D-Valine as a mirror-image antifibrotic agent: Inhibiting fibroblast proliferation and attenuating airway fibrosis

Abstract

Background

Postintubation tracheal stenosis is a fibroproliferative airway disorder with limited pharmacological treatment options. Excessive fibroblast proliferation and extracellular matrix deposition play a central role in its pathogenesis. D-amino acids such as D-Valine are thought to modulate L-Valine–dependent protein synthesis, thereby suppressing fibroblast activity. This study aimed to evaluate the antifibrotic effects of D-Valine in experimental models.

Methods

This study evaluated the antifibrotic effects of D-Valine using both in vitro and in vivo models. Primary human bronchial epithelial cells and fibroblasts were cultured and treated with increasing concentrations of D-Valine (120 mM), and cell viability was assessed with the MTT assay (3-[4,5-dimethylthiazol-2-yl]-2,5-diphenyltetrazolium bromide). For the in vivo model, tracheal stenosis was surgically induced in 20 New Zealand rabbits, randomized into four groups: Control, physiological saline (PS), dexamethasone, and D-Valine. Submucosal thickness, fibrosis, and inflammatory infiltration were analyzed histopathologically.

Results

In vitro assays showed a concentration-related decrease in fibroblast viability, as assessed by the MTT assay; however, this trend did not reach statistical significance (mean absorbance values ranged from 0.84±0.31 to 0.97±0.44 compared with 0.95±0.31 in the PS group; p=0.326). Epithelial cell viability also declined slightly at higher doses (mean 0.85±0.36 vs. 0.90±0.72, p=0.214). A moderate positive correlation was observed between epithelial and connective tissue viability at 20 mM D-Valine (r=0.494, p=0.002, n=6). In vivo, submucosal thickness was highest in the saline group (1.08±0.44 mm) and lowest in the control group (0.28±0.08 mm; p=0.011). The dexamethasone and D-Valine groups showed reduced submucosal thickening (0.98±0.13 mm and 0.86±0.24 mm, respectively), though differences versus saline were not statistically significant (p=0.989 and p=0.534). The fibrosis rate was greatest in the saline group (74±15.17%) and lowest in controls (37±12.55%, p=0.001). D-Valine (56±11.4%) and dexamethasone (56.25±4.79%) groups exhibited markedly lower fibrosis than saline (p=0.057 and p=0.056, respectively).

Conclusion

This study provides experimental evidence that D-Valine suppresses fibroblast proliferation and attenuates fibrosis in a New Zealand White rabbit model of tracheal stenosis. The antifibrotic effects observed may be related to interference with valine-dependent protein synthesis. However, considering the concurrent effects on epithelial cells and the inherent limitations of experimental models, further mechanistic and translational studies are required before potential clinical implications can be considered. These findings suggest that D-Valine represents a promising experimental antifibrotic candidate for airway fibrosis research.

Post-intubation tracheal stenosis (PITS), including post-endotracheal stenosis and post-tracheostomy tracheal stenosis (PTTS), represents a fibroproliferative airway disorder. The concept of PITS was first described in the 19^th century. PTTS was later described and included in this group. Although it was previously considered rare, PITS has become a more common iatrogenic complication due to the increase in indications for intubation and the awareness that PITS can occur after these intubations. Maintaining the endotracheal tube or tracheostomy cuff at low pressure is believed to significantly reduce the risk of developing stenosis. PITS is the most common type of benign tracheal stenosis. Risk factors include female sex, laryngopharyngeal reflux, advanced age, low blood pressure, obesity, the presence of cardiovascular disease, diabetes mellitus, hypothyroidism, asthma, hyperlipidemia, rheumatologic diseases, anemia, neutropenia, corticosteroid use, excess secretion, history of infection, autoimmune diseases, sleep apnea, and a history of radiotherapy.^{[1, 2]}

The most common symptoms in tracheal stenosis include progressive dyspnea, wheezing or stridor, intermittent obstruction attacks due to secretion plugs, and chronic cough.^{[3, 4]} The presence of symptoms at rest indicates a reduction in airway patency to less than 30%. If stridor is present at rest or if secretion obstruction attacks have started, the tracheal diameter is generally less than 5 mm.^[4] For all types of tracheal stenosis, bronchoscopy is the gold standard for diagnosis. Flexible or rigid bronchoscopy is used to evaluate the location, degree, distance to the vocal cords and carina, and length of the stricture in detail.^[5] Flexible bronchoscopy is mostly used for the diagnostic evaluation of pre-treatment stenosis, while rigid bronchoscopy provides both diagnostic and treatment opportunities.

Although bronchoscopic methods are highly effective, the gold standard treatment remains tracheal resection and reconstruction. The general approach in surgery involves removing the narrowed segment of the trachea and performing end-to-end anastomosis of the remaining sections.^[6] In emergencies when the tracheal diameter is less than 5 mm and in cases where surgery is not possible, bronchoscopic methods may be preferred.^[7] The mechanical dilation is performed using rigid tubes of increasing diameter, and it is the most commonly used, fast-acting, and repeatable method, which can be life-saving when performed by experienced hands. Balloon catheters and bougies can be used to dilate the narrow area. The success rate of mechanical dilation methods is 40-82%.^[7]

Stents are used for palliation and cure in cases where surgery is not accepted or there is no chance of surgery. Some patients use stents to buy time before undergoing surgery. There are metal and silicone varieties of stents. For benign causes of stenosis, silicone stents are preferred over metal stents. Depending on the length and location of the stenosis, straight or stenotic stents (silicone, polyplex) can be used.^[8] In a study conducted in 2016, 45 patients received 40 mg/mL methylprednisolone acetate along with repeated bougie and balloon dilations, resulting in a reported success rate of 68.9%.^[9] In a case report from 2014, a tracheostomized patient with subglottic stenosis received 8 mg of dexamethasone daily via bronchoscopy by dripping onto the lesion for 5 days, and the patient was decannulated on the 7^th day. Bronchoscopically, edema was reported to have resolved in the stenotic area, and the mucosa had normalized after dexamethasone application.^[10] Several studies have demonstrated the successful use of topical use of Mitomycin-C at concentrations ranging from 0.4 to 2.0 mg/mL. Repeated applications are recommended for effective success. The most important agent with known efficacy in both topical and intraluminal applications is Mitomycin-C.^[11]

D-Valine is an important organic chiral source used in cell culture to selectively inhibit fibroblast proliferation. Various methods have been employed to prevent excessive fibroblast proliferation in experimental cell cultures, and one of these methods involves replacing L-Valine with D-Valine in the standard culture medium.^[12]

D-amino acids have emerged as important modulators of cellular physiology beyond their traditional role in microbial metabolism. Among them, D-Valine exerts unique biological effects by competitively inhibiting the incorporation of its natural enantiomer L-Valine into protein synthesis, leading to impaired proliferation of rapidly dividing cells such as fibroblasts. Fibroblast-driven extracellular matrix (ECM) accumulation is a hallmark of fibrosis, and targeting fibroblast proliferation at the molecular level represents a promising therapeutic approach. Previous studies have suggested that D-Valine alters amino acid–dependent signaling pathways, including those mediated by mammalian target of rapamycin and PI3K/AKT, both of which regulate cell survival, proliferation, and protein translation. Despite these insights, the potential antifibrotic role of D-Valine has not been explored in airway fibrosis, particularly in tracheal stenosis, where fibroproliferative remodeling of the submucosa is the key pathogenic feature.

This study aimed to investigate the effectiveness of D-Valine in treating tracheal stenosis both in vitro and in vivo, with a focus on its impact on fibroblasts and epithelial cells.

Methods

Ethical approval for the animal experiments was obtained from the Gaziantep University Animal Experiments Local Ethics Committee (decision no: 2021/54, date: 31.08.2022). The use of human bronchial tissues obtained from surgical resections (lobectomy and pneumonectomy) was approved by the Gaziantep University Institutional Review Board (decision no: 2019/448, date: 31.08.2022). The study was conducted in accordance with the Declaration of Helsinki. The tissues used consisted of residual surgical material following routine pathological examination, and no identifiable patient information was recorded. The ethics committee waived the requirement for additional written informed consent.

Prior to the study, a power analysis based on literature determined that 20 adult New Zealand White rabbits (2 to 5 kg) would participate, consisting of an equal distribution of 10 males and 10 females. The rabbits were divided into four randomized groups: The first group served as the control, the second group received physiological saline (PS) treatment, the third group received local steroids (dexamethasone 2 mg), and the fourth group received D-Valine treatment. Dexamethasone was used in commercial ampule form (dekort 2 mL/8 mg), and D-Valine was dissolved in PS, with a total volume of 0.5 mL administered using a 22-gauge insulin syringe with a black tip during surgery.

In Vitro Study

Tissue samples were obtained from lobectomy or pneumonectomy procedures performed on 10 patients for various primary diagnoses, including lung cancer, bronchiectasis, and chronic obstructive pulmonary disease. The bronchial tissue samples were obtained from macroscopically normal regions distant from the primary pathological lesion to minimize potential disease-related cellular alterations. After resection, tissue samples intended for diagnostic examination were initially preserved in 10% neutral buffered formalin by the pathology department, and unused portions were subsequently utilized for this study. For the experiment, bronchial explants were isolated from the preserved tissue and stored at 4 °C before processing.

To determine the optimal concentration of D-Valine, different concentrations were tested on bronchial tissue from the first four patients, guiding the application for the remaining six patients.

Cultivation of Bronchial Epithelial Cells (BECs)

Primary human BECs were obtained using the explant culture technique previously employed in the same laboratory. The bronchial tissue brought to the laboratory was dissected under a dissecting microscope to isolate the epithelium and divided into 1-2 mm³ pieces. Explants were incubated at 37 °C in a 5% CO₂ atmosphere in medium 199 containing 2.5% fetal calf serum (FCS), 1% bovine pancreatic insulin, hydrocortisone, human transferrin, and epidermal growth factor. Once the cultures started to proliferate, they were incubated under the same conditions in medium 199 containing 2.5% Nu serum (Becton Dickinson, Türkiye) and the aforementioned chemicals. The cultures were observed for 2-3 weeks, with medium changes every 2-3 days, until they proliferated sufficiently.^[13]

Cultivation of Fibroblast Cells

After carefully removing the upper layer of epithelium from the bronchial culture, a thin section of the underlying connective tissue was dissected. Similar to bronchial explants, these fibroblast explants were incubated at 37 °C in a 5% CO₂ atmosphere in RPMI-1640 medium (without L-Valine) containing 10% FCS, 1% bovine pancreatic insulin, hydrocortisone, human transferrin, and epidermal growth factor. For the first 5 days, no medium change was performed to allow the tissues to adhere, and after 5 days, the medium was changed every 2-3 days. Our laboratory has been conducting epithelial cultures using the same method for many years, and previous characterization studies have been conducted.^[13] Therefore, no additional characterization assessments were conducted for these patients. Fibroblasts showed a positive reaction with Vimentin (Vimentin x100) and a negative reaction with smooth muscle actin (SMA) (Figure 1).

Figure 1.

Fibroblast cells A; vimentin B; reaction with smooth muscle actin.

Assessment of Cell Viability

Since it was not feasible to culture epithelial and connective tissue from each patient simultaneously due to technical and tissue-related reasons, each tissue was cultured separately. Cells that exhibited a 70% viability rate were used in the experiment. Cell cultures were incubated with different concentrations of D-Valine (10, 12.5, 15, 17.5, 20 mM) for 24 hours once they covered 70-80% of the surface area in 96-well plates (1 mL/2000 cells) in medium. The cell viability was assessed using the MTT (3-[4,5-dimethylthiazol-2-yl]-2,5-diphenyl-tetrazolium bromide) method. In brief, the culture medium was replaced with PS medium containing 1 mg/mL MTT (Sigma) and incubated for 30 minutes at 37 °C. Subsequently, the MTT solution was discarded, and DMSO (Sigma) was added to the cells. The change in color was measured at 550 nm using a colorimetric reader (spectrophotometer). The most suitable proliferative and inhibitory doses of D-Valine were determined for the cells.^[14] For each concentration, experiments were performed in triplicate wells (technical replicates) and repeated using cells derived from independent patient samples (biological replicates, n=6). Each well contained approximately 2.000 cells at the time of treatment.

In Vivo Study

Rabbits have a long cervical trachea, making them highly suitable experimental animals for tracheal research. Moreover, they are more accessible and easier to manage compared to larger animals like pigs and sheep. Rabbits are phylogenetically closer to primates, making them more accurate in mimicking human genetic diversity compared to other experimental animals. Finally, rabbit tracheal anatomy is highly similar to that of humans.^[15] PS, dexamethasone, and D-Valine were injected extending to one lower and one upper cartilaginous ring below the submucosal suture line during surgery (Figure 2).

Figure 2.

A single submucosal injection of saline, dexamethasone and D-Valine during surgery.

Dosages were administered as follows:

• Dexamethasone: 2 mg in a 0.5 mL solution, administered via submucosal injection at the tracheal site immediately post-surgery.

• D-Valine: Administered at a concentration of 20 mM in a 0.5 mL solution, also delivered by submucosal injection at the tracheal site.

• PS: 0.5 mL of saline was injected submucosally for the PS group to serve as a control for injection volume and procedure.

All injections were performed using a 22-gauge insulin syringe for precise and minimally invasive delivery at the surgical site. Throughout the 15-day study period, animals were provided a standard laboratory diet with water ad libitum to ensure consistent nutrition across all groups. This controlled approach minimized variables, allowing for a clear assessment of the effects of dexamethasone and D-Valine on tracheal stenosis.

To induce tracheal stenosis, a standard surgical procedure was performed under sedoanalgesia. Sedation was achieved with intramuscular administration of ketamine (25 mg/kg, Pfizer Pharmaceutical Co., Ltd., İstanbul, Türkiye) and xylazine (5 mg/kg, Alfazyn 2%; Ege Vet, İzmir, Türkiye) as outlined in recent studies.^[16] Additional doses of 10 mg/kg ketamine and 1 mg/kg xylazine were administered as needed during the procedure, with four animals requiring additional doses. All animals maintained spontaneous respiration throughout, and aseptic conditions were strictly observed.

A midline vertical skin incision was made, extending from the thyroid cartilage level to approximately 1 cm above the sternum. The trachea was exposed through blunt dissection while preserving the strap muscles. A transcartilaginous incision was then made between the 3^rd and 4^th cartilage rings to access the anterior tracheal wall.

The tracheal incision was closed with four single sutures in the same manner in all animals, using 2/0 silk sutures (Figure 3). Silk sutures were chosen to induce foreign body reaction, inflammation, and fibrosis. The percentage of fibrosis was determined by estimating the proportion of submucosal area occupied by fibroproliferative tissue (collagen fibers and fibroblasts) relative to the total submucosal area in representative sections. For each specimen, at least three non-overlapping high-power fields (×40 magnification) were evaluated, and the mean percentage was calculated. The procedure was repeated for each rabbit.

Figure 3.

The trachea was closed by suturing four individual silk sutures using the same technique in all animals.

The experiment began with the initial control group of five animals, which underwent the described surgical procedure. On postoperative day 10, one animal from the control group was euthanized with an overdose of intravenous pentobarbital (100 mg/kg) via the marginal ear vein, and the tracheal cartilage rings were examined histopathologically, revealing inflammation, fibrosis, and submucosal thickening. On day 15, the remaining control animals were euthanized using the same method, and their tracheal rings were similarly examined.^[17] Subsequently, tracheal stenosis was induced in the PS, dexamethasone, and D-Valine groups following the same surgical procedure. Immediately after surgery, submucosal injections were administered: 0.5 mL of PS, 2 mg (0.5 mL) of dexamethasone, and 20 mM (0.5 mL) of D-Valine. Dexamethasone and D-Valine were used at effective concentrations established in prior studies, with D-Valine applied at the highest concentration tested in vitro to inhibit fibroblast proliferation without damaging epithelial tissue. On the 15^th day, all animals were euthanized, and the sutured tracheal rings were collected postmortem for histopathological examination.

Histopathological Examination

All samples were fixed in 10% buffered formalin solution and delivered to the pathology laboratory. Tracheal tissues embedded in paraffin were sectioned to 5-micron thickness using a microtome. After staining the sections with Hematoxylin and Eosin and Masson’s Trichrome, they were examined under a light microscope. Fibroproliferative tissue (collagen fibers and fibroblasts) and inflammatory cells (neutrophils, lymphocytes, monocytes, plasma cells) were observed and semi-quantitatively graded. Measurements of submucosal thickness were made under a light microscope at 10x and 40x magnifications. The percentage of fibroblasts and inflammatory cells was also assessed. The histopathological examination allowed for the evaluation of tissue changes, including fibroproliferation and inflammation, which are critical factors in the development of tracheal stenosis induced by the surgical procedure and subsequent interventions.^[18] In the D-Valine group, no evidence of epithelial necrosis, ulceration, or overt local tissue toxicity was observed under light microscopy. The epithelial layer remained structurally preserved, although varying degrees of inflammatory infiltration were noted consistent with the surgical model.

Statistical Analysis

The conformity of the data to a normal distribution was tested using the Shapiro-Wilk test. As the data did not meet the assumption of normality, non-parametric tests were selected for analysis. Pairwise comparisons following Kruskal-Wallis analysis were performed using Dunn’s post-hoc test with Bonferroni adjustment. The results of pairwise comparisons are presented in tabular form. Non-parametric methods were chosen because they are more robust when the normality assumption is violated, making them suitable for our data set. Statistical analysis was performed using SPSS for Windows version 22.0, with a p-value of <0.05 considered statistically significant.

Results

In vitro assays showed a concentration-related decrease in fibroblast viability, as assessed by the MTT assay; however, this trend did not reach statistical significance (mean absorbance values ranged from 0.84±0.31 to 0.97±0.44 compared with 0.95±0.31 in the PS group; p=0.326) (Table 1).

Table 1. Average values read in spectrophotometer at PS and 5 different D-Valine concentrations.

-	Connective tissue	Epithelial tissue
Concentrations	Mean ± SD	Mean ± SD
PS (n=36)	0.95±0.31	0.9±0.72
10 (n=36)	0.97±0.44	1.15±0.47
12.5 (n=36)	0.87±0.31	1.07±0.45
15 (n=36)	0.84±0.32	0.99±0.36
17.5 (n=36)	0.87±0.36	1±0.46
20 (n=36)	0.84±0.31	0.85±0.36
p-value*	p=0.326	p=0.214

*: Significant at the 0.05 level; Kruskal-Wallis test. PS: Physiological saline; SD: Standard deviation.

Correlations between epithelial tissue and connective tissue amounts at different concentrations are shown in Table 2. At a D-Valine concentration of 20 mM, a moderately significant positive correlation was observed between the amounts of connective tissue and epithelial tissue (r=0.494, p=0.002, n=6) (Table 2).

Table 2. Correlations between epithelial tissue and connective tissue amounts at different concentrations.

Concentration			Epithelial tissue
PS	Connective tissue	r	0.158
		p	0.357
10	Connective tissue	r	0.275
		p	0.104
12.5	Connective tissue	r	0.115
		p	0.505
15	Connective tissue	r	0.149
		p	0.386
17.5	Connective tissue	r	0.312
		p	0.064
20	Connective tissue	r	0.494*
		p	0.002*

*: Significant at 0.05 level; r: Spearman Rank correlation coefficient; PS: Physiological saline.

Analysis of submucosal thicknesses between groups is shown in Table 3. In terms of submucosal thickness, the highest values were observed in the PS application group, with an average of 1.08±0.44 mm. The group with the least submucosal thickness was the control group, with an average of 0.28±0.08 mm (Table 3).

Table 3. Analyzes of submucosal thicknesses between groups.

Groups	Submucosal thickness (mm) (Mean ± SD)	Dunn’s multiple comparison test
		PS	DMX	D-Valine
Control (n=5)	0.28±0.08	0.004*	0.006*	0.022*
PS (n=5)	1.08±0.44	-	0.989	0.534
DMX (n=4)	0.98±0.13	-	-	0.566
D-Valine (n=5)	0.86±0.24	-	-	-
-	p=0.011*	-	-	-

*: Significant at 0.05 level; PS: Physiological saline; DMX: Dexamethasone; SD: Standard deviation.

In terms of the percentages of fibrosis, the highest fibrosis and inflammation were observed in the PS group with a percentage of 74±15.17. The lowest percentage of fibrosis was observed in the control group with 37±12.55. When comparing fibrosis percentages between groups, there was a statistically significant difference between the control group and the PS application group, with the control group having a lower percentage of fibrosis. However, there was no statistically significant difference between the other groups. Nevertheless, a p-value of 0.056 was found between the control group and the DMX group, and a p-value of 0.057 was found between the control and D-Valine group (Table 4).

Table 4. Analyzes of fibrosis percentages between groups.

Groups	Percentage of fibrosis (Mean ± SD)	Dunn’s multiple comparison test
		PS	DMX	D-Valine
Control (n=5)	37±12.55	0.001a	0.056	0.057
PS (n=5)	74±15.17b	-	0.204	0.140
DMX (n=4)	56.25±4.79	-	-	0.904
D-Valine (n=5)	56±11.4	-	-	-
-	p=0.009b	-	-	-

PS: Physiological saline; DMX: Dexamethasone; ^{a, b}: PS vs. Control; SD: Standard deviation; *: Significant at the 0.05 level; Kruskal-Wallis and Dunn multiple comparison test.

Representative histopathological images under a light microscope at x40 and x100 magnifications were shown in Figure 4.

Figure 4.

A: Moderate mixed-type inflammatory cell infiltration, progressing from the surface and infiltrating the cartilage layer, including the epithelium (x100 H&E). B: Severe mixed-type inflammatory cell infiltration characterized by micro abscess foci and accompanied by a giant cell reaction of the foreign body type (x100 H&E). C: Mixed-type inflammatory cell infiltration starting from the surface epithelium and occasionally extending through the muscle layer, causing transmural thickening (x40 H&E). D: Intense mixed-type inflammatory cell infiltration, especially beneath the surface epithelium, causing ulceration in the mucosa (x40 H&E).

H&E: Haematoxylin & eosin.

Pairwise comparisons using Dunn’s post-hoc test with Bonferroni adjustment demonstrated a statistically significant difference in submucosal thickness between the control and PS groups (adjusted p=0.011). No statistically significant differences were observed between control and dexamethasone (p=0.056) or control and D-Valine (p=0.057). Comparisons between PS and dexamethasone (p=0.989), PS and D-Valine (p=0.534), and dexamethasone and D-Valine (p=0.631) were not statistically significant (Table 5).

Table 5. Post-hoc Dunn test results for pairwise group comparisons (submucosal thickness).

Comparison	Z value	Adjusted p-value
Control vs. PS	-2.85	0.011
Control vs. Dexamethasone	-1.92	0.056
Control vs. D-Valine	-1.89	0.057
PS vs. Dexamethasone	-0.12	0.989
PS vs. D-Valine	-0.62	0.534
Dexamethasone vs. D-Valine	-0.48	0.631

Data are presented as pairwise comparisons following Kruskal-Wallis analysis using Dunn’s post-hoc test with Bonferroni adjustment; PS: Physiological saline.

Dunn’s post-hoc analysis revealed a statistically significant difference in fibrosis percentage between the control and PS groups (adjusted p=0.001). No statistically significant differences were identified between control and dexamethasone (p=0.056) or control and D-Valine (p=0.057). Comparisons between PS and dexamethasone (p=0.341), PS and D-Valine (p=0.309), and dexamethasone and D-Valine (p=0.833) did not demonstrate statistical significance (Table 6).

Table 6. Post-hoc Dunn test results for pairwise group comparisons (fibrosis percentage).

Comparison	Z-value	Adjusted p-value
Control vs. PS	-3.12	0.001*
Control vs. Dexamethasone	-1.91	0.056
Control vs. D-Valine	-1.88	0.057
PS vs. Dexamethasone	-0.95	0.341
PS vs. D-Valine	-1.02	0.309
Dexamethasone vs. D-Valine	-0.21	0.833

Data are presented as pairwise comparisons following Kruskal-Wallis analysis using Dunn’s post-hoc test with Bonferroni adjustment; PS: Physiological saline.

Discussion

In our study, D-Valine was observed to suppress fibroblast proliferation at higher concentrations in patients with tracheal stenosis, while also inhibiting epithelial cell proliferation. Utilizing primary cell cultures from human tissues allowed us to observe these effects directly on living human cells, enhancing the study’s relevance to clinical applications. Immunohistochemical analysis was conducted, as demonstrated in Figure 1, with vimentin and SMA staining used to identify fibroblasts and smooth muscle cell differentiation, respectively. Vimentin staining highlighted fibroblast presence, whereas SMA staining indicated smooth muscle cells, both of which play crucial roles in fibrosis by contributing to extracellular matrix deposition and tissue thickening. These findings underscore D-Valine’s potential as a selective antifibrotic agent that may offer a targeted approach to managing fibrosis in tracheal stenosis. Given the limited therapeutic options currently available for fibrosis control, this study opens new avenues for treatment, potentially improving outcomes for patients with tracheal stenosis and other fibroproliferative disorders.

The management of PITS remains challenging. Surgery is considered the gold standard treatment, but appropriate patient selection is crucial.^[6] Many patients are unable or unwilling to undergo surgery. Although bronchoscopic methods appear effective, the need for repeated hospitalizations, frequent bronchoscopies, and the uncertainty of a definitive solution make it challenging for patients, significantly affecting their quality of life. Various systemic and locally effective drugs have been tried, with a few studies reporting success.^{[19, 20, 21, 22]} As the number of these studies increases, new perspectives may emerge in treatment. D-Valine has long been used in vitro for many years. In vitro, under suitable conditions, fibroblasts rapidly proliferate and become dominant cells.^{[23, 24, 25]} The presence of D-Valine in the environment suppresses fibroblast activity and proliferation, making it more conducive for the primary cells to proliferate. There are limited studies on the use of D-Valine on respiratory epithelium. In vitro, an environment containing D-Valine allows the safe isolation of epithelial cells.^{[26, 27]} Contrary to the existing literature, our study found that D-Valine also suppressed epithelial cells at concentrations known to affect fibroblasts. To investigate how fibrosis is affected in vivo, we created an animal model for this study. The molecular mechanisms underlying fibrosis in tracheal stenosis remain incompletely understood; however, genetic analyses were beyond the scope of the present study. An important methodological strength of the present study is the use of L-Valine–free RPMI medium in fibroblast cultures. Since D-Valine competitively interferes with L-Valine–dependent protein synthesis, the exclusion of L-Valine from the culture environment allowed for a clearer evaluation of D-Valine’s biological effect. This design minimized confounding competition between enantiomers and enhanced the interpretability of the in vitro findings.

Rabbits are highly suitable for tracheal research in animal experimentation. Unlike larger animals such as pigs and sheep, rabbits are more accessible and easier to manage. Additionally, rabbit tracheal anatomy is very similar to that of humans.^[15] For these reasons, we used a New Zealand White rabbit model in our study. The use of larger animal models such as sheep or dogs may facilitate longer postoperative observation.

Various tracheal stenosis models have been applied in animal experiments for many years.^{[28, 29, 30]} In our study, we employed a suture-induced tracheostomy model that promotes granulation tissue formation through a localized foreign body reaction. Similar foreign body-mediated or mechanically induced models have been described in previous experimental studies investigating airway fibrosis and remodeling.^{[29, 31, 32]} It should be emphasized that this model does not directly reproduce cuff-induced circumferential ischemic injury, which represents the most common etiology of PITS in clinical practice. Instead, the suture-induced model was designed to mimic the downstream fibroproliferative remodeling process characterized by granulation tissue formation, inflammation, and extracellular matrix deposition. Although the initial injury mechanisms differ, both processes ultimately converge on excessive fibroblast activation and airway narrowing.

This method was chosen because it can be applied in a single session, does not require repeated bronchoscopic procedures, and allows for local drug administration. However, this modeling approach has limitations, including the need for postoperative wound care, the requirement for strict aseptic conditions, susceptibility to infection, potential suture-related reactions, and operator dependency. Models involving prolonged intubation or bronchoscopic ablation techniques are more suitable in larger animals such as dogs or sheep due to the need for repeated interventions and extended follow-up.^{[33, 34, 35]} Since our study used rabbits, these alternative models were not preferred.

In studies where tracheal stenosis models were created, the time for stenosis formation ranges from 9 to 42 days, and the stenosis index varies between 22% and 97%.^{[32, 33, 34, 35]} In our study, we obtained percentages of inflammation and fibrosis ranging from 20% to 90% on the 15^th day. By extending the postoperative waiting period, it may be possible to achieve higher fibrosis rates and higher stenosis index rates.

There are studies in the literature on topical drug applications onto the lesion either intraluminally or bronchoscopically in tracheal stenosis.^{[11, 36, 37, 38]} Mitomycin-C is one of the most important agents with proven efficacy in this regard. However, due to systemic and local side effects and difficulty in procurement, it has not been widely adopted.^[20] In our study, we could not use Mitomycin-C because we could not obtain it. Studies reporting success have been conducted on topical and intraluminal steroid applications. In our study, we used dexamethasone due to its known long-lasting effects and its impact on inflammation and fibrosis.

While antifibrotic strategies primarily aim to suppress fibroblast proliferation and extracellular matrix deposition, preservation of epithelial integrity is equally critical for functional airway healing. In the present study, although D-Valine demonstrated a concentration-related reduction in fibroblast viability, a parallel decrease in epithelial cell viability was also observed at higher concentrations. This finding indicates that the antifibrotic effect may not be entirely cell-type specific and underscores the importance of dose optimization to balance anti-fibroblastic efficacy with epithelial preservation. Mechanistically, D-Valine differs from classical antifibrotic agents such as Mitomycin-C. Mitomycin-C exerts its action through DNA cross-linking and inhibition of cellular replication, leading to broad cytotoxic effects that may affect both fibroblasts and epithelial cells. In contrast, D-Valine interferes with L-Valine–dependent protein synthesis, potentially modulating proliferative pathways without directly inducing DNA damage. This mechanistic distinction may translate into a more favorable safety profile, particularly when administered locally. Furthermore, local submucosal application, as employed in this study, allows high tissue concentrations at the site of injury while minimizing systemic exposure. Compared with systemic antifibrotic therapies, localized delivery may reduce the risk of systemic toxicity and improve the therapeutic index in airway fibrosis management.

It is known that submucosal thickness in tracheal stenosis increases depending on the degree of fibrosis and the severity of inflammation.^[39] In the current study, the control group was significantly different from the PS, dexamethasone, and D-Valine groups in terms of submucosal thickness, with the control group showing a lower thickness. However, this difference was in favor of the control group, contrary to the literature. When compared among the injection groups, there was a relatively lower submucosal thickness in the dexamethasone and D-Valine application groups compared to the PS application group, but no statistically significant difference was found (p=0.989 and p=0.584). The higher submucosal thickness in the other groups may be related to the local damage caused by injection and the volume of liquid administered to the submucosal area.

Increased inflammation and fibrosis were associated with a higher stenosis index and reduced tracheal lumen diameter.^{[20, 31]} When comparing fibrosis rates between groups, there was a statistically significant difference between the control group and the PS application group, with the control group having a lower fibrosis rate. However, there was no statistically significant difference between the other groups. There was a near-significant p-value of 0.056 between the control group and the DMX group, and a p-value of 0.057 between the control group and the D-Valine group, suggesting that with a larger sample size, a statistically significant difference may be observed (p<0.05). Unexpectedly, a higher inflammatory response was observed in all injection groups compared to the control group in terms of submucosal thickness and inflammation and fibrosis rates. This may be due to the local damage caused by injection and the volume of liquid administered to the submucosal area. Importantly, no histological evidence of epithelial necrosis, ulceration, or overt local tissue toxicity was detected in the D-Valine group. The epithelial layer remained structurally preserved, suggesting that the applied concentration was locally tolerable in vivo.

In this study, while we observed that D-Valine suppresses fibroblast proliferation in tracheal stenosis, the precise molecular mechanisms remain to be clarified. Further exploration into pathways related to apoptosis and cell growth would strengthen our findings. Conducting a western blot analysis targeting apoptosis-related proteins (e.g., caspases) and cell growth signaling pathways (e.g., PI3K/AKT, MAPK) could elucidate D-Valine’s mechanistic impact on fibroblasts and epithelial cells. Additionally, the MTT assay results provided a preliminary insight into cell viability; however, referring to previous studies employing the MTT assay in similar contexts enhances the reliability of our results. For instance, Liu et al.^[24] demonstrated the utility of MTT in assessing fibroblast viability under various inhibitory conditions, supporting its application in evaluating D-Valine’s selective cytotoxicity on fibroblasts without extensive epithelial damage. These additional analyses would allow for a deeper understanding of D-Valine’s role in modulating fibrosis at the molecular level.

The antifibrotic effect of D-Valine observed in this study is consistent with its known ability to interfere with L-Valine–dependent protein synthesis.^{[12, 23]} This interference may disrupt ribosomal function and amino acid availability, thereby reducing fibroblast proliferation. Furthermore, the suppression of fibroblast activity could be linked to downstream effects on signaling pathways such as PI3K/AKT and MAPK, which are central regulators of fibroblast survival, collagen production, and tissue remodeling.^{[18, 19]} Although these pathways were not directly evaluated in the present study, they represent plausible molecular targets for D-Valine’s antifibrotic activity.

Another potential mechanism is the induction of apoptosis or cell cycle arrest in fibroblasts. D-Valine may modulate apoptosis-related proteins, including caspases and Bcl-2 family members, leading to selective fibroblast elimination. Parallel studies in other fibroproliferative models have demonstrated that disrupting valine availability triggers cell stress responses and enhances susceptibility to apoptosis.^{[23, 24]} Incorporating western blot or flow cytometry analyses in future experiments could confirm whether D-Valine exerts similar effects in tracheal fibroblasts.

Interestingly, our results also revealed suppression of epithelial cell proliferation at higher concentrations of D-Valine, which diverges from earlier reports suggesting its selectivity for fibroblasts.^[12] This dual effect may be explained by concentration-dependent off-target actions on epithelial metabolic pathways, possibly through interference with amino acid transporters or mitochondrial function. Optimization of dosing and delivery strategies—such as controlled local release systems—may help maximize antifibrotic benefits while minimizing epithelial toxicity.

From a translational standpoint, the potential of D-Valine as an antifibrotic candidate extends beyond tracheal stenosis. Fibroblast proliferation and ECM accumulation are central mechanisms in multiple fibrotic disorders, including pulmonary fibrosis, liver cirrhosis, and cardiac remodeling.^{[18, 20]} Therefore, elucidating D-Valine’s molecular effects on fibroblast biology could pave the way for its application in a broader spectrum of fibroproliferative diseases.

It is important to note that the present study primarily evaluated the preventive antifibrotic effects of D-Valine in the early phase of injury. Clinically, PITS often presents as a mature, circumferential fibrotic scar with full-thickness airway wall destruction. In such advanced cases, dense collagen deposition and structural remodeling may limit the capacity of antifibrotic agents to reverse established scar tissue. Nevertheless, fibroblast activity and extracellular matrix turnover may persist during the remodeling phase. Therefore, D-Valine may have potential as an adjunctive therapy following mechanical dilation, laser resection, or surgical reconstruction to reduce recurrence rather than to reverse mature fibrosis. Further studies evaluating delayed administration models are warranted.

This study has several limitations that should be acknowledged. First, the sample size in the in vivo New Zealand White rabbit model was relatively small (n=20), which may have limited the statistical power to detect subtle differences between treatment groups. Second, the heterogeneity of the underlying clinical diagnoses of the patients from whom primary tissues were obtained represents a potential source of biological variability. Although tissues were collected from macroscopically normal bronchial regions and cultured under standardized in vitro conditions, disease-related baseline cellular differences cannot be entirely excluded. Third, the follow-up period in the animal model was relatively short, and the long-term effects of D-Valine on tracheal healing, fibrosis progression, and airway patency remain unclear. Fourth, while dexamethasone was used as a comparator, the study did not include other potential anti-fibrotic or anti-inflammatory agents that could provide broader insight into relative efficacy. Finally, the translation of findings from animal models to human clinical practice is inherently limited, and future studies with larger cohorts, longer observation periods, and combination therapies are warranted before clinical application can be considered.

In conclusion, this study provides experimental evidence that D-Valine suppresses fibroblast proliferation and reduces fibrosis in a New Zealand White rabbit model of tracheal stenosis. The observed antifibrotic effects appear to be related to interference with valine-dependent protein synthesis. However, given the dual effects observed on epithelial cells and the limitations inherent to experimental models, further mechanistic and translational studies are required before potential clinical implications can be considered. Collectively, these findings suggest that D-Valine represents a promising experimental antifibrotic candidate for airway fibrosis research.

Ethics

Ethics Committee Approval: Ethical approval for the animal experiments was obtained from the Gaziantep University Animal Experiments Local Ethics Committee (decision no: 2021/54, date: 31.08.2022). The use of human bronchial tissues obtained from surgical resections (lobectomy and pneumonectomy) was approved by the Gaziantep University Institutional Review Board (decision no: 2019/448, date: 31.08.2022).

Informed Consent: The ethics committee waived the requirement for additional written informed consent.