Renal and Vascular Effects of Selective FGFR4 Inhibition in Folic Acid-induced Kidney Injury in Rats

Abstract

Background/Aim

Fibroblast growth factor-23 (FGF-23) has been implicated in chronic kidney disease (CKD)-associated cardiovascular dysfunction, potentially through fibroblast growth factor receptor 4 (FGFR4). This study investigated renal and vascular alterations in folic acid-induced CKD and evaluated the effects of selective FGFR4 inhibition.

Materials and Methods

Male Sprague-Dawley rats were assigned to healthy control, CKD control, vehicle, and roblitinib-treated groups. CKD was induced by a single intraperitoneal dose of folic acid. Serum creatinine, FGF-23, tumor necrosis factor-α, interleukin-1β, and asymmetric dimethylarginine levels were measured. Vascular reactivity was evaluated in isolated aortic rings using phenylephrine, carbachol, and sodium nitroprusside. Renal fibrosis, carotid calcification, and medial thickness were assessed histologically.

Results

Folic acid administration increased serum creatinine and relative kidney weight and impaired both endothelium-dependent and endothelium-independent vascular relaxation. Relative kidney weight was lower in the roblitinib-treated group than in the CKD control group. However, changes in vascular responses observed in both the treatment and vehicle groups prevented these effects from being attributed specifically to FGFR4 inhibition. FGF-23 and inflammatory marker levels did not differ significantly among the groups. Carotid medial thickness was significantly greater in the treatment group than in the CKD-control group (p<0.05), whereas no other significant between-group differences were detected.

Conclusion

Folic acid-induced renal injury was associated with early functional vascular impairment despite limited structural and inflammatory alterations. Selective FGFR4 inhibition did not demonstrate a clear therapeutic effect. Further invivo studies using complementary CKD models, direct assessment of FGFR4 signaling, and vehicle formulations without intrinsic vascular activity are required to validate these findings.

Introduction

Chronic kidney disease (CKD) is a major public health problem affecting millions of people worldwide, characterized by a progressive and irreversible loss of kidney function (1). Beyond affecting the kidneys, CKD also leads to a complex multi-system disorder by disrupting the cardiovascular, endocrine, and immune systems (2, 3). While early stages are often asymptomatic, as the disease progresses, it is accompanied by hypertension, anemia, mineral metabolism disorders, and cardiovascular disease, and patients progressing to end-stage renal failure require serious treatments such as dialysis or kidney transplantation to survive (4, 5). In CKD patients, cardiovascular diseases are the leading cause of mortality, stemming not only from traditional risk factors but also from mechanisms closely associated with CKD, such as uremic toxins, chronic inflammation, and mineral imbalance (6-9). Among the relevant molecular mediators, fibroblast growth factor-23 (FGF-23), whose circulating levels increase significantly as renal function decreases, stands out as an important regulator of phosphate and vitamin D metabolism (10, 11). Beyond its physiological role, high blood FGF-23 levels contribute to left ventricular hypertrophy, vascular calcification, endothelial dysfunction, and inflammation, demonstrating its role as an active trigger of cardiovascular pathology associated with CKD (12, 13).

FGF-23 acts through four fibroblast growth factor receptors (FGFR1 to FGFR4) that are differentially expressed across tissues and linked to different pathological processes in CKD (10). FGFR4 is associated with left ventricular hypertrophy and uremic cardiomyopathy, while FGFR1, FGFR2, and FGFR3 are more involved in vascular calcification, inflammatory responses, and impaired bone mineralization (14-17). Importantly, FGFR4 is expressed in human vascular smooth muscle and endothelial cells, and FGF-23 activation of FGFR4 has been shown to trigger inflammatory signaling via the phospholipase C gamma (PLCγ)/ nuclear factor of activated T cells (NFAT) pathway and contribute to vascular remodeling (18-20).

Chronic low-grade inflammation is one of the conditions associated with CKD and is sustained by uremia, oxidative stress, dyslipidemia, infection, and impaired clearance of inflammatory mediators (21, 22). Low-grade systemic inflammation is closely linked to cardiovascular disease and is characterized by persistently elevated cytokines such as TNF-α and IL-1 (23-26). Although FGFR4 activation is known to be associated with inflammatory cytokine production, its specific contribution to inflammation and cardiovascular dysfunction associated with chronic kidney disease has not been sufficiently defined (27-29).

Therefore, this study aimed to investigate the role of FGF-23 and FGFR4 in cardiovascular and renal complications of chronic kidney disease and to evaluate whether selective FGFR4 inhibition can modulate inflammation, endothelial dysfunction, and disease-related biomarkers in an experimental model of chronic kidney disease. A better understanding of this pathway may clarify its contribution to CKD-associated vascular dysfunction and its potential as a therapeutic target.

Materials and Methods

Experimental animals and study design. Male Sprague-Dawley rats aged 10-12 weeks were used in this study, a strain commonly employed in CKD and cardiovascular research (30, 31). Animals were housed under controlled environmental conditions (22±2°C, 55±5% relative humidity, 12 h light-dark cycle) with free access to standard laboratory chow and water.

The study consisted of four experimental groups. Five animals were assigned to the healthy control and CKD-control groups, whereas nine animals were allocated to the vehicle and treatment groups due to the length of the protocol and reported mortality rates in similar CKD models (32, 33). The number of animals in the groups was determined in accordance with the recommendations of the Institutional Animal Ethics Committee. One animal was lost in the vehicle group during the experimental period, resulting in a final sample size of 27 animals. The experimental protocol lasted 43 days. Because organ bath experiments could be performed on a maximum of four animals per day, sacrifices were conducted on consecutive days following completion of the protocol.

Ethics. This study was designed as a randomized, prospective, controlled experimental study and was conducted in accordance with the ethical approval granted by the Gülhane Animal Experiments Local Ethics Committee of the University of Health Sciences (approval no. 24/13). All experimental procedures were carried out in compliance with the Regulation on the Welfare and Protection of Animals Used for Experimental and Other Scientific Purposes (Official Gazette, 13 December 2011, No. 28141).

Material support. In this study, the folic acid used to create the CKD model was supplied by Abdi İbrahim Pharmaceutical Industry (Istanbul, Turkey). The company had no role in the study design, data collection, data analysis, interpretation of the findings, manuscript preparation, or the decision to publish.

CKD induction and pharmacological treatments. Chronic kidney disease was induced in all groups except healthy control by a single intraperitoneal injection of folic acid (250 mg/kg; DSM, Basel, Switzerland). Folic acid was freshly dissolved in 0.2 M NaHCO₃ immediately before administration and injected at a volume of 2 ml/kg. CKD establishment was verified on day 28 by measuring serum creatinine levels from tail vein blood samples (34, 35).

Beginning on day 28, the vehicle group received intraperitoneal injections of 10% dimethyl sulfoxide (DMSO) in injectable water (2.5 ml/kg) once daily for 14 days. The treatment group received the selective FGFR4 inhibitor roblitinib (FGF401) (Macklin, Shanghai, China) at a dose of 0.63 mg/kg intraperitoneally for 14 days. The administered dose was determined based on previously reported pharmacokinetic data and interspecies dose conversion approaches (36-40).

Tissue collection. On day 43, animals were anesthetized with ketamine (90 mg/kg) and xylazine (10 mg/kg). Blood samples were collected via cardiac puncture under general anesthesia. The right and left carotid arteries, thoracic aorta, and both kidneys were carefully removed for further analysis. Blood samples were centrifuged to obtain serum and stored for biochemical analysis. The obtained carotid artery, thoracic aorta, and kidney tissues were used for histopathological evaluation and vascular reactivity experiments.

Organ bath experiments. Thoracic aortas were isolated and examined in vitro using an organ bath system. Excised aortas were placed in cold Krebs-Henseleit solution, cleaned of surrounding connective tissue and fat without damaging the endothelium, and cut into approximately 5 mm rings. Two rings from each animal were mounted in a four-channel organ bath system (MAY IOBS 99, COMMAT Communication Co., Ankara, Turkey) containing 20 ml Krebs-Henseleit solution per chamber. The bath temperature was maintained at 37 °C and continuously aerated with a gas mixture of 95% O₂ and 5% CO₂. After a 45-min equilibration period under a resting tension below 1.5 g, with the bathing solution replaced every 15 min, tissue viability was confirmed using 80 mM KCl. Cumulative concentration-response curves to phenylephrine (10⁻⁷–3×10⁻⁵ M) were obtained to evaluate contractile responses. Endothelium-dependent relaxation was assessed by cumulative addition of carbachol (10⁻⁸–3x10⁻⁵ M) in rings precontracted with phenylephrine (10⁻⁶ M), whereas endothelium-independent relaxation was evaluated using sodium nitroprusside (10⁻⁸–3x10⁻⁵ M). Relaxation responses were expressed as a percentage of phenylephrine-induced contraction. Isometric tension was recorded using force displacement transducers (FDT 10-A, May IOBS 99; COMMAT Communication Co.) connected to a data acquisition system (MP36, BIOPAC Systems Inc., Santa Barbara, CA, USA) with BSL PRO v3.6.7 software.

Histopathological analysis. At the end of the experimental period, kidney and carotid artery tissues were collected from all animals for histological evaluation. Kidney samples were used to assess renal fibrosis. Tissues were fixed in 10% neutral buffered formalin, incubated in 15% and 30% sucrose, and cryosectioned at 8 µm thickness using a cryostat with a 1/10 sampling strategy. Sections were stained with Masson’s trichrome and scanned using a fully motorized microscope (Leica DMI6000, Leica Microsystems, Wetzlar, Germany). Fibrosis was quantified using open-source image analysis software (41) based on mean intensity, percentage area, and a composite fibrosis index calculated as the multiplication of these two parameters. Carotid artery samples were analyzed for vascular calcification and medial thickness. After fixation and sucrose incubation, carotid tissues were cryosectioned (10 µm; 1/5 sampling) and stained with Von Kossa method. Calcification was evaluated as present or absent using the same image analysis software (41). Medial thickness measurements were performed under light microscopy (40× magnification) by averaging measurements obtained from 10 different points in each section.

Biochemical analysis. On day 28 following folic acid administration, blood samples were collected from the tail vein of the rats, and serum creatinine levels were measured using a spectrophotometric method. Animals showing elevated creatinine levels, confirming the establishment of renal injury, were included in the continuation of the experiment. At the end of the experimental period, serum creatinine levels were measured again from blood samples collected during euthanasia using an automated clinical chemistry analyzer (Mindray BS-400, Mindray Bio-Medical Electronics Co., Shenzhen, China). Blood samples and tissue extracts (aortic homogenates) collected during euthanasia were used to determine the levels of ADMA, TNF-α, IL-1β, and FGF-23 using rat-specific sandwich enzyme-linked immunosorbent assay kits (TNF-α: E0764Ra; FGF-23: E0170Ra; ADMA: EA0008Ra; IL-1β: E0119Ra; BT LAB Bioassay Technology Laboratory, Shanghai, China) according to the manufacturers’ instructions. Absorbance values were measured using a microplate reader (Bio-Tek ELx800, BioTek Instruments, Winooski, Vermont, USA), and washing steps were performed with an automatic microplate washer (Bio-Tek ELx50, BioTek Instruments).

Statistical analysis. Statistical analyses were performed using GraphPad Prism version 8.0 (GraphPad Software, San Diego, CA, USA). Normality of continuous variables was assessed using the Shapiro-Wilk test. For independent single-time-point outcomes, normally distributed data were analyzed using one-way analysis of variance (ANOVA) followed by an appropriate multiple comparisons test, whereas non-normally distributed data were analyzed using the Kruskal-Wallis test followed by Dunn’s multiple comparisons test. Normally distributed data are presented as mean±standard error of the mean (SEM), whereas non-normally distributed data are presented as median with interquartile range (IQR).

Body weight measurements obtained repeatedly from the same animals over time were analyzed using two-way repeated-measures ANOVA to assess the effects of group and time and their interaction, followed by appropriate multiple comparisons testing. Serum creatinine levels measured in the same animals on days 28 and 43 were analyzed using two-way repeated-measures ANOVA to evaluate the effects of group, time, and the group × time interaction, followed by Bonferroni’s multiple comparisons test.

Organ-bath concentration-response curves for phenylephrine (Phe), carbachol (CCh), and sodium nitroprusside (SNP) were analyzed using two-way repeated-measures ANOVA, with experimental group and agonist concentration as factors, followed by Tukey’s multiple comparisons test. EC₅₀ and pD₂ values were additionally derived from the concentration-response curves by nonlinear regression analysis, with pD₂ calculated as −log₁₀(EC₅₀). A p-value <0.05 was considered statistically significant.

Results

Changes in body and relative kidney weight (kidney weight/body weight ratio). Body weights were meticulously monitored on a weekly basis throughout the experimental period (Figure 1A). The initial measurements obtained on day 0 were designated as week 0, and subsequent measurements were recorded at 7-day intervals. There was a significant increase in body weight in all groups until week 3 (p<0.05). At week 4, the increase in body weight persisted in the healthy control group, whereas a decline was observed in the CKD-control and vehicle groups compared with week 3 (p<0.05). Prior to sacrifice, there was a slight but non-significant increase in body weight in the healthy control and treatment groups relative to week 4 (p>0.05). In contrast, body weight continued to decrease in the CKD-control and vehicle groups. This reduction was not statistically significant in the CKD-control group (p>0.05) but reached statistical significance in the vehicle group (p<0.05).

Figure 1.

Changes in body weight and relative kidney weight across experimental groups. (A) Body weight changes during the experimental period. S denotes the measurement obtained at sacrifice. Data are presented as mean±SEM. *p<0.05, **p<0.01, and ***p<0.001 versus the previous time point (B) Relative kidney weight (kidney weight/body weight ratio) in the experimental groups. Data are presented as median (IQR). Sample sizes were n=5, 5, 9, and 8 for the Healthy control, CKD-control, Treatment, and Vehicle groups, respectively. *p<0.05 versus the Healthy control group and ^#p<0.05 versus the CKD-control group. SEM: Standard error of the mean; IQR: interquartile range.

Following sacrifice, both kidneys were excised, carefully cleared of surrounding connective tissues, and weighed. Relative kidney weight (kidney weight/body weight ratio) was calculated for each animal (Figure 1B). The CKD-control and vehicle groups exhibited significantly higher relative kidney weights compared with the healthy control group (p<0.05). Notably, the treatment group showed significantly lower relative kidney weights compared with the CKD-control group (p<0.05).

The viability of isolated rat aortic rings was confirmed by stimulation with 80 mM potassium chloride (KCl), which induces smooth muscle contraction via membrane depolarization (42, 43). KCl-induced contractions did not differ between the healthy control and CKD-control groups, whereas significant differences were observed between the CKD-control group and both the treatment and vehicle groups (Figure 2A). Contractile responses to cumulative phenylephrine concentrations (10⁻⁷–3×10⁻⁵ M) were significantly reduced in the CKD-control group compared with the healthy control group at 10⁻⁵ M and 3×10⁻⁵ M. In contrast, the treatment and vehicle groups showed significantly greater contractile responses than the CKD-control group from 3×10⁻⁶ M onward (p<0.05) (Figure 2B). Endothelium-dependent relaxation, assessed by cumulative carbachol concentrations in phenylephrine-precontracted rings, was significantly attenuated in the CKD-control group compared with the healthy control group from 10⁻⁶ M onward (p<0.05), whereas relaxation responses increased significantly in the vehicle group from 10⁻⁶ M and in both the vehicle and treatment groups from 3×10⁻⁶ M onward compared with the CKD-control group (Figure 2C). Similarly, endothelium-independent relaxation induced by sodium nitroprusside (SNP) was significantly reduced in the CKD-control group between 3×10⁻⁸ M and 3×10⁻⁶ M compared with the healthy control group (p<0.05). In contrast, relaxation responses increased significantly in the treatment group at 10⁻⁶ M and 3×10⁻⁶ M, and in the vehicle group at 3×10⁻⁵ M compared with the CKD-control group (Figure 2D).

Figure 2.

Vascular reactivity of isolated rat aortic rings across experimental groups. (A) Contractile responses to 80 mM KCl. (B) Concentration-response curves for Phe-induced contraction. (C) Concentration-response curves for CCh-induced endothelium-dependent relaxation. (D) Concentration-response curves for SNP-induced endothelium-independent relaxation. Contractile responses in panels A and B are expressed as absolute tension (mg), whereas relaxation responses in panels C and D are expressed as percentages of Phe-induced precontraction. Data are presented as mean±SEM. *p<0.05, **p<0.01, and ***p<0.001 versus the Healthy control group; ^#p<0.05, ^##p<0.01, and ^###p<0.001 versus the CKD-control group. KCl: Potassium chloride; Phe: phenylephrine; CCh: carbachol; SNP: sodium nitroprusside; CKD: chronic kidney disease; SEM: standard error of the mean.

In addition to the concentration-response curve analyses, EC₅₀ and pD₂ values were determined by nonlinear regression analysis to further characterize agonist potency. The calculated EC₅₀ and pD₂ values for Phe, CCh, and SNP across the experimental groups are presented in Table I. For CCh, the fitted EC₅₀ values for the CKD-control and treatment groups exceeded the maximum experimentally tested concentration (3 × 10⁻⁵ M) and were therefore considered extrapolated estimates. As reliable EC₅₀ values could not be determined within the experimental concentration range, the corresponding EC₅₀ and pD₂ values were reported as “not reliably estimable.” Significant differences in the reliably estimable pharmacological parameters were observed between experimental groups for some responses.

Table I. EC₅₀ and pD₂ values obtained from the concentration–response curves.

EC₅₀ and pD₂ values for phenylephrine (Phe), carbachol (CCh), and sodium nitroprusside (SNP) were determined by nonlinear regression analysis. pD₂ was calculated as −log₁₀(EC₅₀). *p<0.05 vs. Healthy control group; ^#p<0.05 vs. CKD-control group. For CCh, EC₅₀and pD₂ values in the CKD-control and treatment groups were not reliably estimable because the fitted EC₅₀ values exceeded the highest concentration tested (3×10⁻⁵ M).

Biochemical parameters. Serum creatinine levels were measured 28 days after folic acid administration using tail vein blood samples to confirm CKD model induction. The mean creatinine level in the healthy control group was 0.59±0.03 mg/dl, whereas higher levels were observed in the CKD-control (0.79±0.15 mg/dl), treatment (0.73±0.05 mg/dl), and vehicle (0.77±0.13 mg/dl) groups. All CKD-induced groups showed significantly higher creatinine levels than the healthy control group (p<0.05), exceeding the reference range reported for rats (0.58±0.24 mg/dl) (44), confirming successful CKD induction.

At the end of the in vivo experimental period (day 43), serum creatinine levels were re-evaluated at sacrifice. Creatinine levels showed an increasing trend in the CKD-control group and a decreasing trend in the treatment and vehicle groups compared with day 28; however, these changes were not statistically significant (p>0.05). No significant differences were observed between days 28 and 43 within any experimental group (Figure 3).

Figure 3.

Changes in serum creatinine levels during the experimental period. Serum creatinine levels measured on days 28 and 43 in the experimental groups are shown as absolute concentrations (mg/dl). Data are presented as mean±SEM (n=5, 5, 9, and 8 for the Healthy control, CKD-control, Treatment, and Vehicle groups, respectively). *p<0.05 versus the Healthy control group at day 28. No significant differences were observed between days 28 and 43 within the experimental groups (p>0.05). SEM: Standard error of the mean; CKD: chronic kidney disease.

At the end of the experiment, serum FGF-23 levels measured from blood samples collected at sacrifice did not differ significantly among the experimental groups (p>0.05), although a non-significant decrease was observed in the treatment group. Similarly, TNF-α and IL-1β levels measured in both serum and aortic tissue homogenates showed no statistically significant differences among the groups (p>0.05). Consistent with these findings, ADMA levels measured in serum and aortic tissue homogenates were also comparable across all experimental groups, with no statistically significant differences detected (p>0.05). These findings are illustrated in Figure 4.

Figure 4.

Levels of biochemical markers in serum and aortic tissue across experimental groups. Serum FGF-23, ADMA, IL-1β, and TNF-α levels, as well as their levels in aortic tissue homogenates, are shown as absolute concentrations in the units indicated on the respective y-axes. Data are presented as median and IQR. No statistically significant differences were observed among the experimental groups (p>0.05) (n=5, 5, 9, and 8 for the Healthy control, CKD-control, Treatment, and Vehicle groups, respectively). FGF-23: Fibroblast growth factor-23; ADMA: asymmetric dimethylarginine; IL-1β: interleukin-1β; TNF-α: tumor necrosis factor-α; IQR: interquartile range; CKD: chronic kidney disease.

Histological findings. Quantitative analysis of renal sections revealed no significant differences among the experimental groups, indicating that renal fibrosis had not yet developed at this stage. Similarly, no evidence of vascular calcification was observed in carotid artery sections. Due to the extremely thin intima layer, thickness measurements were performed only on the media layer. Carotid medial thickness was significantly greater in the treatment group than in the CKD-control group (p<0.05), whereas no other significant between-group differences were detected. The results of renal fibrosis analysis are presented in Figure 5, and carotid medial thickness measurements are shown in Figure 6.

Figure 5.

Quantitative assessment of renal fibrosis across experimental groups. (A) Mean staining intensity. (B) Fibrosis area expressed as a percentage of the analyzed tissue area. (C) Fibrosis index calculated as fibrosis area × mean staining intensity. Data are presented as mean±SEM. No statistically significant differences were observed among the experimental groups (p>0.05) (n=5, 5, 9, and 8 for the Healthy control, CKD-control, Treatment, and Vehicle groups, respectively). SEM: Standard error of the mean; CKD: chronic kidney disease.

Figure 6.

Carotid artery media thickness across experimental groups. Carotid artery media thickness is shown as absolute values in micrometers (μm). Data are presented as mean±SEM. p<0.05 versus the CKD-control group (n=5, 5, 9, and 8 for the Healthy control, CKD-control, Treatment, and Vehicle groups, respectively). SEM: Standard error of the mean; CKD: chronic kidney disease.

Discussion

Various experimental models have been developed to study the pathophysiology of CKD, including subtotal nephrectomy, adenine diet, streptozotocin-induced diabetic nephropathy, ischemia-reperfusion injury, and folic acid administration (35, 45-48). In this study, the folic acid-induced model was preferred because high-dose folic acid can mimic CKD development by causing proximal tubule damage and tubulointerstitial changes (35, 49). In our study, the significant increase in serum creatinine levels in the CKD-induced groups compared to the healthy control group demonstrates the successful establishment of the model and is consistent with previous studies (35, 46, 50-52).

During the experiment, a tendency towards weight loss was observed in the CKD-induced groups after the third week, whereas this change was not significant in the treatment group. Similarly, relative kidney weight (kidney weight/body weight ratio) was found to be higher in the CKD-control and vehicle groups than in the healthy control group. These findings are consistent with weight loss and renal weight gain reported in experimental CKD models (53-57). However, histopathological examinations revealed no significant differences in renal fibrosis or vascular calcification. Carotid medial thickness was significantly greater in the treatment group than in the CKD-control group, whereas no other significant between-group differences were detected. The limited renal fibrotic and vascular calcification findings despite the increase in serum creatinine levels suggest that the renal injury had not yet progressed to an advanced fibrotic stage. Indeed, renal fibrosis is characterized by chronic inflammation, TGF-β1 activation, extracellular matrix accumulation, and fibrogenic processes involving epithelial-to-mesenchymal transition and usually becomes apparent in the later stages of the disease (58-63).

In CKD, a significant portion of mortality is due to cardiovascular disease, and this is associated not only with traditional risk factors but also with uremia-specific mechanisms (6, 64-67). In this context, FGF-23 has been associated with cardiac hypertrophy, vascular dysfunction, and increased mortality, in addition to its role in regulating mineral metabolism (11, 68, 69). It has been reported that FGF-23 acts through FGFR subtypes, and that FGFR4 may play a role in cardiovascular pathophysiology (10, 70, 71). Although no significant difference was found between the groups in terms of serum FGF-23 levels in our study, a non-significant decrease in FGF-23 was observed in the treatment group; this finding should not be interpreted as evidence of FGFR4 target engagement.

In vascular function analyses, it was shown that both endothelium-dependent and endothelium-independent relaxation responses were impaired in the CKD model created with folic acid. The decrease in endothelium-dependent relaxation assessed with carbachol and endothelium-independent relaxation assessed with SNP suggests that not only endothelial function, but also vascular smooth muscle responses are affected. Oxidative stress and inflammatory signaling are also important components of experimental renal injury, with alterations in ROS-related pathways, NF-κB-associated inflammatory signaling, and nitric oxide-mediated signal transduction reported in renal injury models (72). These findings are consistent with experimental studies reporting that FGF-23 may lead to endothelial dysfunction by reducing nitric oxide bioavailability (73). However, the increased relaxation responses observed in both the treatment and vehicle groups in our study raise concerns about the vascular effects of DMSO used as a solvent. Indeed, various studies have reported that DMSO can exhibit antioxidant, vasodilator, and anti-inflammatory properties and affect vascular tone via nitric oxide pathways and Rho-kinase (74-78). Therefore, the biological effect of DMSO constitutes a significant limitation in interpreting the observed vascular changes. FGFR4 activation has been shown to trigger hypertrophic, fibrotic, and inflammatory processes by stimulating signaling pathways such as PLCγ/NFAT, transient receptor potential canonical 6-Ca²⁺-reactive oxygen species, and glycogen synthase kinase 3 beta /β-catenin (13, 71, 79-81). Roblitinib (FGF401), used in this study, is a potent inhibitor exhibiting high selectivity for FGFR4 (36).

The vascular reactivity results obtained from our study show that functional impairment has already begun in the vascular bed, despite the “early stage” CKD picture indicated by biochemical markers. The significant decrease in both endothelium-dependent and endothelium-independent relaxation responses suggests that the pathology affects not only endothelial nitric oxide mechanisms but also the relaxation capacity of smooth muscle cells in the tunica media. While the absence of vascular calcification in histopathological examinations is consistent with early-stage CKD, the functional vascular alterations indicate that vascular dysfunction may precede pronounced structural remodeling. The fact that the improvement observed in the treatment group was not statistically different from the vehicle (DMSO) group suggests that the known antioxidant and free radical scavenging properties of DMSO may have partially compensated for this early-stage vascular stress. In the literature, thickening in the intima and media portions of the vascular structure, especially in the carotid artery, has been frequently described in CKD models (82, 83). However, degenerative vascular remodeling, characterized by medial thinning and loss of elastic fibers, has also been reported in some cardiovascular pathologies (84-86). In the present study, carotid medial thickness was significantly greater in the treatment group than in the CKD-control group, whereas no significant difference was observed between the healthy control and CKD-control groups. Therefore, the observed difference cannot be interpreted as evidence that CKD itself induced medial thinning or that roblitinib reversed a CKD-associated structural alteration. Moreover, because the treatment group did not differ significantly from the vehicle group, a specific effect of FGFR4 inhibition on carotid medial thickness cannot be established. Together with the absence of vascular calcification, these findings suggest that pronounced structural vascular remodeling was not consistently established in this model, despite the clear functional alterations observed in vascular reactivity.

Although no significant changes were detected in terms of inflammatory markers in our study, a tendency towards a decrease in TNF-α and aortic IL-1β levels was observed in the treatment group. However, the focus solely on FGFR4, the use of a single dose and duration protocol, and the limited sample size are the main limitations of the study. In conclusion, renal dysfunction and vascular disorders were observed in the folic acid-induced CKD model.

Study limitations. This study has several limitations. The relatively small and unequal group sizes, use of male animals only, and evaluation of a single roblitinib dose and treatment duration limit the generalizability of the findings. In addition, although folic acid administration produced a significant increase in serum creatinine and altered vascular reactivity, the absence of significant renal fibrosis, vascular calcification, inflammatory-marker changes, and increased circulating FGF-23 suggests that the model primarily represents early renal injury rather than advanced CKD. Thus, the findings should not be extrapolated to established chronic renal disease. Interpretation of FGFR4 inhibition is further limited by the vascular activity of the DMSO vehicle and the absence of direct measurements of FGFR4 expression, activation, downstream signaling, or roblitinib target engagement. Consequently, vascular changes in the treatment group cannot be attributed specifically to FGFR4 inhibition. Future studies using alternative vehicle formulations, complementary kidney-injury models, longer follow-up, direct assessment of FGFR4 signaling, and broader characterization of renal and mineral-metabolism markers are required.

Conclusion

Folic acid-induced renal injury was associated with early functional vascular impairment despite the absence of pronounced renal fibrosis, vascular calcification, or inflammatory alterations. Selective FGFR4 inhibition did not produce a clearly distinguishable therapeutic effect, as the vascular responses observed in the treatment group could not be consistently separated from those in the vehicle group. Further in vivo studies using complementary CKD models, longer treatment periods, direct assessment of FGFR4 signaling, and vehicle formulations without intrinsic vascular activity are warranted to validate these findings and clarify the contribution of FGFR4 to CKD-associated vascular dysfunction.

Conflicts of Interest

The Authors declare no conflict of interest in relation to this study.

Authors’ Contributions

ZR and IMV contributed to the conceptualization and methodology of the study, performed formal analyses, and wrote the original draft of the manuscript. ZR conducted the investigation, including animal experiments, sample collection, organ bath experiments, and data curation. EA performed, evaluated, and interpreted the biochemical analyses, with support from ÜİY. Histopathological analyses were carried out by BB and BK, with support from ZR. OAA and BAG provided support for in vivo animal experiments. IMV was responsible for funding acquisition. All authors reviewed and approved the final version of the manuscript.

Acknowledgements

The Authors gratefully acknowledge the support of the Scientific Research Projects Coordination Unit (BAP) of the University of Health Sciences. The authors would also like to thank Abdi Ibrahim Pharmaceutical Industry (Istanbul, Türkiye) for kindly providing folic acid used in this study.

Funding

This study was supported by the Scientific Research Projects Coordination Unit (BAP) of the University of Health Sciences under Project No. 2024/080. The study constitutes a part of the PhD thesis of the first author Zeliha RUMANLI conducted under the supervision of Prof. Dr. İsmail Mert VURAL. The funding body had no role in the study design, data collection, data analysis, interpretation of the results, manuscript preparation, or the decision to publish.

Artificial Intelligence (AI) Disclosure

During the preparation of this manuscript, a large language model (ChatGPT, OpenAI) was used solely for language editing and stylistic improvements in select paragraphs. No sections involving the generation, analysis, or interpretation of research data were produced by generative AI. All scientific content was created and verified by the authors. Furthermore, no figures or visual data were generated or modified using generative AI or machine learning-based image enhancement tools.