European Urology

1. Introduction

Seminal vesicles were first described in 1561 by the Italian anatomist Gabriel Fallopius. In the late 19th^. century, they became the focus of considerable interest because of their involvement in inflammatory diseases such as tuberculosis and gonorrhea [1]. At that time, however, their physiologic role was still not entirely understood.

In one of the earliest observations, the physiologist Ivan R. Tarchanoff noticed in 1887 that the seminal vesicles of frogs were enlarged during mating season. He observed that their removal halted the coitus of mating frogs. Tarchanoff concluded that in frogs, and possibly, therefore, in mammals, the seminal vesicles were the starting point of sexual activity [2]. In 1924, Floyd H. Allport, the father of experimental social psychology, made a similar postulation for humans: “The original stimulus for sex responses … in the male is the gradual distention of the seminal vesicles, a condition requiring a fairly periodic discharge of their contents”[3]. Later in the 20th century, an increasing number of experiments were performed, which, despite occasional trends, were not able to prove an influence of internal genital organs on sexual activity and sex drive (Table 1) [4], [5], [6], [7], [8], [9], and [10].

Little also is known about the physiologic influence of seminal vesicles in humans beyond their fertility function. Driven by a persistent lack of evidence that distended seminal vesicles may increase sex drive, we sought to take a fresh approach to the issue by combining the experience of various previous experiments in the design of one novel animal model [11].

2. Materials and methods

Naive Crl:CD1(ICR) 10-wk-old male and female mice were purchased from Charles River Laboratories (L’Arbresle, France). The animals were housed in individually ventilated cages: males in groups of four in Makrolon II cages (23 × 17 × 14 cm) (Bayer MaterialScience LLC, Sheffield, MA, USA) and females in groups of eight in Makrolon III cages (39 × 23 × 15 cm). The day/night cycle was maintained with lights on at 06:00 and off at 18:00. The cages were equipped with sawdust, transparent red plastic nest boxes, and pieces of wood and paper. Food and water were available ad libitum. After arrival, the animals were allowed to acclimatize for at least 5 d. Male mice were randomly assigned to one of three experimental groups: (1) bilateral seminal vesicle occlusion (SVO), (2) bilateral seminal vesicle resection (SVR), and (3) sham operation (SO). The study was approved by the animal ethics committee of the canton of Bern, Switzerland.

2.2. Surgery

Surgery was performed under sterile conditions using a binocular microscope (Carl Zeiss AG, Feldbach, Switzerland). In the SVO group, the seminal vesicle outlets were occluded by titanium clips (Fig. 1A–1C); in the SVR group, both seminal vesicles were resected (Fig. 1D–1F); and in the SO group, both seminal vesicles were simply exposed (Fig. 1G–1I).

Fig. 1 Surgical procedures of the three experimental groups. A median laparotomy of 15–20 mm was done in all groups. (A–C) In the seminal vesicle (SV) occlusion group, all tissue around the exposed SV was cleaved and spared on both sides, enabling gentle placement of a small (3.1 mm) titanium ligating clip (Aesculap AG, Tuttlingen, Germany) at the base of each SV. (D–F) In the SV resection group, the anterior lobe of the prostate was cleaved off the exposed SV on both sides. After placing a small titanium clip at the base of each SV, both SVs were resected, leaving a stump of 2–3 mm. (G–I) In the sham operation group, both SVs were only exposed. In all three groups, the laparotomy was closed using a running suture (Vicryl 5-0) for the rectus muscles and single stitches (Vicryl 4-0) for the skin. AP = anterior lobe of the prostate; B = bowel; UB = urinary bladder.

2.4. Evaluation of sexual activity

Tests of sexual activity were conducted during the dark phase of the nonreversed day/night cycle and begun at least 1 h after lights off. The floor of the rectangular glass apparatus (60 × 40 × 36 cm) housing the mice was covered with cardboard and sawdust. A clay bowl (10.5 × 5 × 3 cm) contained food and water ad libitum. The apparatus was placed in a box made of double-fluted corrugated cardboard (77 × 56.5 × 57 cm, 0.7-cm thick) allowing complete darkness and reduced noise. An infrared camera (1/4 Sharp CCD 450TVL; Sharp Electronics Corp., Osaka, Japan) was fixed 55 cm above the floor of the apparatus (Fig. 2). Twelve identical settings were used. Before each session, the apparatus was cleaned thoroughly and completely reinstalled. Males were allowed to adapt to the environment for 15 min before a female was introduced into the apparatus.

Fig. 2 Schema of the test apparatus: a rectangular glass box measuring 60 × 40 × 36 cm. The apparatus was placed in a box of 0.7-cm-thick, double-fluted, corrugated cardboard measuring 77 × 56.5 × 57 cm. Observations were recorded by infrared cameras fixed above the apparatus and transmitted to a personal computer for storage and later evaluation.

The experiment was conducted in two rounds of identical design, except for the number of sessions per mouse. In the first round, each naive male was observed with a naive female during a single session of 3 h. In the second round, each male was observed during three sessions of 3 h each on three consecutive days (Fig. 3). At the first session, the males were naive, at the second and third session they were considered non-naive. Naive females were used in the first session, and either naive or non-naive females in the second and third sessions. A female was considered non-naive when used for another session earlier the same day. After termination of the experiment, all mice were euthanized with carbon dioxide and a regional necropsy was performed to assess the status of seminal vesicles.

Fig. 3 Schema of the experimental design: After arrival, naive male and female mice were housed separately for at least 5 d of acclimatization. Then 10–14 d after male surgery, female mice were brought into estrus by exposure to male-originating pheromones during 3 d (Whitten effect). Thereafter, observations were made in sessions of 3 h each. SO = sham operation; SVO = seminal vesicle occlusion; SVR = seminal vesicle resection.

3. Results

Seventy-seven males completed 141 sessions for a total observation time of 423 h. Due to higher perioperative mortality, the number of mice in the SVO (n = 24) and SVR (n = 23) groups were smaller than in the SO group (n = 30) (Table 2). Naive males were used in 77 sessions (55%) and non-naive males in 64 sessions (45%). Naive females were used in 97 sessions (69%), non-naive females in 44 (31%).

A logistic regression model sequentially showed that treatment by session-rank interaction, female experience, and the constant term were not statistically significant. These parameters were dropped from the model at each step. The SVO males had 20 of 42 sessions (48%) with intromission, which was significantly higher than both the 8 of 39 (21%) achieved by SVR males (p = 0.001) and the 18 of 60 (30%) achieved by SO males (p = 0.004) (Fig. 5). In the second round, with three consecutive sessions for each male, the percentage of sessions with intromission was lower in the first sessions than in the second (p = 0.038), but there was no statistically significant difference between the first and third sessions (p = 0.813). This model, with resection group and session order, was statistically significant compared with a null model with only a constant term (p = 0.0017). It also showed that the difference between this model and an ordinary logistic regression model, which did not take into account the clustering (mouse ID) as a random effect, was not significant (p = 0.2379); that is, it did not provide a better fit than an ordinary logistic regression model.

The two secondary end points (number of intromissions and the latency time until the first intromission per session) did not differ statistically among the three groups (p = 0.303 and 0.450, respectively) (Table 2).

Regional necropsies confirmed enlarged seminal vesicles (2.5- to 3.0-cm long) in all SVO, absent seminal vesicles in all SVR, and normal-sized seminal vesicles (1.5-cm long) in all SO mice (Fig. 4).

Fig. 4 Findings at regional necropsy: Left (*): a mouse after sham operation (SO); right (**): a mouse after seminal vesicle occlusion (SVO). After SO, the SVs were of normal size with a stretched length of approximately 1.5 cm. After SVO, the SVs were enlarged and measured a stretched length of 2.5–3.0 cm. B = bowel; F = fat; UB = urinary bladder.

Fig. 5 Percentage of sessions with intromission related to the total number of sessions observed per group. Males after seminal vesicle occlusion were statistically significantly more often sexually active (48%) than males after seminal vesicle resection (21%) or sham operation (30%).

4. Discussion

Over decades, many vain attempts have been made to elucidate the influence of various genital organs on sexual activity in animal models. We think our present mouse experiment finally reveals a novel aspect of the physiologic understanding of the seminal vesicles. Showing that males with occluded, and thus engorged, seminal vesicles were significantly more often sexually active than males with SVR or who underwent SO suggests that seminal vesicles affect sex drive.

Unlike previous animal experiments on the role of seminal vesicles on sexual activity, ours uses a mouse model, which is now recognized, together with the rat and some other small animal models, as one of the most important mammalian models for the study of male and female sexual function [11], [13], [14], and [15]. We attribute our ability, in contrast to others (Table 1), to show significant differences in sexual activity depending on seminal vesicle status to three major factors. First, the experimental design was based on maintaining a natural environment and keeping animal stress as low as possible by, for example, preserving the natural day/night cycle, use of noise-reducing boxes, and allowing sufficient time for males to habituate to the apparatus. We also applied the Whitten effect, which allows for natural and reliable female priming by male pheromones with up to 80% females reaching estrus [12]. In mice, unlike in rats, the Whitten effect is possibly even more effective than hormone priming [18] and [19]. Second, to optimize assessment of natural mouse sexual activity, each observation was extended to 3 h, since short observation times (eg, 15 min) often do not capture later components of sexual activity, such as intromission and ejaculation, and therefore might underestimate sexual activity [15]. Third, the study design focused on a single genital organ, namely seminal vesicles.

The persistent lack of standardized tests for assessing sexual activity in animals (ie, mice) has led to heterogeneous experimental designs and inconclusive results [10] and [15]. The number of intromissions and latency times until first intromission were often used as end points; however, the number of intromissions is considered to be mainly an indicator of sexual endurance [15]. Moreover, intromission ending in ejaculation usually terminates male sexual interest and thus sexual activity for several hours. This affects reliability [13]. The latency time until first intromission is influenced by the receptiveness of the female and so may inaccurately reflect male sexual activity [14] and [15]. Therefore, we chose to assess male sexual activity by distinguishing between sessions with and without intromission.

Sexual experience and natural variance within the same male can also influence sexual activity [14] and [15]. The female is also a relevant confounding factor when assessing male sexual activity [14] and [20]. To render the outcome of the experiment as independent of sexual experience as possible, both naive and non-naive males and females were used. By adding these data into a logistic regression model, sexual experience was excluded as a confounding factor for both males and females.

In our experiment, males with SVO had the highest sexual activity, with intromission occurring in 48% of sessions. Their sexual activity was significantly higher compared with that of SVR and SO males, with intromission in those mice occurring in 21% and 30% of sessions, respectively. The secondary end points did not differ statistically among the three experimental groups. We hypothesize, therefore, that occluded, and thus engorged, seminal vesicles act as an intrinsic stimulus for sexual activity, eventually resulting in greater sexual activity. Possibly reflecting a general physiologic preparedness of enough seminal fluid for a successful sperm transfer, the intrinsic stimulus is consequently highest in males with SVO and absent in males with SVR. The mechanism by which sexual activity is influenced when occluding the seminal vesicles’ outlet is unknown and was not the subject of this experiment. The identification of a possible mechanism, such as neural or humoral substrates, must be a subject of future experiments.

The results achieved in the present mouse model cannot necessarily be translated to other mouse strains, or other species, including humans, although some of them have numerous common components of sexual behavior that may reflect similar physiologic actions [15] and [21]. That male sexual activity is also influenced to varying degrees by factors other than SVO (eg, by extrinsic stimuli emitted by the female) is best reflected by the fact that seminal vesicles are not present in all mammals. They are absent in marsupials, monotremes, and carnivores, and they are smaller in some insectivores, rodents, lagomorphs, bats, and primates [22], [23], and [24].

The extent to which the seminal vesicles play a role in sexual activity in humans is unknown. However, subjective observations in clinical practice suggest that men who have undergone seminal vesicle-sparing radical cystoprostatectomy tend to report a higher sex drive than men who have undergone radical cystoprostatovesiculectomy (ie, after seminal vesicle removal). Knowing from our mouse experiment that seminal vesicle sparing may positively influence sex drive should be an additional incentive to question the need for standard seminal vesicle removal during cystoprostatectomy, especially because several trials have shown seminal vesicle sparing to be oncologically safe in selected patients [25] and [26].

A potential limitation of our experiment is its design involving two different rounds, which was necessary because of limited housing and testing capabilities. The differences between the two rounds were taken into account in the logistic regression model. The criticism can also be made that after the Whitten effect, selection of females was not based on a positively detected fertility status. We think, however, that the inclusion of all females is also a strength of the experiment, as any selection by means of seemingly obvious criteria inherently bears the risk of a selection bias. Instead, we accounted for the female factor in the logistic regression model and eventually ruled it out as a confounding factor. Finally, our method for assessing sexual activity as such had not been previously standardized. We think, however, that our method straightforwardly reflects sexual activity by scoring each session as being with or without intromission, leading to even more conservative results compared with counting the absolute numbers of intromissions and the latency times until first intromission, which are more susceptible to biasing factors.

5. Conclusions

Male mice with SVO were significantly more often sexually active than males undergoing SVR or SO. These findings suggest that occluded, and thus engorged, seminal vesicles have a significant effect on sex drive in mice. The applications of these results in men, especially the potential benefits of seminal vesicle-sparing techniques in uro-oncologic surgery, warrant further investigation.

Author contributions: Urs E. Studer had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Study concept and design: Studer, Birkhäuser.

Acquisition of data: Birkhäuser, Schumacher, Seiler, de Meuron, Wetterwald, Cecchini.

Analysis and interpretation of data: Birkhäuser, Studer, Schumacher, Seiler, de Meuron.

Drafting of the manuscript: Birkhäuser, Schumacher.

Critical revision of the manuscript for important intellectual content: Studer, Roth, Zehnder, Thalmann.

Statistical analysis: None.

Obtaining funding: Birkhäuser.

Administrative, technical, or material support: Zehnder, Thalmann, Roth.

Supervision: Studer.

Other (specify): None.

Financial disclosures: Urs E. Studer certifies that all conflicts of interest, including specific financial interests and relationships and affiliations relevant to the subject matter or materials discussed in the manuscript (eg, employment/affiliation, grants or funding, consultancies, honoraria, stock ownership or options, expert testimony, royalties, or patents filed, received, or pending), are the following: None.

Funding/Support and role of the sponsor: Max und Hedwig Niedermaier foundation, Zurich, Switzerland, was involved in the design and conduct of this study.

Acknowledgment statement: The authors are grateful to Richard J. Sylvester, ScD, Department of Biostatistics, European Organisation for Research and Treatment of Cancer (EORTC), Brussels, Belgium, who performed all of the statistical analyses. They also thank Ursula Gerber for her excellent technical help and Hans Peter Käsermann, PhD, Institute of Laboratory Animal Science, University of Zurich, Switzerland, for his valuable advice.