3.533, P<0.001) and solute glycoprotein 130 AR ( Z=-8.670, P<0.001), and lower serum albumin ( t=2.595, P=0.010) and residual renal function ( t=2.133, P=0.033). Multiple linear regression analysis showed that body mass index ( β=0.194, P=0.005), serum albumin ( β=-0.215, P=0.002) and dialysate lg[IL-6 AR] ( β=0.197, P=0.011) were independently correlated with sIL-6R AR. By the end of the study, 57 (28.8%) patients developed ultrafiltration insufficiency. Kaplan-Meier analysis showed that high sIL-6R AR group had a significantly inferior ultrafiltration insufficiency-free survival rate than that in low sIL-6R AR group (log-rank χ 2=5.375, P=0.020). Multivariate Cox regression analysis and multivariate competing risk models showed that high dialysate sIL-6R AR (>2 094.7 pg/min) was an independent influencing factor of ultrafiltration insufficiency ( HR=2.286 , 95% CI 1.254-4.165 , P=0.007 ; SHR=2.074, 95% CI 1.124-3.828, P=0.020) in PD patients. Conclusions:Dialysate sIL-6R level was associated with body mass index, serum albumin and dialysate IL-6 level. Dialysate sIL-6R may be a predictive factor of ultrafiltration insufficiency in PD patients.

Objective To investigate the factors affecting the efficacy of leflunomide combined with medium/low dose corticosteroids in the treatment of progressive IgA nephropathy (IgAN).Methods Clinical and pathological parameters were collected retrospectively in patients of primary IgAN with proteinuria＞ 1.0 g/24 h and chronic kidney disease (CKD) stage 1-3 treated with leflunomide combined with medium/low dose corticosteroids in Ren Ji Hospital,School of Medicine,Shanghai Jiao Tong University from Jan 2005 to Dec 2010.According to the treatment effects,patients were divided into complete remission group and non-complete remission group.The biochemical and pathological indexes of the two groups were compared.Results A total of 42 patients were included.The remission rates at 3,6,9 and 12 months were 62％,64％,67％ and 74％,respectively.Seventeen (40.5％) and fourteen (33.3％) patients achieved complete and partial remission after one-year treatment,and the remission rate remained stable within one year after withdrawal of drugs.The 24hour proteinuria was 1.50 (0.67,2.66) g,which was significantly reduced compared with the baseline 2.44 (1.36,3.74) g (P ＜ 0.01).The decrease rate was 31.3％.There was a slight decrease in proteinuriawithin one year after withdrawal of drugs.Estimated glomerular filtration rate (eGFR) remained stable during the treatment and a year of follow-up.No serious adverse event was observed during the followup period.Among 31 responder patients,6(19.4％) patients relapsed.Logistic multivariate regression analysis suggested that the degree of renal interstitial inflammatory infiltration was an independent predictor of complete remission with one-year treatment of leflunomide combined with medium / low dose corticosteroids (HR=0.067,95％ CI 0.008-0.535,P=0.011).Conclusions IgAN treated with leflunomide and medium/low dose corticosteroids can achieve remission in early stage,and the remission rate remains stable after withdrawal of drugs.It is a safe option for the treatment of IgAN.Renal interstitial inflammatory infiltration is an independent predictor of complete remission.

Objective To compare the efficacy and safety of leflunomide (LEF) combined with medium/low dose corticosteroids and full dose of corticosteroids in the treatment of IgA nephropathy. Method Primary IgAN patients diagnosed by renal biopsy with 18?65 years old and eGFR≥30 ml·min?1·(1.73 m2)?1 and proteinuria>0.5 g/24 h were enrolled in a prospective controlled clinical study. They were randomly divided into leflunomide combined with medium/low dose corticosteroids (LEF group) and corticosteroids alone (steroid group). The primary outcomes were (1) end stage renal disease or dialysis (2) 50% increase in serum creatinine above the baseline. Secondary outcome was the remission of proteinuria. Results Ninety patients completed the follow?up. The 24?hour proteinuria at baseline were 2.00(1.10, 2.88) g and 1.87(1.13 ,3.08) g in LEF group and steroid group respectively. Compared with baseline, it was significantly decreased in both groups at 6 months [0.30(0.11, 0.93) g, 0.30(0.14, 1.33) g] and 12 months [0.30(0.09, 0.82) g, 0.32(0.14, 0.66) g], (P<0.05). Estimated glomerular filtration rate (eGFR) at baseline, 6 months and 12 months were (80.39 ± 28.56), (87.12±28.70) and (88.20±30.26) ml·min-1·(1.73 m2)-1. It was decreased in steroid group (P<0.05), while no significant difference was detected in LEF group[baseline (87.63 ± 27.35), 6 months (86.91 ± 32.45), 12 months (90.06 ± 30.00) ml·min-1·(1.73 m2)-1, P>0.05]. At 6 and 12 months, there was no significant difference in terms of 24?hour proteinuria, serum creatinine and eGFR (CKD?EPI) between groups (P>0.05). There was no statistically significant difference in adverse events between groups during the treatment (9/40 cases in LEF group and 11/50 cases in steroid group, P>0.05). The average follow?up was 79 months, and there was no difference in the renal prognosis between the two groups. Multivariate Cox regression analysis revealed that serum creatinine at baseline and renal interstitial inflammatory cell infiltration predicted the risk of the progress of IgA nephropathy. Conclusion Leflunomide plus medium/low dose corticosteroids has a similar effect as full dose of corticosteroids in IgA nephropathy and does not increase the risk for adverse events during the treatment.

Objective To explore the relationship between resveratrol and Notch 1 signalling pathway in podocytes.Methods Interference RNA (RNAi) and doxycycline (Dox) were used to inhibit the Sirtuin (SIRT) 1 expression in the wild-type and inducible SIRT1 shRNA (CAGGS) podocytes respectively.Recombinant mouse delta-like ligand 4 (DLL4) was used to activate Notch1 signalling.The message RNA of SIRT1,Notch1 downstream gene Hes1 and Hey2,as well as the key enzymes of Notch1 signalling pathway were detected by using real-time PCR.Western blotting was used to detect intracellular domain of Notch 1 (ICD1),SIRT1,and metalloprotease (ADAM) 10 and components of γ-secretase complex protein expression.Results In WT murine podoytes,resveratrol up-regulated ICD1 protein production,as well as the mRNA of Hes1 and Hey2 in a dose-dependent manner.Treatment with resveratrol resulted Nicastrin mRNA and protein increase in podocytes (P ＜0.05),as well as inhibit ADAM10 expression (P ＜ 0.05),but all these changes were prevented after the use of SIRT1 RNAi(P ＜ 0.05).DLL4 up-regulated the expression of mRNA of Hes1 and Hey2,as well as ICD1 protein production in a dose-dependent manner.Treatment with doxycycline resulted decrease of SIRT1 gene and protein expression in CAGGS podocytes after 24 h and 48 h respectively(P ＜ 0.05),which weakend the role of DLL4 significantly(P ＜ 0.05).Conclusion Resveratrol induces Nicastrin expression,as well as activation of Notch1 signalling pathway in a SIRT1-dependent manner.

Objective To investigate the role of activated cylic AMP(cAMP) signaling in chemical-induced podocyte injury.Methods Eight-weeks-old male BalB/C mice were randomly divided into three groups:control group,Adriamycin (ADR) group and Forskolin+ADR group.ADR nephropathy models were established by tail intravenous injection,and part of them were injected Forskolin,an agonist of adenylate cyclase,intraperitoneally.Phosphorylation of cAMP response element binding protein (CREB) was detected by laser confocal microscopy,morphology of foot processes were determined with transmission electron microscope,and WT-1 expression in glomeruli were detected by immunohistochemistry.Conditionally immortalized podocytes were treated with puromycin aminonucleoside (PAN),Exchange protein directly activated by cAMP (Epac) agonist 8-pCPT-2-O-Me-cAMP (2Me),protein kinase A (PKA) antagonist H89 and its agonist pCPT-cAMP(pCPT).Western blot was used to detect the expression levels of Epac,caspase3 and cleaved caspase3.PKA activity was assayed using cAMP-dependent protein kinase detection system.Cell viability was determined by a cell count kit and podocyte apoptosis was estimated by TUNEL staining.Mitochondrial membrane potential was evaluated by JC-1 staining.Results (1)Compared with ADR group,the urine albumin decreased significantly (P ＜ 0.05) among Forskolin + ADR group and the WT-1 positive cells per glomerulus increased obviously (P ＜ 0.05).(2)PAN decreased podocyte number in a time-dependent manner (P ＜ 0.05),pre-treatment with pCPT obviously inhibited PAN induced podocyte decrease (P ＜0.05),but H89 prevented the effect of pCPT in a dose-dependent manner (P ＜ 0.05).(3)JC-1 staining showed that the percentage of podocyte with green fluorescence for control,PAN and pCPT+PAN group were (12.67±2.15)％,(31.35±4.60)％ and (16.96 ± 2.51)％ respectively (P ＜ 0.05),and pretreatment with H89 inhibited the effect of pCPT (P ＜ 0.05).(4) PAN promoted podocyte apoptosis and cleaved caspase3 expression (P ＜ 0.05),and pretreatment with pCPT significantly prevented PAN-induced podocyte apoptosis and cleaved caspase3 expression (P ＜ 0.05).Conclusions cAMP signaling activation ameliorated podocyte injury in ADR nice and PAN-induced podocyte apoptosis,and cAMP/ PKA pathway may mediate these processes.