A cytogenetic investigation of the effects of cryopreservation on human sperm.

The effects of cryopreservation on the frequency and type of chromosome abnormalities in human sperm have been investigated for the first time. With a technique which enables direct visualization of human sperm chromosomes following in vitro penetration of hamster oocytes, sperm samples from 13 normal men were examined before and after being frozen in liquid nitrogen. The overall abnormality frequencies of 17.8% for fresh semen and 13.4% for previously frozen semen were not significantly different (chi 2(1) df = 3.04, p = 0.08). When specific abnormality types were analyzed, only the category of hypohaploidy was significantly different (chi 2(1) df = 6.75, p = 0.009) before (7.5%) and after (3.4%) freezing. Hypohaploidy was significantly higher than hyperhaploidy both prefreeze and postfreeze, and chromosome loss was random. Because the observed excess of hypohaploid cells may be attributable to technical artifact, the aneuploidy levels were estimated by doubling the number of hyperhaploid cells. Neither the adjusted numerical abnormality frequencies (1% prefreeze vs. 0% postfreeze) nor the overall abnormality frequencies (11.8% prefreeze vs. 10.4% postfreeze) were significantly different. The types and distributions of karyotypically abnormal sperm complements (numerical, structural, or combined) observed before and after freezing were not different. Interdonor variability in sperm chromosome abnormality frequencies and a possible donor-dependent response to cryopreservation were suggested by the data. The sex ratios were not affected by cryopreservation and did not differ significantly from the theoretical 50%. It is concluded that cryopreservation does not affect the type or frequencies of chromosome abnormalities or alter the sex ratio in human sperm.     

Comparison of chromosomal abnormalities in hamster egg and human sperm pronuclei

One thousand human sperm and hamster egg haploid karyotypes were analyzed at the pronuclear stage after in vitro penetration. The frequency of abnormalities in human sperm was 8.5%, with 5.2% aneuploidy and 3.3% structural abnormalities. The hamster egg complements had an abnormality rate of 3.8%, with 3.3% aneuploidy and 0.5% structural abnormalities. In both human and hamster complements, chromosome abnormalities were observed in all chromosome groups, demonstrating that all chromosomes are susceptible to nondisjunction, not just acrocentric or small chromosomes. There is an intriguing difference between the frequency of hyperhaploid and hypohaploid complements in human sperm and hamster eggs. In the human complements, 2.4% were hyperhaploid and 2.7% hypohaploid. This is very close to the theoretical 1 to 1 ratio expected from nondisjunction. The hamster egg complements had more hypohaploid (2.2%) than hyperhaploid (0.9%) complements, despite identical treatment. Higher rates of hypohaploidy are generally ascribed to artificial loss of chromosomes, but may in fact reflect a predisposition of oocytes to anaphase lag during meiosis. The frequency of abnormalities (both numerical and structural) is higher in human complements than in hamster. This may reflect an innate propensity for meiotic chromosome abnormalities in humans or may result from greater exposure of humans to mutagenic agents.     

Variation in the frequency and type of sperm chromosomal abnormalities among normal men

Summary The chromosomal constitution of 1582 human sperm from 30 normal men of proven fertility was investigated after sperm penetration of hamster eggs. A minimum of 30 sperm chromosome complements were analysed per donor so that the distribution and variation in the frequency and type of sperm chromosomal abnormalities could be assessed. The mean frequency of sperm chromosomal abnormalities in individual men was 10.4% (±6.0%) with a range of 0–24.7%. For numerical abnormalities the mean was 4.7% (±2.9%) with a range of 0–10% and for structural abnormalities the mean was 6.2% (±6.0%) with a range of 0–23.1%. The 95% confidence intervals for the mean of an individual male were 0–10.5% for numerical abnormalities, 0–18.2% for structural abnormalities, and 0–22.4% for total abnormalities. There was a significant excess of hypohaploid complements compared with hyperhaploid complements. Since hypohaploid complements could be caused by technical artefact, a conservative estimate of aneuploidy was obtained by doubling the frequency of hyperhaploid sperm, yielding an estimate of 2.4% aneuploidy. The proportion of X-bearing (53%) and Y-bearing (47%) sperm did not differ significantly. These results were compared to the other two large studies of sperm chromosome complements from normal men.     

The effect of age on the frequency of sperm chromosomal abnormalities in normal men.

It has been suggested that advanced paternal age (independent of maternal age) is associated with an increased incidence of trisomy. However, studies of human liveborn offspring and of data from prenatal diagnosis have yielded conflicting results. To investigate this possible paternal age effect, we have studied sperm chromosome complements from 30 normal men of proven fertility stratified by age, with five males in each of six age categories (20-24, 25-29, 30-34, 35-39, 40-44, and 45+ years). Sperm chromosome complements were visualized after penetration of golden-hamster oocytes. A minimum of 30 complements were analyzed for each male. The analysis was performed blindly, without knowledge of the donor's age. The mean frequency of sperm chromosomal abnormalities in the individual men was 10.4% with means of 4.7% for numerical abnormalities and 6.2% for structural abnormalities. There was no relationship between age and the frequency of numerical abnormalities in sperm. Since there was a significant difference between the frequency of hyperhaploid and hypohaploid complements, these two types of numerical abnormalities were analyzed separately. There was no correlation between the frequency of hypohaploid complements and age. There was a significant negative correlation between age and the frequency of hyperhaploid complements. For structural abnormalities, there was a highly significant positive correlation with age. Thus, our results do not support the hypothesis of an increased risk of trisomy with paternal age.     

Cytogenetic analysis of Q-banded pronuclear chromosomes in fertilized Syrian hamster eggs

The frequency and type of chromosome abnormalities were analyzed in 917 female pronuclei in Syrian hamster eggs fertilized by human sperm. Analysis at this stage allows detection of errors which have occurred during meiosis I and II. The chromosomes were Q-banded to identify individual chromosomes and detect subtle alterations. Thirty-three (3.6%) of the hamster egg complements were abnormal: 19 (2.1%) were hypohaploid, seven (0.76%) were hyperhaploid, two (0.2%) had double aneuploidy, and five (0.5%) had a structural chromosome abnormality. Since there were significantly more hypohaploid than hyperhaploid complements, a conservative estimate of aneuploidy can be derived by doubling the frequency of hyperhaploid complements. Thus a minimal estimate of aneuploidy (single, 1.5%, and double, 0.2%) is 1.7% and a minimal estimate of the total frequency of abnormalities is 2.2%. All chromosome groups were represented among the aneuploid complements suggesting that all chromosomes are susceptible to non-disjunction.     

The frequency of aneuploidy among individual chromosomes in 6,821 human sperm chromosome complements

The human sperm/hamster egg fusion technique has been used to analyse 6,821 human sperm chromosome complements from 98 men to determine if all chromosomes are equally likely to be involved in aneuploid events or if some chromosomes are particularly susceptible to nondisjunction. The frequency of hypohaploidy and hyperhaploidy was compared among different chromosome groups and individual chromosomes. In general, hypohaploid sperm complements were more frequent than hyperhaploid complements. The distribution of chromosome loss in the hypohaploid complements indicated that significantly fewer of the large chromosomes and significantly more of the small chromosomes were lost, suggesting that technical loss predominantly affects small chromosomes. Among the autosomes, the observed frequency of hyperhaploid sperm equalled the expected frequency (assuming an equal frequency of nondisjunction for all chromosomes) for all chromosome groups. Among individual autosomes, only chromosome 9 showed an increased frequency of hyperhaploidy. The sex chromosomes also showed a significant increase in the frequency of hyperhaploidy. These results are consistent with studies of spontaneous abortions and liveborns demonstrating that aneuploidy for the sex chromosomes is caused by paternal meiotic error more commonly than aneuploidy for the autosomes.     

The chromosome constitution of 1000 human spermatozoa

Summary Chromosomal analysis of 1000 spermatozoa from 33 normal men was performed using in vitro fertilization of zonafree golden hamster eggs. The frequency of abnormal sperm complements was 8.5%: 5.2% were aneuploid and 3.3% had a structural chromosome abnormality. The frequencies of hyperhaploid (2.4%) and hypohaploid (2.7%) sperm complements were not significantly different and all chromosome groups were represented among the aneuploid complements. The majority (22/33) of structurally abnormal complements had a chromosome break. The percentages of X and Y-bearing sperm were 53.9% and 46.1%, which is significantly different from the expected one to one ratio.     

An increased frequency of human sperm chromosomal abnormalities after radiotherapy

13 cancer patients were studied before radiotherapy (RT) and at regular intervals after RT to determine the effect of RT on chromosomal abnormalities in sperm. The men were 19-47 years old and received testicular radiation doses of 0.4-5.0 Gray. Human pronuclear sperm chromosomes were analysed after penetration of zona-pellucida-free hamster eggs. Unfortunately the hamster egg penetration rates were exceedingly low, both before and after RT and this limited the number of sperm chromosome complements which could be analysed. Before RT, the frequency of abnormal sperm chromosome complements was 0% (0/9). After RT, the majority of men were azoospermic for 24 months but complements could be analysed from 4 men. In the first 12 months the frequency of abnormalities was 13% (1/8) and at 24 months it was 13% (7/55). By 36 months after RT, most men had recovered sperm production and the frequency of abnormalities in 8 men was 21% (18/86), which is significantly higher than the rate in control donors (8.5%). For individual men the range was 6-67%, and there was a significant correlation between testicular radiation dose and the frequency of sperm chromosomal abnormalities. The frequencies of both numerical and structural abnormalities were significantly increased after RT. This is the first evidence that radiation may increase the frequency of chromosomal abnormalities in human gametes.     

Chromosomal analysis of unfertilized human oocytes

Unfertilized human oocytes were obtained from women in an in-vitro fertilization programme. The women had a mean age of 29.4 years (range 24-35 years). Chromosomal complements could be analysed in 50 oocytes. Q-banding of the chromosomes facilitated identification of individual chromosomes: 34 oocytes (68%) had the normal haploid chromosomal complement, 14 complements were hypohaploid (28%), 1 complement was hyperhaploid (2%) and 2 had structural abnormalities (4%). (One oocyte had numerical and structural abnormalities). The 16 abnormal oocytes were obtained from 15 different women. A conservative estimate of aneuploidy in this sample is 4%; however, the frequency of aneuploidy may be higher if there is a predisposition to chromosome loss during oogenesis. This study provides information on the largest series of karyotyped unfertilized human oocytes published to date.     

Analysis of the chromosome complement in outbred mouse sperm fertilizing in vitro

The chromosome complements in a population of mouse sperm from random-bred ICR donors were analyzed at first-cleavage metaphase after in vitro fertilization (IVF) of oocytes from females of the same strain. The sperm were aged as donations occurred within an average of 31 days, either since last mating or at arrival at the animal facility in the case of virgin males. Of a total of 598 sperm complements studied from 22 sexually mature males aged 10–26 weeks old, there was one diploid complement (0.17%). The frequencies of hyperhaploidy and structural aberrations that were studied in 338 complements were 4.4% and 3.6%, respectively, giving an overall frequency of 8.0%. The hyperhaploid complements consisted of n + 1, n + 2, n + 3, and n + 7 counts, while the structural abnormalities were of the chromosome type and included large and small fragments and a possible translocation. This is the highest frequency of sperm chromosome abnormalities reported for mouse sperm obtained from males under physiological conditions and fertilized in vitro or in vivo. Sperm aging, strain, and/or technique differences are among the factors that may be responsible for this high frequency. Since the 8.0% frequency of hyperhaploidy and structural abnormalities is similar to the frequency reported for human sperm after IVF, the outbred murine in vitro fertilization system may be a useful model to study the origin of human sperm chromosome abnormalities.     

Effect of culture conditions and media on the frequency of chromosomal abnormalities in human sperm chromosome complements

Human sperm chromosomes can be visualized after fusion with hamster eggs. Most laboratories use one of two methods of sperm treatment for capacitation: incubation in a modified Krebs-Ringer medium (BWW) for 5-7 h at 37 degrees C or storage in a TES-Tris yolk buffer (TYB) for 24-72 h at 4 degrees C. To determine whether data from the two methods were comparable, we performed a series of controlled experiments on one normal donor in which ejaculates were split and one aliquot of sperm was capacitated in BWW for 5-7 h at 37 degrees C (fresh) and the second aliquot was capacitated in TYB for 48 h at 4 degrees C (TYB). After capacitation, the technique used to obtain human sperm chromosome complements was identical for both aliquots. Both fresh and TYB sperm were further subdivided into two groups, which were subjected to either a short (1 h) or a long (3 h) gamete coincubation in BWW. This experiment was performed to determine if the longer incubation in BWW might induce chromosomal fragile sites and breaks because of nutritional depletion of the medium. A total of 458 human sperm chromosome complements was analysed. There was no significant difference in the frequency of sperm chromosomal abnormalities or in the sex ratio in the sperm coincubated with eggs for a short (1 h) or long (3 h) time in BWW. When sperm pretreatments were compared, there was a significant increase in the frequency of total sperm chromosomal abnormalities after TYB storage compared to fresh treatment.(ABSTRACT TRUNCATED AT 250 WORDS)     

Human sperm chromosome complements after microinjection of hamster eggs

A technique was developed for microinjection of human spermatozoa into golden hamster (Mesocricetus auratus) eggs to obtain human pronuclear chromosome complements. Before microinjection the spermatozoa were treated by brief sonication or incubation in TEST-yolk buffer to reduce motility. Very few sperm chromosome complements developed after sperm treatment with sonication and the frequency of spermatozoa with structural chromosomal abnormalities was exceedingly high (91%). The majority of sperm chromosome complements analysed had multiple breaks and rearrangements. Sperm incubation in TEST-yolk buffer before microinjection provided more analysable sperm karyotypes with a significantly lower frequency of structural chromosomal abnormalities (39%, P less than 0.001). Our results therefore suggest that sonication induces structural chromosomal abnormalities in spermatozoa. Since the frequency of chromosomal abnormalities after microinjection was higher than after sperm fertilization of hamster eggs, it appears that microinjection per se may also increase the frequency of chromosomal abnormalities in spermatozoa. These results are based on small numbers and must be confirmed on larger sample sizes, but our study suggests that microinjection of spermatozoa into eggs should not be recommended for clinical use until fully evaluated.     

A comparison of chromosomal aberrations induced by in vivo radiotherapy in human sperm and lymphocytes

Chromosomal aberrations in human sperm and lymphocytes were compared before and after in vivo radiation treatment of 13 cancer patients. The times of analyses after radiotherapy (RT) were 1, 3, 12, 24, 36, 48 and 60 months. The median total radiation dose was 30 Gy and the testicular dose varied from 0.4 to 5.0 Gy. Human sperm chromosome complements were analysed after fusion with golden hamster eggs. There were no abnormalities in sperm or lymphocytes before RT. Following RT there was an increase in the frequency of numerical and structural chromosomal abnormalities in both lymphocytes and sperm. For structural abnormalities there were more rejoined lesions (dicentrics, rings) in lymphocytes and more unrejoined lesions (chromosome breaks, fragments) in sperm. After RT there was a dramatic increase in the frequency of chromosomal abnormalities in lymphocytes: at 1 mo. the frequency was 42%, at 3 mo. 25%, at 12 mo. 14%, at 24 mo. 11%, at 36 mo. 9%, at 48 mo. 7% and at 6 mo. 4%. Since the majority of men were azoospermic after RT, there is little data on sperm chromosome complements before the analyses performed at 24 mo. post-RT. At 24 mo. the frequency of abnormalities was 13%, followed by 21% at 36 mo., 12% at 48 mo. and 22% at 60 mo. Thus it appears that the frequency of lymphocyte chromosomal abnormalities had an initial marked increase after RT followed by a gradual decrease with time whereas the frequency of sperm chromosomal abnormalities was elevated when sperm production recovered and remained elevated from 24 to 60 mo. post-RT. This difference in the effect of time makes it very difficult to compare abnormality rates in lymphocytes and sperm and to use analysis of induced damage in somatic cells as surrogates for germ cells since the ratio between sperm and lymphocytes varied from 1:1 (at 24 mo. post-RT) to 5:1 (at 60 mo. post-RT).     

Effect of cryopreservation protocol on postthaw characteristics of stallion sperm

Three ejaculates from each of eight stallions were subjected to cryopreservation in a milk/egg yolk-based freezing extender or an egg yolk-based freezing extender. Semen was exposed to a fast prefreeze cooling rate (FAST; semen immediately subjected to cryopreservation) or a slow prefreeze cooling rate (SLOW; semen pre-cooled at a controlled rate for 80 min prior to cryopreservation). Postthaw semen was diluted in initial freezing medium (FM) or INRA 96 (IMV Technologies, L'Aigle, France) prior to analysis of 10 experimental end points: total motility (MOT; %), progressive motility (PMOT; %), curvilinear velocity (VCL; μm/s), linearity (LIN; %), intact acrosomal and plasma membranes (AIMI; %), intact acrosomal membranes (AI; %), intact plasma membranes (MI; %), and DNA quality. Eight of 10 experimental endpoints (MOT, PMOT, average-path velocity [VAP], mean straight-line velocity [VSL], LIN AIMI, AI, and MI) were affected by extender type, with egg yolk-based extender yielding higher values than milk/egg yolk-based extender (P < 0.05). Exposure of extended semen to a slow prefreeze cooling period resulted in increased values for six of eight endpoints (MOT, PMOT, VCL, AIMI, AI, and MI), as compared with a fast prefreeze cooling period (P < 0.05). As a postthaw diluent, INRA 96 yielded higher mean values than FM for MOT, PMOT, VCL, average-path velocity, and mean straight-line velocity (P < 0.05). Treatment group FM yielded slightly higher values than INRA 96 for LIN and MI (P < 0.05). In conclusion, a slow prefreeze cooling rate was superior to a fast prefreeze cooling rate, regardless of freezing extender used, and INRA 96 served as a satisfactory postthaw diluent prior to semen analysis.     

Cytogenetic studies in motile sperm from normal men

Summary Most studies on human sperm chromosomes from normal men involve the heterologous fertilization of zona free hamster eggs by unselected human sperm. In this work, we have performed cytogenetic studies of highly motile sperm, selected by a swim-up method. A total of 505 motile human sperm complements from three normal donors was analysed. The total frequency of sperm with chromosomal abnormalities (10.9%; 6.9% structural aberrations and 4.0% aneuploidy) and the sex ratio (50.4% X49.6% Y) were similar to those obtained from whole semen samples. Our results indicate that the selection of motile sperm does not imply chromosomal selection.     

Sequence analysis of long FMR1 arrays in the Japanese population: insights into the generation of long (CGG) n tracts

Sperm chromosome abnormalities were assessed in testicular cancer patients before and after treatment with BEP (bleomycin, etoposide, cisplatin). The frequencies of disomy for chromosomes 1, 12, X, Y and XY were assessed along with diploid frequencies and sex ratios by multicolour fluorescence in situ hybridization (FISH). For each cancer patient, a minimum of 10 000 sperm was assessed for each chromosome probe before and after chemotherapy (CT). Data was analysed “blindly” by coding the slides. A total of 161 097 sperm were analyzed, 80 445 before and 80 642 after treatment. The mean disomy frequencies were 0.11% pre-CT vs 0.06% post-CT for chromosome 1, 0.18% vs 0.15% for chromosome 12, 0.10% vs 0.9% for the X chromosome, 0.13% vs 0.10% for the Y chromosome and 0.25% vs 0.20% for XY sperm. There was no significant difference in the frequency of disomy pre-CT vs post-CT for any chromosome except that chromosome 1 demonstrated a significant decrease after CT. The “sex ratios” and frequency of diploid sperm were also not significantly different in pre- and post-CT samples with 50.2% X-bearing sperm pre-CT and 50.5% X post-CT and 0.14% diploid sperm pre-CT vs 0.15% diploid sperm post-CT. There was no significant donor heterogeneity among the cancer patients. None of the values in the cancer patients differed significantly from 10 normal control donors. Thus our study suggests that BEP chemotherapy does not increase the risk of numerical chromosomal abnormalities in human sperm. Received: 11 June 1996 / Revised: 8 August 1996     

Chromosome analysis of human sperm

Summary A modified technique has been developed for the visualization of the chromosomes in human sperm. The cytogenetic analysis of 129 G-banded human sperm metaphases of 6 normal donors showed an incidence of structural and numerical chromosome abnormalities of 7.8%. Two out of 129 spermatozoa were aneuploid (1.6%). The frequency of sperms with chromatid-type aberrations was 2.3% (3/129). Chromosome-type aberrations were found in 5 out of 129 (3.9%) spermatozoa. X to Y ratio did not differ significantly from the expected one-to-one ratio. Twenty-six sperm complements from a patient 18–20 months after testes exposure to 30 Gy were examined. A significant increase of numerical and structural chromosome abnormalities was not observed. Chromatidtype aberrations were found in two sperm complements (7.7%) and chromosome-type aberrations in one sperm complement (3.9%). The cytogenetic analysis of 15 human sperms from a cancer patient 26 months after chemotherapy showed an increased frequency of aberrant sperm complements (33.4%). One chromatid-type (6.7%), three chromosometype aberrations (20.0%) and one (6.7%) hyperploid sperm complement could be observed. The sample size is still too small to answer the question whether chemical mutagens may increase the frequency of chromosomal abnormalities in human sperm.     

Human sperm chromosome complements in chemotherapy patients and infertile men

Our studies of human sperm karyotypes and interphase sperm analyzed by fluorescence in situ hybridization (FISH) have both yielded estimates of disomy frequencies of approximately 0.1% per chromosome with an overall aneuploidy frequency in human sperm of approximately 5%–6%. However, the distribution of aneuploidy in sperm is not even, as our data from sperm karyotypes and multicolour FISH analyses both demonstrate a significant increase in the frequency of aneuploidy for chromosome 21 and the sex chromosomes. We have studied men at increased risk of sperm chromosomal abnormalities including cancer patients and infertility patients. Testicular cancer patients were studied before and 2–13 years after chemotherapy (CT) with BEP (bleomycin, etoposide, cisplatin). Sperm karyotype analysis on 788 sperm demonstrated no significant difference in the frequency of numerical or structural chromosomal abnormalities post-CT vs pre-CT. Similarly, multicolour FISH analysis for chromosomes 1, 12, XX, YY and XY in 161,097 sperm did not detect any significant differences in the frequencies of disomy before and after treatment. However, recent evidence has suggested a significant increase in the frequency of disomy and diploidy during CT. We have found that infertile men, who would be candidates for intracytoplasmic sperm injection, have an increased frequency of chromosomally abnormal sperm karyotypes. Also, FISH analysis for chromosomes 1, 12, 13, 21, XX, YY and XY in 255,613 sperm demonstrated a significant increase in chromosomes 1, 13, 21, and XY disomy in infertile men compared with control donors. Received: 4 July 1998; in revised form: 7 September 1998 / Accepted: 8 September 1998     

Cryopreservation of epididymal sperm collected postmortem in the Tasmanian devil (Sarcophilus harrisii)

Effective sperm cryopreservation protocols are limited to a small number of marsupial species. In this study, postmortem gamete rescue (PMGR) epididymal sperm samples from Tasmanian devils (N = 34) euthanized due to the fatal Devil Facial Tumor Disease were used to develop long-term sperm storage techniques for the species. Cryoprotectant toxicity associated with equilibration of sperm samples in a TEST yolk diluent (TEST; 189 mM N-Tris(hydroxymethyl)methyl-2-aminoethanesulfonic acid, 85 mM Trizma base [Tris], 11 mM glucose, 20% vol/vol egg yolk; pH 7.1, and 315.0 ± 5.0 mOsm/kg) with a final concentration of 0.06 M trehalose, or 4%, 10%, and 18% vol/vol of either glycerol or dimethyl sulfoxide (DMSO), was examined over 12 h at 15 °C. Trehalose supplementation resulted in an immediate decline (P < 0.05) of total motility. After 12 h, total motility was reduced (P < 0.05) in treatments containing 18% glycerol, and 10% and 18% dimethyl sulfoxide. The effects of final glycerol concentration (4% and 10%), glycerol equilibration duration (10 min 1 h, or 3 h) prefreeze, freezing rate and the addition of 0.10 M lactose or a combination of 0.10 M lactose and 0.11 M raffinose were assessed during three experiments on the cryopreservation of postmortem gamete rescue samples in TEST. In all experiments, motility and viability were reduced (P < 0.01 postthaw). Samples cryopreserved in TEST supplemented with lactose or lactose with raffinose using a fast freezing rate (−8 °C/min from 4 to −40 °C, then −65 °C/min until −165 °C) produced the highest (P < 0.05) postthaw motility (18.6 ± 5.5% and 16.9 ± 8.5%, respectively), which represented 35% to 48% retention of prefreeze motility. These results apparently were the best postthaw results of dasyurid sperm reported to date and will help lay the foundations for developing assisted reproductive technologies for marsupial species.     

Sperm chromosome complements in a 47,XYY man

Summary Human sperm chromosomes from a 47,XYY male were examined using the direct method of sperm chromosome analysis with two modifications in the semen processing. A total of 75 sperm complements was karyotyped and all of these contained one sex chromosome. The percentages of X-and Y-bearing sperm were 53% and 47%, respectively. There were 10 sperm with autosomal chromosomal abnormalities. The frequencies of numerical (4.0%), structural (10.6%), and total (13.3%) abnormalities were not significantly different from the frequencies observed in normal donors in our laboratory. Our results do not support the suggestion that XYY males have an increased risk of aneuploid progeny as a result of secondary non-disjunction or interchromosomal effects. They do support the hypothesis that one Y chromosome is eliminated in the germ cells of XYY males. However since our study provides the first information on sperm chromosomes in an XYY male, further studies on other XYY men are required.