The Toxicology of
Male Reproduction

In the field of toxicology, the adverse effects of greatest concern are those of chronic toxicity, cancer and reproductive dysfunction. For this reason, it has not been unusual to exclude women of childbearing age from certain occupational environments, but only recently have employers given serious consideration to potential adverse effects on male reproduction (Foster, P.M.D., Lamb, J.C. IV, 1988). Estimates of couples who desire but fail to have children range from 10 to 12% (Field, B., Selub, M.S., Hughes, C.L., 1992), though the proportion of these failures which are due to a failure in the male system is unknown. Recent evidence, however, suggests that the mean sperm count in normal Western men has fallen by 40 to 50% over the last 50 years (Sharpe, R.M., 1992).

Total reproductive capacity can be described as a function of four factors;

A) Fecundity or rate of fertility from individual mating.
B) Sexual Behaviour that influences frequency and timing of mating,
C) Reproductive life span,
D) Capability of carrying the conceptus to term,

so the number of viable offspring can be described as f(ABCD).

Of these, only A, B, and C are relevant when assessing male reproductive capacity, and for practical purposes only A and B are generally included in an assessment - except when a toxin causes permanent azoospermia. Male reproductive toxins are encountered in many ways; industrially and environmentally, therapeutically, and self-administered as recreational drugs. To properly evaluate a chemical for potential effects on male reproductive function, the entire integrated male reproductive system must be considered. Figure 1 outlines the major organ systems or processes that are essential for normal reproduction and gives selected examples that effect them. Reproductive effects are often not confined to a single system. For example a deleterious effect on hormone production will affect sperm production, and vice versa.

The purpose of the following review is to summarise scientific knowledge in the area of male reproductive toxicology, with particular emphasis on work performed since 1980. Firstly, some of the basic methods of investigation will be reviewed, followed by information regarding individual toxicants, categorised with respect to each target system of the male. A short introduction to the physiology of each target system is included, but for a more comprehensive overview of the physiology of male reproduction, see Waller et al. (Waller, D.P., Killinger, J.M., Zaneveld, L.J.D., 1985).

The full range of investigative procedures involved in male reproductive toxicology is complex and involved. For a detailed examination, the reader is referred to the overview provided by Lamb (Lamb, J.C. IV, 1988).

It is very difficult to assess human male fertility - any disturbance in sperm production is probably asymptomatic, and any attempt to assess the actual fecundity of a population (for example by assessing how many children are fathered over a given period of time) is bedevilled by the number of non-toxicological factors which could affect the outcome (see Sharpe, R.M., 1992).

Male reproductive capacity in the human can be assessed via the following parameters:

1. Reproductive history. Primarily infertility, parity of sex in offspring, birth defects, spontaneous abortions, prenatal mortality, developmental disabilities. Also useful is information on factors known to affect reproductive ability, such as maternal age, paternal age, smoking habits alcohol use.

2. Sperm analysis. A variety of tests can be used to evaluate the quality of sperm produced, including sperm count, motility, morphology (seminal cytology), and double-Y body ( a fluorescence technique thought to detect Y-chromosome non disjunction) (Wyrobeck, A.J., Gordon, L.A., Burkhart, J.G., Francis, M.W., Kapp, R.W. Jr., Letz, G., Malling, H.V., Topham, J.C., Whorton, M.D. (2), 1983). No one test is more biologically responsive; all may be needed for unknown agents. For the very few agents studied with both human and mouse sperm tests, similar responses are seen, and there is strong evidence that human sperm tests can be used to identify chemicals which affect sperm production (Wyrobeck, A.J. et al. (2), 1983). This is, however, made difficult by the variation in semen characteristics within the normal population. The sperm count in individual normal men can vary between 20-150 million/ml of ejaculate, and even in any individual man the count can range over a period of months from 7-170 million/ml (Sharpe, R.M., 1992). Added to these difficulties is that the effect of a toxicant will not be uniform between men, and that the collection of semen samples by masturbation can pose problems of compliance, as well as ethical, moral and religious questions (Sharpe, R.M., 1992).

3. Hormone evaluations. Luteinising Hormone (LH), Follicle Stimulating Hormone (FSH), and testosterone are commonly analysed (for an overview of the action of these hormones, see endocrinology, below). The sensitivity of this approach is limited, however, since serious disturbances in spermatogenesis are often observed with normal FSH (Sever, L.E., Hessol, N.A., 1985). Similarly normal testosterone levels are very broad, and this is further complicated by the fact that Leydig cells often increase production to compensate for damage.

4. Testicular histology. A highly invasive procedure, and so not often used. When a biopsy is performed however, valuable information can be gleaned (see, for example, DBCP under spermatogenesis, below)

By correlating mouse sperm tests with other germ cell mutation tests, Wyrobeck, A.J., et al. demonstrated that it was of high specificity but low sensitivity (Wyrobeck, A.J., Gordon, L.A., Burkhart, J.G., Francis, M.W., Kapp, R.W. Jr., Letz, G., Malling, H.V., Topham, J.C., Whorton, M.D. (1), 1983).

A comprehensive method for the analysis of a potential reproductive toxicant in a fast breeding laboratory animal such as the mouse is the Reproductive Assessment by Continuous Breeding protocol (RACB) (Fail, P.A., George, J.D., Seely, J.C., Grizzle, T.B., Heindel, J.J., 1991). This is a three generation study which involves dosing at three levels, plus a control. There is firstly a pre-mating dosing period of 13 weeks. The animals are then paired and dosing continues until the pups are weaned (a total of 27 weeks). The F1 generation is crossed with controls. Various parameters are measured including percentage of cohabited pairs having at least one litter (fertility index), number, weight and sex of pups, and samples from all stages and dosing regimes are sent for necropsy. The reproductive consequences of the test compound on spermatogenesis (as well as the consequences of in utero, lactational and postnatal exposure on male reproductive development) can therefore be followed through a full reproductive cycle. The disadvantage of such a comprehensive study is the time and cost involved.

Williams et al. have sugested using the rabbit as the animal of choice to be evaluated in risk extrapolation studies for male reproductive toxicology, since it has several advantages over rodent species (Williams, J., Gladden, B.C., Turner, T.W., Schrader, S.M., and Chapin, R.E., 1991). For example, semen can be repeatedly collected over time and analyzed enabling longitudinal monitoring of testicular sperm production. In addition, the biologically sensitivity of male fertility assessment is potentially increased in the rabbit compared with rodents, since the ability to use artificial techniques allows reduced numbers of sperm to be inseminated.

Nuclear magnetic resonance (NMR) has been reported to be able to identify different stages of spermatogenesis within the developing testis, and has been used to evaluate the integrity of the blood testis barrier (Farghali, H., Williams, D.S., Gavaler, J., Van Thiel, D.H., 1991). It can also be used to measure the concentrations of various metabolites (particularly those containing phosphorus, such as adenosine triphosphate, ATP), and measurement of the degree of protonation of inorganic phosphate (Pi) can be used to give a good indication of intracellular pH.

The adenohypophysis of the male produces three peptide hormones that modulate reproductive function: luteinizing hormone (LH), follicle stimulating hormone (FSH), and prolactin. Release of both FSH and LH is induced by the hypothalamic hormone, gonadotrophin releasing hormone (GnRH). The levels of LH and FSH are controlled by negative feedback mechanisms which appear to be independently controlled (figure @).

LH stimulates the Leydig (interstitial) cells to synthesize and secrete the steroid hormone testosterone, which in turn inhibits the release of LH. The pathway leading to testosterone production is shown in figure #. Other steroids are also synthesised by the Leydig cells such as dehydroepiandosterone and androstenedione, but these are only weakly androgenic. Androgens are also synthesized in other nonreproductive tissues such as the adrenal glands but normally in much smaller amounts.

FSH stimulates the Sertoli cells of the seminiferous tubules to produce androgen binding protein (ABP) which is structurally related to the blood protein testosterone-oestradiol binding protein. The ABP-testosterone steroid-protein complex probably moves via the Sertoli cell to other germ cells and to the epididymis, where the testosterone is released to exert its physiological effects. FSH also causes the Sertoli cells to release a poorly characterised protein, inhibin, which decreases the release of FSH from the adenohypophysis.

Testosterone, the primary androgen, controls the functional activity of all male reproductive tract structures. Spermatogenesis in seminiferous tubules, sperm maturation in the epididymis, amd the secretory activity of the accesory sex glands all require adequate levels of testosterone, athough spermatogenesis in the rat can be maintained with intratesticular testosterone concentrations as low as 20% of normal values (Mably, T.A., Bjerke, D.L., Moore, R.W., Gendron-Fitzpatrick, A. Peterson, R.E., 1992). Testosterone also acts at many extragonadal tissues involved in the expression of secondary sex characteristics, such as facial hair, voice, and muscle mass. Aggresion, libido, and male behaviour are probably mediated by testosterone receptors in the central nervous system (CNS). The anabolic action of testosterone includes effects on liver, kidney, bone, and muscle.

When ethanol is fed to mice as 5% or 6% in drinking water, testosterone levels are depressed. (Robert, B.R.W., Anderson, R.A. Jr., Oswald, C., & Zanveld, L.J.D., 1983). This effect appears to be reversible, but the degree of reduction is only poorly correlated with blood ethanol. Similarly, when ethanol is given as a single, acute administration to normal healthy men, the result is a fall in testosterone and a rise in LH (Mendelson, J.H., Mello, N.K., Ellingbee, J., 1977). They suggested that the timing of these changes indicated a suppresion of plasma testosterone via peripheral mechanisms which regulate the biosynthesis and/or biotransformation of the steroid. Ethanol at levels commonly seen in the blood of chronic alcohol-ingesting men has been shown to inhibit the activity of 17a-hydroxyprogesterone aldolase (the enzyme which forms dehydrotestosterone from 17a-hydroxyprogesterone, see figure #) in a concentration dependent manner (Johnston, D.E., Chiao, Y.B., Gavaler, J.S., & Van Thiel, D.H., 1981). It is also shown to markedly inhibit gonadotrophin and cyclic-AMP (cAMP) stimulated testicular steroidogenesis both in vivo and in vitro (Cicero, T.J., Bell, R.D., 1982). The mechanism in vitro is quite specific, blocking the last stage in the biosynthetic pathway - that of the conversion of androstenedione to testosterone. In vivo, however, the effect is likely to be more wide-ranging, blocking all stages in the pathway (Cicero et al, 1982). Since the metabolism of ethanol involves the production of NADH (Stryer, 1988), it has been considered that an alteration of the testicular NAD+/NADH ratio may be responsible for the non-specific effects on steroidogenesis. However, the ratio in the testes was found to be unaltered (Cicero et al., 1982) and so other, as yet unidentified mechanisms must be responsible. These could be;

1. Inhibition of the uptake by the testes of cholesterol.

2. Blockage of gonadotrophin binding.

3. Blockage of the consequences of receptor occupation.

The chronic hormonal effects of ethanol in man are difficult to ascertain due to the heterogeneity of the study population with regard to drinking history, age and alcoholic liver disease. However, it is apparent that most hypogonadal alcoholics have circulating testosterone and oestradiol levels within the normal range (Anderson, R.A. Jr., Willis, B.R., Oswald, C., Zaneveld, L.J.D., 1983). LH and FSH levels are frequently higher than normal, however there is no relationship between elevated gonadotrophin levels and testicular atrophy.

Oral administration of D9-tetrahydrocannabinol, the major active ingredient of cannabis (Marijuana), increases both plasma testosterone and LH concentrations in male mice (Daltero, S., Bartke, A, Mayfield, D., 1981). The effect appears to be a direct response of the the glandular tissues involved, as the release of testosterone does not require the presence of the pituitary, and the administration of testosterone does not inhibit the release of LH. The effect on testosterone is biphasic, causing rapid sustained increases in plasma levels at low doses and subsequent decreases at higher doses. Daltero et al suggested that this may be responsible for the reported aphrodisiac effects of cannabis, which are arousal at low doses or in occasional users, followed by suppresion of libido at high concentrations.

The nematocide dibromdichloropropane (DBCP) causes increased FSH levels in man (Potashnik, G., Yanai-Inbar, I, 1987), but this effect is probably secondary to its deleterious effects on spermatogenesis. High FSH levels are indicative of a poor prognosis as the decreased feedback from compromised sperm production allows greater release of FSH.

Boric acid is thought to raise LH and FSH, and lower testosterone, in treated rats, which is possibly a cause of its known reproductive toxicity (Fail, P.A., George, J.D., Seely, J.C., Grizzle, T.B., Heindel, J.J., 1991 - but see below). TCDD when given to pregnant rats lowers testosterone levels in the male offpsring, but there was little or no effect on plasma FSH concentrations (Mably, T.A., Bjerke, D.L., Moore, R.W., Gendron-Fitzpatrick, A., Peterson, R.E., 1992).

Ethane 1,2-dimethane sulphonate (EDS) is a specific toxin for the Leydig cells of the rat testis. A single injection of EDS is cytotoxic to Leydig cells, resulting in a loss of plasma testosterone and increased LH hormone, followed 7 to 21 days later by disruption of spermatogenesis (Sprando, R.L., Santulli, R., Awoniyi, C.A., Ewing, L.L. & Zirkin, B.R., 1990). Under the influence of LH the Leydig cells recover after 6 weeks, even with chronic administration (Morris, I.D., 1985), suggesting that the regenerating Leydig cells are functionally different from the mature cell.

Cadmium has a toxic effect on many enzymes dependent on iron as a cofactor, one of these being cytochrome P-450 (Maines, M.D., 1984). The Leydig cells contain ten times more of this than the Sertoli cells, and are very sensitive to low levels of Cd2+ - again unlike Sertoli cells. Since cytochrome P-450 is required for the functioning of 17-a-hydroxylase and 17-20-lyase, its disruption may well interfere with testicular steroidogenesis.

Cimetidine, an antihistamine which specifically blocks H2 receptors, is used in the treatment of peptic ulcers. A know side effect in males is impotence and the loss of libido. This was examined in three cases (Peden, N.R., Cargill, J.M., Browning, M.C.K., Saunders, S.H.B., Wormsley, K.G., 1979), and the subjects were found to have abnormally high gonadotrophin levels but normal testosterone levels, suggesting an anti-androgen effect (though no mechanism was suggested).

The male genital tract consists of the testicles and epididymides (suspended in the scrotal sac), the vas deferens that leads to the penile urethra, and the accesory sex glands. The testical consists of interstitial tissue and seminiferous tubules. Spermatogenesis, the formation of the spermatozoon, occurs in the seminiferous tubules of the testis.

The only nongerm cell of the seminiferous tubules is the Sertoli cell. The primary function of this cell appears to be to maintain a biochemical milieu that is optimal for spermatogenesis. The seminiferous tubules are limited by a basement membrane on which the Sertoli cells rest as well as the most basic germ cells, the type A spermatogonia. Sertoli cells extend from the basement membrane to the lumen of the tubules and surround the germ cells (figure #). Where they come into contact with each other, they are interconnected by tight junctions.

The physical presence of the basement membrane, the tight junctions, and the biochemical activity of the Sertoli cells form an effective blood-testes barrier so that only a limited number of substances can enter the seminferous tubules.

Significant species variations are present in spermatogenesis. Much of our knowledge is based on the rat, but insight into other species, including man, is growing. In all animals, type A spermatogonia divide mitotically into a type B spermatogonium and another type A spermatogonium. Intermediate type spermatogonia have been identified in the rat but appear to be absent in man. Type B spermatogonia are the precursors of a series of primary spermatocytes - initially resting (preleptotene) spermatocytes which differentiate sequentially into leptotene, zygotene, pachytene, and diplotene spermatocytes. During this development the first meiotic division is undergone, and finally two pseudodiploid secondary spermatocytes are formed. Primary spermatocytes have a long life span (3 weeks or more in man), but secondary spermatocytes undergo the second meiotic division within a few hours to become spermatids, which are haploid.

The differentiation of spermatids into spermatozoa is called spermiogenesis, and requires about 3 weeks in man. It proceeds in several stages;

1. Condensation of the nuclear DNA

2. The Golgi apparatus condenses to a single acrosomal granule situated at the anterior portion of the nucleus.

3. Centrioles develop into the axoneme and other fibrillar elements of the tail.

4. The volume of cytoplasm is reduced, and the mitochondria become situated round the anterior portion of the tail elements (mitochondrial sheath).

5. The spermatid cell membrane migrates caudally and forms the cytoplasmic droplet near the neck of the spermatozoon.

6. The fully formed spermatozoon is then released into the seminiferous tubule.

At a given point in the seminiferous tubule, several germ cell generations develop simultaneously; however, only certain germ cells are present at the same location. The succesion of a complete series of germ cells up to the reappearance of the next association at a certain point in the tubule is called the seminiferous epithelial cycle (Fig. #). Each cycle is divided into stages, depending on the various cell types involved. In the rat, 14 stages have been identified in each cycle, 12 in the mouse and monkey and 6 in man. The duration of the seminiferous epithelial cycle is precisely timed and varies from species to species:

In the rodent various stages of the spermatogenic cycle are arranged in consecutive order along the length of the seminiferous tubule so that adjacent areas are either more or less advanced by a single stage. Thus a co-ordinated series of stages occur along a part of the tubule, referred to as a spermatogenic wave. This feature appears to be absent in man.

When an abnormal sperm is observed in the ejaculate, it is important to know the target cells that were initially exposed. This can be done by consulting a table which, by assuming that the histological frequency of each stage in the testis is proportional to its duration, and the duration of transition through the epididymus, enables the back calculation of the stage the sperm was at when exposed to the toxicant. A computer program is now available to calculate the affected cell stage automatically, and has proved to correspond reasonably well with results obtained by histological analysis of the testis (Hess, R.A. &. Chen, P., 1992). Basic information on the type of lesion induced can be determined by the mouse specific locus test (Russell, L.B., 1990). A review of 14 chemicals which produce germ-line mutations has found that the effect is clearly weighted towards post-stem cell stages (Russell, L.B., 1990). He has found that post-stem cell mutagens fall into three basic groups, and that the nature of the lesion is chiefly determined by the stem cell stage affected, rather than the chemical nature of the mutagen.

Sperm count is one of the most sensitive tests for spermatogenesis since it gives the cumulative result of all stages in sperm production, and it is highly correlated with fertility (Meistrich, M.L., Finch, M., da Cunha, M.F., Hacker, U., Au, W.W., 1982). Since the very earliest (spermatogonial) phase is when nearly all cell multiplication occurs, chemicals which interfere with this phase will probably have a disproportionately greater effect on sperm output than s chemical acting during the spermatid phase (Sharpe, R.M., 1991). A large number of spermatozoa are produced daily, ranging from about 0.2 to 0.5 billion in man to 2 to 3 billion in the bull. Extrapolation of data from animals to man is often therefore difficult. In rats epidymal sperm counts can be reduced as much as 90% without loss of fertlity (Mori, K., Kaido, M., Fujishiro, K., Inoue, N., Koide, O., Hori, H., Tanaka, I., 1991), whereas man is far more sensitive with the number of sperm per ejaculate often being close to that required for fertility (Mably et al., 1992). In general, a equipotent dose level extrapolation from mice to men is of a factor of around 2.6 to 7. The reduction in sperm counts in mice at these equipotent doses, however, is between 11 to 44 times that in man (Meistrich, M.L., Samuels, R.C., 1985). Additionally, a much higher proportion (20%-40%) of abnormal sperm forms are found in the human than in other species (5%-10%)

At body temperature sperm production cannot occur, although testosterone production occurs normally.

Exposure of adult male mice to high levels of cannabinoids (the equivalent of 3 cigarettes 3 times a week for 5 weeks) is associated with a reduction in fertility and an increased incidence of chromosomal aberrations (Daltero, S., Badr, F., Bartke, A., Mayfield, M., 1982). These effects are evident not only in the treated mice, but also in their untreated male offspring. The decreased fertility was measuredby the ability of treated males to impregnate fertile females and so, although there was a decrease in spermatogenesis, the infertility detected could be due to a depressed libido (see below) - especially as male mice have a large spare sperm capacity. There was also an increase in pre- and post-implantation loss.

Ethylene oxide is a highly reactive alkylating agent widely used in chemical synthesis and in sterilisationand fumigation. It produces dominant lethal mutations and testicular atrophy when inhaled by rats at 200-1000ppm. There is also an increase in the number of immature sperm produced at 250ppm (Mori, K., Kaido, M., Fujishiro, K., Inoue, N., Koide, O., Hori, H., Tanaka, I., 1991). It exerts an effect with doses as low as 50ppm, causing an increased dose-dependent production of teratic sperm in numbers which may have an effect on fertility. Ethylene oxide is a small molecule (RMM 44), and is thought to pass easily through the blood-testis barrier. It is known to be an alkylating agent of DNA and protein of germ cells, which may be responsible for the increase in teratic sperm. The sloughing of immature cells may be a result of Sertoli cell damage. Testicular atrophy is probably a result of the decrease in sperm and Sertoli cell numbers.

Acrylamide has been shown to produce dominant lethal mutations and translocations in early spermatozoa and late spermatids of the mouse, but not in earlier stages (Sega, G.A., Generosa, E.E., Brimer, P.A., 1990). It has also been shown by Sega et al. to produce unscheduled DNA synthesis (UDS) in mouse early spermatids, with the maximum effect occuring six hours after a testicular injection. The response is linearly proportional to the dose of acrylamide over a range of up to at least 125 mg/kg. This was an unexpected result, since the acrylamide contains a double bond which should be able to alkylate nucleophilic sites in DNA (see figure #) without the need for metabolic conversion. They concluded that the most likely explanation is that it is metabolised to the epoxide glycidamide which then goes on to form adducts with the germ cell DNA and induce UDS (similar to ethylene oxide, above). Late spermatids and early spermatozoa also suffer from DNA lesions when exposed to this chemical, but they are incapable of repairing them and so go on to produce dominant lethal mutations.

The fungicide benomyl (methyl 1-(butylcarbamoyl)-2-benzimidazolecarbamate) also induces sloughing and testicular atrophy. It produces hypospermatogenesis and multinucleated germ cells in rats and dogs (Hess, R.A., Moore, B.J., Forrer, J., Linder, R.E., Abuel-Atta, A.A., 1991). Chronic and subchronic exposures causes testicular atrophy and decreased fertility, but a single acute dose induces testicular swelling. The swelling is caused by occlusion of the efferent ductules of the testis within two days of treatment. The most sensitive short-term effect is a premature release of germ cells (sloughing), which was detected at the lowest dosage (25 mg/kg) and apparently occurs before the ductal occlusions. At low doses, the primary cell stage affected was Stage VII, but at higher doses all stages were affected except Stages VII-XI. This was followed at 70 days post-treatment by testicular atrophy due to degeneration of the seminiferous tubules. Benomyl's mechanism of action in fungus is an inhibition of microtubule formation. Since the Sertoli cell contains an abundance of microtubules, the authors considered that a similar mechanism may be responsible for the germ cell sloughing observed. Circumstantial evidence also links benomyl with the condition anopthalmia. In an article in the Observer newspaper it was claimed that clusters of this disease exist (Paduano, M., McGhie, J., Boulton, A., 1993) and in a follow up article (McGhie, J., Paduano, M., 1993) it was stated that one of the cases was born to a farmer who had used benomyl in his work - a possible male effect.

While EDS is cytotoxic to Leydig cells (see above), another effect of EDS is seminiferous tubule atrophy, which could be due to either lack of testosterone or a direct effect of EDS on the seminiferous epithelium. To investigate this, Sprando et al used polydimethylsiloxane (PDS) capsules containing testosterone and oestradiol to supress LH production in rats given EDS, thereby also preventing leydig cell repopulation (Sprando et al, 1990). Surprisingly, the testicular atrophy caused by this combination was even greater. They suggested that the factors behind this enhanced effect could be either a synergism of EDS with oestradiol or testosterone, or that other factors produced by the Leydig cell are required for spermatogenesis. By reperforming the experiment with and without oestradiol they were able to demonstrate that it was largely responsible for the the enhanced toxicity, rather than testosterone, although they were unable to rule out an unknown Leydig cell effect. They suggested that it may be due to;

1. Stimulation of proliferation by oestradiol, thus allowing the seminiferous epithelia to become more sensitive to the effects of EDS.

2. A direct effect on Sertoli cell function. This is attractive because there was an observable decrease in spermatids within two weeks, too early for a direct depletion of spermatogonia (the primary germ cell -see below) to have an effect.

Fluoride is a common environmental contaminant. Acute fluorosis is known to cause increased FSH and reduced testosterone in man, and the application of an acute dose of sodium fluoride (NaF) to the fruit fly drosphila has been shown to increase the susceptibility of the male to x-ray induced sex-linked recessive lethal mutations (Susheela, A.K., Kumar, A., 1991). In the environment, fluorine exposure takes the form of low levels of fluoride contained in the drinking water. The water in some areas of India naturally contains high levels of fluoride, and infertility has been reported in men in these areas. Chronic dosing of mice with high levels of chlorine in the water results in lack of maturation and differentiation of spermatocytes and loss of spermatogenesis. Similarly, in rabbits chronic dosing with 10 mg/kg/day of NaF will stop spermatogenesis after 29 months. This can be compared with levels of 1 mg/l of Fluoride in the drinking water which can cause the first symptoms of fluorosis (mottled teeth) in man. The blood-testis barrier is relatively effective in preventing damage to the germ cells, with levels in the testes only 10%-20% of that of the plasma.

Cadmium produces sterility when administered to rats at a dosage of 10 mg/kg (Dwivedi, C., Singh, D.N., Crump, E.P., Harbison, R.B., 1977). These levels also impair spermatogenesis by about 50%. No significant effects were noticed at lower doses, but a chronic dosage of 1 mg/kg for one month produced similar effects. The damage is due to capillary stasis followed by massive thrombosis, and is highly specific to the testes (Parizek, J., Zahor, Z., 1956). The primary insult seems to be an increase in capillary permeability and a breakdown of the blood testes barrier (Setchell, B.P., Waites, G.M.H., 1970). Exposure to cadmium causes an accumulation of haem within the Sertoli cells (Maines, M.D., 1984). This is probably a result of its ability to interfere with the haem and haemoprotein degradative pathway. The selective destruction of testicular micro-vasculature in response to CD2+ at 20-40 mmoles/kg would facilitate this accumulation (Maines, M.D., 1984). The net effect of decreased spermatogenesis is likely to be a combination of diminished blood flow and the ability of cadmium to compete with zinc for binding to enzymes essential for cell replication (Zinc defficiency also causes azoospermia - Elcoate, P.V., Fischer, M.I., Mawson, C.A., Millar, M.J., 1955).

Cadmium also causes testicular tumours (mostly Leydig cell adenomas), the incidence of which is highly dependent on the degree of testicular degeneration (Waalkes, M.P., Rehm, S., Riggs, C.W., Bare, R.M., Devor, D.E., Poirier, L.A., Wenk, M.L., Henneman, J.R., Balaschak, M.S., 1988). Low dose pretreatment (5.0 mmol/kg) reduced or prevented the testicular degeneration and tumour formation that would otherwise result from a subsequent higher dose of cadmium (20 mmol/kg). Waalkes et al. noted that this may be related to the previously demonstrated protection provided by pretreatment with zinc.

Together with the therapeutic use of x-rays, there is considerable public concern over environmental exposure to ionising radiation, particularly in regard to nuclear power stations. Gardner et al. compared the children of workers at the Sellafield nuclear plant in West cumbria, and found a significantly higher levels of leukemia and non-Hodgkin's lymphoma when compared with a reference population (Gardner, M.J., Snee, M.P., Hall, A.J., Powell, C.A., Downes, S., Terrell, J.D., 1990). They found that the important risk factor was paternal employment at the plant and recorded external dose of whole body radiation. They found that other factors, including exposure to x-rays, maternal age, employment elsewhere, eating seafood and playing on the beach did not explain the high incidence, although they admitted that the quality of information on these potential confounding factors was relatively low. However, doubt has been cast on this scenario for two reasons (Miller, S.K. (2), 1992). Firstly, tens of thousands of children of the survivors of the atomic bomb blasts in Japan in 1945 have been followed, and they do not show any predisposition to cancer. Secondly, clusters of leukemia are extremely frequent and seem to occur spontaneously, and the US Center for Disease Control has studied more than 100 similar leukemia clusters and found no apparent cause for them.

Methyl chloride is a colourless gas which is primarily used as a chemical intermediate in the production of organosilicate compounds and petrol anti-knock agents. Acute exposure to MeCl causes toxicity in both the testis and the epididymus (Chellman, G.J., Hurtt, M.E., Bus, J.S., Working, P.K., 1987). The initial testicular lesion appears to be a delayed release of spermatids from the semiferous epithelium, followed by generalised disruption and disorganisation. It is cytotoxic to stem cells, primary spermatocytes and spermatids. Dominant lethal lethal mutations are produced, with post-implantation losses being caused 1-2 weeks after exposure, and pre-implantation losses caused 2-8 weeks after exposure. The post-implantation losses are caused by a genotoxic effect on germ cell DNA while the sperm are in the epididymus (see below), whereas the pre-implantation losses are cause by cytotoxicity to the developing sperm.

Certain bromine compounds are known to have an adverse effect on male reproductive capability (such as DBCP - see below - and ethylene dibromide - see Sperm Maturation and Epididymal Function below). Accidental exposure of six men to bromine vapour occured when a semi-trailer loaded with bromine overturned in Israel (Potashnik, G., Carel, R., Belmaker, I. & Levine, M., 1992). During follow up studies, there was some suggestion of a mild degree of spermatogenic suppression and impaired reproductive performance, including an increased rate of spontaneous abortion. However, the study cohort was too small, and the conditions of exposure too poorly defined, for a definate cause and response relationship to be defined.

Dibromochloropropane (DBCP), an industrial nematocide, produces acute toxicity in male mice. The fact that the toxicity is age related (LD50 for prepubertal mice is 180.7 mg/kg, that for adult mice is 123 mg/kg) suggests an interaction with the reproductive system (Lee, I.P. & Suzuki, K., 1979). An injection of a single maximally tolerated dose (100 mg/kg) intraperitoneally induces significant unscheduled DNA synthesis in premiotic cells but not in spermatozoa (Lee et al. 1979). DBCP given to male rats produces dominant lethal mutations in the post-miotic stage of spermatogenesis, an effect not shared by the chemically related dibromoethane (Teramoto, S., Saito, R., Aoyama, H. & Shirasa, Y., 1980). This occurs especially in the early spermatid phase, but the effect was not seen in mice.

DBCP is one of the few compounds in which the toxic spectrum in man is well characterised. Its suppresive effect on spermatogenesis was first detected in 1977 (Whorton, D., Kraus, R.M., Marshall, S., Milby, T.H., 1977). Epidemiological studies have indicated that factory employees exposed to this compound during its production developed oligozoospermia or azoospermia (Potashnik, G., Yanai-Inbar, I., 1987). Follow up studies have revealed that the effect is reversible to an extent, and the reversibility is directly related to normal FSH levels (Potashnik, G., Yanai-Inbar, I., 1987). Biopsies conducted by Potashnik & Yanai-Inbar in one severely affected worker in 1977 revealed selective atrophy of the germinal epithelia, with the great majority of tubules lined only with Sertoli cells. Occasional spermatogonia were identified in the tubules, and large groups of Leydig cells were observed. A follow up biopsy eight years after exposure showed a marked improvement. More than one third of the seminiferous tubules of both testes showed signs of active spermatogenesis, but large groups of Leydig cells were still present, located largely around the empty tubules. They suggested that the Leydig cell hypertrophy occurs because their functioning is modulated by the seminiferous tubules, which either exert a restraining influence on Leydig cells or, when damaged, produce an agent which stimulates them.

Tetrachlorodibenzodioxin (TCDD) is a contaminant of the defoliant 2,4,5-T, commonly known as Agent Orange, which was used extensively during the Vietnam conflict (Hodgson, E., Levi, P.E., 1987). In overtly toxic doses it can alter testicular morphology, decrease spermatogenesis and generally inhibit male reproductive capability in laboratory animals. Since compounds which cause testicular peturbations could be expected to have their greatest effect if exposure occurs early in development, Mably et al. performed an experiment to discover how the male reproductive system would be affected by in utero and lactational exposure (Mably, T.A., Bjerke, D.L., Moore, R.W., Gendron-Fitzpatrick, A. & Peterson, R.E., 1992). They found that the lowest dose to cause a disruption of spermatogenesis (as measured by cauda epididymus sperm number and daily sperm production) in offspring when given to dams was only 0.064 mg/kg in a single dose. This compared with a previously reported dose of 1.0 mg/kg at 5 days a week for 13 weeks (total 65 mg/kg) in adult males. The number of leptotene spermatocytes to Sertoli cells was normal, suggesting that the effect was due either to;

1. The degeneration of intermediate in development between leptotene spermatocytes and final stage spermatocytes,

2. Decreased postleptotene spermatocyte celldivision, and/or

3. Decreased number of Sertoli cells per testis.

This was partially recovered as the rats aged, but nevertheless they were able to conclude that either there was a permanent reduction in production, or the age at which maximum production was acheived was raised.

Boric acid (BORA) is widely used for both medicinal and nonmedicinal purposes and is best known for its use as an insecticide powder. It has been thoroughly investigated for its potential as a reproductive toxicant using the Reproductive Assessment by Continual Breeding protocol (RACB - see Methods of Investigating Reproductive Toxicity) (Fail, P.A., George, J.D., Seely, J.C., Grizzle, T.B., Heindel, J.J., 1991). They found that fertility was reduced at 4500 ppm and was totally eliminated at 9000 ppm, and confirmed that the male was the effected sex. A No Observed Adverse Effect Level (NOAEL) for the F0 generation of 1000 ppm was found, in which the motility of epididymal sperm was affected, but not fertility. This was well below the level at which the first the first clinical signs appeared (a reduction in body weight at 9000 ppm). F1 males appeared to be even more sensitive, since they produced 25% lower sperm concentrations and, when mated, produced pups with slightly lower birth weights. They hypothesised that the diminished sperm production could be due to a testicular effect of BORA to alter germ cell, Sertoli cell, or Leydig cell function or an induced effect on the pituitary-hypothalamic axis. Although they were unable to distinguish between these possibilities, they noted that BORA is known to inhibit cell division and possibly ATP synthesis and that men exposed to the boron containing tranquilizer, Methyl-5-n-propyl-5-R-tolyl-2-dioraborane, had germinal aplasia. This suggests a possible effect on the germinal epithelium.

The germ cells and spermatozoa are shielded from the immune system by the blood-testis barrier from an early age (Guyton, A.C., 1988). This is required because the recombination of DNA within spermatozoa produces new proteins which are regarded as foreign by the immune system. A consequence of this is that any compound causing a hypersensitivity response in the testis can result in the selective destruction of spermatozoa due to an autoimmune reaction. For example, intratesticular injection of Bacillus Calmette-Guérin (BCG, used in tuberculosis vaccination) causes azoospermia with apparently no effect on Sertoli cells or other testicular components, with no loss of libido (Talwar, G.D., Naz, R.K., Das, R.P., 1979). The same effect can be achieved by a BCG injection in the cauda epididymus or vas deferens (Suri, A., Shaha, C., Talwar, G.P., 1988). Gossypol is a yellowish phenolic compound which occurs naturally in certain species of cotton. It has been thoroughly investigated as a male reproductive toxicant because of its potential as a male contraceptive. It is antispermatogenic, and there is a pronounced species variation - among the laboratory animals tested, hamsters seem to be the most sensitive, followed by rats, monkeys and dogs in decreasing order (Qian, S.Z., Wang, Z.G., 1984). In clinical trials, gossypol given orally at a dose of 60-70 mg per day for 35-42 days caused a gradual increase in the percentage of nonmotile spermatozoa in the ejaculate, followed by oligospermia, necrospermia, and azoospermia. In a large scale trial in 1980 with 8806 volunteers, the antifertility efficacy was found to be 99.07%, with a total of 266 conceptions (Qian, S.Z., Wang, Z.G., 1984). Of the subjects, around 10% remained azoospermic six months to 4.5 years post-regimen, indicating a possibly irreversible effect. Additionally, the severe side-effect of hypokalemic paralysis occured in 0.75% of the study population. In an effort to identify a chemical which circumvents these problems, Hoffer et al. studied the effects of 16 optical isomers and chemical analogues of gossypol on sperm and testes characeristics both in vitro and in vivo (Hoffer, A.P., Agarwal, A., Meltzer, P., Herlihy, P., Naqvi, R.H., Lindberg, M.C., Matlin, S.A., 1987). They found that the analogues were antispermicidal in vitro, but had no effect on fertility in vivo. Gossypol itself also had differing effects in vitro and in vivo, and they suggested that protein binding was important in determining activity. The mechanism of action of gossypol is uncertain, but it is capable of uncoupling oxidative phosphorylation (Qian S.Z. et al., 1984), and it induces a pathygnomic defect in the mitochondrial sheath (Hoffer, A.P. et al., 1987), which may be related to its antifertility effects.

Radiation has particularly severe adverse effects on spermatogenesis, exposure of rat testis to 3 grays of x-rays kills most cycling spermatogonia (Schally, A.V., Paz-Bouza, J.I., Schlosser, J.V., Karashima, T., Debeljuk, L., Gandle, B., Sampson, M., 1987). Damage can be prevented by inhibiting spermatogenesis, and Schally et al. found that antagonists of LH-RH were able to protect against testicular damage induced by up to 415 rads of x-radiation. Jegou et al. have shown a similar protective ability is provided by medroxyprigesterone acetate plus testosterone (MT) (Jegou, B., De La Calle, J.F.V., Bauche, F., 1991). They were also able to show that that MT also maintained fertility and protected against germ line mutations.

Radiation is of particular concern as a reproductive toxicant due to the high possibility of the induction of inheritable abnormalities. In mice, parental exposure to X-rays induces a higher rate of tumours and other anomalies in the offspring (Nomura, T., 1982). These abnormalities are inherited in a dominant fashion, and the susceptibility of males and females is identical. The fact that a germ line mutation has occured is indicated by the fact that the effects are passed to the F2 generation. The effect is organ specific, with 90% of tumours occuring in the lung, but other kinds of tumours (such as ovarian and lymphocytic leukemias) are also increased.

Radiomimetic drugs are designed to kill rapidly dividing cells - the target being neoplastic growth - and have a similar spectrum of biological effects to x-radiation. A consequence of their mode of action is damage to other rapidly dividing body cells, and permanent azoospermia and sterility is now recognised as a common side effect of cancer chemotherapy in humans. Different chemotherapuetic drugs differ in their toxicities but it is difficult to discover in man which drugs are the most dangerous, since they are often used in combination. Meistrich et al. used a sperm cell count assay in mice to assess the effects of 14 chemotherapeutic drugs (Meistrich, M.L., Finch, M., da Cunha, M.F., Hacker, U., Au, W.W., 1982). They looked for potential mutagenicity using the spermatocyte translocation assay (which involves visually scoring post-meiotic spermatocytes for chromosome aberrations, and noted that previous studies had shown that chemotherapeutic agents did not produce clinically significant damage in the nongerminal cells of the testis. They found that only two (prednisone and 6-mercaptopurine) produced little or no cytotoxicity, and of the rest 8 (especially triethylenethiophosphoramide, thio-TEPA) showed significant stem cell killing. Thio-TEPA was also highly effective in producing chromosome aberrations. Perhaps the most surprising (and controversial) result was that the cytotoxic effects of these drugs in mice did not seem to correlate with the ability to cause long-term azoospermia in humans - although this study was weakened by the fact that it was performed with a single injection, in contrast to the chronic exposure of human cancer patients. The potential for germ-line mutations caused by radiomimetics has been underlined by the demonstration that cyclophosphamide causes an increased number of tumours in the offspring of male rats dosed with it (Miller, S.K. (1), 1992).

Exogenous oestrogens applied in the first trimester of pregnancy are known to cause cryptorchidism. It was reported in 1984 that the annual number of hospital diagnoses of cryptorchidism has risen by a factor of 2.3 between 1962 and 1981 (Chilvers, C., Pike, M.C., Forman, D., Fogelman, K., Wadsworth, M.E.J., 1984). Although the authors noted that the rise in diagnosis may be a artefact due to the modern tendency to operate at an early age rather than wait for possible spontaneous recovery at puberty, they argued that the increase could not be ascribed to any known risk factors. Field et al. suggested that the unascribed increase may be due to xenobiotic oestrogens such as pharmaceuticals and livestock growth promoters (Field, B., Selub, M.S., Hughes, C.L., 1992).

Diethylstilbestrol (DES) is an oestrogen analogue which has been used as a food additive for cattle and to treat threatened abortion. The number of abnormal sperm on male offspring in women treated with DES in pregnancy is significantly higher than normal, with the average sperm density and total motile sperm less than 50% of that which is usually found (Field et al., 1992). It has been shown to cause developmental lesions in male mice - 60 % of the male offspring from pregnant mice treated with DES during gestation were sterile, as indicated by their fertility index (McLaclan, J.A., Newbold, R.R., 1975). The affected animals had gonadal changes which included intra-abdominal or fibrotic testes, or both. Although no malignancies have been observed in man, 25% of males exposed to DES in utero exhibit genital lesions and abnormal spermatozoa (Beckman, D.A., Brent, R.L., 1984).

After spermiation, the spermatozoa move through the seminiferous tubules and collect in the rete testis from where they migrate via the vasa (ductuli) efferentia to the epididymis. Their movement through these ducts appears to be primarily accomplished by the flow of fluid secreted by the Sertoli cells. Contraction of the tubules and the testicular capsule may also aid in sperm transport. The gross morphology of the epididymus is shown in figure #.

The first portion is called the caput (head), the middle part, the corpus (body), and the last part the cauda (tail), which fuses with the vas deferens. Almost all of the testicular fluid is absorbed in the caput resulting in a large increase in the concentration of spermatozoa as they pass through.

Testicular spermatozoa are immotile and incapable of progressive movement. They gain these properties during their transport through the caput and corpus. A primary indication of epididymal function and sperm maturation is the loss of the cytoplasmic droplets as the sperm pass through. Epididymal transport requires several days to weeks, depending on the species. Over frequent ejaculation can result in the presence of immature sperm in the ejaculate. Long periods of abstinence may also decrease viability. Passage through the epididymus takes about 8 days in the rat, although transit time can be up to 25% faster in actively mating animals (Chellman, G.J., Hurtt, M.E., Bus, J.S., Working, P.K., 1987).

Storage of the spermatozoa occurs primarily in the cauda of the epididymis. The fate of unejaculated spermatozoa is unknown. They may disintegrate and be reabsorbed, or be phagocytized.

Indirect evidence of a deleterious effect of ethanol on the epididymus is provided by the observation of increased frequency of caudal epididymal spermatozoa which have retained their cytoplasmic droplets (Anderson, R.A. Jr., Willis, B.R., Oswald, C., Zaneveld, L.J.D., 1983). Inhibitors of epididymal 5a-reductase activity has been shown to reduce the fertilizing capacity of spermatozoa, and ethanol is known to reduce the hepatic activity of this enzyme - but its effect on the epididymal enzyme is unknown (Anderson et al., 1983).

Ethylene dibromide (EDB), extensively used as an antiknock agent and a fumigant, is a highly reactive alkylating agent and rodent carcinogen. It is toxic to male reproduction in a number of species, resulting in decreased sperm counts, increased abnormal sperm, and testicular and epididymal atrophy (Williams, J., Gladden, B.C., Turner, T.W., Schrader, S.M., and Chapin, R.E., 1991). The effects of EDB on the human male has also been evaluated, with results varying from no adverse effects, through decreased sperm quality, to reduced fertility in male workers in one of four manufacturing plants. In a comprehensive study, Williams et al. analysed the applicability of a rabbit model for the effects of EDB in man. They found that although there was an effect on fertility, this was only significant at doses which also caused systemic toxicity. Of the parameters measured which were affected in man, only some were altered in their rabbit model:

The results suggest a direct effect of EDB on rabbit sperm. The kinetics of the effects (occuring primarily 2-3 weeks post-dosing), compared to the duration of spermatogenesis and epidiymal transit time (around 10 weeks), suggest that the epididymal spermatozoa are most sensitive. Further support for an epididymal effect is provided for by the observed increase in alkaline phosphatase (AP) in seminal plasma 2 weeks post dosing. In the rabbit, this enzyme is actively secreted by the epididymis. The effects could also be due an effect on accesory sex glands (see below).

Methyl chloride produces a severe inflammatory response against the epididymal tubular epithelium, followed a week later by a persistant chronic inflammatory focus (sperm granuloma) in the interstitial tissue (Chellman et al, 1987). The post-implantation losses caused by this compound (see above) are thought to arise from a genotoxic effect while the sperm are in transit through the epididymus. This is possibly a result of superoxide and peroxide production during the inflammatory response, and the formation of a granuloma would explain why this effect is not seen after week 2.

These organs show a wide variation between species. Their role is to produce a range of secretions which complement the function of the main sex organs, but they do not appear to be essential for fertility. No reproductive toxicant is known to act exclusively on the accessory sex glands, but decreased steroidogenesis often associated with atrophy of these glands. They are illustrated diagramatically in figure @.

Glands located at the terminal portion of the vas deferens are called the ampulla. Information concerning the the compositions of secretions is scarce, but they are typically ergothioneine, citric acid, lactic acid, glycerolphosphoryl-choline, calcium, sodium, potassium and chlorine. The seminal vesicles are large, paired glands which generally empty into an ejaculatory duct which empties into the urethra. Vesicular fluid contains several compounds related to sperm maintainence, notably fructose (a metabolic fuel), and an alkaline pH which neutralises the acid content of the vagina. The prostate gland is located around the urethra. It may aid in maintaining sperm viability and other reproductive tract functions. In humans, prostatic enzymes have the effect of lysis of the coagulum formed by the seminal vesicles. In rodents, they have the opposite effect and catalyse the formation of a vaginal plug. The bulbourethral (Cowper's) glands are mucous secreting glands which empty into the urethra. Their role is probably to lubricate the urethra and tip of the penis before emission and ejaculation takes place.

Chronic ethanol administration will also reduce prostate and seminal vesicle size in both man and animals (Farghali et al., 1991). Analysis of semen from mice chronically treated with ethanol suggested impaired seminal vesicle and prostatic function, as measured by decreased fructose and alkaline phosphatase content (Anderson, R.A. Jr., Willis, B.R., Oswald, C., Zaneveld, L.J.D., 1983).

Ethylene dibromide (EDB) has a deleterious effect on fertility and sperm motility sperm motility in many species (see above) (Williams, J., Gladden, B.C., Turner, T.W., Schrader, S.M., and Chapin, R.E., 1991). This could be due to a change in the composition of seminal plasma, and there is evidence of perturbed accesory sex gland function in the rabbit, in that semen pH, ejaculate volume, and alkaline phosphatase activity were changed. Accesory sex gland function is also perturbed in the rat, as evidenced by atrophy of the seminal vesicles and prostate. The effects on semen cannot be ascribed exclusively to the accesory sex glands, however, since secretions from all parts of the reproductive tract contribute to the pH and volume of semen.

Cadmium treatment was found to be associated with prostatic tumours (mostly adenomas of the ventral lobe), but the effect is not dose related and a saturation of the tumourigenesis occured at a dose of only 2.5 mmol/kg (Waalkes, M.P., et al., 1988). Preneoplastic lesions occurred in a dose related fashion up to and including 20 mmol/kg, suggesting that prostatic hyperplasia induced by cadmium only develops into tumours at doses well below those causing marked degeneration of the testes and atrophy of the prostate. This is probably due to the requirement of testosterone for the maintenace of the prostatic epithelium.

Boric acid was found to produce decreased prostate weight when fed at 4500 ppm for 27 weeks (Fail et al, 1991). This is probably an effect secondary to decreased testosterone production. Cadmium was found to cause degeneration of sex vesicles at concentrations of 10 mmol/kg (Dwivedi, C., et al., 1977)

Treatment of pregnant mice with diethylstilbestrol (DES) resulted, amongst other changes (see above), in nodular masses in the ampullary region of the reproductive tract in 6 out of 24 animals; one of these appeared to be preneoplastic (MacLachlan, J.A., et al., 1975). Carcinoma of male rodent accesory sex glands is normally very rare.

Libido is under pyschosomatic, neurogenic, vascular, and hormonal (primarily testosterone) control. Laferla et al. found that sexual arousal induced in healthy adult males during viewing of erotic film material was highly correlated with increments in plasma LH (Laferla, J.J., Anderson, D.L., Schalach, D.S., 1976). If the rise of LH is not merely correlated with arousal, but causal, then any agent that affects LH levels will affect libido. Erection is produced when blood fills the erectile tissue of the penis, due to vasoconstriction under mostly parasympathetic (but also sympathetic) control. When erections are not obtained, it is often difficult to differentiate whether the cause is physiological or psychological.

The process of ejaculation can be divided into two phases:

1. Emission, during which the spermatozoa are transported from the epididymus into the vas.

2. Ejaculation, during which the spermatozoa and the fluids from the genital tract are voided.

Very little is known about agents which specifically alter the fertilization process (Waller, D.P., Killinger, J.M., Zaneveld, L.J.D., 1985). A spermatozoon may appear morphologically normal but lack the biochemical properties to undergo fertilisation, but the effect of chemical agents on these biochemical properties is only rarely studied.

Cannabis is reported to produce either sexual arousal or depressed libido at high and low doses respectively (Daltero et al., 1981). Daltero et al. suggested that this may be due to the reported effects on testosterone and LH release, although cannabis is known to exert a direct effect on the central nervous system (Bowman, W.C., Rand, M.J., 1980). This may be the cause of the depressed fertility detected in male mice exposed to cannabinoids (Daltero et al., 1982). Extracts of cannabis (but not D9-tetrahydrocannabinol) are reported to compete with oestrogen for binding to the oestrogen receptor in vitro (Sauer, M.A., Rifka, S.M., Hawks, R.L., Cutler, G.B. Jr., Loriaux, D.L., 1983). This may have an influence on the reported effects on libido, but Sauer et al. noted that cannabis is not oestrogenic in vivo.

Loss of libido and impotence occur in as many as 70-80% of chronic alcoholics (Anderson et al., 1983). The factors underlying this are largely undetermined, but it could be a secondary manifestation of alcoholic liver disease resulting in altered steroid metabolism. Ethanol has a direct effect on the hypothalamus and other brain areas (Bowman, W.C., Rand, M.J., 1980), and also interferes with the pituitary-gonadal axis. It is likely that its effect on reproductive behaviour are due to a combination of all three of these possible causes.

Ethanol also exerts an effect on the vas deferens in vitro, by increasing the spontaneous release of noradrenaline. Depletion of noradrenergic stores within the vas deferens may result in decreased ejaculation volume, and may be responsible for the accumulation of epididymal sperm subsequent to chronic ingestion of relatively low levels of ethanol (Anderson et al., 1983). Anderson et al. also note that there is an increased frequency of damaged spermatozoa found in ejaculates from non-alcoholic volunteers subsequent to acute ethanol intoxication. They speculated that this may be due to increased shear forces in the epididymus due to enhanced release of noradrenalin.

Organophophorus pesticides, which act by inhibiting acetylcholinesterase and thereby facilitating cholinergic transmission, are suspected of causing impotence. Espir et al. reported a study on five farm workers who had been using a variety of pesticides in intensive agriculture, of which four had become impotent (Espir, M.L.E., Hall, J.W., Shireffs, J.G., Stevens, D.L., 1970). All four recovered after further contact with chemicals was stopped and hormone therapy given. No specific chemical was implicated, though dieldrin, paraquat, and malathion were used. Impotence was the only clinical sign and there was no loss of libido. Espir et al. suggested that the effect could due to the the organophosphate malathion, although dieldrin has an oestrogenic activity which could have been partly responsible.

In a preliminary study into infertility caused by DBCP, Whorton et al. noted that none of the men affected had a loss of libido, difficulty with erection or ejaculation, or altered distribution of facial or body hair despite having up to 8 years exposure (Whorton et al., 1977). Similarly, in their eight year follow up study, Potashnik & Yanai-Inbar did not mention any effect on libido (Potashnik, G., Yanai-Inbar, I., 1987. But ...

Many antihypertensives are anticholinergic, and act by antagonising the parasympathetic nervous system (Bowman, W.C., Rand, M.J., 1980). As a result they often cause impotence. A study of 27 males treated with methyldopa for essential hypertension revealed that seven experienced some disorder of sexual function which started within a few days of commencing therapy (Newman, R.J., 1974). The disorders observed included decreased libido, an inability to maintain an erection, and difficulty in ejaculation. These effects disappeared within two weeks of discontinuing treatment. The effects were not related to age or daily dose of the drug.

The most striking feature made apparent by this review is the diversity of compounds which are capable of reducing male reproductive capacity. They vary widely in their specificity. With some there is a very narrow range of effects - for example gossypol which is directly cytotoxic to spermatozoa with apparently no other reproductive side effects. Others, notably ethanol, appear capable of striking at every stage.

As for the the somewhat artificial categories imposed on the reproductive system by this review, there are numerous incidences of disimilar toxicants causing similar effects. Of the agents for which there is good evidence of an effect male endocrinology, only two (ethanol and cadmium) have a demostrable mechanistic basis. This is in part due to the the complicating factor of neurological control over hormone production and the difficulty in finding a root cause for the hormone changes which can often be observed. Many studies are also directed towards an analysis of spermatogenesis and other testicular effects, and so often an effect on hormones is either passed over or simply not detected. This may be because spermatogenesis is more sensitive to a toxic insult - for example, chemicals which disrupt spermatogenesis often cause raised FSH, but no toxins described here reduce spermatogenesis by lowering FSH production (a theoretical possibility).

Of the reproductive toxins which have been evaluated, the majority have an effect on spermatogenesis. This is partly an artificial effect due to the ease with with sperm count and morphology can be assessed as an endpoint. But it is also a genuine reflection of the sensitivity of the rapidly dividing seminiferous epithelium. Effects on spermatogenesis can generally be divided into 5 areas;

Many of the compounds have effects on more than one part of the reproductive system. These are summarised in the table below: