Retracted: Effects of exposure to a mobile phone on testicular function and structure in adult rabbit

Summary

The accumulating effects of exposure to electromagnetic radiation emitted by a conventional mobile phone (standby position) on the testicular function and structure are not yet fully investigated. To study these effects longitudinally, a total of 24 adult male rabbits were randomly and equally divided into three groups. Rabbits in the first (phone) group were exposed, in specially designed cages, to radio frequency emitted from the mobile phone (800 MHz) in a standby position opposite to that of testes for 8 h daily for 12 weeks. The second group consisted of the stress controls which were kept in the same kind of cages to appreciate any cage-induced anxiety. The third group included the ordinary controls which were kept in the conventional roomy cages. Semen analysis and sperm function tests (viability, hypo-osmotic swelling and acridine orange) were conducted weekly. Histological testicular sections and serum total testosterone were also evaluated. A drop in the sperm concentration appeared in the phone group at week 6. This became statistically significant at week 8, compared with the two control (stress and ordinary) groups (133, 339 and 356 × 106/mL, respectively) and to the initial sperm count (341 × 106/mL) of this group. Motile sperm population showed similarity amongst the three study groups until week 10 when it declined significantly, and thereafter in the phone and stress control groups, with more significant decline in the phone animals (50, 61 and 72.4%, respectively). Histological examination showed also a significant decrease in the diameter of seminiferous tubules in the phone group vs. the stress and ordinary controls (191 μm vs. 206 and 226 μm, respectively). The other study points did not show any difference. In conclusion, low intensity pulsed radio frequency emitted by a conventional mobile phone kept in the standby position could affect the testicular function and structure in the adult rabbit.

Introduction

Mobile phones (MP) have become popular devices worldwide. It was estimated that more than 2 billion people would now be MP carriers, according to the documented growth rate drifts (Wearden, 2004). This was associated with raised public concerns regarding human safety related to electromagnetic waves (EMW) emitted from these devices. Recently, an increasing number of studies related to the potential adverse impacts of EMW on several body systems of humans and animals have been conducted. However, results of these studies appeared contradictory (Malyapa et al., 1997; Freude et al., 1998) and the question of adverse health impacts provoked by MP is unresolved as yet. Therefore, an additional research is needed.

As men usually lug MP in their pockets, close to testes in the standby position most of the day, assessment of the consequences of MP handling on male reproductive function seems of great importance. Although studies in humans highlighted negative impacts of MP on semen quality (Fejes et al., 2005; Agarwal et al., 2008), the natural history of these impacts have not been extensively addressed and the studies are still lacking.

Animal studies addressing the same issue of the adverse effects of MP on male reproductive function were numerous when compared with human studies. However, results of these studies were conflicting. Whilst some showed negative findings (Dasdag et al., 2003; Ribeiro et al., 2007), others indicated that EMW may have a wide spectrum of detrimental effects on sperm parameters (Yan et al., 2007), testicular structure (Dasdag et al., 1999) and male germ line (Aitken et al., 2005).

Rabbit is an important animal model used during research on semen quality (Veeramachaneni et al., 2001). Perhaps, one of the main advantages it can offer during this research is its easy manipulation to collect readily, the serial semen samples. These can be well suited for any longitudinal study to define precisely the timing the EMW may affect the reproduction. The aim of this work was therefore to describe and label the accumulating effects of exposure to MP emission, kept in the standby position, on testicular function and structure through longitudinal evaluation of rabbits.

Material and methods

Animals

A total of 24 adult male White New Zealand rabbits (Nihon SLC, Hamamatsu, Japan), 20-weeks old and weighing 3.15–3.25 kg represented this study’s material. Rabbits were individually caged and kept in a temperature-controlled environment (22–24 °C) under a 12 h light/dark cycle (light on at 7:00 am). Animals were provided standard rabbit chow and water ad libitum. The protocol for this experimental study was approved by the Ethics Committee of Experimentation on Animals at the University of Tokushima.

Exposure

Animals were randomly divided into three groups. The first group was the MP group (n = 8), where the rabbits were placed individually in specially designed cages with sizes of 50 cm × 25 cm × 35 cm. These cages could accept placing plastic partitions according to the animal dimensions which averaged to 30 cm × 16 cm × 18 cm to restrict its movements. Hence, it sits down throughout the period of the daily phone exposure with its testes opposite to the phone antenna as the MPs were fixed at the cage bottoms (Fig. 1). Phones used were conventional GSM (global system for mobile communications) MP handsets (800 MHz) which were turned to the standby position with an average strength of the electric field of 2.92 V/m, estimated at 0.5 cm away from the phone and 0.487 V/m at the most distant region inside the cage. In both control groups where the cages were positioned 7 m away, average strength of the electric field detected was equivalent to background radiation (0.18 V/m). The whole-body average specific absorption rate (SAR) was estimated using the Finite Difference Time Domain method (Kunz & Luebbers, 1993) and it was 0.43 W/kg. No bedding material was provided in the exposure cages.

Figure 1.  Position of the mobile phone.

Phones were applied for 8 h (9:00 am–5:00 pm) daily during the course of this current study which extended for 12 weeks. Following this daily exposure to MP, the animals were returned to their individual standard cages (90 cm × 60 cm × 40 cm) available in the animal house. Because of the restriction of animal movements and the possibility of obtaining a stress-related outcome (Linares et al., 2005), two control groups were taken. Rabbits in the first group were the sham or stress controls (n = 8) which were placed in the same kind of the small cages daily during the 8-h study but with the phone switched off whilst rabbits in the second group represented the ordinary controls (n = 8) which were left throughout the time of the study in the conventionally available cages provided by the animal house.

Rectal temperature assessment

This was carried out for all animals in this study twice per week. The check-up was made in the morning before and in the evening after phone exposure.

Semen retrieval and manipulation

Semen was taken from rabbits twice per week over the 12 weeks of the study via an artificial vagina constructed from the local materials in presence of a teaser doe (Bredderman et al., 1964). The volume and pH of each ejaculate was recorded after removal of the gel mass using Pasteur pipette and forceps. The numbers of spermatozoa in samples were determined in duplicate by direct microscopic examination with a haemocytometer. Sperm motility was assessed by placing a droplet of semen on a glass slide on a microwarm plate (Kitazato, Shizuoka, Japan) at 37 °C and examined by light microscopy (Nikon, Tokyo, Japan) at ×100. Sperm motility was defined as the percentage of sperms that showed any sperm head movement. Morphological features of sperm were evaluated using a light microscope equipped with differential interference contrast optics (Olympus, Tokyo, Japan). Two hundred sperms per ejaculate were evaluated in wet smears for abnormalities of the acrosome, head, mid- and principal pieces, retention of cytoplasmic droplet and presence of residual cytoplasm using criteria previously established for rabbits (Veeramachaneni et al., 2001).

Sperm function tests

Eosin-nigrosin viability test

Assessment of live and normal spermatozoa were performed using an eosin–nigrosin blue staining mixture according to method by Blom (1950). In brief, the stain was performed by dissolving 0.5 g eosin Y and 3 g nigrosin in 30 mL distilled water. Two drops of the prepared solution were added to a drop of the fresh semen sample on a clean microscopic slide. A smear of this mixture was then made on another microscopic slide and allowed to air dry. This slide was then examined under a phase contrast microscope at ×400. The percentage of normal live viable sperms (sperm heads unstained) and non-viable sperms (sperm heads stained) was assessed by counting a minimum of 100 sperms.

Hypo-osmotic swelling (HOS) test

This was performed by mixing 100 μL of each semen sample with 900 μL of 60 mosmol fructose solution, differing from that originally used by Jeyendran et al. (1984) because studies showed a form of sodium citrate toxicity on rabbit spermatozoa (Ducci et al., 2002). The mixture was incubated for 60 min at 37 °C in 5% CO₂ and 95% air. Then 0.2 mL of the mixture was placed on a slide and mounted with a coverslip and immediately examined at a magnification of ×400 under a phase contrast microscope. The percentage of reacted sperms (curled tails) and non-reacted sperms (non-curled tails) was assessed by counting a minimum of 100 sperms.

Acridine orange (AO) test

The test was used to assess sperm chromatin DNA as originally described by Tejada et al. (1984). Briefly, after air-drying the sperm smears prepared from semen, fixation in Carnoy’s solution (three parts of methanol and one part of glacial acetic acid) was carried out for at least 3 h or overnight at 4 °C. After fixation, the slides were allowed to air-dry for few minutes before staining with acridine orange (AO) solution (Sigma Chemical Co, St. Louis, MO, USA) for 5 min. Then the slides were gently rinsed with distilled water and mounted. The percentage of sperms with normal DNA was determined by randomly scoring 200 sperms under a fluorescence microscope (Nikon) with magnification of ×400 and excitation of 450–490 nm. Sperms with normal (double-stranded) DNA fluoresce with a green colour, and those with denatured or single-stranded DNA fluoresce with a red or yellow colour.

Testosterone measurements

Blood samples were collected from the ear vein of each buck at 10:00 am. The serum total testosterone concentration was determined by DSL-4000^® radioimmunoassay (SRL, Osaka, Japan). The limit of sensitivity was 0.1 ng/mL, and intra-assay and inter-assay coefficients of variation were 8.5%.

Histopathological evaluation of the testis

Seminiferous epithelium and interstitium were evaluated using 5-μm-thick sections stained with haematoxylin and eosin and counter-stained with periodic acid Schiff under light microscopy (Nikon FXA light microscope; Nikon). The diameters of 100 randomly selected, essentially round seminiferous tubules from each animal, were estimated and classified into one of eight different grades (Veeramachaneni et al., 2006): grade 0, normal intact seminiferous epithelium; grade 1, seminiferous epithelium with pyknotic cells and desquamation or focal vacuolation; grade 2, seminiferous epithelium intermediate between grades 1 and 3; grade 3, seminiferous epithelium with pre-meiotic germ cells and Sertoli cells; grade 4, Sertoli cells only; grade 5, no seminiferous epithelium, leaving only the basement membrane; grade 6, seminiferous tubule with sperm stasis, sperm granuloma, or mineralization; and grade 7, fibrotic seminiferous tubule.

Statistics

Data were expressed as mean ± SD. Statistical analysis was carried out using spss statistical software (SPSS for Windows; SPSS Inc, Chicago, IL, USA), and non-parametric analysis with Mann–Whitney test was performed to examine the difference between each of the two groups and the longitudinal change within each group. A p-value <0.05 was considered to be significant.

Results

Mean rectal temperature (37.6 ± 0.3 °C) did not alter following exposure (37.5 ± 0.27 °C). Testicular (2.8 ± 0.4 g) and epididymal weights (0.8 ± 0.2 g) showed similarity in the different groups whilst total body weights were not significantly different (Table 1). A drop in the sperm concentration/mL started to appear in the MP group at week 6. This became statistically significant at week 8, and thereafter compared with the initial sperm count (341 × 106/mL) in this group (Fig. 2) and in the two control (stress and ordinary) groups which showed no difference in between (Table 1). Motile sperm population showed similarity amongst the three study groups (Fig. 3) until week 10 when it declined significantly and thereafter in the MP and stress control groups, with more significant decline in the MP animals (Table 1). Measured seminiferous tubular diameters of animals in the MP group were significantly lower than both control groups (Table 1). All tissue sections from the different animals showed normal intact seminiferous epithelium (grade 0). No difference was observed between the MP and the two control groups in terms of normal sperm forms or live sperm percentages, HOS test, AO test, or mean serum total testosterone throughout the course of the study (Table 1).

Table 1.   Total body weight, sperm parameters, sperm function tests, seminiferous tubular diameter and serum total testosterone (mean ± SD) in the animal groups at the end of the study

End point	MP group	p-Value (MP vs. S)	S group	p-Value (S vs. O)	O group	p-Value (MP vs. O)
S, stress (sham) controls; O, ordinary controls; MP, mobile phone; HOS, hypo-osmotic swelling test; AO, acridine orange test.
Total body weight (g)	3583 ± 233	0.092	3763 ± 190	0.36	3590 ± 144	0.36
Sperm count (m/mL)	133 ± 46.1	0.004	339 ± 17.2	0.17	356 ± 12.9	0.006
Total motile sperm (%)	50 ± 2.5	0.003	61 ± 2.2	0.006	72.4 ± 3.2	0.005
Normal sperm forms (%)	72 ± 3	0.62	72.5 ± 2.7	0.71	73 ± 3.2	0.46
Normal live sperm (%)	81.3 ± 1.2	0.6	81.6 ± 1.1	0.7	81.8 ± 1.2	0.46
HOS (%)	92.5 ± 1.04	0.15	91.7 ± .8	0.45	91 ± 1.2	0.11
AO (%)	96.2 ± 0.8	0.43	95.8 ± 0.8	0.92	95.8 ± 0.8	0.43
Seminiferous tubular diameter (μm)	191 ± 8.6	0.045	206 ± 10.9	0.055	226 ± 19.8	0.006
Serum total testosterone (ng/mL)	1.2 ± 0.08	1	1.2 ± 0.08	0.6	1.1 ± 0.05	0.6

Figure 2.  Sperm count during the course of the study.

Figure 3.  Sperm motility during the course of the study.

Discussion

This study is an experimental study which evaluated longitudinally the biological effects of MP exposure kept in the standby position on sperm parameters, functional integrity of sperm membranes, sperm DNA status, testis histology, testosterone concentration and rectal temperature in adult rabbits. Using experimental studies with animals makes one capable of controlling the lifestyle factors that are known to influence semen quality. Herein, we used the rabbit as an animal model which can easily yield semen directly at the time we want and thus helping us to design a longitudinal study. This is in contrast with the rat where the sperm is mostly obtained via epididymal retrieval (Dasdag et al., 1999, 2003). Besides, if we consider the non-pendulous nature of rat’s scrotum and the fact that its testes can migrate easily and frequently between the abdomen and scrotum in the inguinal canal because of its small body size (Cairnie & Harding, 1981), this will enable us to conclude rat as a poor model to study the impact of MP on reproductive function.

Because of the existence of different protocols to apply the EMW, controversial results appeared in the literature. So, some studies addressed significant testicular and sperm parameters changes (Varma & Traboulary, 1975; Dasdag et al., 1999; Yan et al., 2007) whilst others did not acknowledge any changes (Cairnie & Harding, 1981; Dasdag et al., 2003). In this current study, using the MP in the standby position, we were able to reveal several positive changes including longitudinal decline in sperm count as well as decrease in sperm motility and diameter of seminiferous tubules in the MP animals with lack of any change in rectal temperature. To our knowledge this is the first longitudinal study of the testicular non-thermal effects of MP performed after an adequate exposure time.

As experimental data in animals (Linares et al., 2005) and humans (Morelli et al., 2000) suggest that chronic or severe stress leads to decrease in sperm count, motility and morphology in males, we added two control groups, sham (stress) and ordinary controls, to avoid the possibility of obtaining stress-related outcome. In addition, this study demonstrated that stress-induced cage restriction could affect significantly the motile sperm population in the stress control group. However, the study showed that MP animals had significantly profound drop in motile sperm population than stress controls. Therefore, we think that it was not only the stress that produced these evident changes in motility in the MP animals, but it was also the MP itself, and this stress may have a cofactor effect working with the MP side by side to produce ultimately a greater negative impact on the motile sperm population.

These positive findings addressed in the current study may indicate an EMW-specific effect or non-thermal effect. However, a heat effect in the testes is another possibility even though there was no change in rectal temperature. Three lines of evidence exist to support this possibility. First, testes under this study were relatively more exposed than other body organs of the animals to EMW and for long durations. Thus the chance to absorb more radiation by testes is greater. Second, MP used in this study has carrier frequency of 800 MHz. It is known that frequencies from about 300 MHz to several GHz can result in significant local, non-uniform absorption (ICNIRP Guidelines, 1998). According to these two evidences, we think that testis SAR value may be relatively higher than the estimated whole-body SAR. This higher SAR could lead to this heat effect in the testis. Third, testes rely mainly on surface conduction for any thermal changes and seem likely to be easily heated in contrast to other body parts which depend on their local blood flow for heat control. Our two suggestions of non-thermal and thermal effects, therefore, agree with other workers who suggested a combination of both effects as probable mechanism for the pathogenesis of MP on biological systems (Blackwell, 1979).

This non-thermal effect causing the positive changes, as suggested in the present study, was previously discussed by previous workers who related it to different mechanisms. So Wang et al. (2003) proposed in their study on mice that Leydig cells are amongst the most susceptible cells to EMW, and injury to these cells may affect spermatogenesis. Other workers (Burch et al., 1998) indicated an EMW-dependent decrease in melatonin, an antioxidant, which can predispose the sperm to oxidative stress and hence cause a decline in its motility (Aitken & Baker, 1995a) and count (Whittington et al., 1999). Recent studies suggested also that low-level radio frequency (RF) fields may cause alterations in cellular behaviours as that seen in Merkel cells upon exposure to EMW, which showed higher exocytotic activity and depletion of its intracellular granules (Irmak et al., 2003). Pace et al. (2000) suggested in their study that EMW may exert an effect on the state of polarization of the cellular membranes in the human body which is responsible for the process of spermatogenesis and for the properties of a sperm cell enabling its penetration into the egg cell (Darszon et al., 2001). Lastly, a number of in vivo and in vitro studies (Zhao et al., 2007) indicated that RF fields could interact with charged intracellular macromolecular structures. Especially sensitive could be the cytoskeleton which has an important role in the sperm motility (Chemes et al., 1987), and is actively involved in the morphological changes that occur during mammalian spermiogenesis (Fouquet et al., 1998).

This study has indicated significant decrease in sperm count and motility at weeks 8 and 10 because of exposure to MP emission, respectively. This exposure time is nearly equivalent to or exceeding six seminiferous epithelium cycles where the duration of one cycle of seminiferous epithelium in the rabbit is about 10 days (Swierstra & Foote, 1965). This time lag of six cycles covers the period necessary for spermatogonia to change into sperm and reach the cauda epididymis, and this is recommended by several workers doing research to study subtle reproductive toxicity alterations in rodents (Creasy, 1997).

Acridine orange test used in this experiment is a simple microscopic procedure to indicate sperm DNA abnormalities. However, we could not detect any changes in AO test outcome in the different animal groups. Ono et al. (2004) used DNA sequence technique and indicated that RF of small bursts is not mutagenic in mouse when exposed in utero. Aitken et al. (2005), using power electrophoresis, found a statistically significant genotoxic effect on epididymal sperm with damage to mitochondrial DNA and the nuclear ß-globin locus. Therefore, the nature and range of mutation that may be induced by RF exposure is not yet clear. Although AO test is a less expensive procedure and can be performed in a short period of time, previous studies showed its low clinical significance for infertility testing (Eggert-Kruse et al., 1996). Therefore, it is hypothesized that the microscopic evaluation under AO test only provides reliable results when there is a high degree of DNA fragmentation (Chohan et al., 2006). This may represent one limitation in the current study, and hence we think that it was a less sensitive tool to evaluate any possible sperm DNA changes.

In conclusion, the current study shows that the low intensity pulsed RF emitted by a conventional MP, kept on the standby position, could affect the testicular function and structure in the adult rabbits. More studies, however, are needed to identify precisely the mechanism involved in the reduction of sperm quality and testicular structure.