Effect of Experimental Diabetes and STZ on Male Fertility Capacity. Study in Rats


Complejo Hospitalario Universitario de Albacete, Hermanos Falco 37, Albacete, Spain 02006 (e-mail: lnavarro@sescam.jccm.es).


ABSTRACT: To assess the effect of experimental Type 1 diabetes on male fertility, male Sprague Dawley rats were injected with either streptozotocine (STZ) to induce diabetes or with citrate buffer as controls. Diabetic animals and 2 control groups (STZ-resistant and buffer-injected rats) were sacrificed at 2 different times after injection: 6 weeks (6W) and 20 weeks (20W). We analyzed serum testosterone (sTT), epididymal sperm parameters, and weight of testicles and epididymides, and carried out a histological evaluation of testicular tissue. Diabetic animals presented a significant increase in teratozoospermia (20W, P < .01) and a decrease in sTT (P < .01), tubular diameter (6W, P < .05), and testicular (6W, P < .01) and epididymal (P < .01) weight. STZ-resistant animals showed significantly decreased sTT (6W, P < .01), epididymal weight (6W, P < .05), and sperm count (6W, P < .01) compared with buffer–injected controls. Experimental STZ diabetes increases teratozoospermia and decreases sTT, testicular weight (reverting at medium-term), and epididymal weight.

It is well established that diabetes can produce erectile dysfunction and retrograde ejaculation in men (Rodriguez-Rigau, 1980). Nevertheless, many studies continue to provide conflicting results with respect to the direct effects of diabetes on male fertility potential. Although Andersson et al (1994), observed microscopic abnormalities in testicular morphology in 10 Type 1 diabetic young men with erectile dysfunction, this has not been found by others. However, with the introduction of new analytical techniques, evidence is now emerging of previously undetectable effects of diabetes on male fertility. Also an increase in advanced glycated end products in diabetic men suggests that these compounds could play a hitherto unrecognized role in male infertility, as proposed by Mallidis et al (2009).

Conversely, reviewing the literature (Padron et al, 1984; Cameron et al, 1985; Murray et al, 1988; Pitteloud et al, 2005), significant differences between the levels of testicular and pituitary hormones and changes in seminal quality in diabetic men compared with nondiabetic men have not been found consistently. The findings regarding semen quality have not been consistent, showing variation from normal to altered sperm count, sperm morphology, motility, or a combination of conditions (Bartak, 1979; Vignon et al, 1991; Ali et al, 1993). With regard to recently introduced new analysis techniques, specifically, a recent study has found a significantly higher level of sperm nuclear DNA fragmentation in diabetic men (Agbaje et al, 2008).

Similarly, controlled experimental studies performed in spontaneous or post-streptozotocin diabetic rats have described divergent data. They have shown serum testosterone impairment linked to increased or decreased serum levels of gonadotropins (Anderson et al, 1987; Escrig et al, 2002; Ballester et al, 2004). In some studies (Orth et al, 1979), slight injuries of testicular and epididymal structures have been detected by means of an electron microscope. In other reports, severe macroscopic testicular and epididymal structural lesions were present in diabetic compared with nondiabetic animals. Diabetes significantly decreased seminiferous tubule diameter, increased testicular blood vessel numbers, and altered tubule stage distribution compared with controls (Paz and Homonnai, 1979; Anderson and Thliveris, 1986, 1987; Anderson et al, 1986; Cai et al, 2000; Amaral et al, 2006). In addition, very limited data are available in the literature regarding the effect of streptozotocin on male reproductive function independent of diabetic status.

Streptozotocin, a monofunctional nitrosourea derivative isolated from Streptomyces achromogenes, is a potent alkylating agent known to methylate DNA directly and to be highly genotoxic, producing DNA strand breaks, chromosomal aberrations, and cell death. This antibiotic was found to be mutagenic in bacterial assays and in eukaryotic cells. STZ is also carcinogenic: a single administration can induce tumors in rat kidney, liver, and pancreas. Several lines of evidence indicate that free radicals are involved in the damage of DNA and chromosomes by this compound (Bolzan and Bianchi, 2002). Despite all these facts, few studies exist of the influence of STZ on fertility in male rats. Anderson et al (1987) found that STZ had no significant effect on cytochemistry, morphometry of Leydig cells, or testosterone serum levels 3 months after its administration.

Because of the discrepancy in data, we conducted this study to analyze the effect of streptozotocin-induced diabetes on different parameters of male reproduction in rats. The second aim of this study was to differentiate between toxic effects of STZ and the effect of hyperglycemia on fertility status. We further evaluated whether STZ alters male fertility independently from induced diabetes.

Materials and Methods


The experiments were performed in immature male Sprague Dawley rats (6–7 weeks old) reared in our laboratory. Animals were housed under controlled conditions of light cycle (12 hours:12 hours light:dark) with free access to water and rat chow. All experimental procedures complied with the National Regulations for the Care and the Use of Laboratory Animals (similar to the US National Research Council guidelines) and were approved by the local Ethics Committee.

Diabetic Induction

A total of 86 rats were injected with either streptozotocin (STZ, n = 64; Sigma Chemical Co, St Louis, Missouri) or vehicle (0.1 M citrate buffer, pH 5; n = 22). Two different STZ doses were administered (45 and 60 mg/kg body weight) given in a single intraperitoneal injection. STZ was injected in freshly prepared citrate buffer and protected from daylight.

All animals were housed in standard cages in groups of 2 or 3. Ten days after injection, animal glucose metabolism was evaluated. Those animals that had urine glucose levels higher than 1000 mg/d were considered diabetic; they manifested typical polyuria, polyphagia, and polydipsia. STZ-injected animals that exhibited urine glucose levels lower than 50 mg/d were considered nondiabetic; they did not suffer from diabetic symptoms at any time. A second metabolic evaluation was performed 1 week before sacrificing to confirm the diabetic status and to measure diuresis and food-water intake. All animals were maintained without insulin until sacrifice. The sacrifice was performed in non-fasting rats, therefore basal serum glucose levels were not used for animal classification.

Experimental Design

Two groups were established. Group A were nondiabetic control rats divided into 5 subgroups. Animals in subgroups A1, A2, and A3 were nondiabetic vehicle-injected rats evaluated at 6 (n = 8), 20 (n = 6), and 60 (n = 8) weeks after injection, respectively. Animals in subgroups A4 and A5 were STZ-injected rats that did not develop diabetes (STZ-resistant), evaluated at 6 (n = 18) or 20 (n = 7) weeks after injection, respectively.

Three groups A4a, A4b, and A5 were additionally differentiated according to the STZ dose administered and time of evaluation. Animals in group A4a were rats belonging to subgroup A4 that received 60 mg/kg of STZ (n = 4); animals in group A4b were rats belonging to subgroup A4 that received 45 mg/kg of STZ (n = 14); animals in group A5 were rats that received a single STZ dose of 60 mg/kg and were evaluated after 20 weeks.

Group B were diabetic rats divided into 2 subgroups: B1 and B2. Animals in subgroup B1 were STZ-injected animals with developing diabetes evaluated at 6 weeks (6W, n = 25). Animals in subgroup B2 were STZ-injected animals with developing diabetes evaluated at 20 weeks (20W, n = 6) after injection. Subgroup B1 was further divided into groups B1a and B1b and received 60 mg/kg of STZ (n = 18) or 45 mg/kg of STZ (n = 7), respectively.

Testicular hormones, testicular structure, and seminal quality were evaluated at short term (6W) from groups A1 and B1 and at medium-long term (20W) from groups A2 and B2. Groups A1, A2, and A3 were used to investigate the effect of age on testicular hormonal production, testicular structure, and seminal quality. Groups A4 and A5 were included to distinguish the STZ-dependent from the diabetes-dependent effects at short- and medium-long–term periods. Finally, groups B1a and B1b were compared to examine the effects of the severity of diabetes. (The experimental design is shown in Table 1.)

Table 1. . Scheme of groups. A1–A3 are vehicle-injected control rats. A4 and A5 are streptozotocine (STZ)-resistant control rats. B1 and B2 are diabetic rats (STZ-sensitive). Subgroups a and b depend on STZ dose of 60 or 45 mg/kg of body weight, respectively
 ControlSTZ-ResistantSTZ-Sensitive (Diabetic)
Sample size868187256
Time of sacrifice, wk62060620620
Injected dose of STZ, mg/kg00060 (A4a, n = 4)6060 (B1a, n = 18)60
    45 (A4b, n = 14) 45 (B1b, n = 7) 

Surgical Technique

We performed median laparotomy under deep combined ketamine-xylazine anesthesia to remove testes and epididymides. Unilateral testis and epididymis from each rat were weighed before and after testicular decapsulation (testicular parenchyma), and semen was obtained from the epididymis. The remaining testis and epididymis were fixed in a B5 medium (Panreac, Barcelona, Spain) for 60 minutes and then placed into 10% formol fixative for later morphological examination. Blood was collected from rat hearts, and the animals were sacrificed immediately after with a lethal injection of sodium pentothal.

Analytic Procedures in Serum Samples

We measured serum glucose by an automated glucose oxidase method and serum fructosamine by a method based on the capacity of the ketoamines to reduce the nitroblue tetrazolium to formazan (Hitachi analyzer; Roche Diagnostics, Barcelona, Spain). Serum testosterone levels were assessed by a chemiluminescence assay (Elecsys 2120, Roche).

Histological Techniques for Light Microscopic Examination

Fixed testis tissue samples were embedded in paraffin, mounted on glass slides, and stained with hematoxylin-eosin. Ten consecutive seminiferous tubules from each animal were selected according to a systematic method of microscopic analysis following a line from the edge to the center of the testis cross-section. The short axis diameters of the selected tubules were measured under ×100 magnification with the aid of an ocular micrometer calibrated by means of a stage micrometer scale. Seminiferous tubules and the interstitial space were examined under light microscopy to evaluate basal membrane thickening, vascular changes, Leydig cells, and spermatogenesis.

Sperm Reserve in Epididymis

Sperm reserve in the epididymis was determined by the procedure of Cooke et al (1991). After the epididymis was dissected out, a small cut was made in the caudal epididymis and the epididymal fluid was aspirated from the vas deferens. The epididymal fluid was placed into a preweighed Pasteur pipette, and the pipette was reweighed to measure the epididymal fluid weight. An aliquot was subsequently placed into 5 mL of Ham F-10 medium. Sperm density was assessed by a hemocytometer. Sperm motility was evaluated under ×400 magnification with a phase-contrast microscope and classified into 4 groups: a, linear motility; b, no linear motility; c, no propulsive motility; d, nonmotile.

Sperm morphology was analyzed under ×1000 magnification with a Diff-Quick stained preparation. Head, medial piece, and tail abnormalities were recorded. An aliquot was frozen for later enzymatic l-carnitine determination after deproteinization with 0.6 M perchloric acid and neutralization with 1.2 M potassium carbonate (enzymatic method, Roche).

Statistical Analysis

Data are expressed as x̄ ± standard error (SE). Single comparisons were analyzed with Student's or Mann-Whitney testing where appropriate. For multiple comparisons, analysis of variance followed by post hoc analysis was used where appropriate (Bonferroni test when the variances are matching or Tamhane when not). A value of P < .05 was considered to be statistically significant.


Table 2 shows the initial general characteristics of the different groups and the first metabolic evaluation (food intake and urine output). Table 3 depicts the second metabolic evaluation (water ingestion, urine volume, food, and urine glucose levels) performed 1 week before sacrifice, as well as body weight, serum glucose, serum fructosamine, and serum testosterone at the end of the study. Diabetic rats had significantly higher nonfasting serum glucose and urine volume and significantly lower body weight compared with controls. The increase in serum glucose was observed at short (6W) and medium-long terms (20W). The levels of fructosamine in the diabetic animals were higher than in the nondiabetic STZ-injected rats at both times (P < .001) and in the nondiabetic vehicle-injected group at short term (P = .025). Fructosamine levels overlapped between groups because, in addition to an increase in diabetic animals, fructosamine also rises with age, although not significantly. Testosterone levels were significantly lower in diabetic rats compared with controls at both times (P < .01), as shown in Table 3. It appears that the short-term effect on testosterone level is related to STZ, this effect being transient, whereas the long-term effect would be attributable to diabetes. Dose-related lower testosterone levels were significant in STZ-resistant animals compared with vehicle-injected rats at 6W (P = .006) and depending on the dose (A4a vs A4b, P = .042). In diabetic rats, the effect was not evident, as shown in Figure 1.

Table 2. . Comparison of initial physiological profiles (10 days after streptozotocine [STZ] or vehicle injection) in control and diabetic rats (x̄ ±SE)
GroupaAge, wkWeight, gFirst Control: Water Ingestion, mLDiuresis, mLFood, g
  1. a Group A: vehicle-injected rats sacrificed after 6 (A1), 20 (A2), and 60 weeks (A3); STZ-resistant rats sacrificed after 6 (A4) and 20 weeks (A5). Group B: diabetic rats sacrificed after 6 (B1) and 20 weeks (B2).

  2. b Significantly different from control group (P < .05).

  3. c Significantly different from STZ-resistant group (P < .05).

A1–A36.4 ± 0.07288 ± 1439.5 ± 1.415.0 ± 3.523.9 ± 0.6
A4, A56.5 ± 0.07236 ± 834.1 ± 1.511.9 ± 3.525 ± 0.6
B1, B26.6 ± 0.15225 ± 10184.3 ± 8.1b,c140.5 ± 4.2b,c44.8 ± 1.4b,c
Table 3. . Comparison of physiological parameters (x̄ ±SE) in control and diabetic rats at time of second metabolic control, and at time of sacrifice
 Second Metabolic ControlSacrifice
GroupaAge, wkWater Ingestion, mLFood, gDiuresis, mLUrine Glucose, mg/dAge, wkWeight, gTestosterone, ng/mLNonfasting Glucose, mg/dLFructosamine, μmol/L
  1. a A1–A3, vehicle-injected control rats; A4 and A5, streptozotocine (STZ)-resistant control rats; B1 and B2, STZ-injected diabetic rats. Subgroups a and b depend on STZ dose of 60 (a) or 45 (b) mg/kg of body weight.

  2. b Significantly different from control group (P < .05).

  3. c Significantly different from subgroup with STZ dose of 45 mg/kg (P < .05).

  4. d Significantly different from STZ-resistant group (P < .05).

A112.439.7 ± 622.5 ± 2.216.5 ± 3.13 ± 113.4489 ± 162.90 ± 0.60347 ± 26167 ± 16
A225.441.1 ± 222.5 ± 1.314.3 ± 3.62 ± 126.4527 ± 364.82 ± 0.66365 ± 35250 ± 98
A365.435.6 ± 222.7 ± 1.319.6 ± 0.85 ± 166.4>6003.73 ± 0.34335 ± 25204 ± 25
A412.532.6 ± 223.8 ± 113.6 ± 0.94 ± 013.5411 ± 91.11 ± 0.3b303 ± 22126 ± 3
A4a 25.5 ± 3c18.9 ± 1.2c10.5 ± 0.6c3 ± 1 379 ± 210.45 ± 0.13c316 ± 33115 ± 6c
A4b 34.6 ± 225.3 ± 114.5 ± 14 ± 0 420 ± 91.31 ± 0.36300 ± 27129 ± 3
A525.536.6 ± 123.8 ± 19 ± 0.99 ± 726.5540 ± 164.07 ± 0.44254 ± 48205 ± 13
B112.6216 ± 10b,d51.8 ± 1.6b,d164.8 ± 14b,d12 488 ± 1551b,d13.6278 ± 17b,d0.25 ± 0.13b,d758 ± 43b,d244 ± 10b,d
B1a 229 ± 552.6 ± 1.4183.4 ± 15c15 636 ± 1532c 238 ± 9c0.038 ± 0.006835 ± 40c259 ± 10c
B1b 181 ± 3349.9 ± 4.9116.9 ± 224380 ± 1451 383 ± 290.73 ± 0.43583 ± 77210 ± 20
B225.6233 ± 7b,d51.8 ± 7b,d150 ± 17b,d11 677 ± 491b,d26.6321 ± 21b,d2.28 ± 0.26b,d914 ± 34b,d319 ± 10d
Figure 1.

. Effect of streptozotocine (STZ) dose on serum testosterone: groups A4a and A4b are STZ-resistant control rats depending on STZ dose, 60 or 45 mg/kg of weight. Groups B1a and B1b are diabetic rats injected with STZ at 60 or 45 mg/kg of weight. Dashed lines join groups with significant differences between them.

The levels of glucose and fructosamine were higher in animals with more severe diabetes (group B1a vs B1b, P = .004 and P = .019, respectively; Table 3).

Diabetic animals exhibited around a 20% decrease in testicular raw weight at 6W (not shown when expressed per unit of body weight), both when compared with STZ-resistant (P < .01) and vehicle-injected rats (P < .001). However, this effect was found to be transient because no significant differences were shown at 20W. The higher dose of STZ was associated with an increase in testicular weight in STZ-resistant rats (P = .015), whereas the opposite effect was seen in diabetic rats (P = .003). The apparent increase in testicular weight was not maintained when the comparison was made with the parenchyma weight (A4a vs A4b, P = .142). Diabetic animals showed decreased epididymal weight at both times, this effect being correlated with the severity of diabetes. STZ nondiabetic rats exhibited decreased epididymal weight at 6W (P = .002) and unaffected testicular weight when the analysis was performed with the A4a and A4b groups together. Table 4 summarizes testicular and epididymal weights and shows the comparison between the different groups. Both diabetes and STZ produced a significant decrease at 6W, whereas with diabetes, the effect lasted up to 20W. Diabetic rats that received a higher dose of STZ had a significantly lower epididymal weight (P < .001), although this did not occur in STZ-resistant animals (data not shown).

Table 4. . Physiological parameters (x̄ ±SE) for experimental groups at sacrifice
GroupaTesticle Weight, gParenchyma Weight, gEpididymis Weight, gSeminiferous Tubule Diameter, μmSperm Quantity, mgSperm Count, ×106/mgMotility c + b, %b
  1. a A1–A3, vehicle-injected control rats; A4 and A5, streptozotocine (STZ)-resistant control rats; B1 and B2, STZ-injected diabetic rats. Subgroups a and b depend on STZ dose of 60 (a) or 45 (b) mg/kg of body weight.

  2. b b, No linear motility; c, no propulsive motility.

  3. c Significantly different from control group (P < .05).

  4. d Significantly different from subgroup with STZ dose of 45 mg/kg (P < .05).

  5. e Significantly different from STZ-resistant group (P < .05).

A12.08 ± 0.071.87 ± 0.070.64 ± 0.02283.4 ± 3.948 ± 42.30 ± 0.3127 ± 5
A21.99 ± 0.071.79 ± 0.060.70 ± 0.02297.8 ± 3.067 ± 42.03 ± 0.1831 ± 1
A32.22 ± 0.131.97 ± 0.060.77 ± 0.03 71 ± 51.69 ± 0.1322 ± 5
A41.89 ± 0.061.68 ± 0.060.51 ± 0.02c286.6 ± 4.449 ± 80.33 ± 0.01c 
A4a2.18 ± 0.04d1.83 ± 0.110.49 ± 0.04 28 ± 4d0.38 ± 0.04 
A4b1.80 ± 0.051.62 ± 0.060.52 ± 0.06 55 ± 100.32 ± 0.02 
A52.03 ± 0.051.88 ± 0.060.69 ± 0.01302.5 ± 4.061 ± 31.97 ± 0.2121 ± 5
B11.73 ± 0.05ce1.56 ± 0.050.35 ± 0.02ce261.8 ± 5.1ce17 ± 5ce0.34 ± 0.03c 
B1a1.65 ± 0.05d1.49 ± 0.05d0.30 ± 0.03d 8 ± 10.35 ± 0.03 
B1b1.90 ± 0.071.71 ± 0.100.48 ± 0.02 40 ± 140.30 ± 0.03 
B21.85 ± 0.041.67 ± 0.040.53 ± 0.02ce309.8 ± 3.730 ± 4ce1.99 ± 0.1232 ± 4

On the other hand, diabetic rats did not, in practice, show structural testicular defects: their seminiferous tubules showed all the distinct developmental stages of spermatogenesis in lumen. We analyzed spermatogenesis in 3 control and 3 diabetes medium–long term cases in detail without finding differences. Therefore, no in-depth structural analysis was performed on the rest of the cases because no further abnormalities were found. It was apparent that interstitial tissue space, basal membrane thickness, vessels injuries, or Leydig cells did not change and that number of empty seminiferous tubules did not increase (see Figure 2, micrographs 1 and 4). Sperm density and seminiferous tubule diameter was reduced in diabetic rats sacrificed at 6W when compared with controls (P < .05), and correlation in degree of reduction between tubular diameter and testicular weight was positive (P < .01). Despite the epididymal weight decrease, sperm, as seen in Figure 2 (micrograph 3), filled the epididymis.

Figure 2.

. (Micrographs 1–6) H&E light microscopic images of testicles from diabetic rat tubules; efferent cones and epididymis at short term (1–3) and long term (4–6). No abnormal basal membrane thickening or empty seminiferous tubules are observed. There is no evidence of vascular change in the interstitial space, and abnormal Leydig/germinal cells are not identified. The epididymis shows an adequate number of spermatozoa.

Diabetic rats presented a decrease in seminal volume and sperm density with no change in semen motility (Table 4). Conversely, teratospermia in diabetic mice increased, mainly head abnormalities (Figure 3, micrographs 7 and 8), when compared with the whole control group (P < .001). But this effect was less evident when compared with its age-matched vehicle-injected group (P = .031), as seen in Figure 4.

Figure 3.

. (Micrographs 7, 8) Light microscopic image of nonstained sperm preparation from diabetic rats under ×100 magnification (left). Diff-Quick–stained sperm preparation from diabetic rats under ×1000 magnification (right). Note the common feature of rat hook-shaped sperm heads. A detached sperm head is observed in the image (arrow). Sperm in rats is characterized by a hook-shaped head and long tails that allow sperm to group and move faster. The fragile sperm in rats could explain the detached heads observed. However, a higher proportion of detached sperm heads were seen in the diabetic group compared with nondiabetic animals.

Figure 4.

. Proportion of normal forms. A1–A3 vehicle-injected controls rats. A5 streptozotocine (STZ)-resistant control rats. B2-injected STZ diabetic rats. There were significant differences between diabetic rats compared with grouped controls (P < .001). When age-matched, vehicle-injected rats were compared, the difference was significant as well (P = .031).

l-Carnitine levels did not vary (Figure 5). Nondiabetic STZ-injected rats had oligozoospermia at short term (P = .007). Normal sperm density was observed at medium-long term (Table 4).

Figure 5.

. Distribution of l-carnitine concentration among the different groups. A1 and A2 vehicle-injected controls rats. A5 streptozotocine (STZ)-resistant control rats. B2-injected STZ diabetic rats. Differences were not significant.

As an effect of age, epididymal weight (P = .033) increased significantly, but no significant changes were shown in serum testosterone, testicular weight, sperm concentration, percent motility, or morphology.


The experimental model of STZ-induced diabetes has been used frequently in the literature, describing changes in the male reproductive system as part of the disease. Authors postulate that diabetes produces changes in the male reproductive system, excluding a possible direct toxic effect of STZ or one of its metabolites, although nitrosourea and its compounds are known to have cytotoxic and carcinogenic properties (Anderson et al, 1987; Bolzan and Bianchi, 2002; Escrig et al, 2002). We studied age-matched, nondiabetic, STZ-injected rats to discard the possible interference of STZ.

We found higher serum glucose levels in all groups because the sacrifice was performed on nonfasting animals. Diabetic animals showed further elevation of serum glucose, another characteristic sign of the disease.

Regarding the effect of diabetes on male fertility, we found lower serum testosterone levels in diabetic rats when compared with nondiabetic, vehicle-injected and STZ-resistant controls. This effect was permanent for more than 20 weeks of follow-up after injection. Our results concur with previous reports but differ from other papers that report changes in serum testosterone were not seen (Jackson and Hutson, 1984; Murray et al, 1988; Pitteloud et al, 2005). A decrease in testosterone production with time might be attributed to diabetes status per se; however, higher levels of testosterone in diabetic animals at 20 weeks suggest an accumulative effect of STZ in a short time period.

In diabetes, the mechanism of testosterone reduction might be a direct effect of glucose or its metabolites, of defective gonadotropins, or of resistance to these hormones. The fact that the measurement of total testosterone by chemiluminescence assay depends on the levels of albumin and other binding proteins should be taken into account when analyzing serum testosterone in diabetics. Because no studies exist concerning free testosterone in experimental diabetes, controlled studies will be necessary to evaluate free testosterone in diabetic animals compared with nondiabetic animals.

Additionally, our experiments demonstrated a decrease in testicular weight accompanied by an initial tubule diameter reduction that was not evident over time. We concurred in this matter with Jackson and Hutson (1984) and Ford and Hamilton (1984), but differed from Oksanen (1975), who found more evident decreased testicular weight over time. Other authors found an increase in testicular weight at short term (of follow-up 1 month after onset of diabetes) or a decrease in testicular weight at long term (follow-up 6 months after onset of diabetes) (Anderson and Thliveris, 1987).

The reduction in tubular diameters was demonstrated by several assays (Anderson and Thliveris, 1986, 1987). This reduction was generally accompanied by an increase in the number of empty testicular tubules and in vascular density, results not supported by our findings.

We agree with Soudamani et al (2005) that the epididymal weight reduction was dependent on diabetes. In addition, we observed that the epididymal lumen was full of spermatozoa, as shown in micrographs 3 and 6, a result that differs from other authors who did not find spermatozoa (Soudamani et al, 2005).

Data provided by this study indicate that diabetes affects mainly the epididymal reserve of semen and promotes a teratozoospermic effect without changes in motility or l-carnitine concentration. This is the first description of impairment in sperm morphology in experimental diabetic rats, but it has been found in diabetic men (Bartak, 1979). In all groups, we have found a lower sperm motility than other studies (Cooke et al, 1991). The differences could be due to the spermatozoa being stored in extratesticular ducts in an immotile state and their motility activated by dilution once released from the ducts. Differences in composition of dilution medium—that is, changes of Ca2+, K+, Na+, and H+ concentrations (Wade et al, 2003)—could be the reason for disagreement between our results and those of others.

l-Carnitine is an essential molecule involved in mitochondrial metabolism, controlling the transport of acetyl and acyl groups across the mitochondrial inner membrane. Carnitine and acetylated carnitine (l-acetylcarnitine) are found in high concentrations in the epididymis, where they also act as antioxidants, protecting spermatozoa against damage caused by reactive oxygen species (Agarwal and Said, 2004). In this study, we demonstrated the absence of motility impairment in the diabetic or STZ-resistant rats, which could be related to the similarity in levels of l-carnitine among the different groups studied.

Regarding the possible effect of STZ on male fertility testosterone production, some previous studies concluded that there are no significant differences between serum testosterone in STZ-resistant animals and controls, although up to a 50% reduction in testosterone levels has been observed (Anderson et al, 1987).

Interestingly, we found transitory dose-related lower serum testosterone in STZ-resistant controls compared with vehicle-injected rats.

STZ injection by itself produces no permanent changes in epididymal weight and sperm count.

In this paper, we have demonstrated the presence of diabetes-dependent effects on male reproductive capability. We defined “diabetic testicular dysfunction” as the direct effect of diabetes on testicular function seen in our study. Thus, our data suggest transient (testicular weight) and permanent (serum testosterone, epididymal weight, and teratozoospermia) testicular dysfunction, depending on the degree and duration of the disease. Moreover, this is the first study that shows no permanent changes in semen quality and serum testosterone produced by STZ injection. This issue should be clarified in future studies.


The skilful technical assistance of Dr M. D. García-Olmo is gratefully acknowledged. We also thank Dr M. Belilty for her help in preparing the final edition of the text.


  1. This work was supported by a grant from Junta de Comunidades de Castilla La Mancha, project 98197.