spermatid with intracytoplasmic sperm injection (ICSI) was first reported by Vanderzwalmen .... was performed 36 h later under guidance of vaginal ultrasound.
Human Reproduction vol.13 no.1 pp.104–110, 1998
Multiple pregnancies obtained by testicular spermatid injection in combination with intracytoplasmic sperm injection
S.Kahraman1,3, G.Polat1, M.Samli2, E.So¨zen1, ¨ zbic¸er1 ¨ zgu¨n1, K.Dirican1 and T.O O.D.O 1Assisted
Reproductive Techniques and Reproductive Endocrinology Unit and 2Urology Department, Sevgi Hospital, Ankara, Turkey
3To
whom correspondence should be addressed at: Department of Reproductive Endocrinology and Infertility, Tunus Caddesi, No: 28, Kavaklidere, Ankara, 06680, Turkey
Recent studies have shown that the injection of spermatid cells into the human oocyte can result in normal fertilization, embryo development and even delivery of live, healthy offspring. In our study, 23 azoospermic cases with severe spermatogenetic defects in their testicular biopsy are presented. The serum follicle stimulating hormone (FSH) concentrations and histopathological results of these males have been documented and compared in terms of fertilization and embryo development. The mean FSH value of the azoospermic males was 15.8 K 2.3 mIU/l, ranging from 1.6 to 39 mIU/l. Elongated spermatids were used in three cases only, as these more mature forms were mostly present in the testicular sample. In the remaining 20 cases, only round spermatids were found for use in intracytoplasmic sperm injection (ICSI). The fertilization rate with two pronuclei was 31.3%. The fertilization rate was found to be as high as 71% in three patients in the elongating and elongated spermatids group and as low as 25.6% in the round spermatid group. A few immature, non-motile spermatozoa were seen in only two cases from the elongated spermatid group. However, in the remaining cases, no spermatozoa were observed. The number of pronuclear (PN) arrest was quite high when only round spermatids were used (36.1%). Total fertilization failure was observed in two cases from the round spermatid group with Sertoli cell only and germ cell aplasia. A total of three pregnancies was achieved in 23 cases (13.0%), two from the elongated spermatid group and one from the round spermatid group. One biochemical pregnancy with a round spermatid resulted in an early spontaneous abortion and surprisingly, the remaining pregnancies were achieved with elongated spermatids resulting in multiple pregnancies. One twin and one triplet pregnancy were established following four embryo transfers in each patient. The twin pregnancy resulted in a live birth with two healthy babies; unfortunately, the triplet pregnancy ended in an abortion at 11 weeks. The use of testicular spermatids in the treatment of non-obstructive azoospermia may give hope by offering a novel treatment model. In cases with very severe spermatogenetic defect, even multiple pregnancies can be 104
achieved with elongated spermatid cells by yielding a high implantation rate. However, the efficiency of round spermatids in achieving fertilization and pregnancy was disappointing. Key words: intracytoplasmic sperm injection/multiple pregnancy/testicular spermatid injection
Introduction Although it is understood in animal studies that fertilization and pregnancy can be achieved using round spermatids and even with secondary spermatocytes, implementation of this technique in human beings is a very novel concept (Kimura et al., 1995a,b). The ability of spermatid cells to fertilize a mature oocyte with elongated or round spermatid injection (ELSI, ROSI) procedures and delivery of live healthy offspring have been shown in several animal studies in both mouse and rabbit models (Ogura, 1993, 1994; Sofikitis, 1994a,b; Ogura and Yanagimachi, 1995). Successful fertilization of a human oocyte by a late stage spermatid with intracytoplasmic sperm injection (ICSI) was first reported by Vanderzwalmen (1995). The first two live births after intrauterine transfer of embryos obtained by injection of spermatids from semen were reported by Tesarik et al. (1996). Fishel et al. (1996) reported the first live birth after intrauterine transfer of embryos accomplished by utilization of testicular elongated spermatids. These pregnancies introduced the concept of using spermatids to alleviate severe cases of male infertility. Sofikitis (1994b) was the first to report a pregnancy achieved with an injection of round spermatids into rabbit oocytes resulting in a complete gestation. Spermatogenetic arrest is an interruption of the complex process of germ cell differentiation leading to the formation of spermatozoa, and it may result in either oligospermia (partial arrest) or azoospermia (total arrest). Various factors play a role in the aetiology of spermatogenetic arrest. Varicocele and bilateral cryptorchidism are the most common causes of nonobstructive azoospermia. Other factors such as infections and hormonal disturbances (gonadotrophin insufficiency or late onset adrenogenital syndrome, etc.) should be scrutinized in cases of incomplete arrest, as such conditions are treatable. Some pre-testicular causes may be due to hepatic and renal dysfunction, exposure to heat and toxic substances (chemotherapy, alcohol, some antibiotics) and nutritional disturbances (zinc, vitamin A, etc.). In cases of complete arrest, unless these factors are present, the arrest is usually at the primary © European Society for Human Reproduction and Embryology
Multiple pregnancies and spermatid injection
spermatocyte level and has a genetic origin due to chromosomal anomalies either in somatic cells or in germ cells (Martin du Pan et al., 1993). In this study, we report the results attained by intracytoplasmic sperm injection of round and elongated testicular spermatids in our centre in which micromanipulation techniques have been widely used. We aim to present our experience on identification, recovery and utilization of testicular spermatids and most importantly to emphasize that multiple pregnancies can be achieved by using these round cells. Materials and methods This study was carried out between March 1996 and September 1996 at Sevgi Hospital Assisted Reproductive Technologies and Reproductive Endocrinology Unit. A total of 23 couples was included in the study. On demonstration of azoospermia in two semen samples given at least 3–4 weeks apart after a sexual abstinence of 3–5 days, more advanced examinations were performed. After physical and genital examinations, ultrasonographic evaluation of testicular volume and hormonal studies were performed. Previous and simultaneous biopsy results were evaluated. Genetic karyotype determinations were executed in three males, who were found to have a normal karyotype. However, no pregnancy was achieved in these couples. Spermatid recovery Testicular tissue samples were obtained by open biopsy under local anaesthesia on the same day that the oocytes were retrieved. If an adequate number of spermatozoa or spermatids were recovered then oocyte retrieval was performed. Otherwise, the procedure was cancelled. The couples were informed that the number of pregnancies achieved by ROSI/ELSI procedures to date is limited, and they signed a consent form. The amount of resected testicular tissue varied due to the number of biopsies executed from a minimum of two to a maximum of eight. The testicular tissue obtained was placed in a Falcon 2003 tube (Becton Dickinson Labware, Lincoln Park, NJ, USA) containing 2 ml of Gamete 100 medium (Scandinavian IVF Science AB, Go¨teberg, Sweden) and progressively divided into small segments. It was gently crushed between two pincettes in a petri dish (Falcon Primaria 3802; Becton Dickinson) containing Gamete 100 to obtain a suspension. The suspension was then transferred into a Falcon tube 2003 and rotated for 50–60 s with a vortex mixer before being examined at 3200–400 magnification under a Hoffman modulator mounted inverted microscope, in a Falcon 3802 Petri dish containing 5 ml medium. If spermatid cells (early or occasionally late) were observed, the suspension was centrifuged on a two-layer Percoll gradient (Sigma Chemical Co., St Louis, MO, USA) (70– 90% was used depending on the concentration of spermatids). The testicular tissue solution was kept in an incubator (5% CO2 in air) at 37°C until the ICSI procedure. Discrimination of spermatids in the testicular sample was accomplished by differentiation of these round cells from the other round cells. This is a problematic process that requires expertise and proficiency. In the testicular sample, elements of germ cell series were distinguished from other round cells such as Sertoli cells, spermatogonia, primary and secondary spermatocytes, erythrocytes, leukocytes, monocytes and lymphocytes. Spermatids were recognized as mostly being smaller but occasionally slightly bigger than the erythrocytes, 8–9 µm in diameter with a round nucleus and a regular zone of cytoplasm surrounding the nucleus (Figure 1). A developing acrosomal granule could be seen in some round spermatids as a bright spot adjacent to the spermatid nucleus. Round spermatids have a circular appearance, unlike other flat shaped germ
Figure 1. A round spermatid indicated by two large arrows next to a spermatozooon (small arrow). A dark spot (developing acrosomal granule) is clearly seen in the periphery of the round cell.
Figure 2. A mature spermatid contains a residual body around the mid-piece and is generally observed between the Sertoli cells. cells such as Sertoli cells, and are observed as three-dimensional entities. They can easily be distinguished from a Sertoli cell nucleus by their size and shape. Furthermore, a Sertoli cell nucleus appears to be flat. They can only be confused with lymphocytes as far as their size is concerned, as lymphocytes have a larger nucleocytoplasmic ratio by virtue of larger nuclei. Spermatids can easily be distinguished from primary and secondary spermatocytes by their size. A primary spermatocyte is twice as big as a spermatid. A round spermatid is a pure round cell without any tail. However, late or elongated spermatids are cells that exhibit a slight tail closer to the stage of spermatozoa. Such late spermatids are occasionally encountered by virtue of their shorter life cycle. An elongating spermatid is a transitional form between a round and an elongated spermatid. An oval shape with a visible head but with no well defined tail was classified in the ELSI group. Elongated and elongating spermatids are discriminated according to the length of their tail. An elongated spermatid has a more defined tail in comparison to the elongating spermatid. It is usually difficult to discriminate a mature spermatid from a mature spermatozoon. Mature spermatids contain a residual body around the mid-piece and are generally observed to be immotile. They very rarely have twitching tail motility between the Sertoli cells (Figure 2). However, the discrimination criteria can be complicated.
105
S.Kahraman et al.
Ovarian stimulation Gonadotrophin releasing hormone analogue (Suprefact®; Hoechst, Frankfurt, Germany) was given as a nasal spray, starting in the luteal phase, and was continued until the third day of menses, which was regarded as a sign of sufficient down-regulation. Follicular development was then stimulated with an injection of follicle stimulating hormone (FSH) (Metrodin®; Serono, Rome, Italy) and human menopausal gonadotrophin (HMG) (Humegon®; Organon, Oss, The Netherlands). Luteinization was induced by 10 000 IU HCG (Pregnyl®; Organon, Istanbul, Turkey) injection and oocyte aspiration was performed 36 h later under guidance of vaginal ultrasound. Oocyte preparation After oocyte retrieval, the cumulus cells and the corona radiata were removed by a brief exposure to HEPES buffered Earle’s medium containing 80 IU/ml hyaluronidase (type VIII, specific activity 320 IU/mg, H 3757®; Sigma Chemical Co.). A cumulus mass was removed completely by aspiration of the cell complex in and out of a handdrawn glass Pasteur pipette. The oocytes were rinsed several times in IVF 50 and were assessed at 3200 magnification under an inverted microscope to determine their stage of nuclear maturity and polar body existence. Oocytes were incubated in IVF 50 at 37°C in an atmosphere of 5% CO2 in air covered by paraffin oil (Ovoil 150®; Scandinavian IVF Science). ICSI (ELSI, ROSI) procedure The spermatids were aspirated from the pellet or Percoll suspension with a 10 µm pipette and transferred into the injection dish (Falcon 1006®) containing a droplet of 5 µl Gamete 100 medium and kept in the incubator until the ICSI procedure. They were then aspirated one by one from the droplet and transferred into the polyvinylpyrrolidone droplet (ICSI 100®; Scandinavian IVF Science). A 7 µm micropipette was used for ICSI. In the case of late spermatids, no mid-piece could be observed and the beginning of the tail was very rigid when touched. The head of the round spermatid is compressed and takes on an oval shape when entering the pipette during the ICSI procedure as previously described by Vanderzwalmen et al. (1995). The spermatid was injected into the oocyte after vigorous aspiration of ooplasm to facilitate activation. After intracytoplasmic injection, spermatids can be followed clearly within the ooplasm due to their large size. Injected oocytes were placed into individual droplets in a Petri dish. Further evaluation of injected oocytes and establishment of pregnancy Fertilization was checked at 16, 18 and 24 h after injection. Fertilization was assessed as normal when two clearly distinct pronuclei containing nuclei were present. The state of embryo cleavage and quality was assessed after a further 24 h of in-vitro culture. The embryos were evaluated according to the blastomere size equality and the relative proportion of anucleate fragments. Figure 3 represents middle and low quality embryos attained by spermatid injection. One pregnancy was achieved with these mediocre quality embryos with a round spermatid injection (ROSI); unfortunately, this pregnancy ended in a spontaneous abortion. A maximum of four embryos was transferred, and assisted hatching was applied to each embryo using acidified Tyrode’s solution. All patients underwent an initial test for serum β-HCG on day 12 after embryo transfer. Clinical pregnancy was diagnosed by ultrasonography at 7 weeks of gestation. Statistical analysis Fisher’s exact test was used to compare FSH concentrations with fertilization and pregnancy. No statistical evaluation of the variation
106
Figure 3. Low (right) and medium (left) quality embryos obtained with round spermatids demonstrating uneven blastomeres and fragmentation.
Table I. Distribution of follicle stimulating hormone (FSH) concentrations, testicular biopsy results and infertility aetiology of azoospermic males Patient
FSH (mIU/ml)
Results of testicular biopsy Aetiology
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 Mean Range
3.7 39 7 12 8.2 1.6 11 28 23 29 6.4 18.4 2.2 29.1 10.1 10.4 20.3 5.3 7.2 38.6 19.4 27.5 7 15.8 6 2.3 1.6–39
MA MA, HS MA, HS Complete MA Complete MA Complete MA GCA, MA Complete MA GCA GCA Complete MA MA, HS MA, HS MA, HS MA, HS SCO SCO Complete MA HS GCA HS Complete MA SCO
PTF Mumps PTF PTF PTF PTF PTF PTF Cryptorchidism PTF PTF PTF PTF PTF PTF PTF PTF PTF PTF PTF PTF PTF PTF
MA 5 maturation arrest, HS 5 hypospermatogenesis, GCA 5 germ cell aplasia, SCO 5 Sertoli cell only syndrome, PTF 5 primary testicular failure.
in fertilization rates of elongated and round spermatids was possible because of the low numbers involved.
Results The mean ages of females and azoospermic males were 30.3 6 1.8 and 30.6 6 0.9 years respectively. The mean duration of infertility was 8.8 6 0.9 years. The mean testicular volume was found to be 8.1 6 0.6 ml. Serum FSH levels, biopsy results and aetiology distribution of 23 azoospermic cases are presented in Table I. In most cases (n 5 21) primary testicular failure was diagnosed. Mumps orchiditis was diagnosed in one
Multiple pregnancies and spermatid injection
Table II. Number of oocytes retrieved and injected, fertilization rates and pregnancy outcomes in couples treated with ROSI and ELSI procedures Patient no.
No. of oocytes injected
Technique
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 Total
14 8 3 16 13 13 5 10 5 7 6 10 14 9 10 5 6 18 9 11 11 18 9 230
ROSI ROSI ROSI ROSI ROSI ROSI ROSI ROSI ROSI ROSI ROSI ROSI ROSI ROSI ROSI ROSI ROSI ROSI ELSI ELSI ELSI ROSI ROSI
No. of fertilized No. of 2 PN No. of embryos oocytes with arrested oocytes transferred 2 PN 1 2 1 3 5 5 – 2 2 1 2 1 3 2 2 2 3 8 8 6 8 6 – 72 (31.3%)
– – 1 3 3 3 – – 1 1 – – 2 – 1 – 2 1 1 2 3 2 – 26 (36.1%)
1 2 – 3 3 3 – 2 1 – 2 1 1 2 1 2 2 3 4 4 3 3 – 43 (59.7%)
Outcome
– – – – – – – Pregnant* – – – – – – – – – – Pregnant** – Pregnant*** – – 3 (13%)
ROSI 5 round spermatid injection; ELSI 5 elongated spermatid injection; PN 5 pronuclear. *Biochemical pregnancy. **A triplet pregnancy which ended in abortion at 11 weeks of gestation. ***A twin pregnancy which resulted in healthy twins.
case and cryptorchidism in another. In the testicular biopsy samples, complete maturation arrest (n 5 7), incomplete maturation arrest (n 5 7), Sertoli cell only (SCO) (n 5 3), germinal cell aplasia (n 5 4), hypospermatogenesis (n 5 2) diagnoses were histopathologically diagnosed. In six cases, maturation arrest was found with various degrees of hypospermatogenesis. However, the rate of hypospermatogenesis was mostly ,5% in most cases. It was interesting to observe ~70% hypospermatogenesis in two cases biopsied simultaneously with multiple testicular tissue samples for ICSI, where no spermatozoa were identified. In 23 patients, spermatids were utilized for microinjection. In three of these cases elongated spermatids were used and a few immotile spermatozoa were identified in the basal sample. However, they were not found during the sperm collection period. Table II summarizes the distribution of cases according to the number of retrieved and microinjected oocytes, normal fertilization rate with 2 PN, and abnormal fertilization with two different size or triple PN and outcome of cycles. The total number of injected oocytes was 230. The fertilization rate with two normally developed pronuclei (2 PN) was 31.3%. A total of 43 embryos were transferred. In two cases where round spermatids were injected, no embryo was transferred, as fertilization was achieved with 2 PN, but no cleavage was observed. The rate of 2 PN arrested oocytes was found to be as high as 36.1%. Furthermore, a total fertilization failure was observed in two cases from the round spermatid group. In two pregnancies achieved by round and elongated
spermatids, the male FSH levels were 38.9 and 28 mIU/ml respectively. Neither the FSH levels nor a histopathological diagnosis were found to be significant factors in terms of fertilization and pregnancy (P , 0.05). In one triplet pregnancy, the male FSH level was found to be 20 mIU/ml. In another pregnancy achieved by round spermatids the FSH level was 28 mIU/ml, while in a third pregnancy, the FSH level was found to be normal. Thus, 19 embryo transfer procedures were performed and three pregnancies were achieved (15.7% per embryo transfer and 13% per started cycle). One biochemical pregnancy was obtained with round spermatids and ended in abortion. The remaining two pregnancies, surprisingly, resulted in twins and triplets. The triplet pregnancy unfortunately resulted in abortion in the first trimester, while the twin pregnancy resulted in healthy female babies. Discussion Spermatogenetic arrest is one of the most problematic diagnostic situations for both andrologists and patients. Firstly, no significant improvement has occurred in the successful treatment of such cases. Secondly, in cases defined as being in maturation arrest, normal spermatozoa can still be recovered from the testes by virtue of focal spermatogenesis. Diagnostic testicular biopsies are still informative in cases of azoospermia as observed in germinal cell aplasia, SCO syndrome and maturation arrest of spermatogenesis. However, these biopsy results should not exclude such cases from treatment, as 107
S.Kahraman et al.
spermatozoa can be recovered from testicular tissue in cases with spermatogenetic arrest and it is still possible to obtain ¨ zgu¨r et al., 1996; Tournaye et al., spermatozoa for ICSI (O 1996). Conversely, even in patients with normal testicular volumes where obstructive azoospermia is suspected due to normal gonadotrophin values, spermatogenetic arrest can still occur. No direct correlation has been observed between testes volumes, serum levels of pituitary gonadotrophins and the success of sperm retrieval via testicular sperm extraction (TESE) (Kahraman et al., 1996, Tournaye et al., 1996). Early (round) or late (elongated, elongating) spermatids isolated either from the ejaculate or from testicular tissue for ICSI seem promising as fertilization and pregnancies can be achieved by this method. Edwards et al. (1994) recommended the injection of spermatids into oocytes as an alternative and more direct approach in uniting the two gametes containing haploid genetic material. He proposed that the round spermatid, in which meiosis has been completed, and the histone– protamine transition, where it has hardly begun, might offer the best means of transferring haploid male DNA-associated proteins to enable the nuclear modifications to be preserved in the male gamete. In 23 cases with non-obstructive azoospermia, multiple testicular tissues were taken in the hope of finding spermatozoa. Multiple biopsy samples were obtained in view of focal spermatogenesis. However, in most of these cases spermatogenetic arrest was discovered in their testicular tissues which were sent for histopathological evaluation at the same time. Testicular biopsies were performed prior to oocyte retrieval. This strategy may be incorrect as couples with cystic follicles only or germinal vesicle/metaphase I (GV/MI) oocytes only may have undergone an unnecessary testicular biopsy. However, in our study a multiple testicular biopsy was performed to make both a diagnostic biopsy and to evaluate the presence of spermatozoa for use with ICSI. Complete maturation arrest and severe hypospermatogenetic arrest (mostly ,2% in at least 100 tubules) were observed histopathologically. Interestingly, although the percentage of hypospermatogenesis was relatively high in two cases, only a few spermatozoa were observed in the whole testicular sample. These were the cases in which elongated spermatids were used for ICSI and multiple pregnancies were achieved. Although only two to three spermatozoa were observed in the whole sample, we were unable to find any during the ICSI procedure. However, it is very important to note that the presence of a few immotile spermatozoa in the sample is a promising prognostic factor in terms of both fertilization and pregnancy. Different authors have reported spermatogenetic arrest in 4– 30% of testicular biopsies with severe oligozoospermia or azoospermia. In the largest study, maturation arrest was observed in 27 of the 147 cases (18%) occurring consistently at the end of the meiotic process (Soderstrom et al., 1980; Wong et al., 1997). In our previous study, although the results of testicular biopsies showed complete maturation arrest, SCO or germinal cell aplasia, sufficient spermatozoa were obtained for ICSI (Kahraman, 1996). The existing discrepancy between results is mostly due to the assessment of the biopsies. Some authors interpret the same biopsies as spermatogenetic arrest 108
while others refer to them as hypospermatogenesis. In hypospermatogenesis, all germ cells are decreased in number and similar proportion, whereas in spermatogenetic arrest there is a halting of germ cell differentiation at the level of a specific cell type (Levin et al., 1979). In recent studies, the significance of oocyte activation after spermatid injection has been emphasized (Kimura and Yanagimachi, 1995b; Sofikitis, 1996) While Souza et al. (1996) and Yamanaka et al. (1997) have claimed that an oocyte has acquired an activity factor during spermiogenesis, Sofikitis (1996) has reported the necessity of oocyte activation by electrical stimulation for oocyte calcium ossilation. However, it has been suggested that an oocyte can be activated by a vigorous aspiration of cytoplasm without further electrical stimulation (Vanderzwalmen et al., 1997). It has been stated that the cytoplasm of a round spermatid is not as mature as its nucleus. There may be a deficiency in the oocyte activating factor due to a lack of calcium ossilation response. Tesarik (1994, 1995) has reported that an artificial activation of the oocyte a few hours before injection is a beneficial procedure. In our study, neither electrical stimulation nor artificial activation was employed. However, a vigorous aspiration of oocyte cytoplasm was executed. Nevertheless, the fertilization rate was found to be as low as 31.3%, suggesting that the calcium ossilation response may still have been insufficient. Mostly, one large pronucleus was observed in the injected oocytes. However, a normal embryo morphology was observed in these zygotes. In most of the zygotes, a large single nucleus was observed 16–18 h after injection, giving the impression that the oocytes were only activated but not fertilized. However, according to Tesarik, pregnancy can be achieved by transferring these developing embryos with one large PN when spermatid cells are used (personal communication). Fertilization should be checked after 9–12 h after injection, as the speed of the extrusion of the second polar body, the appearance of 2 PN and embryonic development is faster after a ROSI procedure. However, it was not practical in our study to check the presence of pronuclei during the night, as the ICSI procedure is usually performed at 1400 or 1500 h. The injection of round or elongated spermatids into human oocytes can result in fertilization and embryo development (Vanderzwalmen, 1995; Fishel, 1996; Tesarik, 1996). Implantation of such embryos after uterine transfer suggests that viable human embryos can be obtained by injection of spermatids into oocytes. Thus, a new perspective for the treatment of non-obstructive azoospermia has been introduced. It was suggested by Tesarik (1996) that intra-ooplasmic injection of spermatids obtained from the ejaculate may become the first treatment choice in patients with non-obstructive azoospermia. However, mature spermatozoa can still be obtained with testicular multiple samples, as sperm recovery is still possible in azoospermic patients with spermatogenetic arrest (Devroey, 1995; Kahraman, 1996). Further studies are necessary to decide which sample, ‘ejaculated’ or ‘testicular’, should be selected for azoospermic patients. Very recently, Pisum sativum agglutinin (Sigma) and
Multiple pregnancies and spermatid injection
anti-acrosin antiserum were used to recognize spermatids in ejaculated samples by Mendoza and Tesarik (1996). The fertilization rate with two normally developed PN was found to be as low as 31.3% in our study. Relatively higher fertilization with elongated or elongating spermatids shows that the fertilization rate increases depending on the stage of the spermatids closer to the spermatozoa. In the two cases in which elongated and elongating spermatids were used, the peripheral serum FSH levels were found to be abnormally high. No correlation was found between the FSH level and the fertilization rate in both the elongated and round spermatid groups. In two pregnancies achieved by round and elongated spermatids, the FSH levels were 38.9 and 28 mIU/ml respectively. As the number of cases in the ELSI group is very low (n 5 3) no statistical comparison was performed. High fertilization, cleavage, embryo development and pregnancy rates were achieved with late stage spermatids, although the number of patients was very low in this group. An abnormal fertilization rate with two distinct size PN and triple PN was high in the round spermatid group and the cleavage rate was slow. As emphasized by Fishel et al. (1996), the efficiency of spermatid fertilization is much lower than in conventional ICSI with spermatozoa. However, this is probably due to technical difficulties rather than other factors. Transmission electron microscopy (TEM) is the most accurate quantitative criteria to identify round spermatids. Recently, Yamanaka et al. (1997) applied quantitative criteria for the characterization of human round spermatids. However, as this facility is unavailable in most of the assisted reproductive techniques units, qualitative criteria are more widely used although highly susceptible to intra-observer and intra-procedure variation. To identify round spermatids using qualitative criteria may result in recovery of round cells rather than the round spermatids as TEM demonstrated that only 84% of selected cells were round spermatids. This may suggest one reason for a low fertilization rate with round spermatids. Computer-assisted micromanipulators or a confocal scanning laser microscope–micromanipulator system may be the solution to the problem of spermatid recognition. With a round spermatid, normal fertilization can be achieved as the second meiotic division of the spermatid cell has already been completed. The nuclei of round spermatids have been used for intracytoplasmic sperm injection (Sofikitis et al., 1994). Hannay (1995) has reported four pregnancies with an injection of round spermatid nuclei. Unfortunately, all pregnancies resulted in spontaneous abortions. Although the number of cases injected with round spermatid nuclei was very low, it may be concluded that the isolation of nuclei may be hazardous to the nuclear envelope and can result in a loss of genetic material (Hannay, 1995). Furthermore, significant cytosolic components will be lost. Poor embryo development was observed in the round spermatid group. It would be extremely beneficial if these poor quality embryos were examined by fluorescence in-situ hybridization techniques. The implantation rate was high in the late spermatid group. One triplet and one twin pregnancy were obtained with four embryo transfers in two cases where elongated spermatids
were used. The number of transferred embryos was high due to a mediocre embryo quality. Possibly, a high implantation was achieved due to the high number of embryos transferred and a selective assisted hatching procedure by acid Tyrode’s solution. Unfortunately, the pregnancy obtained by round spermatids ended in a biochemical abortion. Currently, it is not clear which source is preferable to obtain spermatids in testicular or ejaculated samples. However, Fishel indicated that elongated spermatozoa from the testicles achieve fertilization more efficiently than those from the ejaculate, whereas similar rates of fertilization have been achieved with round spermatids from both the ejaculate and testicular samples (unpublished data). The presence of a normal centrosome in human elongated and round spermatids is supported by the work of Fishel et al. (1995) and Tesarik et al. (1995). However, this cannot be assumed for earlier developmental stages from the work of Kimura and Yanagimachi (1995b) because of a maternal inheritance of the centrosome in rodents. Studies on conception using human spermatocytes would be needed to evaluate the normality of the centrosome at this stage of spermatogenesis. Genetic imprinting during mouse spermatogenesis has been examined by assessing methylation sites in a number of cases. A recent study by Fishel et al. (unpublished) confirms completion of gametic imprinting in the testes after the second meiotic division. Similarly, other studies indicate that human spermatogenetic cell imprinting is complete by the round and elongated spermatid stage. Paternal inheritance of the human centrosome was published by Simerly et al. (1995). Abnormal centrosome and consequently abnormal spindle formation can explain an arrest during mitotic cell cycle progression and anomalies arising in embryos after the ROSI procedure. Extreme care should be taken to prevent unfortunate serious genetic results. Thus, if there is a possibility that immature cells are going to be used the patient should be evaluated carefully for chromosomal abnormality or genetic disease. Indeed, all patients with severe male infertility should be assessed in this way. The use of testicular spermatids in the treatment of nonobstructive azoospermia could be a novel treatment model. In cases with severe spermatogenetic defects, even multiple pregnancies can be achieved with elongated spermatid cells by yielding a high implantation rate; however, the fertilization rate with round spermatids has yet to be fully explored.
References Devroey, P., Liu, J. and Nagy, Z. (1995) Pregnancies after testicular sperm extraction and intracytoplasmic sperm injection in non-obstructive azoospermia. Hum. Reprod., 10, 1457–1460. Edwards, R.G., Tarin, J.J., Dean, N. et al. (1994) Are spermatid injections into human oocytes now mandatory? Hum. Reprod., 12, 2217–2219. Fishel, S., Green, S., Bishop. M. et al. (1995) Pregnancy after intracytoplasmic injection of spermatid. Lancet, 345, 1641–1642. Fishel, S., Aslam, I. and Tesarik, J. (1996) Spermatid conception: a stage too early, or a time too soon? Hum. Reprod., 7, 371–375. Hannay, T. (1995) New Japanese IVF method finally made available in Japan. Nature Med., 1, 289–290.
109
S.Kahraman et al. ¨ zgu¨r, S., Alatas, C. et al. (1996) Fertility with testicular Kahraman, S., O sperm extraction and intracytoplasmic sperm injection in non-obstructive azoospermic men. Hum. Reprod., 11, 756–760. Kimura, Y. and Yanagimachi, R. (1995a) Development of normal mice from oocytes injected with secondary spermatocyte nuclei. Biol. Reprod., 53, 855–862. Kimura, Y. and Yanagimachi, R. (1995b) Mouse oocytes injected with testicular spermatozoa or round spermatids can develop into normal offspring. Development, 121, 2397–2405. Levin, H.S. (1979) Testicular biopsy in the study of male infertility. Its current usefulness, histologic techniques, and prospects for the future. Hum. Pathol., 10, 569–584. Martin du Pan, R. and Campana, A. (1993) Physiopathology of spermatogenetic arrest. Fertil. Steril., 6, 937–946. Mendoza, C. and Tesarik, J. (1996) The occurrence and identification of round spermatids in the ejaculate of men with non-obstructive azoospermia. Fertil. Steril., 5, 826–829. Ogura, A. and Yanagimachi. R. (1993) Round spermatid nuclei injected into hamster oocytes from pronuclei and participate in syngamy. Biol. Reprod., 48, 219–225. Ogura, A. and Yanagimachi. R. (1995) Spermatids as male gametes. Reprod. Fertil. Dev., 7, 165–169. Ogura, A., Matsuda, J. and Yanagimachi, R. (1994) Birth of normal young offsprings after electrofusion of mouse oocytes with round spermatids. Proc. Natl Acad. Sci. USA, 91, 7460–7462. ¨ zgu¨r, S., Kahraman, S. and Samlu¨, M. (1996) Sperm recovery rate and O testicular histology in absolute azoospermic patients with primary testicular failure. Intracytoplasmic Sperm Injection: Basic and Clinical Aspects of ICSI, Thessaloniki, Greece, September 20–21. Simerly, C., Wu, G., Zoran, S. et al. (1995) The paternal inheritance of the centrosome, the cell’s microtubule-organizing center, in humans, and the implications for infertility. Nature Med., 1, 47–52. Soderstrom, K.O. and Suominen, J. (1980) Histopathology and ultrastructure of meiotic arrest in human spermatogenesis. Arch. Pathol. Lab. Med., 104, 476–482. Sofikitis, N., Miyagawa, I., Agapitos, E. et al. (1994a) Reproductive capacity of the nucleus of the male gamete after completion of meiosis. J. Assist. Reprod. Genet., 11, 335–341. Sofikitis, N., Zavos, P., Koutselinis, A. et al. (1994b) Achievement of pregnancy after injection of round spermatid (RS) nuclei into rabbit oocytes and embryo transfer: a possible mode of treatment for men with spermatogenic arrest at the spermatid stage. J. Urol., 5, 151. Sofikitis, N., Toda, T., Miyagawa, I. et al. (1996) Beneficial effects of electrical stimulation before round spermatid nuclei injections into rabbit oocytes on fertilization and subsequent embryonic development. Fertil. Steril., 65, 176–185. Souza, M., Mendoza, C., Barros, A. and Tesarik, J. (1996) Calcium responses of human oocytes after intracytoplasmic injection of leucocytes, spermatocytes and round spermatids. Mol. Hum. Reprod., 11, 853–857. Tesarik, J. and Mendoza, C. (1996) Spermatid injection into human oocytes. I. Laboratory techniques and special features of zygote development. Hum. Reprod., 4, 772–779. Tesarik, J. and Testart, J. (1994) Treatment of sperm-injected human oocytes with Ca21 ionophore supports the development of Ca21 oscillations. Biol. Reprod., 51, 385–397. Tesarik, J., Mendoza, C. and Testrat, J., (1995) Viable embryos from injection of round spermatids into oocytes. N. Engl. J. Med., 33, 525. Tesarik, J., Rolet, F., Brami, C. et al. (1996) Spermatid injection into human oocytes. II. Clinical application in the treatment of infertility due to nonobstructive azoospermia. Hum. Reprod., 4, 780–783. Tournaye, H., Goossens, A., Ubaldi, F. et al. (1996) Are there any factors for successful testicular sperm recovery in azoospermic patients? 12th Annual Meeting of the European Society of Human Reproduction and Embryology, Maastrich, The Netherlands, June 30–July 3, 1996. Vanderzwalmen, P., Lejeune, B., Nijs, M. et al. (1995) Fertilization of an oocyte microinseminated with a spermatid in an in-vitro fertilization programme. Hum. Reprod., 8, 502–503. Vanderzwalmen, P., Zech, H. and Birkenfeld, A. (1997) Intracytoplasmic injection of spermatids retrieved from testicular tissue: influence of testicular pathology, type of selected spermatids and oocyte activation. Hum. Reprod., 6, 1203–1213. Wong, T.W., Stauss, F.H. and Worne, N.E. (1997) Testicular biopsy in the study of male infertility: testicular causes of infertility. Arch. Pathol. Lab. Med., 95, 151–159.
110
Yamanaka, K., Sofikitis, N. and Miyagawa, I. (1997) Ooplasmic round spermatid nuclear injection procedures as an experimental treatment for nonobstructive azoospermia. J. Assist. Reprod. Genet., 14, 55–62. Received on February 17, 1997; accepted on September 30, 1997