Methoxsalen deteriorative effects on the testicular parenchyma and testosterone synthesis gene expression in male rats

Abstract

Studies show that psoralens have adverse effects on the reproductive system. This study investigated the impact of methoxsalen in combination with UVA on testicular parenchyma and the expression level of testosterone synthesis pathway genes as well as sperm quality and quantity, in rats. 24 adult male rats were randomly divided into 4 groups: control, sham group receiving the same volume of the corn oil + ultraviolet radiation (UVA): 0.046 J/cm2, for 20 min, and experimental groups receiving 37.5, and 75 mg/kg doses of Methoxalen + UVA, respectively (MTX 37.5 and MTX 75). After 48 days’ serum samples were separated to measure the concentrations of LH, FSH, and Testosterone hormones. Spermatozoa were obtained from the epididymis for analysis of quality, quantity, morphology, viability and motility. Additionally, testicular tissue was subjected to stereological analysis to evaluate structural changes, and mRNA expression levels of STAR, Cyp11a1, Hsd17b3, Bcl-2, and Caspase-9 were measured. The results show detrimental effects of MTX in testosterone, sperm parameters, weight, and volume of the testis. Also, the volume of sperm and testosterone-producing cells such as Leydig cells decreased after treatment. The expression level of STAR, Cyp11a1, Hsd17b3, and Bcl-2 mRNAs were diminished, though the mRNA expression of caspase-9 and serum level of LH, and FSH in the high dose of the methoxsalen group increased compared to the sham group. Administering methoxsalen along with UVA can cause more severe damage to testicular parenchyma at high doses and reduce the mRNA expression level of genes involved in the testosterone synthesis pathway. Therefore, administering this drug is not recommended for people who are planning to become pregnant, although more studies are needed in this field.

Introduction

Infertility is defined as the inability to reproduce in living organisms, and according to the World Health Organization, it represents a failure of the reproductive system resulting in the inability to conceive after 12 months or more of consistent, unprotected sexual intercourse¹. Clinical and epidemiological studies indicate that male-related reproductive disorders contribute to approximately 50% of infertility cases². Overall, infertility affects nearly one in five couples worldwide, and male factors alone account for about 20–30% of all reproductive problems³. These abnormalities may result from a variety of genetic conditions—including chromosomal aneuploidies, structural rearrangements, microdeletions, and single-gene defects—or may be induced by acquired factors such as medications or environmental toxicants⁴. A broad spectrum of pharmaceutical agents has been documented for their deleterious, and at times irreversible, effects on male fertility⁵. To date, approximately 65 FDA-approved medications have been identified as having direct or indirect negative impacts on human spermatogenesis, including hormonal drugs, anti-inflammatory agents, and antineoplastic compounds⁶. Therefore, elucidating the cellular and molecular mechanisms underlying drug-induced impairment of male fertility is essential for developing strategies to minimize these adverse effects.

Psoralens are planar furocoumarin compounds classified as secondary metabolites found in various plant species such as Psoralea corylifolia and Ammi majus. Their concentrations in natural sources have occasionally been reported at levels that pose risks to both human and animal health, mainly through dermal contact or oral consumption of plant tissues containing psoralens^7,8. Among them, methoxsalen (8-methoxypsoralen) is a well-known photosensitizing agent widely used in the management of epidermal proliferative disorders (EPDs). Despite its therapeutic value, methoxsalen has demonstrated mutagenic and genotoxic properties in biological systems⁹. In clinical practice, methoxsalen is frequently combined with UVA radiation (320–400 nm) in PUVA therapy, particularly for treating psoriasis, a chronic immune-mediated inflammatory skin disease¹⁰. Experimental studies have shown that oral administration of methoxsalen to male rats significantly reduces reproductive capacity, decreases body and testicular weights, and alters testicular histoarchitecture. Its influence on the pituitary–gonadal axis leads to reductions in spermatogenic cells, sperm disorganization, and decreases in the numbers of spermatocytes, spermatids, and mature spermatozoa¹¹. Furthermore, methoxsalen exposure reduces circulating levels of sex hormones such as estrogen and progesterone¹². In male Wistar rats, significant decreases in testosterone concentration and sperm count have been documented following methoxsalen treatment, indicating its disruptive effects on the pituitary–testicular hormonal axis and consequent declines in gonadotropin and testosterone secretion⁷. In addition to the effects of methoxsalen itself, UVA irradiation has been associated with folate degradation—a crucial factor for fertility in both sexes—and reduced folate levels have been reported in the sera of psoriatic patients treated with PUVA¹³.

Androgens, particularly testosterone, are essential for the development and function of the male reproductive system, fertility maintenance, and manifestation of secondary sexual characteristics. Testosterone biosynthesis is a tightly regulated process involving multiple checkpoints, ranging from cholesterol transport into mitochondria to the negative-feedback regulation of androgen levels at the hypothalamic and pituitary levels¹⁴. Key steroidogenic genes—including Steroidogenic Acute Regulatory (Star), Cytochrome P450 family 11 subfamily A polypeptide 1 (Cyp11a1), and 17-beta hydroxysteroid dehydrogenase 3 (Hsd17b3)—play central roles in this pathway. Star mediates cholesterol transfer into the inner mitochondrial membrane; Cyp11a1 catalyzes the conversion of cholesterol to pregnenolone; and Hsd17b3 converts androstenedione to testosterone while also catalyzing the reverse reaction^15,16.

Several mechanistic studies have explored methoxsalen’s broader cellular effects. Methoxsalen alone has been shown to induce apoptosis in human gastric cancer cells through the upregulation of p53 and its downstream target p21¹⁷. PUVA therapy induces apoptosis primarily by forming DNA adducts, an effect particularly pronounced in highly proliferative malignant cells¹⁸. Arabzadeh et al. (2002) further demonstrated that methoxsalen interacts with DNA in a concentration-dependent manner: at low concentrations, it intercalates into double-stranded DNA, whereas at higher concentrations it binds to DNA grooves, generating compaction stress and precipitating DNA, thus contributing to cellular DNA damage¹⁹.

Despite these observations, the exact intracellular pathways through which methoxsalen affects the male reproductive system remain largely uninvestigated. Given the extensive clinical use of methoxsalen, especially in combination with UVA for treating epidermal proliferative disorders, as well as its documented toxic effects on reproductive and developmental processes, it is crucial to elucidate how methoxsalen influences testicular parenchyma, spermatogenic cell populations, and the expression of key genes involved in testosterone biosynthesis. Additionally, assessing the hormonal profile associated with fertility and infertility markers is essential for understanding the systemic consequences of methoxsalen exposure. Therefore, in addition to molecular analyses, the present study also evaluates circulating sex hormone levels and stereological features of reproductive organs in rats exposed to UVA alone and to the combination of UVA and methoxsalen—an area that, to date, has not been comprehensively examined.

Method

Chemical compounds

8-Methoxsalen was purchased from sigma aldrich (cats number: 298-81-7). ELISA kits to measure rats’ testosterone (EA0023Ra), LH (EA0013Ra) were acquired from Biotech (Shanghai, China), and FSH(CSB-E06869r) was prepared from Cusabio Biotech Co., Wuhan, China. The RNA isolation reagent, cDNA synthesis kit, qPCR Master Mix Reagent Kit, and agarose for gel electrophoresis were obtained from Aria Toos (Mashhad, Iran. YT9051). Safe stain for gel electrophoresis was bought from SMOBio (USA, NS10001504991-8).

Animals

The animals in this study included 24 healthy adult male Sprague-Dawley rats, randomly selected from the Endocrinology and Metabolism Center of Shiraz University of Medical Sciences¹. Special rodent food from Behparvar Company (Tehran, Iran) was used to feed the animals. The animals were given tap water and kept in transparent Makrolon cages in a photoperiod of 12 h of light and a dark period of 12 h. The ambient temperature was 22 ± 2 °C and the humidity was 55 ± 2%. An institutional animal ethics committee approved the study. Research animals were also cared for accordance to ARRIVE standards²⁰. The study’s procedures were authorized by Zarghan University of Medical Sciences’ Institutional Animal Ethics Committee (IR.IAU.M.REC.1403.508).

Experimental design

24 adult male Sprague-Dawley rat) three months old at the start of the experiment(were randomly divided into the following groups each containing 6 rats: (A) Healthy control group: animals only consumed water and regular food (without any treatment). (B) Sham group: In this group, animals received the same volume of solution given to the experimental groups, the drug solvent (corn oil). 1 h after administration, the animals were exposed to ultraviolet radiation at a dose of 0.046 J/cm2 for 20 min¹¹. UVA exposure was performed generally, with the UV light positioned above the cages. (C) Experimental groups 1: In these groups, animals received methoxsalen at a dose of 37.5 mg/kg by gavage. an hour after administration, the animals were exposed to ultraviolet radiation at a dose of 0.046 J/cm2 for 20 min. (D) Experimental groups 2: In these groups, animals received methoxsalen at a dose of 75 mg/kg²¹. 1 h after administration, the animals were exposed to ultraviolet radiation at a dose of 0.046 J/cm2 for 20 min. All doses were applied orally every morning. the duration of treatment of the rats was 48 days.

Following the protocol for working with laboratory animals, the rats were anesthetized using a ketamine (10%) and xylazine (2%) mixture at a dose of 80/5 mg/kg (Elfasan, Netherlands), at the end of the experiment, and blood was drawn from their hearts). Subsequently, 5 ml of blood samples were taken from the heart of rats as well and centrifuged for 10 min at 3500 rpm to isolate the serum, which was then stored at −70 °C for further assessment of biochemical parameters. 50–100 milligrams of testis tissue were separated for RNA extraction and molecular analysis.

LH, FSH, and testosterone measurement

Blood was collected from the hearts of anesthetized rats to measure hormone levels. Serum was then separated using a centrifuge at 3500 rpm for 10 min and stored at −80 °C until biochemical tests were conducted. Hormone concentrations were evaluated using commercially available ELISA kits. Briefly, the collected serum samples, standards, and controls were incubated with anti-LH, anti-FSH, or anti-testosterone in a 96-well plate coated with antibodies and incubated at 37 degrees Celsius. Next, an enzyme-labeled reagent called a conjugate was added to label the antigen-antibody reactions. The wells were then washed with the kit’s washing solution to remove non-specific antigens and antibodies. Subsequently, a substrate reagent was added, which reacted with the conjugate to produce a color change in the wells. Finally, a stop solution was added, and the absorbance of the wells was measured using an ELISA reader at specific wavelengths.

RNA isolation and cDNA synthesis

After the rats were sacrificed, their testis tissues were separated immediately and kept at − 80 °C. Samples’ RNA was isolated using an RNX-Plus solution following the company’s instructions. NanoDrop (Thermo Fisher Scientific, Waltham, MA, USA), and horizontal gel electrophoresis were used to quantitatively and qualitatively examine the extracted RNA. The A260/A280 ratio was used to determine the purity of the RNA. Samples with appropriate OD (range 1.8 to 2) and proper integrity according to the electrophoresis results were selected for further steps. To remove DNA from RNA samples before cDNA synthesis, the DNase I digestion was included in the RNA isolation method based on the kit’s manual. Finally, 500 nanograms of the isolated DNA-free RNA was used for cDNA synthesis, in an ultimate volume of 10 µl according to the kit’s specifications.

The Star, Cyp11a1, Hsd17b3, Bcl-2, and Caspase-9 mRNA expression evaluation in the testes

Real-time PCR reaction was performed on the synthesized cDNA using the specific primer pairs (Table 1.). this reaction took place in the Applied Biosystems 7500 Real-Time PCR device by using SYBR Green I dye solution in the volume of 10 µl to measure the PCR reaction product fragments. In this experiment, GAPDH was considered as the housekeeping gene and the expression levels of STAR, Cyp11a1, hsd17b3, Bcl-2, and Caspase-9 were examined using Livak’s rule by the 2^−ΔΔCT formula¹⁵.

Table 1.

Gene-specific forward and reverse primer sequences.

Primer	Sequences (5’->3’)	PCR Product length
STAR: F	AGATGAAGTGCTAAGTAAG	150 bp
STAR: R	TTGATTTCCTTGACATTTG
GAPDH: F	AAAGAGATGCTGAACGGGCA	100 bp
GAPDH: R	ACAAGGGAAACTTGTCCACGA
Cyp11a1:F	AAAGTATCCGTGATGTGGG	112 bp
Cyp11a1:R	TTTCTGGGCATAGTTGAGC
Hsd17b3:F	ATTGAGAGGACCACTGGAAGC	193 bp
Hsd17b3:R	AGTGGATGACACTCTGGCTCT
Bcl-2:F	GGAGGATTGTGGCCTTCTTT	100 bp
Bcl-2:R	GTCATCCACAGAGCGATGTT
Caspase-9:F	ACATCTTCAATGGGACCGGC	85 bp
Caspase-9:R	TCTTTCTGCTCACCACCACAG

Analysis of sperm quality and quantity parameters

Sperm motility

After opening the abdominal muscles in the lower region, approximately 1 cm of the proximal vas deferens was excised and immediately transferred into 5 mL of Ham’s F-10 medium supplemented with 0.35% fatty acid–free bovine serum albumin (BSA). Using fine scissors and forceps, the vas deferens was repeatedly sectioned and gently compressed to release the spermatozoa contained within. For optimal incubation and capacitation, the sperm sample obtained from the rat vas deferens was then placed in Ham’s F-10 culture medium and incubated at 37 °C with 5% CO₂ for 15 min. Immediately thereafter, a drop of the sperm suspension was mounted on a glass slide, and sperm motility was assessed under a light microscope at ×40 magnification. The results were reported as percentages. Based on motility characteristics, the spermatozoa were classified into four categories:

1. Rapidly progressive motile spermatozoa.

2. Slowly progressive motile spermatozoa.

3. Non-progressive spermatozoa (exhibiting only flagella movement without forward progression).

4. Immotile spermatozoa (completely lacking any movement).

Analysis of sperm count

To count sperms, one drop of the sperm-containing fluid was applied to the neobar lam. The average number of sperms was calculated by counting the number of sperms on each of the four large squares. The total number of extricated sperms from one centimeter of the cauda epididymis was estimated using the formula below:

A: The total number of sperms extracted from the cauda epididymis.

B: The number of counted sperms per 0.1 mm3 of solution.

C: Depth factor equals 10.

D: Dilution factor = 5000 mm3 (sperm from one centimeter of cauda epididymis discharged in 5 ml of solution).

Evaluation of sperm viability

Eosin-nigrosin staining was used to determine sperm viability. Eosin and nigrosin were produced using distilled water (both manufactured by Merck in Darmstadt, Germany). One volume of sperm suspension was combined with two volumes of 1% eosin. After 30 s, add an equal volume of nigrosin to the mixture. Create thin smears and inspect them with a light microscope (Nikon E-200, Japan) at a magnification of 40×. In this fashion, viable sperms remained colorless whereas non-viable sperms turned scarlet²².

To measure sperm morphology, sperm were stained with a 5% eosin Y solution. In this staining, the dye can only penetrate dead sperm and turn red, while live sperm remain colorless. Then, the results were reported as a percentage.

Sperm morphology analysis & estimated percentage of normal sperms

The Diff-Quik Kit Rapid Sperm Staining Kit REV 112,401(Ideh varzan farad, IVF Co, Company, Iran) was used to assess sperm morphology. The observation was performed according to the kit protocol. To determine the normality of sperm appearance, 200 sperm were studied, and the result was reported as a percentage. The sperm anomalies included tailless sperm, short-tailed sperm, pinhead sperm, and sperm with coiled tails were evaluated.

Stereological study

For preparation of testicular parenchyma sections and H\&E staining, and stereological evaluation, testes were excised, and their initial weight and volume were measured using a digital scale (Scaltec, Germany) with an accuracy of 1 mg. The water immersion method was employed to determine the testicular volume. In this method, the testicle was immersed in a beaker with thread so that it was fully submerged in water. The reading on the digital scale indicates the initial volume of the testicular parenchyma in cubic centimeters (cm³)^22,23 (Fig. 1. A).

Fig. 1.

Total testis volume was estimated using the immersion method (A). For isotropic uniform random (IUR) sectioning, the testis was placed on a circle, and a random number between 0–9 (here 6) was chosen to guide the initial cut, dividing the testis into two parts (above circle). The head section was oriented at 0°, a second random number (here 3) was selected for the next cut, and the remaining half was sectioned accordingly (B). The testis was further sliced into 8–12 parallel slabs, and circular samples were obtained from the slabs using a trocar (arrow) to estimate tissue shrinkage (front circle). The slices and circular samples were then blocked, sectioned, and stained with Hematoxylin & Eosin (C). Volume parameters were evaluated using the point-counting method, counting points hitting the top-right corner of each target cross (arrow) (D). The optical disector method was applied to count cells (arrowheads) (E), and length density was assessed by counting tubules (wavy arrow) (F).

Staining method

The samples were fixed in Bouin’s solution for 24 h, followed by a standard process of dehydration, clearing, and paraffin embedding. Sections were cut at thicknesses of 5 μm and 20 μm. The sections were mounted on glass slides and subjected to the standard H\&E staining protocol. Briefly, slides were stained with Mayer’s hematoxylin for 10 min, rinsed under running tap water to achieve nuclear bluing, and differentiated in acid alcohol. Counterstaining was performed with 1% eosin solution for 30 s. Finally, sections were dehydrated, cleared, and mounted with coverslips using a permanent mounting medium. This standardized protocol provided clear differentiation of tissue components: cell nuclei appeared blue to violet, whereas cytoplasmic and extracellular structures exhibited varying shades of pink to red under light microscopy. To avoid the reference trap in stereological studies, the tissue’s reference volume must be assessed. According to the Fig. 1. B & C, the “Orientator method” was used to obtain sections with the same shape. For this technique, the testes were randomly placed on the Phi clock with equal divisions and cut along the designated number¹⁹. Random cutting, fixation, and slab preparation followed protocols established beforehand. The samples were fixed in Bouin’s solution for 24 h, followed by a standard process of dehydration, clearing, and paraffin embedding (Fig. 1. C). In other words, Tissue processing was performed using a LEICA TP 1010 (Shandon, Germany) device. The tissue samples were dehydrated through increasing concentrations of alcohol. For paraffin embedding, a flat surface such as glass was used. Molten paraffin was poured onto the glass, and square metal molds were secured on it. The sample was positioned so that the IUR slices of each rat testicle formed a circle inside the mold. Then, melted paraffin was carefully poured into the mold, and each sample was labeled with a code inscribed in the paraffin²⁴. The prepared paraffin blocks were sliced with a microtome (MICRO HM 325, Germany). Initially, 20-micron sections were prepared for cell counting, followed by 5-micron sections for volume analysis. A stereological evaluation was conducted using a Nikon Eclipse microscope. To determine the volumetric density of the samples, a cross-dot grid was used along with the following Delesse’s formula^15,25(Fig. 1. D):

: The sum of the points that hit the desired structure

The sum of the points that hit the entire testicle field

The Optical Disector method was used to calculate the number of cells. The unbiased test grid principle was used to count cells per unit volume. For counting, after placing 20-micrometer slides under the microscope, the section image was transferred to a 19-inch Flatrun monitor screen (LG, Madiran, Iran) using a 60× objective lens and a Sony video camera (Sony, Japan). Then, the counting frame was placed on the image using stereology software. No counting was performed in the upper five micrometers of the section. From this level down, the nuclei of the cells were counted. This counting continued up to a depth of fifteen micrometers of the section using a Heidenhainmicrocator (MT12, Germany). No counting was performed in the lower part of the section. The following formula was used to estimate the average numerical density of each cell type (Fig. 1. E)^15,22.

Nv is the numerical density in terms of number per unit volume (cubic millimeters), ΣQ is the total number of cells counted, ΣP is the number of dissectors (with accompanying dots) used, a(f) is the area of each dissector in square millimeters, and h is the height of the dissector in millimeters (0.01). To estimate the total number (N) of cells, the numerical density¹ is multiplied by the reference volume (Vref).

The number of tubules counted in the unbiased counting frame was evaluated in all fields to calculate the length of the seminiferous tubules. Then, the length density per unit volume was calculated using the following formula (Fig. 1. F)²².

In this formula, ∑Q is the number of tubes counted in the counting frame, ∑F is the total number of fields in which the count was performed, and a/f is the area of the counting frame in all microscopic fields.

Stereological Estimation of Epithelial Height.

The height of the seminiferous epithelium was determined using an indirect stereological method. The mean epithelial height (H) was calculated from the ratio of its volume density (Vv) to its luminal surface density (Sv) using the following equation:

This method provides an unbiased estimation of mean epithelial height, taking into account the natural curvature of the seminiferous tubules. The volume density (Vv) of the epithelium was determined using the Delesse principle through point-counting stereology. Luminal surface density (Sv) was estimated via the test line intersection method. A line grid of known length per point (l/p) was overlaid on the same systematically uniformly random (SUR) sampled fields, and the number of intersections (I) between the test lines and the luminal surface of the tubules was counted²⁵. The surface density was then calculated using the formula:

The total absolute surface area of the surface can be obtained by multiplying SV by the testis volume, (V testis).

Statistical analysis

The data obtained from the experiments were analyzed using SPSS version 23 software. Initially, the Shapiro-Wilk test was used to assess the normality of data distribution. Since the data were normally distributed, one-way analysis of variance (ANOVA) followed by Tukey’s post hoc test was applied. All experimental values are presented as mean ± standard deviation (SD), and differences were considered statistically significant at p < 0.05. Relevant histograms were plotted using GraphPad Prism to visually represent the data.

Results

Average body weight

The MTX 37.5 and MTX 75 groups’ mean body weights were significantly lower than those of the control group at the p = 0.034 and p < 0.001 levels, respectively. In comparison to the MTX 37.5 and sham groups, the mean body weight of the MTX75 group decreased significantly at the levels of (p = 0.001) and (p < 0.001), respectively (Fig. 2).

Fig. 2.

Data are presented as mean ± SD. Different superscript letters (a, b, c) above columns indicate statistically significant differences between all experimental groups (p < 0.05). Groups sharing the same letter are not significantly different, while groups with different letters are significantly different.

The weight and volume of testes

In comparison to the control and sham groups, the average testicular weight in the MTX 37.5 and MTX 75 groups decreased significantly at the level (p < 0.001). When compared to the MTX 37.5 group, the average testicular weight in the MTX75 group decreased significantly (p = 0.001). Comparing the MTX 37.5 and MTX 75 groups to the control and sham groups, the testicular volume dramatically dropped (p < 0.001). Furthermore, the MTX75 group’s mean testicular volume decreased significantly at the level of the MTX 37.5 groups(p < 0.001) (Fig. 3).

Fig. 3.

The average weight and volume of testes tissue in the experimental groups after treatment. Data are presented as Mean ± SD. Superscript letters (a, b, c) above columns indicate the results of post hoc multiple comparisons following one-way ANOVA. Groups sharing the same letter are not significantly different, while groups with different letters are significantly different at p < 0.05.

LH, FSH, and testosterone serum level

In comparison to the control and sham groups, the mean LH hormone concentration in the MTX 75 group rose significantly at levels of (p = 0.011) and (p = 0.021), respectively. Compared to the control and sham groups, the mean FSH concentration in the MTX 75 group increased significantly at (p = 0.002) and (p = 0.003) levels, respectively. At the level, the MTX 37.5 and MTX 75 groups’ mean testosterone concentrations were significantly lower than those of the control and sham groups (p < 0.001). When compared to the MTX 37.5 group, the mean testosterone level of the MTX75 group decreased significantly (p = 0.028) (Table 2).

Table 2.

Hormones concentration in different groups.

Group	LH (mIU/ml)	FSH (mIU/ml)	Testosterone (ng/ml)
Control	19.45 ± 2.05 a	18.95 ± 2.41 a	0.87 ± 0.12 a
Sham	20.00 ± 3.58 a	19.28 ± 5.69 a	0.84 ± 0.06 a
MTX 37.5	21.33 ± 4.41 ab	23.23 ± 2.81 ab	0.52 ± 0.07 b
MTX 75	26.33 ± 3.20b	27.67 ± 1.75b	0.38 ± 0.043c

Data are presented as Mean ± SD. Superscript letters (a, b, c) indicate the results of post hoc multiple comparisons following one-way ANOVA. Groups sharing the same letter are not significantly different, while groups with different letters are significantly different at p < 0.05.

The levels of Star, Cyp11a1, and Hsd17b3 mRNA expression

Melting curve analysis was utilized to investigate the specificity of the reactions. The expression level of the Star mRNA was significantly lower in the MTX 75 group than in the control, sham, and MTX37.5 groups (p ≤ 0.001). Furthermore, the Star mRNA expression level of the MTX 37.5 groups was significantly lower than that of the control group (p = 0.006). Cyp11a1 mRNA expression levels were significantly lower in the MTX 37.5 and MTX 75 groups compared to the control and sham groups (p = 0.027, p = 0.036) and (p < 0.001, p < 0.001), respectively. In comparison to the MTX 37.5 groups, the Cyp11a1 mRNA expression level was significantly lower in the MTX75 group (p = 0.004). When comparing the MTX 37.5 and MTX 75 groups to the control and sham groups, the expression of the Hsd17b3 mRNA decreased dramatically at the levels of (p = 0.006, p = 0.003) and (p < 0.001, p < 0.001), respectively (Fig. 4).

Fig. 4.

The average level of Star, Cyp11a1, and Hsd17b3 mRNA expression in the experimental groups after treatment. Data are presented as SD ± Mean. Superscript letters (a, b, c) above columns indicate the results of post hoc multiple comparisons following one-way ANOVA. Groups sharing the same letter are not significantly different, while groups with different letters are significantly different at p < 0.05.

The mRNA expression levels of apoptotic genes Bcl-2 and Caspase-9

The Bcl-2 mRNA expression was significantly lower in the MTX 37.5 and MTX 75 groups than in the control and sham groups (p < 0.001). When comparing the MTX 75 group to the MTX 37.5 groups, there was a significant increase in the Bcl-2 mRNA expression level (p < 0.001). In contrast, the Caspase-9 mRNA’s expression level was considerably higher after MTX 75 therapy than it was in the control, sham, and MTX37.5 groups (p < 0.001, p < 0.001, p = 0.002). Additionally, compared to the control, MTX 37.5 administration markedly elevated the expression of the Caspase-9 (p = 0.015). The Bcl-2 and Caspase-9 mRNA expression levels in the other groups, however, did not differ significantly (Fig. 5).

Fig. 5.

The average level of Bcl-2 and Caspase-9 mRNA expression in the experimental groups after treatment Data are presented as SD ± Mean. Superscript letters (a, b, c) above columns indicate the results of post hoc multiple comparisons following one-way ANOVA. Groups sharing the same letter are not significantly different, while groups with different letters are significantly different at p < 0.05.

Sperm quality and quantity parameters evaluation

The percentage of progressive fast-moving sperm was significantly lower in the MTX 37.5 and MTX 75 groups than in the control and sham groups (p < 0.001). The proportion of progressive fast-moving sperm in the MTX75 group was significantly lower than that in the MTX 37.5 group (p < 0.001). In contrast, the average percentage of slow-moving sperm in the MTX 37.5 and MTX 75 groups was significantly higher than in the control and sham groups (p < 0.001, p < 0.001 and p < 0.001, p = 0.018, respectively). In comparison to the MTX 37.5 group, the average number of slow-moving sperm was significantly lower in the MTX75 group (p = 0.009). The mean proportion of sperm with non-progressive movement increased significantly in both the MTX 37.5 and MTX 75 groups as compared to the control and sham groups (p = 0.003, p = 0.001) and (p < 0.001, p < 0.001), respectively. In comparison to the control and sham groups, the sperm count was considerably reduced in the MTX 37.5 and MTX 75 groups (p < 0.001). The mean number of sperm in the other groups did not differ significantly (p < 0.05). In comparison to the control, sham, and MTX 37.5 groups, the average percentage of sperm with normal morphology was significantly lower in the MTX 75 group (p < 0.001). In comparison to the control group, the MTX 37.5 group also showed a noticeably reduced percentage of properly transformed sperm (p = 0.047). In comparison to the control and sham groups, the mean sperm viability % significantly decreased in both the MTX 37.5 and MTX 75 groups (p < 0.001). The other groups’ criteria did not differ significantly from one another (Table 3).

Table 3.

Sperm’s parameter in groups.

Group	Sperm Progressive Movement %	Slow Movement Sperm %	Non- Progressive Movement %	Immotile sperm %	Sperm count ×106	Sperm normal morphology %	Sperm Viability %
Control	83.36 ± 5.23 a	6.25 ± 1.89 a	6.19 ± 1.95 a	4.19 ± 2.46 a	13.82 ± 0.98 a	92.33 ± 1.37 a	93.00 ± 1.55a
Sham	78.70 ± 4.30 a	7.38 ± 2.27 a	7.46 ± 2.73 a	6.46 ± 1.96 a	13.14 ± 0.98 a	91.00 ± 4.10 ab	91.67 ± 3.93 a
MTX 37.5	26.90 ± 3.28 b	19.35 ± 1.76 b	15.99 ± 5.30 b	37.76 ± 5.42 b	5.23 ± 1.73 b	81.17 ± 3.06 b	53.17 ± 10.30 b
MTX 75	6.29 ± 1.81c	12.48 ± 5.63c	21.80 ± 5.77b	59.43 ± 6.06c	4.41 ± 0.97b	53.33 ± 12.63c	42.50 ± 14.35b

Data are presented as SD ± Mean. Superscript letters (a, b, c) indicate the results of post hoc multiple comparisons following one-way ANOVA. Groups sharing the same letter are not significantly different, while groups with different letters are significantly different at p < 0.05.

Testes Stereological analysis

The seminiferous tubule volume in the MTX 37.5 and MTX 75 groups significantly decreased compared to the control and sham groups at the levels of (p < 0.001, p < 0.001) and (p = 0.001, p = 0.002), respectively. Also, its volume in the MTX75 group appeared to be lower compared to the MTX 37.5 group (p < 0.001). Furthermore, a significant decrease in the mean germinal epithelium height in the MTX 37.5 and MTX 75 groups was observed compared to the control and sham groups (p ≤ 0.001). No other significant differences were observed in these testicular factors among other groups at the (P < 0.05) level (Table 4). Our observations also revealed a reduction in the mean volume of seminiferous tubules in the MTX 37.5 and MTX 75 groups in contrast to the control and sham groups at the level of (p < 0.001) and (p = 0.001, p = 0.002), respectively. Also, the mean volume in the MTX75 group significantly reduced compared to the MTX 37.5 groups at the level of (p < 0.001) (Table 4).

Table 4.

Testes Stereological parameters in groups.

Group	Height of the germinal epithelium (µm)	Volume of the seminiferous tubule (mm3)	Length of seminiferous tubules (m)	Volume of the interstitial tissue (mm3)
Control	83.27 ± 6.52a	1364.20 ± 84.11a	18.32 ± 1.24a	207.73 ± 44.92a
Sham	81.67 ± 3.56 a	1340.94 ± 69.99 a	17.24 ± 2.55 a	211.95 ± 50.69 a
MTX 37.5	58.00 ± 6.39 b	1103.00 ± 68.24 b	14.72 ± 3.41 a	219.53 ± 66.07 a
MTX 75	51.00 ± 8.60b	699.48 ± 149.27c	10.39 ± 1.06b	247.25 ± 103.08a

Data are presented as SD ± Mean. Superscript letters (a, b, c) indicate the results of post hoc multiple comparisons following one-way ANOVA. Groups sharing the same letter are not significantly different, while groups with different letters are significantly different at p < 0.05.

The number of sexual lineage cells

Compared to the control and sham groups, the MTX 75 group had significantly fewer spermatogonia (p < 0.001), spermatocytes (p < 0.001), and round spermatids (p ≤ 0.001). The average number of Leydig cells and elongated spermatid cells was significantly lower in the MTX 75 group than in the control, sham, and MTX37.5 groups (p ≤ 0.001). Comparing the MTX 37.5 group to the control group, the former showed a significant drop in the numbers of elongated spermatids, spermatocytes, round spermatids, long spermatids, and Leydig cells. In addition, the mean Sertoli cell count in the MTX 75 group was significantly lower than in the control and sham groups (p = 0.001). In addition, the Leydig cell count was significantly lower in the MTX 37.5 group than in the control group (p = 0.034) (Table 5) (Fig. 6).

Table 5.

Number of sexual cell lineage in different groups.

Group	Spermatogonia ×106	Spermatocyte ×106	Round Spermatids ×106	Long Spermatids ×106	Sertoli cells ×106	Leydig cells ×106
Control	197.48 ± 21.43a	208.77 ± 20.82a	439.85 ± 34.48a	448.32 ± 36.16a	32.20 ± 4.34a	38.80 ± 2.54a
Sham	172.81 ± 20.89 a	173.77 ± 24.89 a	425.68 ± 39.67 a	411.69 ± 60.54 a	31.82 ± 3.73 a	37.46 ± 2.71 a
MTX 37.5	120.120 ± 37.70 b	120.47 ± 38.96 b	199.21 ± 21.38 b	272.14 ± 54.70 b	26.32 ± 1.79 ab	25.78 ± 6.09 b
MTX 75	91.49 ± 11.62b	94.82 ± 7.53b	144.11 ± 36.13c	124.84 ± 50.21c	22.26 ± 4.39b	18.95 ± 3.46c

Data are presented as SD ± Mean. Superscript letters (a, b, c) indicate the results of post hoc multiple comparisons following one-way ANOVA. Groups sharing the same letter are not significantly different, while groups with different letters are significantly different at p < 0.05.

Fig. 6.

Photomicrograph of histopathological changes in testicular parenchyma in the experimental groups. Control group (a, and b), Sham group (UVA radiation) (c, and d), MTX 37.5 group (e, and f), and MTX 75 group (g, and h). Sections were stained with Hematoxylin-Eosin. Objective magnification: ×40 for panels a, c, e, g and ×60 for panels b, d, f, h. In the control and sham groups, germ cells and seminiferous tubules were normal and no pathological damage was observed. The lumens were regular and all cell lines including spermatogonia¹, spermatocytes (Sc), round and elongated spermatids¹, Sertoli and Leydig cells (L) were normal and spermatogenesis was active. Also, the height of the germinal epithelium (G.E) was normal and the connective tissue was without congestion (abnormal hyperemia caused by dilation and excessive blood filling of the vessels) and edema. However, in the group receiving low-dose methoxsalen, a significant decrease in the height of the germinal epithelium and the number of germline cells was observed. In this group, no significant difference was observed in the number of Sertoli cells (S) compared to the control and sham groups. Methoxsalen’s detrimental effects on testicular parenchyma are demonstrated by the tubules’ atrophy and shrinkage, the germinal epithelium’s disintegration and reduced height, the sharp decline in the number of germ cells, and the existence of empty spots between the tubules (arrowhead) in the group that received a high dose of the therapy.

Discussion

This study demonstrates that methoxsalen in combination with UVA negatively influences testicular structure and androgenic function, primarily through impairments in Leydig cell activity and disruptions in spermatogenic integrity. The observed reduction in Leydig cell number, along with suppressed expression of key steroidogenic genes, provides a mechanistic basis for the decline in testosterone, consistent with earlier evidence that psoralens adversely affect male reproductive function. The reduction in body weight, particularly at higher doses of methoxsalen, may reflect systemic toxicity related to its interaction with ds-DNA and other biomacromolecules, combined with UVA-induced DNA damage such as thymidine dimer formation²⁶. These molecular effects likely contribute to impaired tissue growth and general metabolic decline during prolonged exposure.

Similarly, the decreased testicular weight and volume observed in methoxsalen-treated groups align with previous findings showing psoralen-induced atrophy of male reproductive organs in rats⁷. Stereological evidence in the present study suggests that these structural alterations derive largely from loss of germ cells and disruption of seminiferous tubule morphology. The accompanying reductions in seminiferous tubule dimensions and germinal epithelium height further indicate that chronic methoxsalen exposure compromises spermatogenic capacity. Collectively, these findings highlight the potential reproductive risks of long-term methoxsalen therapy and align with mechanistic pathways previously described for psoralen-induced oxidative and photochemical damage.

The hormonal findings further support a disruptive effect of methoxsalen on the steroidogenic axis. The significant decline in serum testosterone—more pronounced at higher doses—highlights impaired Leydig cell function, consistent with the essential role of androgens in maintaining seminiferous tubule integrity and normal spermatogenesis²⁷. The concomitant rise in LH and FSH in the high-dose group reflects a classic compensatory response to reduced androgen levels. Similar endocrine alterations have been reported in female rodents, where methoxsalen decreased estrogen and increased gonadotropins, indicating that psoralens can broadly disrupt hypothalamic–pituitary–gonadal regulation¹². As testosterone levels declined, the negative feedback mechanism triggered an increase in LH and FSH concentrations in the high-dose group. While not statistically significant, the low-dose group exhibited comparatively higher levels of FSH and LH.

The molecular data provide mechanistic support for these hormonal changes. Suppression of Star, Cyp11a1, and Hsd17b3 mRNA expression in a dose-dependent manner suggests that methoxsalen interferes directly with key steps of testosterone biosynthesis, particularly cholesterol transport into mitochondria and the enzymatic conversions required for androgen production. Together, these findings show that methoxsalen impairs androgen synthesis at both cellular and transcriptional levels, ultimately compromising reproductive endocrine function.

The histopathological and stereological findings indicate that methoxsalen disrupts the structural and functional integrity of the testis. The reduction in Sertoli and germ cells, along with the dose-dependent loss of Leydig cells, aligns with the observed decline in testosterone production²⁸. Structural deterioration of seminiferous tubules and interstitial tissue further compromises Leydig cell function, consistent with reports linking testis size and volume to germ cell and Sertoli cell numbers as well as sperm output^29,30. Testis size, volume, and weight strongly correlate with germ cell and Sertoli cell numbers and sperm production^31,32. Accordingly, the reductions in spermatogonia, spermatocytes, and spermatids in methoxsalen-treated groups reflect decreased testicular volume and seminiferous tubule dimensions. Type A spermatogonia, being less differentiated, are most sensitive to UVA, while spermatozoa are more resistant^33–35. Therefore, methoxsalen and UVA caused a decrease in the number of mitotic divisions in these cells, and the decrease in the number of these cells can also be due to inducing cell death. The results of the expression of mRNAs involved in the apoptosis process showed that the expression of the Bcl-2 mRNA, which acts as an anti-apoptotic³⁶, showed a considerable drop in the MTX 37.5 and MTX 75 groups compared to the control and sham groups. In contrast to the control, sham, and MTX 37.5 groups, the MTX 75 group exhibited a substantial increase in the mRNA expression of the pro-apoptotic Caspase-9³⁷. According to the study’s histological examination, the MTX 75 group had a marked decline in the average number of round and elongated spermatids and Leydig cells as compared to the control, sham, and MTX37.5 groups. These findings make sense given that the groups who got MTX and UVA had decreased serum testosterone levels.

Our findings of a reduced Leydig cell population are consistent with the established cytotoxic profile of methoxsalen. Upon photoactivation by UVA (PUVA therapy), methoxsalen produces reactive oxygen species (ROS), including singlet oxygen and superoxide anions¹¹. These ROS drive oxidative stress, a recognized trigger of cellular apoptosis and necrosis. Given their high metabolic activity in steroidogenesis, Leydig cells are particularly susceptible to oxidative injury, which likely accounts for the observed reduction in their number. A key observation in our study is the downregulation of genes essential for testosterone biosynthesis (StAR, CYP11A1, HSD3B, CYP17A1). This suppression may occur via several mechanisms discussed in the literature. First, methoxsalen is a potent activator of the aryl hydrocarbon receptor (AhR), which induces hepatic cytochrome P450 enzymes such as CYP1A1 and CYP1A2^7,38. Although this effect is primarily hepatic, cross-talk with testicular pathways is plausible. AhR activation in other models has been shown to inhibit steroidogenic enzyme expression, thereby impairing testosterone production. In parallel, oxidative stress generated by methoxsalen may directly compromise the activity of steroidogenic enzymes, such as CYP11A1, which is particularly vulnerable to ROS-mediated damage. Collectively, these pathways explain the decline in testosterone levels through both loss of Leydig cells and suppression of steroidogenic function.

Interestingly, a previous study reported increased serum and testicular testosterone following methoxsalen administration in male Wistar rats, which the authors attributed to compensatory activation of the HPT axis due to reduced negative feedback. Despite this elevation, pituitary shrinkage and impaired fertility were still observed, indicating dysregulation of the axis and compromised reproductive function. Differences in dosage, treatment duration, administration route (intraperitoneal vs. oral), or progression to later-stage toxicity—when compensatory responses are no longer sufficient—may explain the divergence from our findings. Overall, the combined reduction in Leydig cell number, suppression of steroidogenic gene expression, and altered testosterone production provides a coherent mechanistic explanation for methoxsalen-induced male infertility. This conclusion is reinforced by our histological observations and previous reports of seminiferous tubule disorganization, shrinkage, and marked depletion of spermatogonia, spermatocytes, and spermatozoa, which ultimately link Leydig cell dysfunction to impaired sperm quantity and quality¹¹.

Ionizing radiation, including UVA, can delay cell-cycle progression and promote cell death, with sensitivity depending on the cell’s phase and radiation dose. Cells irradiated before entering mitosis may arrest in G2, while radiation can also inhibit DNA synthesis in S-phase cells and disrupt G1–S transition, collectively leading to delayed division^39,40. In our study, the significant decrease in round and elongated spermatids in the MTX 37.5 and MTX 75 groups likely reflects reduced spermatocyte numbers caused by radiation-induced cell-cycle delay and impaired differentiation. As spermatid formation depends on successful meiotic progression, this disruption ultimately lowers spermatozoa output. Consistent with our findings, a previous report showed that oral administration of 8-methoxypsoralen and 5-methoxypsoralen significantly reduced sperm count in rats⁷.

No significant differences were observed between the control and sham (UVA-only) groups for any spermatogenesis parameters. This indicates that UVA exposure alone (0.046 J/cm² for 20 min) did not appreciably affect spermatogenesis, consistent with prior reports showing minor, non-significant germ cell reductions in UVA-only groups of Balb/C mice¹¹. The limited effect of UVA may be due to the relatively low dose, short exposure, and protective properties of rat testes. Importantly, methoxsalen’s cytotoxicity is primarily UVA-dependent, causing DNA cross-linking and ROS generation. Thus, the marked decrease in spermatid numbers in experimental groups is attributable to methoxsalen and its photoactivated metabolites rather than UVA alone.

In the MTX + UVA groups, the presence of displaced primary spermatocytes and spermatids in seminiferous tubules indicates not only a reduction in sperm number but also increased abnormalities and immotility. Compared to control and sham groups, the MTX-treated groups showed a higher percentage of morphologically aberrant, non-viable, and immobile sperm, with the highest dose also reducing Sertoli cell numbers.

These functional impairments are likely mediated by methoxsalen-induced oxidative stress. Excessive ROS generation in the testes and epididymis damages sperm mitochondria, reducing ATP production necessary for motility, leading to sluggish or non-progressive sperm^41,42. Additionally, ROS-mediated lipid peroxidation of the sperm plasma membrane disrupts membrane fluidity and ionic balance, particularly Ca²⁺ and Na⁺ signaling, further impairing motility and contributing to immobile spermatozoa^43,44.

Another critical target is the axoneme, the flagellar apparatus composed of microtubules and dynein motors that depend on ATP. Oxidative stress induced by methoxsalen may compromise these fragile protein assemblies. Even when ATP levels are adequate, mechanical disruption of the axoneme prevents synchronized flagellar beating, thereby limiting movement to weak and inefficient patterns. It is also important to recognize that spermatozoa leave the testes in an immotile state and gradually acquire motility during epididymal transit through capacitation-associated modifications⁴⁵. Methoxsalen may interfere with this maturation process by injuring the epididymal epithelium or by disturbing the secretion of essential factors required for sperm functional development. Inadequate epididymal maturation would therefore hinder the acquisition of rapid progressive motility⁴⁶. More detailed evaluation of these mechanisms would require supplementary experiments and electron microscopy for structural analysis, which represents a limitation of the current study.

In conclusion, methoxsalen impairs sperm function primarily through oxidative stress, which damages the plasma membrane, disrupts ionic signaling, compromises axonemal integrity, and impairs mitochondrial ATP production. These combined effects reduce the energy and mechanical power required for rapid progressive motility, although some residual slow or non-progressive movement may persist. The prolonged duration of exposure likely exacerbates these effects.

Given the relative resistance of Sertoli cells to toxins, further research is needed to explore methoxsalen’s potential interactions with these cells. Additionally, investigations into the effects of methoxsalen on fetal reproductive system development and long-term impacts in male humans are warranted. Finally, other mechanisms underlying methoxsalen-UVA-induced reproductive toxicity should be studied to develop strategies for mitigation.

Conclusion

This research investigated the adverse effects of methoxsalen, a drug used to treat skin conditions, on male rat reproductive health. The study found that methoxsalen, especially when combined with UVA radiation, significantly reduced testosterone levels, impaired spermatogenesis, and decreased testicular size and weight in a dose-dependent manner. These effects were linked to decreased mRNA expression of genes crucial for testosterone synthesis and increased apoptosis in Leydig cells. The findings suggest potential risks of methoxsalen use for individuals planning pregnancy, warranting further research.

Abbreviations

Bcl-2

B-cell leukemia/lymphoma 2

Chol

Cholesterol

Cyp11a1

Cytochrome P450 family 11 subfamily a polypeptide 1

ELISA

Enzyme-Linked Immunosorbent Assay

FSH

Follicle-Stimulating Hormone

GAPDH

Glyceraldehyde-3-phosphate dehydrogenase

Hsd17b3

17-beta hydroxysteroid dehydrogenase 3

LH

Luteinizing Hormone

MTX

Methoxsalen

Star

Steroidogenic Acute Regulatory

UVA

Ultraviolet A

Author contributions

All experiments, statistical analysis, and figure preparation were conducted by Z.Gh, F.K, H.Fs). The initial draft of manuscript was written by Z.Gh, F.K. All tests were set up and second draft of manuscript was written by A.F, M.S. A final manuscript proof was completed by SD who also contributed more funds to the project.

Data availability

This article contains all data created and examined throughout this investigation. The corresponding author will provide datasets used or analyzed during the current work upon reasonable request.

Declarations

Competing interests

The authors declare no competing interests.

Ethics approval and consent to participate

All animal experiments adhere to the ARRIVE standards and are conducted in accordance with the UK Animals (Scientific Procedures) Act, 1986, as well as the EU Directive 2010/63/EU regarding animal experimentation. The Research Ethics Committee of the Islamic Azad University-Zarghan Branch (Fars Province) has ethically confirmed the study (Ethics Code: IR.IAU.M.REC.1403.508).