USE OF ALBUMIN GRADIENTS FOR X AND Y SPERM SEPARATION AND CLINICAL EXPERIENCE WITH MALE SEX PRESELECTION*

Transcription

1 FERTILITY AND STERILITY 1979 The American Fertility Society Vol. 31. No. I, January 1979 Printed in U.SA. USE OF ALBUMIN GRADIENTS FOR X AND Y SPERM SEPARATION AND CLINICAL EXPERIENCE WITH MALE SEX PRESELECTION* W. PAUL DMOWSKI, M.D., PH.D.t LILIANA GAYNOR, M.D. RAMAA RAO, M.D. MARY LAWRENCE, B.S. ANTONIO SCOMMEGNA, M.D. Department of Obstetrics and Gynecology, Michael Reese Hospital and Medical Center, University of Chicago Pritzker School of Medicine, Chicago, Illinois Semen samples obtained from 18 normal males and 37 husbands requesting male child preselection were separated on concentration gradients of human serum albumin. Separated semen obtained from the husbands was then used for artificial homologous insemination (AlH). A significant increase in the sperm motility, progressive drive, and percentage of Y-bearing sperm along with a decrease in the total sperm count and percentage of abnormal forms were observed in separated specimens. Fathers of three or more female children had a slightly smaller but statistically significant percentage of Y-bearing sperm in their semen samples than did normal males. Ten conceptions were achieved with separated semen. Seven pregnancies terminated in normal deliveries of five male and two female infants, one ended in a spontaneous abortion of a male fetus, and two patients are still expecting. The ratio of male to female conceptions in this small study parallels the ratio of Y to X sperm in the final specimen used for AlH. Fertil SteriI31:52, 1979 The development of a successful technique for sex preselection prior to conception could have profound social, demographic, genetic, or economic consequences, although well-known experts in these fields cannot agree on what exactly these consequences would be. In 1973 Ericsson et al. 1 reported the isolation from human semen of a fraction rich in Y sperm after a passage of the semen through concentration gradients of bovine serum albumin (BSA). These results have been confirmed by some 2 but not by others.3. 4 Using a modification of Ericsson's original technique we were able to isolate a Y sperm-rich frac- Received July 27,1978; accepted August 17, *Presented in part at the First International Symposium on Artificial Insemination Homologous and Male Subfertility, Bordeaux, France, May 6 and 7, treprint requests: W. Paul Dmowski, M.D., Ph.D., Department of Obstetrics and Gynecology, Michael Reese Hospital and Medical Center, 29th Street and Ellis Avenue, Chicago, Ill tion from the semen of normal males. Encouraged by these results, we performed artificial homologous insemination (AIR) with the Y spermenriched semen fraction in 37 couples requesting male child preselection. MATERIALS AND METHODS Semen samples were obtained by masturbation after 3 to 4 days of abstinence in 18 normal males of proven fertility and a mean age of32 (range, 25 to 44). This group served as control. In addition, semen samples were similarly obtained from ten husbands whose wives had delivered three or more female children. These two groups were compared to identify a possible deficiency of sperm bearing Y bodies in the total sperm population of men fathering exclusively girls. The characteristics of the 37 couples requesting male child preselection are listed in Table 1. The mean wife's age was 31 and the mean husband's age was 33. All marriages were stable with a mean

2 Vol. 31, No.1 MALE SEX PRESELECTION 53 TABLE 1. Characteristics of Couples Undergoing Sperm Separation for S Preselection Mean wife's age (range) Mean husband's age (range) Mean number of female children (range) Mean age of youngest child (range) Mean duration of marriage (yr) Couples with 3 or more female children Couples with no evident fertility problems Couples with apparent fertility problems 31 (25-41) 33 (21)-48) 2.2 (0--5) 3.4 (1-10) duration of 8 years. The majority of couples had one or more female children, but a few were childless and preferred a male child as the first. Several couples had tried other methods of male child preselection unsuccessfully, and 10 of37 had three or more female children. The mean age of the youngest child was 3.4 years. The experimental nature of the project was explained to all couples and all understood that at best only an increased chance for male child conception was being offered and that a definite possibility for female birth existed. Couples who considered the birth of a female child detrimental totheir marital or family life were not included in the study. Insemination. Ovulation was timed by utilizing basal body temperature (BBT) records, cervical mucus changes, and vaginal cytology. The patients were instructed to use condoms at the beginning of the cycle and to abstain from sexual intercourse for 3 to 4 days prior to the expected ovulation. At the time of expected ovulation the husband was instructed to produce a semen sample by masturbation. Semen analysis and sperm separation were then carried out. Insemination (AIH) was performed with a separat.ed semen specimen in a final volume ranging between 0.5 and 1.0 ml which was injected into the cervical canal at the level of the internal cervical os. In wives with unpredictable ovulation AIH was repeated one or two times until ovulation occurred. On the basis of history, BBT charts, physical examination, and semen analysis the couples were separated into two groups: those with no apparent fertility problems and those with overt fertility problems (Table 1). The latter group, consisting of 17 couples, included 10 with relative male factor, 6 with prior history of infertility,8 with oligomenorrhea, and 6 with grossly abnormal BBT charts indicating ovulatory dysfunction (Table 2). During subsequent cycles of insemina tion, ovulatory dysfunction in this group was treated with clomiphene citrate, human chorionic gonadotropin, or a combination of both. Separation Technique. The technique of sperm separation was a modification of a method described by Ericsson et al. 1 At room temperature immediately after liquefaction, the semen was di luted 1:1 with Tyrode's solution and then centrifuged at 3600 rpm for 15 minutes. The supernatant was discarded and the spermatozoa were resuspended in the Tyrode's solution to contain approximately 60 x 10 6 of spermatozoa in each 1 ml. Aliquots of 0.5 ml were then layered onto isolation columns, the number of columns being determined by the total number of spermatozoa in each sample. Each column contained two discrete layers of human serum albumin (HSA). The lower layer consisted of 0.5 ml of20 HSA and the upper layer consisted of 1 ml of 10 HSA. After 1 hour the top 0.5-mllayer containing the sperm suspension was removed and the remainder of the column was left undisturbed for another 30 minutes. At that time 10 and 20 HSA layers were collected separately and centrifuged at 3600 rpm for 15 minutes. The supernatants were removed and the spermatozoa were resuspended in Tyrode's solution. The spermatozoa obtained from the 20 HSA layer were used for insemination. The total time between production of the semen sample and insemination was approximately 3 to 4 hours. At each step in the procedure the total number of spermatozoa and the percentage of motile spermatozoa were determined in the hemocytometer. The mean drive (that is, the speed of progressive motion across a distance of 0.05 mm) was measured with a stopwatch. The percentage of abnormal forms was determined subsequently on fixed and stained slides. Also, at each step in the procedure smears were prepared for fluorescent Y -body counts. For that purpose the slides were air-dried and fixed in 95 ethanol. Subsequently, they were stained with quinacrine dihydrochloride and the percentage ofy sperm was determined by fluorescence microscopy.5 RESULTS Total sperm count, sperm motility, percentage of abnormal forms, and progressive drive were similar in fathers of exclusively female children and in normal males. However, the percentage of Y TABLE 2. Fertility Problems in Subfertile Patients Requesting Sex Preselection Oligospermia Menstrual irregularity History of infertility Abnormal BBT Endometriosis, status post-unilateral oophorectomy Total Subfertile couples

3 54 DMOWSKI ET AL. January.1979 spermatozoa before separation was statistically >-! Ii:> different (p < 0.02), lower in men fathering three ti +1 C'<I I.l".) or more girls than in normal males who had t"- eo fathered both boys and girls (Table 3). The same data for couples with normal or im- C'<I paired fertility and for couples who conceived dur- 0 0 " > c ti +1 ing the treatment are shown before and after sep- 0 C'<I C'<I m C! aration in Table 4. The mean values of total sperm 0 count, motility, percentage ofabnormalforms, and =. drive before separation varied widely among the groups but were essentially within normal limits. al cy:l I:! i g ti +1!..:: C) -< The mean percentage of Y bodies prior to separa- cy:l.qo tion ranged between 45 and 47. In comparing values before separation with those obtained from c:l Ii: 10 and 20 HSA fractions, progressi ve decreases r:: " :g in the sperm count and percentage of abnormal '" :::. == eo 0 forms were noted, whereas the forward progres sion, motility, and the percentage ofy sperm were increased. Differences between the initial & fi.- -,, specimens and the 20 HSA fraction were statisti ft8 x I.l".) C'<I cally significant (P < 0.05) in all groups..s The amount of debris and the percentage of ab- normal forms declined dramatically in 20 HSA I:! I.l".) e fraction, while the percentage of motile forms inti '" >-! "tl m cy:l I:!.qo.qo creased consistently from an average of 41 to c:l 790/0. Simultaneously, the speed or progressive mol:! c-.i tion as reflected in sperm drive also improved sig- C'<I ] 0 nificantly. The percentage ofy spermatozoa in the " > v c 10 HSA fraction was similar to that present in Q., Z 0 cy:l C"!.. the initial specimen. However, a significant in- I:! as '"!:l crease in the percentage of Y spermatozoa in the Gl..::..s = S 20 HSA fraction was observed (p < 0.001). The r.:"l.oj total sperm count in the final 20 HSA fraction.!: al.qo e. '" S.. m o Ii was drastically decreased as compared with that in os 1; f: ti ti g c:l.fic2 m C'<I 'W v.gv the original specimen. The final sperm count used I::l -< C'<I cy:l I:Q I:! Q.,GlQ., c for insemination represented only between 3 and.-1:1 e 8 of the original count. This percentage is only.qo!:l ;>!:l t"- Gl bil Gl slightly higher if improvement in motility is taken S!:l S C"I.) f == m t"-. e; :a.e; into consideration. I.l".).qo Gl!:l Gl '" The analysis of clinical data is shown in Table 5.!:l t!:l Altogether, 132 AIH procedures were performed in cw:i c:l cycles of 37 patients, resulting in 10 concep- "3 00cY:l 00cY:l fj... mm tl as h x cy:l.6 S S S tions for an over-all pregnancy rate of 27. In 20 0 C'<I o. cf:: cf:: patients with normal fertility, 42 AIH procedures Eo< 1::1::1:: were performed in 32 cycles resulting in 7 concepf f f tions for a pregnancy rate of 35. Among the 17 5Gl :a.!a 't:s't:s subfertile patients, 90 AIH procedures performed >.>.>. f a during 68 cycles resulted in 3 conceptions for a i Gl S S as as as...,...,..., pregnancy rate of aa!:loa l, " Gl Clinical information on the 10 patients who con- S +1 II ceived is shown in Table 6. Seven conceptions oc- Gl Gl Gl OJ... ::s ::s ::s curred in the first cycle of AIH, one in the second, e II a -E Gli:!.6'0 f!»> OJ OJ OJ and one in the third. One patientwith anovulation ol:! as... conceived during the 13th cycle, after ovulation SoO 01:!'Q) Z

4 J, -.: -..,--;,;:,, " x W :""' Couples with normal fertility, n = 20, mean (range) ± SD Subfertile couples, n = 17, mean, (range) ± SD Pregnant, n = 10, mean (range) ± SD Total spenn count x 10" TABLE 4. Sperm Separation Data in Thirty-Seven Couples Requesting Male Child Preselection Before separation Abnonnal Motile fonns ± ± 11 ( ) Drive Bee Y sperm. Total spenn count x 10" 10 HSA traction Abnormal Motile Corms Drive 1.2 ± ± 7 88 ± ± ± ± 0.2 sec After separation Total Y spenn spenn count Motile x 10" 43± 4 16 ± 1{)l 83 ± go 20 HSA traction Abnormal Drive Y forms sperm Bee 17 ± 11b 1.0 ± 0.2" 71 ± 7a ± ± ± ± 6 43 ± ± ± ± ± 9 8 ± &' 77 ± lsa 14 ± &' 1.11 ± ± 11a ( ) ± ± ± ± 4 65 ± ± ± ± ± 7 16 ± 1 77 ± 17a 18 ± ± ± go ( ) I [;; Value significantly different from original specimen atp < bvalue significantly different from original specimen at P < "Value significantly different from original specimen atp <

5 56 DMOWSKI ET AL. January 1979 TABLE 5. Results of AlH with Separated Semen in Thirty-Seven Couples Requesting Sex Preselection No. AIR cycles Total AIR Conceptions Pregnancy rate Couples with normal fertility Subfertile couples Total induction. All couples were instructed to abstain from unprotected intercourse until after the rise in BBT and all but two (couples 8 and 10) claimed compliance. One pregnancy terminated spontaneously in the abortion of a male fetus; seven patients delivered at term (five male and two female infants) and two patients are still expecting. DISCUSSION Throughout history, prospective parents have sought ways to influence the sex of their unborn child and some of the recommendations to this effect date back to antiquity. If the advice of Greek philosophers, such as unilateral orchiectomy, and folk methods, such as coitus with boots on, are disregarded, all sex preselection techniques thus far recommended may be divided into three categories: ('1) techniques designed to increase the chances of egg fertilization by X or Y sperm after coitus, such as coital timing or pre- and postcoital douching; (2) in vitro sperm separation techniques coupled with artificial insemination; and (3) selective abortion. Numerous attempts have been made to select the sex of the offspring prior to conception in both animals and humans using techniques of the first two categories, and many claims as to the effectiveness of such procedures have been published in both professional and lay literature. All methods, including folk recommendations and ancient techniques, can claim at least a 50 success. Two recently published articles on sex preselection review our current knowledge of the subject and discuss techniques of sperm separation applied thus far in both animals and humans. 6, 7 Of all sex preselection techniques recommended, the one reported by Ericsson et al. I claims to isolate from human semen a highly motile fraction rich in spermatozoa containing a fluorescent Y body. By using a three-step isolation procedure the authors were able to isolate a sperm fraction characterized by 98 motility and 85 ofy bodies. Subsequently, Ross et au demonstrated only an increase in motility with this method, while Evans et al. 3 observed neither effect. However, Soupart2 in a brief communication reported an increased Y sperm concentration with the same technique, in agreement with the initial report by Ericsson et au In this report, using a two-layer HSA technique similar to that described by Ericsson, we were able almost to double the initial sperm motility. The concentration of Y sperm increased simultaneously by about 50 of the original value. Both changes were statistically significant. Our results, although convincing, are less spectacular than those reported by Ericsson et al. l It should be pointed out, however, that several parameters in this technique are variable and that their optimal values remain to be established. With two-step isolation the concentration of Y sperm achieved by Ericsson et al. l was only 76 and was closer to the number observed in our studies. Furthermore, fluorescent staining and identification ofy bodies are subject to error, primarily the one of underscoring for Y bodies. In our studies the initial Y sperm concentration was about 46, that is, 4 less than that observed by Ericsson et al. l in their control samples prior to separation. Along with the increase in motility there was a simultaneous increase in the percentage of normal forms with a faster progressive motion. The percentage of abnormal forms decreased from 38 to 15 after separation, and the mean sperm drive increased from 1.19 seconds before separation to 0.93 second after the procedure. The hypothesis that men with three or more female children produce a lower number of Y spermatozoa has been confirmed by this study. In 10 such couples there was a slight, but statistically significant, decrease in the Y sperm population in the semen sample produced by the husband. A group of subfertile couples which included 10 husbands with borderline oligospermia had lower than average total sperm counts, lower motility, and an increase in abnormal forms (Table 4). In this group, the final sperm count after separation was less than 10 million. The significant decrease in total number of spermatozoa after separation was indeed a disturbing finding in all groups. Less than 10 of the initial sperm count was available for AIH after separation. The number is only

6 Vol. 31, No.1 MALE SEX PRESELECTION 57 TABLE 6. Analysis often Conceptions with Separated Semen in Couples Requesting Sex Preselection Day ofbbt ofy sperm Couple Parity No. of Day of Pregnancy no. AIH cycles AIH Dip Rise Before After outcome , , , a IOU a Unprotected intercourse within 1 day of BBT dip Spontaneous abortion (male) Male Male Male Female Male Male Female Due date 11/26/ Due date 12/20/78.. slightly higher if correction for the increase in the percentage of motile sperm after separation is considered. The low final sperm count used for insemination was probably one of the reasons for the rather low conception rate. In 10 patients who conceived the average total sperm count used was 16 x 10 6 Seven of the ten conceptions occurred in the group of couples with normal fertility and three in the subfertile group (Table 5). In the latter group, a significantly smaller total number of spermatozoa, frequeptly in the range of only 2 to 3 millions, was available for insemination. Although the minimal number of spermatozoa necessary for conception has not been determined, the lowest count resulting in pregnancy was 4 x 10 6 in two patients in our study. This is probably the minimal count that can achieve pregnancy. In many couples who did not conceive the AIH specimen contained a much lower count, probably too low to assure success. As demonstrated in Table 6, 10 conceptions in our study resulted in term deliveries of five male and two female infants. There was one spontaneous abortion of a male fetus, and two pa tients have not as yet delivered. This ratio of male to female pregnancies reflects the mean ratio of Y:X sperm as calculated in vitro in the separated specimens used for AIH. In two patients who conceived female infants, the degree of Y sperm enrichment was not substantial. Furthermore, one couple (couple 8) had unprotected intercourse subsequent to AIH and prior to the BBT rise. However, even with the highest concentration of Y sperm (80) which was achieved in some patients in our study, there is still a possibility that the ovum may be fertilized by an X sperm and that a female child will be conceived. In conclusion, the isolation of the highly motile, Y sperm-rich semen fraction is possible using density gradients of HSA. The separation ofy from X sperm, however, is far from complete and the ratio, although increased, is not as high as originally reported by Ericsson et au The very low sperm recovery was probably the reason for the rather low conception rate in the clinical setting. Although the number of pregnancies reported is small, the ratio of male to female conceptions seems to correspond to the ratio of Y to X sperm in specimens used for AIH. REFERENCES 1. Ericsson RJ, Langevin CN, Nishino M: Isolation of fractions rich in human Y sperm. Nature 246:421, Soupart P: MGA-M appearance in ejaculated human sperm (abstr 133). Presented at the Eighth Annual Meeting ofthe Society for the Study of Reproduction, Fort Collins, Colo, July 22-25, Evans JM, Douglas T A, Renton JP: An attempt to separate fractions rich in human Y sperm. Nature 253:352, Ross A, Robinson JA, Evans HJ: Failure to confirm separation of X and Y bearing human sperm using BSA gradients. Nature 253:354, Pearson P: The use of new staining technics for human chromosome identification. J Med Genet 9:264, Rinehart W: Sex preselection-not yet practical. Population Rep, Ser 1, No 2, May Glass RH: Sex preselection. Obstet Gynecol 49: 122, 1977