This review aimed to investigate the protective effects of exogenous melatonin on the parameters of male reproductive function against physical, heat, or ischemic injuries compared to placebo in rodent subjects. We asked if melatonin therapy (whether before or after the induction of stress) prevents oxidative pathways in testicular cells.

Two reviewers (AS and SP) conducted a comprehensive search in PubMed, Scopus, and Web of Science for records from January 1, 1970, until September 9, 2022, for “melatonin”, “testicular function”, and their equivalents as keywords. The search strategies comprised a combination of Medical Subject Headings (MeSH) or their equivalent (where available), keywords, truncations, and boolean operators. A manual backward and forward citation search was also done for all the included studies. The detailed search strategy is provided in Supplementary Material 1 .

Firstly, duplicate records were removed electronically. Then, using the Rayyan online tool for managing systematic reviews (37), four reviewers (NDE, SP, NM, and AS) independently screened titles and abstracts, followed by the full-text screening of the identified records for eligibility criteria. Disagreements were resolved with discussion. Studies were included if they satisfied the following criteria: (1) controlled animal studies, (2) the population was rodents that were exposed to physical, electrical, ischemic, or thermal injuries as oxidative stress to the testicular tissue, (3) at least one intervention group received melatonin regimen, (4) at least one control group with similar stress, and (5) reported major hallmarks of testicular tissue (histopathologic, biochemical, and sperm analyses).

The exclusion criteria were: (1) in-vitro and ex-vivo studies, (2) non-rodent animals with other types of stresses (radiation, chemotherapy, toxins, and metabolic), (3) combination therapy of melatonin with other drugs, (4) treatment using derivatives of melatonin, (5) healthy controls without stress, and (6) reported irrelevant outcomes. Also, reviews, letters, and human trails were excluded from the review. Our search was not restricted by language.

Two reviewers (ET and FSS) independently extracted the favorable data into Excel spreadsheets. Any disagreements were resolved by discussion and involvement of a third author (AS). Structured forms were used for data extraction of the following contents: (1) study characteristics (first author, publication year, and country), (2) population characteristics (species, age, sample size, and type of stress), (3) melatonin dose, duration, and route, (4) time of assessment of outcome, (5) tissue and plasma biochemical indices, and (6) histopathological characteristics. For missing data, we contacted the first or corresponding author and waited for a response for at least one month.

We assessed the risk of bias in the studies using the Systematic Review Centre for Laboratory animal Experimentation (SYRCLE) tool for animal intervention studies (38). Two authors (NDE and ET) independently reviewed each article and classified it as high, low, or unclear for each bias domain. Disagreements were resolved via consensus or by a third (AS) reviewer if necessary.

Data were analyzed using Stata MP Version 16 (StataCorp, College Station, TX, USA), and a p-value < 0.05 was considered statistically significant. Under a random-effects model (DerSimonian-Laird method), the pooled effect sizes were reported as standardized mean difference (Glass’s Δ) and 95% confidence interval (39). The statistical heterogeneity was examined using Cochran’s Q statistic, p-value, and I-squared. I-squared was employed to qualify heterogeneity as “perhaps not important”, “moderate heterogeneity”, “substantial heterogeneity”, and “considerable heterogeneity” if I-squared values were 0-40%, 30-60%, 50-90%, and 75-100%, respectively (40). Subgroup analyses were implemented only if three or more studies were available for each subgroup to identify possible sources of heterogeneity. In case of missing data, if crucial, studies were removed from the analysis. Also, funnel plots were used to detect visual asymmetry in publications when at least ten studies were available for the outcome (40).

The PRISMA flow diagram of the literature search is presented in Figure 1 . The systematic search yielded 10,039 records, while manual citation searching yielded no additional studies. The database searching included PubMed (n=1,375), Web of Science (n=3,838), and Scopus (n=4,826). 1,016 records were removed using automatic duplicate detection. Title and abstract screening was conducted on 9,023 records, and 110 studies were sought for retrieval. With the exclusion of 5 studies that we failed to retrieve (41–45), 105 articles were assessed for eligibility. A total of 64 articles were excluded due to ineligible population (n=53), design (n=3), intervention (n=4), outcome (n=1), and publication type (n=3). Finally, 41 articles were eligible for the study; 12 (46–57) were only included in narrative data synthesis. Two studies were on the same subject population; thus, for data synthesis, they were treated as one (47, 51). Despite our effort to contact the authors, 8 studies contained missing values; therefore, they were excluded from the analysis (47, 52–55, 57–59).

Included studies were published between 2000 and 2022 in English. Rats (n=36) (25, 46–54, 56, 58–82) and mice (n=5) (55, 57, 83–85) were the subjects in the included studies. Studies employed direct heat (n=3) (82, 84, 85), cryptorchidism (n=2) (58, 72), varicocele (n=3) (60, 73, 77), torsion-detorsion (n=18) (25, 46, 48, 49, 54, 56, 59, 62–66, 74, 75, 78–81), testicular vascular occlusion (n=4) (52, 53, 61, 67), hypobaric hypoxia (n=3) (47, 51, 55), ischemia-reperfusion (n=2) (50, 76), stress by excessive (n=3) (68, 69, 71) or restraint (n=1) (83) activity, spinal cord injury (n=1) (57), and trauma (n=1) (70). Studies induced the stress mechanisms bilaterally (n=16) (47, 48, 50, 51, 55, 57, 58, 68, 69, 71, 72, 75, 76, 82–85) and unilaterally (n=25) (25, 46, 49, 52–54, 56, 58–67, 70, 73, 74, 77–81). Mice aged between 6 - 12 weeks and rats aged between 6 – 24 weeks. Studies administered melatonin intraperitoneal (n=31) (25, 46, 48, 49, 52–54, 56–60, 62, 63, 65–71, 73, 75, 77–81, 83–85), oral (n=5) (47, 51, 55, 61, 72, 82), intravenous (n=3) (50, 64, 76), and intramuscular (n=1) (74). Characteristics of the included articles are summarized in Figure 2 . The dose of melatonin therapy, descriptive histopathological findings, duration of stress induction, and outcomes assessment ranged significantly between the studies presented in Table 1 and Supplementary Materials 2 .

18 outcomes were pooled from the included studies which were classified into five groups: (a) sperm parameters, (b) reproductive hormone profile, (c) markers of oxidative stress in testicular tissue, (d) body weight and testicular somatic indices, and (e) exploratory outcomes. Pooled outcomes included: total sperm count, forward progressive motility, normal sperm morphology, sperm viability, Johnson tubular biopsy score, serum testosterone and Inhibin-B levels, testicular tissue SOD, malondialdehyde (MDA), GPx, and catalase (CAT) activity, final body and testes weight, testis to body weight ratio, seminiferous tubular diameter, percentage of tubules with TUNEL-positive cells (TUNEL: Terminal deoxynucleotidyl transferase dUTP nick end labeling), and number of TUNEL-positive cells per tubule. The forest plots of the overall pooled effects sizes are presented in the Figure 3 – 7 .

In a complementary analysis, we assessed the effect of melatonin therapy on TUNEL assay of seminiferous tubular cells (54, 65, 66, 68, 83, 85). The pooled SMD for the effect of melatonin therapy on the percentage of tubules with TUNEL-positive cells (SMD = -3.886, 95% CI: -6.365 to -1.406, p-value = 0.002) was statistically significant. On the other hand, this measure was not statistically significant for the number of TUNEL-positive cells per tubule (SMD = -5.636, 95% CI: -11.495 to 0.222, p-value = 0.059).

Two outcomes were eligible for analysis of publication bias. The funnel plots regarding the effects of melatonin therapy on testicular tissue MDA activity and Johnsen have been analyzed for publication bias. Both plots lack symmetry on visual inspection, suggesting a high risk of publication bias. Egger’s regression and Begg’s tests also showed consistent results: p-value <0.001 for both tests in both outcomes. The funnel plots are presented in Supplementary Material 4 .

Subgroup analyses were done on the side of induction of stress (unilateral vs. bilateral), mechanism of stress (ischemic vs. non-ischemic injuries), and duration of melatonin therapy (> 2 weeks vs. ≤ 2 weeks). Subgroup analyses revealed significant between-group differences for SOD based on the mechanism of stress and duration of melatonin therapy (p-value <0.001 and 0.01, respectively). All the meta-analyses are stratified in Table 2 .

For each domain, studies scored 1 if they were assessed as low risk using SYRCLE tool. Studies scored between 1 and 4. All the studies were labeled as unclear risk on random sequence generation, allocation concealment, random housing, blinding, and random outcome assessment. For other sources of bias, all the studies were assessed as low risk. Studies were low risk based on baseline characteristics (n=16), incomplete outcome data (n=4), and selective outcome reporting (n= 35). All the details are presented in Figure 8 and Supplementary Material 5 .

Male infertility is commonly caused by ejaculatory dysfunction, no or low sperm count, or abnormal morphology or motility of the sperms (86). In the present study, melatonin improved spermatogenesis, total sperm count, sperm viability and morphology, and forward progression sperm motility in animals with infertility induced by physical injuries. However, total testicular weight, testis to body weight ratio, sperm motility, and seminiferous tubule diameter were not affected by melatonin administration. Our results were partially in line with previous studies on other models of male infertility. Melatonin protects the reproductive system against the toxicity of chemotherapeutic agents (87). Zi et al. reported the favorable effects of melatonin on doxorubicin-induced impaired spermatogenesis and sperm quality, except for sperm motility (88). Melatonin also suppressed bleomycin, etoposide, and cisplatin-induced testicular damage by ameliorating histopathological alterations of testes, testicular weight, and sperm motility, viability, and morphology, but not sperm count and its progressive motility (89). These discrepancies may be due to the differences in the type of animals, experimental models of male infertility, and dose, duration, and route of melatonin administration. Melatonin may directly affect the male reproductive system by interacting with its receptors on the testes, epididymis, or spermatozoa (90, 91). Melatonin increased the expression of spermatogenesis-related genes and led bovine Sertoli cells to transit from G1 to S phase (92). In addition to the experimental studies, lower serum and seminal melatonin levels in patients with idiopathic oligoasthenoteratozoospermia compared to the fertile men and a significant positive correlation between serum melatonin and sperm motility were detected in a previous case-control study (93). Due to the conflicting evidence (55, 91), more studies should be performed to clarify the effects of melatonin on sperm and testis parameters.

Testosterone has fundamental roles in male reproductive system development and spermatogenesis (94). Our results revealed that melatonin does not affect plasma testosterone levels, similar to some reports on other animal models of male infertility (89, 95). Nonetheless, transgenic mammals with endogenously elevated melatonin levels revealed elevated testosterone production (96). Evidence also suggests that melatonin induces testosterone activation and inhibits its destruction (97). Although, the low number of included studies may conceal the possible influence of melatonin on testosterone; consequently, further studies are needed to address the issue.

The serum level of inhibin-B, a hormone produced primarily by the Sertoli cells of the testes, is a potential marker of testicular function and spermatogenesis (98, 99). Our findings demonstrated the improvement of inhibin-B levels following melatonin administration, which may be because of melatonin’s beneficial effects on preserving Sertoli cells against injuries. In this regard, melatonin upregulated inhibin-B expression in bovine Sertoli cells via its MT1 and MT2 receptors (92). In a randomized controlled trial by Lu et al., melatonin supplementation increased the peripheral blood inhibin B levels following varicocelectomy that could be attributed to its effects on spermatogenesis function (100).

Testicular oxidative stress could be derived from intrinsic etiologies, including testicular torsion, varicocele, cryptorchidism, infection, inflammation, or aging (101). Some extrinsic factors, such as intense exercise, can also raise oxidative stress levels in the testes (69). Sperms are extremely sensitive to oxidative damage due to the high levels of polyunsaturated fatty acids in their membranes and the low content of enzymatic antioxidants. Oxidative stress is directly correlated with increased apoptosis in germ cells and mature spermatozoa by changing caspase activity and disrupting mitochondrial membrane (101). Therefore, multiple studies indicated the association between oxidative stress and abnormal sperm count, motility, viability, morphology, DNA integrity, and fertilization ability (5, 101, 102). According to our results, melatonin administration significantly increased the SOD, GPx, and CAT activities, essential enzymatic antioxidants, and reduced MDA levels, as a lipid peroxidation product, in the testicular tissue of animals. Antioxidant properties of melatonin are reported in previous literature. Melatonin could directly scavenge oxidants and indirectly increase enzymatic antioxidant levels (103). Morvaridzadeh et al. conducted a meta-analysis of randomized controlled trials on patients with a different health condition. They reported increased total antioxidant capacity, glutathione levels, SOD, GPx, and glutathione reductase, but not CAT activities, and reduced MDA levels following melatonin supplementation (104). Melatonin could upregulate the antioxidant nuclear factor erythroid 2-related factor 2 (Nrf2)/heme oxygenase-1 (HO-1) signaling pathway in the damaged testes (83). This pathway induces the transcription of antioxidant proteins and leads to ROS clearance (105). Melatonin may also exert its beneficial effects through upregulating micro-RNA-34a/silent information regulator 1 (SIRT1)/forkhead transcription factors-class O (type1) (FOXO1) epigenetic axis (60). This pathway stimulates antioxidants’ expression, inhibits pro-inflammatory pathways, decreases apoptosis, improves mitochondrial biogenesis, repairs cell damage, and prevents cells’ dysfunction and infertility (106). Therefore, the effects of melatonin on other oxidative stress-related conditions should be investigated.

Apoptosis, the programmed cell death, takes place during normal spermatogenesis. However, physical testicular injuries could elevate germ cell apoptosis and reduce seminiferous tubule diameter and sperm count (60, 68, 107, 108). We observed that melatonin decreases apoptotic germ cells and ameliorates the detrimental impact of physical injuries on the testes. Consistently, melatonin had protective effects against apoptotic cell damage caused by radiation (109, 110) or drugs (89) in the testes in other studies. Melatonin also inhibited endoplasmic reticulum stress-induced apoptosis in reproductive tissues, and therefore exerted protective effects on diabetes-related reproductive impairment (111). The anti-apoptotic effect of melatonin could be mediated via increasing Bcl-2 gene expression, an anti-apoptotic gene marker, and lowering mitochondrial membrane potentials and pro-apoptotic Bcl-2-associated X protein (BAX), P53, caspase 3, and nuclear factor-κB (NF-κB) expression (83, 89, 112, 113). Melatonin also activated the phosphoinositide 3-kinase (PI3K)/protein kinase B (AKT) pathway in frozen-thawed human sperms (114). Activation of this pathway causes increased sperm motility, suppressed apoptotic cascade and caspase, and decreased membrane permeability and ROS production in spermatozoa (115). Furthermore, melatonin protected human spermatozoa from H₂O₂-induced DNA fragmentation and apoptosis via MT1 and extracellular signal–regulated kinase signaling (116). Some other studies also support the anti-apoptotic effects of melatonin on injuries to the female reproductive system (117, 118), heart (113), kidney (119), liver (120), and nervous tissue (121). Due to the limited evidence, we recommend that future studies assess the mechanisms related to the anti-apoptotic effects of melatonin in physical injuries to male testicular tissue.

To the best of our knowledge, this is the first systematic review and meta-analysis on the protective effects of melatonin against physical injuries to rodents’ testicular tissue. Animal models are indispensable tools for assessing new agents’ effectiveness and side effects for disease management. However, they do not completely imitate human models. Therefore, the interpretation of our findings should be conducted with caution. High statistical heterogeneity, publication bias, and low quality of the eligible studies are other limitations of our meta-analysis. Between-study methodological heterogeneity was also found in our study due to the differences between animals’ characteristics, the dose of melatonin, the intervention schedule, and model of infertility induction. Furthermore, the low number of the included studies prohibited us from doing subgroup analysis for some variables to detect other sources of heterogeneity. Almost all the available studies that have utilized animal models to evaluate the effects of melatonin therapy against male infertility were on rodent subjects. None of the included studies have evaluated and reported possible adverse effects of melatonin therapy. Finally, there are other outcomes that could have been helpful in explaining the mechanisms behind the effects of exogenous melatonin on male rodents’ reproductive system such as dihydrotestosterone, corticosterone, testicular and general immunity which were not investigated by the included studies.