Metal Oxide Nanoparticles: Evidence of Adverse Effects on the Male Reproductive System

Metal oxide nanoparticles (MONPs) are inorganic materials that have become a valuable tool for many industrial sectors, especially in healthcare, due to their versatility, unique intrinsic properties, and relatively inexpensive production cost. As a consequence of their wide applications, human exposure to MONPs has increased dramatically. More recently, their use has become somehow controversial. On one hand, MONPs can interact with cellular macromolecules, which makes them useful platforms for diagnostic and therapeutic interventions. On the other hand, research suggests that these MONPs can cross the blood–testis barrier and accumulate in the testis. Although it has been demonstrated that some MONPs have protective effects on male germ cells, contradictory reports suggest that these nanoparticles compromise male fertility by interfering with spermatogenesis. In fact, in vitro and in vivo studies indicate that exposure to MONPs could induce the overproduction of reactive oxygen species, resulting in oxidative stress, which is the main suggested molecular mechanism that leads to germ cells’ toxicity. The latter results in subsequent damage to proteins, cell membranes, and DNA, which ultimately may lead to the impairment of the male reproductive system. The present manuscript overviews the therapeutic potential of MONPs and their biomedical applications, followed by a critical view of their potential risks in mammalian male fertility, as suggested by recent scientific literature.


Introduction
Nanotechnology is a field of science that studies the properties, design, manipulation, production, and applications of structures and devices at the nanoscale level (10 −9 m). Objects on this scale, such as nanoparticles (NPs), have properties and functions that differ from those with a larger scale [1]. European and other International Committees have defined NPs, as particles of matter in which at least one of their phases has one dimension (length, width, or thickness) within the range of 1 to 100 nanometers (nm) [2,3].
Among the several types of NPs reported in the literature, metal oxide NPs (MONPs) stand out as the category of versatile materials. Being a type of metallic NPs that have controllable features and small size, making them able to easily cross cells and tissues within the body to reach a target location [4,5]. This makes MONPs a valuable tool for biomedical applications, such as anticancer, antidiabetic, antimicrobial purposes, as well as imaging applications, drug delivery, and even in reproductive medicine [6].
Most of these inorganic materials that make up MONPs are typically classified as biocompatible since their metallic precursors are already present in human tissues, whose

Biomedical Applications of MONPs
MONPs are inorganic materials made to modify the properties of metallic elements. These have been subjected to intense biomedical research, mainly due to their unique intrinsic properties, such as good optical, electrical, catalytic, and magnetic behavior, chemical and mechanical stability, simple preparation process, easily engineered for the desired size, shape and porosity, and large surface area for reactions [4,5,31]. In addition, these materials can easily have their surface modified with several chemical functional groups, allowing their conjugation with antibodies, ligands and drugs of interest, which further enhances their potential in the biomedical field [5]. Although there is a wide spectrum of metals available, their use in the medical field is limited to those tolerated by the organism [32]. The fact that some metals exist in appreciable amounts in the body makes most MONPs biocompatible. For example, in the human body, iron (3-4 g) is mainly found associated with hemoglobin, making it the most abundant metal [33,34]. Followed by iron, zinc (~2 g) [35], and copper (~0.1 g) [36] are the second and third most common metals in the human body, and they are essential constituents of several enzymes. Unlike the previous metals, manganese is present in very small amounts in the body (~12 mg). However, it is one of the most important nutrients for human health as it assists in the development of connective tissue, bones, blood-clotting factors, and sex hormones [33].
The use of MONPs to treat cancer, diabetes, and even to eradicate infectious diseases has been extensively studied, which proves the effort that has been made to create a symbiosis between nanoscience and medical science [31,37].
The common biomedical applications of MONPs and their main mechanisms of action are summarized in Figure 2.
size, shape and porosity, and large surface area for reactions [4,5,31]. In addition, these materials can easily have their surface modified with several chemical functional groups, allowing their conjugation with antibodies, ligands and drugs of interest, which further enhances their potential in the biomedical field [5]. Although there is a wide spectrum of metals available, their use in the medical field is limited to those tolerated by the organism [32]. The fact that some metals exist in appreciable amounts in the body makes most MONPs biocompatible. For example, in the human body, iron (3-4 g) is mainly found associated with hemoglobin, making it the most abundant metal [33,34]. Followed by iron, zinc (~2 g) [35], and copper (~0.1 g) [36] are the second and third most common metals in the human body, and they are essential constituents of several enzymes. Unlike the previous metals, manganese is present in very small amounts in the body (~12 mg). However, it is one of the most important nutrients for human health as it assists in the development of connective tissue, bones, blood-clotting factors, and sex hormones [33].
The use of MONPs to treat cancer, diabetes, and even to eradicate infectious diseases has been extensively studied, which proves the effort that has been made to create a symbiosis between nanoscience and medical science [31,37].
The common biomedical applications of MONPs and their main mechanisms of action are summarized in Figure 2.

Antimicrobial, Anticancer, and Antidiabetic Activity
Although in excessive doses many metals are toxic to all cell types, in lower concentrations, MONPs may be able to selectively target bacteria, since their metal transport system and metalloproteins are different from those existing in mammalian eukaryotic cells [38,39]. To exert this microbial function, MONPs need to be in contact with microbial cells. This interaction increases microbes' membrane permeability, and allows the entry of NPs

Antimicrobial, Anticancer, and Antidiabetic Activity
Although in excessive doses many metals are toxic to all cell types, in lower concentrations, MONPs may be able to selectively target bacteria, since their metal transport system and metalloproteins are different from those existing in mammalian eukaryotic cells [38,39]. To exert this microbial function, MONPs need to be in contact with microbial cells. This interaction increases microbes' membrane permeability, and allows the entry of NPs into the cytoplasm [38,40], where NPs induce damage to cellular macromolecules ( Figure 2) [41]. This antimicrobial activity is enhanced for higher concentrations and smaller MONPs sizes [42,43], since smaller sizes allow a closer contact between NPs and the microbial membrane [4].
Besides presenting antibacterial and antifungal activities, some MONPs also exert antiviral properties (Figure 2). MONPs can adhere to the virus envelope, causing its destruction [52], or they can block their mechanism of viral replication [53] or viral entry into a cell [54]. Metal oxides, such as TiO 2 [52] and Cu 2 O [55], have already been shown to be effective antiviral agents against influenza A virus subtype H3N2 and Hepatitis C, respectively. These findings open a new perspective to prevent and treat viral diseases using MONPs.
MONPs can also selectively target cancer cells [56] and exert their anticancer activity mainly through the generation of oxidative stress [57]. This property can be further enhanced with the application of external stimuli such as magnetic fields or lasers, which induce the local production of heat in tumor sites [58]. Additionally, these NPs can also be used as enhancers of standard therapies, acting as co-adjuvants to improve the effect of radiation on radiotherapy, or to facilitate the action of conventional anticancer drugs, reducing the required dose and side effects of such drugs [59]. Therefore, different strategies take advantage of MONPs in the treatment of cancer: alone, conjugated with biological molecules, ligands, and anticancer drugs, or in combination with other conventional therapies to potentiate their therapeutic efficacy [60].
In addition, other MONPs such as MgO, MnO [61], CeO 2 [62], ZnO [63,64], and Fe 2 O 3 [65] have been explored as possible antidiabetic agents, since recent studies have shown promising results. Essentially, the antioxidant ability of MONPs contributes to a decrease in oxidative stress, which is the main cause of β-cell damage [66]. However, concentration determines whether NPs elicit oxidative stress or increase the cell antioxidant capacity. Generally, small doses seem to be related to the antidiabetic potential [14,65].

Drug Delivery Platforms and Imaging
Medical imaging is essential for medical diagnosis. MONPs have been used as nanoparticle-based contrast agents in multiple modern imaging modalities that allow the visualization of abnormalities, such as tumor lesions or other regions of interest [67]. Of all the plethora of available NPs, metal oxides have advantages in imaging applications due to their diverse size-and shape-dependent optoelectronic properties [27,68] and high stability, which are not achievable with traditional lipid or polymer-based nanoparticles [69]. In addition, compared to molecular probes, MONPs are virtually inert, which means that they hardly interact with other cellular molecules and, therefore, their optical properties remain unaffected [70]. Their surface can also be easily functionalized with drugs, targeting or fluorescent molecules, or other components [71,72]. Therefore, these contrast agents can deliver therapeutic agents simultaneously, allowing for a dual diagnostic and therapeutic effect [73].

An Asset for Reproductive Medicine
Although the detrimental effects of NPs on male fertility and sperm cell function have been suggested [16], some research teams have been exploring the properties of these materials to improve assisted reproductive techniques. Falchi et al. reported that the incubation of ram semen with CeO 2 NPs during cryopreservation improved sperm quantity and quality [79]. This study suggests that CeO 2 NPs can have beneficial effects on sperm preservation. Other research teams have functionalized Fe 2 O 3 NPs with lectins and antibodies, to selectively bind to glycans expressed in acrosome reaction, or to ubiquitin, which is present on the surface of defective spermatozoa [79,80]. Then, aberrant spermatozoa can be removed from a sample using a magnetic force. This method of sperm purification may be used to increase conception rates following artificial insemination [80]. Nanoplatforms for the delivery of biological compounds to spermatozoa are another nanotechnology that has been investigated in the field of reproductive medicine [15].
Makhluf et al. described the spontaneous penetration of polyvinyl alcohol (PVA)-Fe 3 O 4 NPs in bovine sperm, without affecting their motility and ability to undergo the acrosome reaction [81]. These interesting results suggest that, in the future, NPs may be conjugated with target nutrients or treatments for direct nutrient supplementation to sperm.
These and other research teams have presented interesting results that highlight the usefulness of MONPs. However, despite these promising results, uncertainty remains about the safety of MONPs. Therefore, it is crucial to investigate in more detail how MONPs interact with the male reproductive system and the consequences of this exposure.

The Impact of MONPs on Male Fertility
MONPs have received a lot of attention, especially in the biomedical field, due to their biological usefulness, as discussed in previous sections. In addition, due to their unique properties and versatility, the application of NPs extends to many other fields, making them ubiquitous in the environment. Consequently, human exposure to nanomaterials has increased dramatically. However, in recent years, the use of NPs of any material has become controversial [82]. On one hand, MONPs can interact with cellular macromolecules, leading to therapeutic effects [83]. On the other hand, cytotoxic effects were found in some tissues, presenting a health hazard [84].
Many studies suggest that human male infertility has increased significantly over the past few decades [85][86][87]. Due to this alarming trend, it has been hypothesized that environmental, dietary, and/or lifestyle changes are interfering with men's ability to produce spermatozoa with a consequent impact on male fertility [88,89]. In addition, the male reproductive system is known to be susceptible to environmental stress, as toxicants, vehicular pollutants, and even NPs [90]. As a result, the impact of MONPs on male reproductive health has become an important subject of study. While several reports suggest that some NPs might have protective effects on sperm cells [91], other reports suggest that they compromise male fertility by interfering with spermatogenesis [92]. In fact, spermatogenesis is prone to errors. Defects in any of its steps can result in the failure of the entire process and, in some cases, can lead to testicular diseases or male infertility [93,94].
Since spermatogenesis is a highly vulnerable process, it occurs in a protected environment, controlled by the BTB, whose purpose is to protect the developing germ cells from external insults [17]. It is formed by tight junctions between Sertoli cells that divide the epithelium of the seminiferous tubules (ST) into two different compartments: basal and adluminal ( Figure 3). Although it is one of the tightest blood-tissue barriers in the mammalian body [95], it was previously reported that NPs could cross this biological barrier due to their ultra-small size [16]. In fact, in mice treated with TiO 2 [96] and Fe 2 O 3 [97], both NPs were able to penetrate the testis, despite the BTB. Takeda et al. even reported that TiO 2 NPs accumulated in the testis of male offspring from pregnant mice who were treated with these NPs [98]. Other animal studies have also demonstrated that NPs can move from the initial absorption site, for example, the lungs and skin, to secondary organs, such as the testis [99]. The integrity of BTB is a concern since NPs can easily permeate cells and their nuclei. This creates favorable circumstances for mutations appearance, which in germ cells may interfere with fertilization, embryogenesis [100], or even generate congenital defects in the offspring [101]. . Schematic representation of spermatogenesis in the cross-section of a seminiferous tubule. Spermatogenesis is initiated at puberty by the hypothalamus, which produces GnRH, which, in turn, stimulates the release of FSH and LH at the reproductive tract. LH stimulates Leydig cells to produce testosterone and FSH stimulates Sertoli cells that provide support and nutrition for sperm survival, proliferation, and differentiation [102]. Sertoli cells then initiate the functional responses required for spermatogenesis. Spermatogenesis starts when type A spermatogonia (2n) commit to differentiating into type B spermatogonia. Then, through mitosis, B-spermatogonia (2n) give rise to primary spermatocytes (2n). The latter undergo a long meiotic phase that originates the secondary spermatocytes (n), which ends with spermatids (n) generation [103]. The round spermatids then go through substantial morphological changes during spermiogenesis originating highly specialized spermatozoa through the reorganization of the entire cell, where the nuclear envelope seems to be crucially involved [104,105]. The next event is spermiation, in which mature spermatids are released from the supporting Sertoli cells into the lumen of the seminiferous tubule, and the remainder of the spermatid cytoplasm, known as the residual body, is phagocytosed by the Sertoli cells [106]. However, at this stage, spermatozoa still lack motility. Immotile spermatozoa are then transported into the epididymis where the final steps of maturation occur [107]. GnRH, gonadotropin-releasing hormone; LH, luteinizing hormone; FSH, follicle-stimulating hormone; BTB, blood-testis-barrier; 2n, diploid cell; n, haploid cell, created with Biorender.com (accessed on 2 July 2021). Therefore, a clear understanding of the impact of MONPs on reproductive health is fundamental. Tables 1 and 2 summarize the adverse effects of different MONPs on the male reproductive system, both in vitro and in vivo. However, it is important to keep in mind that these effects depend on several factors, such as dosage, duration of exposure, administration route, chemical nature of the compound (e.g., method of synthesis, size, shape, surface charge), as well as the biological system involved (e.g., strain and age of animal/cell, cell variability) [15].

In Vitro Studies
Few studies have focused on the adverse effects of NPs on male germ cells in vitro (Table 1).    The summary studies provide valuable information on the outcome of the interaction between MONPs and germ cells, which is useful for establishing the mechanisms of MONP toxicity. Parameters such as cell viability, oxidative stress, DNA damage, nanoparticle internalization, and mechanisms of cell death were assessed.
The in vitro studies reported in Table 1 were carried out with NPs made from Cerium (Ce), Iron (Fe), Manganese (Mn), Titanium (Ti), and Zinc (Zn) oxides. However, TiO 2 and ZnO NPs are, by far, the most explored NPs.
The studies listed were conducted in different reproductive cells at three stages of maturation: spermatogonia, spermatocyte, and spermatozoa. Additionally, the cells responsible for testicular architecture and function, namely Sertoli and Leydig cells, were also used in the listed studies. In addition, most studies have carried out the extensive chemical and physical characterization of NPs, which is crucial for a better understanding of the toxicity mechanisms of NPs on reproductive cells.
A wide range of concentrations of MONPs has been studied, from very low (0.04 µg/mL) to high concentrations (1000 µg/mL). It is crucial to evaluate different concentrations of MONPs to establish their cytotoxic effect. However, the results still were conflicting. Préaubert et al. reported that the lowest concentrations of CeO 2 NPs (0.01 µg/mL) were associated with higher levels of DNA damage in human spermatozoa [108]. ZnO NPs were also highly cytotoxic to mouse Leydig cells, even at low concentrations and incubation times [117]. Those are the exceptions since most studies indicate that MONP cytotoxicity is dose and time-dependent. Other authors even reported that lower MONPs concentrations were inefficient to cause genotoxicity [92,111].
The periods of incubation were also variable, ranging from 15 min to 24 h. From the results summarized in Table 1, it can be deduced that the reproductive toxicity of MONPs depends mainly on the concentration used and on the time of incubation.
The size of the NPs used ranges from ultrafine particles (7 nm) to much larger NPs (177 nm). Previous studies reported that even a small difference in size can make particles up to six times more harmful [119]. Gromadzka-ostrowska et al. also found that the toxicity of NPs is not only dependent on dose and time, but also depends on size, which seems to be inversely proportional to the cytotoxicity of NPs [120]. However, none of the studies reported in Table 1 evaluated the effect of the size of NPs on male germ cells.
The most studied parameters were oxidative stress indexes, cell viability, apoptosis, and genotoxicity. The principal suggested mechanism by which MONPs may exert that their toxic and genotoxic effect is oxidative stress [113,117]. In fact, increased oxidative stress was observed in almost all studies where this parameter was tested, except one [117]. Bara and Kaul reported an increase in the levels of antioxidant enzymes SOD and CAT in Leydig cells after exposure to ZnO NPs [117]. However, it has also been reported by other studies that NPs initially induce antioxidant enzyme activities in response to stress, as a defense mechanism, but, eventually, ROS production overcomes the capacity of the antioxidant response mechanisms [121].
Both exogenous stimuli and endogenous physiological stress can induce ROS production [117]. Oxidative stress is known to induce DNA damage through the oxidation of DNA bases [108] (Figure 4). However, it can also induce injury to biomolecules and organelles in other cells, mainly mitochondria [117]. In addition, under stress conditions, cells activate different cellular processes important for cell adaption to adverse conditions or to activate cell mechanisms of cell death, such as apoptosis or necrosis [117]. Pinho et al. reported an increase in the number of spermatogonia in necrosis (but not apoptosis) after ZnO NP exposure [92], while other studies have reported apoptosis as the preferred mechanism of cell death [110,117,118]. Autophagy is an example of an adaptive mechanism under stress conditions, and it was reported in Leydig cells after ZnO NPs exposure [118].
The mechanism of MONPs internalization by cells was explored in some studies. Pawar and Kaul, using Scanning Electron Microscopy (SEM) and Transmission Electron Microscopy (TEM) images, reported that TiO 2 in both agglomerated and single forms can remain attached to the spermatozoon surface (head and tail) after the addition of NPs to the sperm suspension, even after washing [111]. This indicates that NPs can attach and remain intact on the cell membrane immediately after mixing the NPs with the cell suspension. When in direct contact with cells, NPs cause mechanical damage to the membrane and destabilization of the plasma membrane, allowing NP entrance. The latter will exert pro-oxidant effects. In fact, Mao et al. monitored the internalization of TiO 2 NPs by spermatocytes and Sertoli cells, both by flow cytometry and by TEM [112]. Bara and Kaul TEM results also revealed that ZnO NPs can enter Leydig cells and cross their nuclear membranes [117]. Moreover, Préaubert et al. also found an accumulation of CeO 2 NPs at the spermatozoon plasma membrane [108]. However, in this case, the NPs were not internalized by the cells, but genotoxicity was still present. These authors proposed that MONPs do not need to be internalized to induce cell damage. To date, the exact mechanism by which the NPs induce cell damage is far to be elucidated, and, therefore, more comprehensive studies are needed.
Changes in the cytoskeleton were assessed only by Mao et al. and Pinho et al., using TiO 2 NPs and ZnO NPs, respectively [92,112]. The latter reported disturbances in both microtubules and microfilament networks in spermatogonia cells [92]. Mao et al. also studied the effect of TiO 2 NPs on the cytoskeleton of two different germ cells, spermatocytes, and Sertoli cells. TiO 2 NPs interfered with microtubules of spermatocytes, but Sertoli cells only had their microfilaments altered [112]. These studies indicate that different germ cells respond differently to MONP insults. Additional studies should investigate alterations in the cytoskeleton since changes in the microtubule dynamics affect the formation of sperm flagella and migration abilities, and changes in the microfilament dynamics can affect the formation of tight junctions of Sertoli cells, which altogether interfere with spermatogenesis [112]. Although Liu et al. did not study cytoskeleton dynamics, their results indicate downregulation of tight junction proteins in Sertoli cells, leading to BTB impairment [116]. In addition, the disturbed microfilament arrangement interferes with the phagocytic capacity of Sertoli cells, which makes cells unable to properly phagocytose abnormal sperm cells [112].
Besides studying the cytoskeleton, Pinho et al. also reported, for the first time, the impact of ZnO NP exposure in the nucleoskeleton [92]. These authors reported several nuclear alterations in spermatogonia that may affect the progression of spermatogenesis. Bara and Kaul was the only in vitro study to investigate the effect of NPs on steroidogenesis and testosterone biosynthesis in male reproductive cells. Interestingly, they found that a low concentration treatment with ZnO NPs for short incubation periods enhanced the steroidogenic ability of Leydig cells [117]. However, the exact mechanism is still unclear and therefore should be explored in future studies.
Overall, the interesting data collected indicate that the reproductive toxicity of NPs is not simply a matter of the NP material type, size, concentration, and exposure time, but also the result of intricate interactions at the nano-bio interface, which is influenced by many factors [13].
Since in vitro studies cannot consider tissue distribution, organs accumulation, clearance, or diffusion across biological barriers, such as the BTB, in vivo studies must be considered [122] and are of paramount importance to understand NP cytotoxicity. Table 2 lists the biochemical, molecular, and histopathological evidence of reproductive toxicity of MONPs. All MONPs that have been used in previous in vitro studies were also applied in in vivo studies. Considering the aluminum oxide (Al 2 O 3 ) NPs, they have not been evaluated under cell culture conditions, only in vivo.   -Transcription profiling in the testis -Increase in Mn content in serum and testis (≥60 days); -Reduction of the thickness of germinative layer (≥60 days) and ST degeneration (120 days); -Decline in testosterone and FSH but increase in LH levels (120 days); -Increase in SOD and MDA levels (120 days); -Increase in sperm abnormalities, decrease in sperm concentration and motility (120 days); -Decrease in fertility and fetal survival rate (120 days); -Upregulation of PPAR-signaling pathway and increased expression of cytochrome P450 [110] Table 2 clearly shows that animal models used for the in vivo experiments were mice and rats of different strains. Most of the studies listed have addressed the toxicity of MONPs at concentrations that are far from real-life conditions, even though there is no information available on the concentration of MONPs to which humans are exposed. Lauvås et al. used a lower and more realistic intratracheal dose of TiO 2 NPs (63 µg/week for seven weeks), based on the estimated lung deposition of titanium at the Danish occupational exposure limit [137].

In Vivo Studies
The exposure times used for the studies were highly variable. In some studies, male mammals received MONPs for very short periods, like 4 days [126], and, in other studies, the MONPs were used for much longer periods, namely six months [131]. These studies with these differences in exposure times are crucial since they help to create a better understanding of the acute and long-term effects of MONP administration. Additionally, many experiments have established the duration of treatment at around four weeks, to accomplish the duration of complete spermatogenesis in mice and rats [149].
Different routes of MONPS administration were used in animal experiments, namely, oral, intragastric, intratracheal, intraperitoneal, intravenous, and subcutaneous administration. It has been previously reported that there is very low absorption of MONPs through inhalation or oral administration in animals [62]. This aspect was confirmed by Lauvås et al., which was the only study included in Table 2, which administered MONPs intratracheally, and found that sperm cells are not susceptible to MONP exposure via the airways, at low doses [137]. On the other hand, in oral exposure, MONPs release more ions in the stomach due to the acidic environment. Therefore, this dissolution may be a reason for the cytotoxicity reported in the studies that used this administration approach, although fewer amounts of NPs are absorbed [145]. In contrast, intraperitoneal injections ensure proper absorption of the tested MONPs due to the highly vascularized peritoneal cavity [140]. The intravenous administration of nanomaterials ensures a much higher direct testicular exposure since NPs are administered directly into the bloodstream.
Regarding the parameters observed, most studies measured the weight of the male reproductive organs. Only Tang et al., Yousef et al. and Radhi et al. reported its increase after the oral administration of NPs, which may be attributed to the inflammation and hypertrophy or even accumulation of NPs in those tissues [90,123,147]. In fact, all studies that evaluated the content of MONPs in the testis and epididymis confirmed their presence in these organs. This was the case for cerium [124], iron [97], manganese [110], titanium [131,134,138], and zinc [90] NPs. The only exception was reported by Miura et al. studies, in which TiO 2 NPs administered intravenously were found in the testis, but not in significant amounts [134,138]. This deposition of NPs in the reproductive tissues triggers the harmful events that will be described throughout this section. In fact, the damage has been reported in the testis and epididymis. Al 2 O 3 [123], F 2 O 3 [125], Fe 3 O 4 [126], Mn 3 O 4 [110,128], MnO 2 [129], TiO 2 [131,132,135,136,139,140], and ZnO [123,140,141,143,144,146] NPs all caused histopathological changes in the testis, mainly due to degeneration of the seminiferous tubules. Furthermore, Morgan et al. studied the histopathological changes induced by TiO 2 NPs in the prostate and seminal vesicle, and reported that these reproductive organs were also affected by NPs, since they caused congestion, hyperplasia, and desquamation of the prostate's epithelial, lining, and congestion in the seminal vesicle [133]. Salman also reported that ZnO NPs caused mild damage in seminal vesicles but severe damage to the prostate [148]. The reduction in the testis cell population has also been commonly reported, which is an indicator of a lack of active spermatogenesis in the testis [150].
The translocation of MONPs from their site of administration to the testicular tissue confirms that these NPs can cross and enter the BTB, where they interfere with normal physiological processes. Then, when in contact with reproductive tissues, these NPs can permeate cell membranes, inducing the overproduction of ROS, which leads to oxidative stress ( Figure 4). This interferes with the balance between the oxidant and antioxidant systems, which causes oxidative damage in biomolecules, such as lipids, proteins, and nucleic acids [97]. To confirm the oxidative damage caused by MONPs, different studies evaluated ROS production and the levels of other oxidant markers, such as Malondialdehyde (MDA), Nitric Oxide (NO), Protein Carbonyl Content (PC), Lipid Peroxidation (LPO), and Total Oxidant Status (TOS). Antioxidant parameters such as Superoxide Dismutase (SOD), Glutathione Peroxidase (GPx), Reduced Glutathione (GSH), Catalase (CAT), and Total Antioxidant Capacity (TAC), were also evaluated. These parameters of oxidative stress were assessed on all types of MONPs, except CeO 2 NPs [124]. The results reported an increase in oxidant markers and a decrease in intracellular antioxidant defenses and TAC. This confirms that MONPs suppress the antioxidant machinery and induce oxidative stress, which can lead to various cellular damages and, consequently, interfere with male fertility. In fact, according to previous studies, 30-80% of male infertility cases can be attributed to oxidative stress-mediated injury to the male reproductive system [110,151,152]. Persistent oxidative stress leads to the downregulation of Bcl-2 and upregulation of Bax, which results in the leakage of cytochrome c from dysfunctional mitochondria, ultimately resulting in apoptosis (Figure 4 [97,118,132,135]. MONPs not only induce apoptosis, but some have also proven to be autophagy activators and inducers of autophagic cell death [118]. The levels of endocrine and reproductive hormones were also evaluated, and the results also suggest an imbalance in reproductive hormones (Testosterone, FSH, LH, GnRH, E2) and thyroid hormones (TSH, T3, T4) that can be attributed to the increase of ROS and the concomitant reduction of antioxidant enzymes. The exceptions were Lauvås et al. and Ogunsuyi et al., who reported that TiO 2 NPs did not trigger alterations in testosterone levels [137,140]. Contrarily, Miura et al. reported that TiO 2 NPs affected testosterone levels, but not FSH, LH, and GnRH [134].
Additionally, some authors explored the influence of MONPs on the expression of genes related to steroidogenesis. Testosterone is produced mainly in Leydig cells by a series of enzymatic reactions. First, the StAR protein transfers cholesterol to mitochondria.
Then, the mitochondrial cytochrome P450scc transforms cholesterol into pregnenolone. Subsequently, other enzymes (3β-HSD, P450c17, 17β-HSD) convert the pregnenolone into testosterone [124]. Interestingly, Nr5A1, a transcription factor that regulates the expression of steroidogenic genes in Leydig cells (such as 3β-HSD), was downregulated after exposure to ZnO NP [144]. The StAR protein was also downregulated by CeO 2 [124] and ZnO NPs [90], which can manifest in their inability to transfer cholesterol to the inner mitochondrial membrane, which stops steroidogenesis and justifies the decline in testosterone levels in most of the results listed. However, Bara and Kaul reported the conflicting results of increased testosterone production and StAR upregulation, but this was only related to small concentrations of ZnO NPs [117]. Ogunsuyi et al. did not report alterations in testosterone levels after intraperitoneal administration of TiO 2 NPs; however, these levels were increased in the same study, under the same conditions, by ZnO NPs [140]. Likewise, Lauvås et al. found no significant alterations in testosterone levels after intratracheal administration of TiO 2 NPs [137].
Sperm parameters, such as sperm number, viability, abnormalities, and motility, have been extensively studied. All studies that analyzed sperm count observed its decline with increasing concentrations of NPs, except for Varzeghani et al., Lauvås et al. and Song et al., who did not report significant alterations [126,136,137]. The results listed in Table 2 also indicate a reduction in motile spermatozoa, which affects their fertilizing potential. This decrease in sperm motility may have been a result of lipid peroxidation [140] ( Figure 4). In addition, Morgan [133,135,142,144,145]. An increase in sperm abnormalities, such as small head, double head, formless head, and double tails, has also been reported, which may be the result of oxidative damage [140] (Figure 4). These results are in agreement with those reported under in vitro conditions (Table 1).
Hong et al. evaluated the activity of metabolism-related enzymes-LDH, SDH, and SODH-that play key roles in the growth and development of testicular cells [130]. The results suggest that there was a decline in their activity, which may be associated with the disturbance of energy metabolism in germ cells. It was also the only study to evaluate the testicular activity of G-6PD, testis-marker enzymes ACP and AKP, and the activity of Ca 2+ -ATPase, Ca 2+ /Mg 2+ -ATPase and Na + /K + -ATPase. G-6PD is associated with androgen biogenesis, and its reduction implies that TiO 2 NPs interfered with androgen secretion. In this study, ACP and AKP were used as markers of impaired spermatogenesis. Since ACP is related to the degeneration of the seminiferous epithelium and AKP is related to the activity of division of germ cells, their increase suggests testicular degeneration. Reductions in ATPases suggest an imbalance in the concentrations of intracellular ions, which could promote spermatogenesis dysfunctions [130].
Due to their small size, MONPs can reach the nucleus and interact directly with DNA, which causes the generation of ROS that further damages DNA (Figure 4) [146]. Not all studies tested the genotoxicity of NPs, but all studies that evaluated DNA damage later confirmed it. Mesallam et al. detected DNA fragmentation in the testis and prostate of rats treated with 422 mg/kg ZnO NPs daily for four weeks [146]. Meena et al. also found DNA strand breaks in spermatozoa of rats treated with 25 and 50 mg/kg TiO 2 NPs weekly for 30 days [132].
Results also indicate elevated levels of TNF-α [123,146], and pro-inflammatory IL-6 cytokine [123], and a decrease in anti-inflammatory IL-4 cytokine [146] in reproductive tissues, which indicates a cellular inflammatory response to the NP exposure.
Zhang et al. evaluated male fertility by assessing the offspring of rats treated with Mn 3 O 4 NPs [110]. The obtained results confirmed that this treatment decreased rats' fertility and reduced the survival rate of their offspring in a time-dependent manner. For these authors, these results are attributed to changes in reproductive hormones and the decline in sperm quality [110].
In summary, most biochemical and molecular results were concomitant with histological findings. Therefore, despite the many benefits of MONPs, the results of the listed in vivo studies confirm the in vitro studies, emphasizing the possibility that exposure to these NPs could have a detrimental impact on male fertility.

MONPs in Human Reproductive Medicine
The recent approval of MONPs-based technologies in clinical medicine allowed an increase in human living standards and an improvement in mankind's healthcare conditions through the prevention, early detection, diagnosis, treatment, and follow-up of multiple diseases [153]. However, their usefulness in human reproductive medicine has yet to be proved.
Considering that 50% of infertile couples, the male partner is affected by aberrations in sperm properties, number, vitality, and morphology [154], there is a clear need to develop novel methodologies for the early identification of infertility causes and its treatment. Some research teams have already developed MONP-based approaches that were tested in vitro and in vivo, with promising results. These include methods to reduce oxidative stress induced by cryopreservation [155], improve the proportion of healthy spermatozoa in semen prior to insemination [156], provide movement to sperm with motility deficit [157], protect the fertility of men who are exposed to fertility disrupters [158], and even treat other male associated disorders, such as erectile dysfunction [159].
Although these and other approaches have shown promising results, most of the literature still suggests uncertainty regarding the risk of MONPs in fertility, which may be one of the main reasons why, to date, there are no trials involving this type of engineered NPs for fertility regulation and treatment of male reproductive diseases. Another limiting factor is that only a few studies tried to identify the exact mechanism and pathways induced by MONPs. Current animal experiments also fail to assess pregnancy rates, and the health of offspring, which is the most relevant outcome parameter of fertility [160]. This gap in literature allows the speculation around the hazard posed by MONPs, which could prevent the translation of the results from the lab to the clinical applications [161]. NPs represent a valuable tool to alleviate much of the suffering arising from many reproductive difficulties and disorders, but further work is required to determine if these NPs can fulfill the needs in reproductive health. Human clinical reproductive trials may help accelerate the commercial availability of these new alternatives.

Conclusions and Future Perspectives
The increased application of MONPs in many industries and scientific fields has made these materials highly present in the environment, resulting in an increased risk of human exposure. Additionally, evidence that keeps emerging suggests that MONPs interfere with the male reproductive system at many biological levels.
The results presented in this review from both in vitro and in vivo studies prove that MONPs can interfere with the male reproductive system, and these results should not be ignored. The collected data show that this reproductive toxicity is achieved due to the MONPs' ability to interfere with cell molecules and reproductive hormones, which often results in DNA damage and altered gene expression. It was also reported that MONPs induce oxidative stress in germ cells, which affects their number, quality, morphology, and activity. At the organ level, MONPs can cross the BTB and accumulate in the testis, resulting in many histological alterations in tissues of the reproductive system. Since the normal physiological processes that occur in the male reproductive system are highly complex and vulnerable, the interference of MONPs at any level can be deleterious and impair male fertility. Whether these harmful effects are reversible or not is still unclear and should be investigated in further research. How these alterations affect pregnancy and offspring is still an unresolved issue and should be addressed in future studies.
In the studies presented, the only conditions considered to evaluate the reproductive toxicity of MONPs were concentration and duration of exposure. However, the size and surface area are two crucial physical properties that affect how MONPs interact with cells and thus greatly determine the cytotoxicity of NPs. In addition, current studies generally focus on individual alterations but fail to establish a relationship between them. This may be partly the reason why the exact mechanism of nanotoxicity is not yet fully elucidated. Therefore, future studies should make a more in-depth examination of the molecular mechanisms of NPs and MONPs, in particular in reproductive toxicity and the interaction between each reported alteration. In addition, the in vivo studies are of significant heterogeneity, mainly due to the difference in the route of administration and the highly variable administered doses and exposure times. All of these factors can potentially be a source of toxicity that may influence the outcome of the studies. In some cases, unrealistically high concentrations of MONPs were used in cell culture and animal studies, which obviously results in cytotoxicity. Those studies lead to discouraging results that affect the accurate estimation of the reproductive health risks and hinder clinical translation.
It is reasonable to conclude that there are still difficulties in evaluating the reproductive toxicity of MONPs and in understanding exactly how they interact with the male reproductive system. The results summarized in this review reinforce the need for further studies with uniform protocols to obtain solid results with real implications in humans.