INTRODUCTION

Dog overpopulation is a worldwide problem which influences public and animal health (^{Rubel & Carbajo, 2019}), because they are a source of health, political, socioeconomic and welfare problems, particularly in developing countries. These problems include roaming the streets causing traffic accidents, barking, aggression, and biting. The growing population of stray dogs in Latin American countries is alarming, several factors contribute to the increase in their population, the most common being abandonment by the responsible parties, which is the main cause of the number of dogs roaming streets or living in dog shelters (^{Mota-Rojas et al., 2021}).

The main concern of dog overpopulation is related to zoonotic diseases, especially canine rabies, an underreported and neglected disease in developing countries, which is responsible for 99% of dog-transmitted human rabies cases worldwide, killing approximately 59,000 [95% confidence interval: 25,000-159,200] people worldwide annually. Efforts are underway to reduce human deaths due to dog-mediated rabies to zero by 2030 (^{Hampson et al., 2015}; ^{Hampson et al., 2019}).

Surgical sterilization is the best method to prevent reproduction and, consequently, dog overpopulation. However, the impact of sterilization programs has been limited, mainly due to high operational costs that make it difficult to reach as many dogs as possible (^{Evans et al., 2022}; ^{Belsare & Vanak, 2020}). Because of this, mass slaughter continues to be used as the main method of population control in much of the world, being little accepted by society, so the World Health Organization published guidelines in 1990 that discourage the use of slaughter and recommend alternative methods such as contraceptive methods (^{Smith et al., 2019}).

Because of this, different research has focused on finding and proposing canine fertility control alternatives. It includes the use of synthetic hormones (e.g., synthetic androgens), chemical sterilization (e.g., zinc gluconate neutralized with arginine, calcium chloride), immunocontraception (e.g., gonadotropin hormone inhibitor vaccine) (^{Sandam et al., 2021}; ^{Root Kustritz, 2018}; ^{Asa, 2018}; ^{Massei et al., 2013}), as well as, the use of therapeutic ultrasound at different frequencies (^{Leoci et al., 2015}), among others. However, some methodologies may present different adverse effects and have not been shown to be fully effective (^{Sandam et al., 2021}; ^{Asa, 2018}; ^{Root Kustritz, 2018}). For such reason, as a complementary method, the use of phytoestrogens is observed.

Phytoestrogens (PE) are plant-derived polyphenolic compounds, which are divided into several groups, including isoflavones (genistein), lignans (enterolactone), flavones (luteolin), stilbenes (resveratrol) and coumestanes (coumestrol) (^{Adler et al., 2015}; ^{Pérez- Rivero et al., 2009}). They have similar structures to endogenous estrogens and can induce or inhibit the response in hormone receptors because they have the ability to bind to estrogen receptors (ER) alpha (ERα) and beta (ERβ) located in tissues in different organs including the reproductive. This binding is with higher affinity to ERβ (^{Rietjens et al., 2017}).

Some research indicates that PEs act as endocrine disruptors by competing with endogenous estrogens for binding to their receptors and can produce positive or negative biological effects on animal health (^{Beszterda & Frański, 2018}). Similar to what happens with any drug, it depends on the dose or route of administration of PEs the intensity of the effect on the target tissue, because they are subjected to different processes of absorption, biotransformation, distribution and excretion. In addition, their effects may vary depending on the phytoestrogen, species exposed, sex, route of administration, dose and duration of exposure, as well as the time of exposure during reproductive development or adult life. Thus, EFs have the ability to act as estrogenic agonists producing classical estrogenic effects or as estrogenic antagonists (inhibiting the effect of estrogens) (^{Domínguez-López et al., 2020}; ^{Mostrom & Evans, 2018}).

Additionally, PEs can modulate the hypothalamus-pituitary axis, they also have the ability to inhibit aromatase (cytochrome P450) mRNA expression, producing the decrease of local estrogen biosynthesis, resulting in cellular inhibition, including spermatogenesis, participating as intracellular regulators of cell cycle and apoptosis (^{Domínguez-López et al., 2020}; ^{Lephart, 2015}; ^{Cerendolo et al., 2009}; ^{Kawakami et al., 2004}).

Coumestrol (COU) is present in alfalfa (Medicago sativa), white clover (Trifolium repens), spinach (Spinacia oleracea) and soybean (Glycine max) sprouts (^{Cerendolo et al., 2009}; ^{Domínguez-López et al., 2020}). It has the ability to act as an endocrine disruptor of ERα and Erβ, which are widely distributed in the male reproductive tract during development and adulthood (^{Cooke et al., 2021}). In the dog, the presence of ER has been detected in interstitial Leydig cells, round spermatids, spermatogonia, and in the connective tissue of the epididymis (^{Serrano et al., 2008}).

A peculiarity of COU is that it can be naturally detected by fluorescence when bound to ER, it is excited at a wavelength of 390-400 nanometers (blue) and emits fluorescence with a wavelength of 532-559 nanometers (green color) (^{Wang et al., 2014}).

COU has been studied as a possible method of reproductive control in dogs. In females when administered orally, it induces a decrease in progesterone levels and an increase in serum estrogen values, without affecting the cellularity of the vaginal epithelium (^{Peña- Corona et al., 2020}). In males, when administered orally, it affects the histoarchitecture of the germinal epithelium of the seminiferous tubes (^{Pérez-Rivero et al., 2009}). The same effect has also been reported in hematophagous bats (Desmodus rotundus) males (^{Pérez-Rivero et al., 2014}).

Spermatozoa that pass from the testes to the epididymis are non-functional gametes. It is during their transit through the epididymis where, through their interaction with proteins, they mature, acquire progressive motility and the ability to fertilize eggs (^{Ali Hassan et al., 2021}; ^{Cornwall, 2009}), so the epididymis can be a target for the reproductive control of species considered to be disease transmitters. Due to the above, the aim of this study was to know the effect of subcutaneous administration of coumestrol in the epididymis of adult dogs as an alternative for their reproductive control.

MATERIAL AND METHODS

RESULTS

Sonographic indicators and gonadal morphology

Gonadal echogenicity (Table 2), showed that the gonadal parenchyma of the TCOU group (69.9±2.9) was increased (p>0.05) with respect to the echogenicity determined in the C group (63.9±2.6).

Table 2 Gonadal echogenicity in control group and coumestrol-treated dogs 

Group/Quartile	Control Median±SE	Q1-Q3 Median	Coumestrol Median±SE	Q1-Q3 Median	Mann Whitney P
Parenchyma	63.9±2.6	45.5-77.5	69.9±2.9	51.5-83.5	>0.05
Mediastinum	67.3±2.0	55.5-77.0	60.0±2.4	46.5-73	<0.01
Epididymal head	53±2.9	35.5-64	44±1.8	39-47	<0.05

Gonadal ultrasound/Gray pixels

In testicular ultrasonography (Figure 1), pixel intensity showed a decrease in echogenicity in the mediastinum region (testicular network) and in the epididymal head region (p<0.05). The evaluation of cellularity in the epididymal head of both groups C and TCOU (Table 3) showed no differences between them; however, ranges of the values determined in each group observed in the quartiles were evident.

Figure 1 Ultrasound of the testis and epididymis. A. Control group. B. OCT group, note the decrease in echogenicity at the height of the testicular hilum (arrows). C. Evident echogenicity in epididymis of dogs in the control group. D. Lower echogenicity in epididymis of dogs in the TCOU group 

Table 3 Cellularity parameters of the epididymis of dogs treated with coumestrol 

Epididymal head	Control Median±SE	Q1-Q3 Median	Coumestrol Median±SE	Q1-Q3 Median	Mann Whitney P
Morphometry
Diameter µm	273±5.7	235-306	314±5.3	279-340	<0.01
Area µm2	60882±2593	43665- 73608	79688±2716	61362- 90872	<0.01
Epithelium thickness µm	59.7±1.0	53.4-64.4	75.9±1.0	67.0-83.6	<0.01
Length of microvilli in µm	23.6±0.4	20.8-25.6	13.0±0.6	8.6-17.9	<0.01
Spermatozoa (Area of the sperm packet in µm2)	12703.9±2026.7	7991.1- 18201.5	8371.8±2605.5	781.0- 14181.5	<0.05

In the histological observations of epididymal cellularity (Figure 2), it is highlighted that the TCOU group presented a decrease in sperm mass and luminal microvilli.

Figure 2 Histological section of a segment of the epididymal head. A. Head of epididymal head control group (40X), inset detail of microvilli B. Head of epididymis coumestrol treated group (40X), inset detail magnification showing absence of the microvilli. Bar 100µm 

Presence of estrogen receptors

The presence of estrogenic receptors (Figure 3 and Table 4), was evidenced by the numerical increase in the mean of the pixels, in the connective tissue in the epididymal structures of the TCOU group of 21.5±0.9 compared to the C group which presented 6±0.5 (p>0.05).

Figure 3 Fluorescence in epididymis showing the presence of estrogen receptors (coumestrol). A. Scanty fluorescence in the epididymis of dogs of the control group, in which the decrease of epithelium stands out. B. Intense fluorescence in the epididymis of dogs treated with coumestrol in which there is evidence of increased connective tissue 

Table 4 Fluorescence intensity associated with the presence of estrogen receptors 

Group/Quartile	Control Median±SE	Q1-Q3 Median	Coumestrol Median±SE	Q1-Q3 Median	Mann Whitney P
Epithelium	1.4±0.1	0-3	3.1±0.4	0-5.7	<0.01
Connective tissue	6.0±0.5	3-8	20.9±0.9	14.2-26	>0.05

Fluorescence intensity/Green Pixels

Hormonal parameters

The hormone levels determined (Figure 4) showed differences between group C and the TCOU group. In-group C, testosterone levels were constant; on the other hand, there was a variation in estradiol levels. In the TCOU group, a decrease in testosterone levels, associated with an increase in estradiol levels, was evident from week four until week five.

Figure 4 Serum testosterone and estrogen values. A. In dogs treated with coumestrol, the decrease of testosterone associated with the increase of estradiol was evident, B. In dogs of the control group, there was no evident relationship between hormones due to the effect of coumestrol 

DISCUSSION

Normal canine ejaculates should have at least 80% morphologically normal and viable spermatozoa (^{Chłopik & Wysokińska, 2020}). In this study, normal sperm morphology of 60% or lower was found from the third week of treatment, which is similar to that reported in dogs considered infertile (^{Barbosa de Souza et al., 2015}). The decrease in sperm concentration and the increase in spermatozoa with cytoplasmic droplet are evidence of the inhibitory effect of coumestrol on gonadal activity.

The sperm motility reported in fertile dogs should be located above 70% (^{Johnston,1991}), in the present study in the dogs of the TCOU group from week 3 was located at 60% or lower, a value similar to that reported in infertile dogs (^{Barbosa de Souza et al., 2015}). In the present study, secondary alterations of sperm morphology were found, which are associated with their maturation in the epididymis or during the preparation of semen samples. Among the main alterations were strongly coiled tail from week 3 and distal cytoplasmic droplet from week 4 post-treatment. A point to consider when interpreting these results is that the average time of passage through the epididymis of spermatozoa in domestic dogs is 15 days (3 weeks) (^{Chłopik & Wysokińska, 2020}), the same time interval at which alterations began to be detected in the spermiogram.

Because the structure of phytoestrogens is similar to that of endogenous estrogens and they have the ability to bind to ER, it is possible to observe physiological effects in tissues expressing the receptor (^{Rietjens et al., 2017}; ^{Beszterda & Frański, 2018}). Once the COU molecule binds to any of the estrogen receptors (ERα and ERβ) it can emit a fluorescent signal (^{Serrano et al., 2008}; ^{Wang et al., 2014}). In the present study, such signal was detected with greater intensity in the connective tissue of the epididymis of dogs in the TCOU group, confirming that COU is absorbed and probably binds to estrogenic receptors present in this tissue. ^{Pérez-Rivero et al. (2014)} previously reported it in adult hematophagous bat (Desmodus rotundus) testes and ^{Serrano et al., (2008)} in male dogs, in both cases when COU was administered orally.

With respect to the echogenicity of the testicular parenchyma, it is homogeneous throughout the periphery, an area with decreased echogenicity is observed in the mediastinal region, which corresponds to the location of the vas deferens and the testicular network. It is consistent with the changes found at histological level related to the dilation of the tubules of the epididymis, loss of microvilli and decreased sperm content in its lumen (^{Mantziaras, 2020}; ^{Lubinus et al., 2006}).

The process of estrogen biosynthesis depends on the enzyme aromatase activity, a member of the cytochrome P450 superfamily (^{Hess & Cooke, 2018}). Aromatase is responsible for converting testosterone and androstenedione to aromatic estrogens, 17β-estradiol and estrone, respectively (^{Hess & Cooke, 2018}). It has been described that COU can bind to the enzyme aromatase, resulting in inhibition of estrogen biosynthesis, so, it is possible to consider that the increase in serum testosterone concentration is due to this effect (^{Wyse et al., 2021}; ^{Lephart, 2015}).

Estrogens are found in high concentrations in epithelial cells lining efferent ducts, their main function being to reabsorb the luminal fluid and thereby increase the concentration of spermatozoa. The biological effect absence of estrogens induces the accumulation of liquid in the efferent ducts, which finally generates dilatation of the same and of the rete testis. Subsequently there is a decrease in the epithelial height, in the number and height of the microvilli in the epididymis and subsequently atrophy of the seminiferous epithelium, all these changes favor semen dilution (^{Hess & Cooke, 2018}; ^{Pérez-Rivero et al., 2009}).

CONCLUSION

The studies performed showed that subcutaneous administration of coumestrol could be a non-invasive alternative to decrease the fertilizing capacity of dogs, since it induced alterations in gonadal activity and cellularity associated with a decrease in sperm production and sperm quality. However, studies are still required to evaluate its possible reversible effect and sperm fertilizing capacity.

Funding

The study was supported by the Metropolitan Autonomous University Xochimilco Unit (UAM), research project DCBS.445.12.

Acknowledgments

National Council of Science and Technology (CONACYT) for the A.J.R. Scholarship for Doctoral Studies in Science.

The authors thank the animal protection shelter Quality of Life Foundation A.C., for providing their facilities and support in the care of the dogs (agreement celebrated with UAM-X dated 27 June 2016).

To Laura Karen Camarillo Rodea for her help in animal handling during the clinical evaluation and collection of biological samples.

To Susana Rojas Maya(†) for her generous support and technical assistance during the hormonal evaluation.

Conflict of interest

The authors declare that they do not have any conflict of interest

Authors' contribution

A.J.R: Designed and conducted the study, collected the data and biological samples, performed the laboratory work, analyzed and interpreted the results, drafted and approved the final manuscript.

J.J.P.R: Conceived the original idea and designed the study, supervised the study, analyzed and interpreted the data, performed critical revision of the manuscript, and approved the final manuscript.

J.A.H.B: Performed critical revision of the manuscript and approved the final manuscript. A.A.S: Performed critical revision of the manuscript and approved the final manuscript. M.P.M: erformed critical revision of the manuscript and approved the final manuscript.