Paper - Sexual differences of the hypophyses and their determination by the gonads
SEXUAL DIFFERENCES OF THE HYPOPHYSES AND THEIR DETERMINATION BY THE GONADS
CARROLL A. PFEIFFER Zoological Laboratory, State University of Iowa
FOURTEEN TEXT FIGURES AND TWO PLATES (FIPTEEN FIGURES)
- Aided by grants from the National Research Council, Committee for Research in Problems of Sex; grants administered by Prof. Emil Witschi.
Quantitative sex differences in the hypophysis have long been claimed. Evans and Simpson (’29) and Ellison, Campbell and Wolfe (’32) reported that pituitary potency of the male rat is two to three times greater than in the female. Simultaneously McQueen-Williams (’34) and Wolfe and Ellison (’34) demonstrated that the acidophiles and basephiles are more numerous and more clearly deﬁned in the male anterior pituitary than in the female. Clark ( ’35 a) has shown by the implantation method that the gonadotropic hormone content of the anterior pituitary gland of the rat is markedly greater in the female from birth to 20 days; it maintains this supremacy to a smaller degree until puberty, at which time the male gland becomes more potent in the majority of cases. At 7 months the male gland has 1.6 times greater potency than that of the female.
Fevold et al. ( ’33) have shown that two gonadotropic hormone fractions, a follicle stimulator and a luteinizer, can be extracted from powdered sheep pituitary; both are necessary for full ovarian function. Moore assumes that both are necessary for testicular function also (Symposium Assoc. Am. Anat., ’35). However, as Goodman ( ’34) demonstrated so beautifully, ovarian grafts in the male do not luteinize. This seems to indicate a failure of the male hypophysis to liberate the luteinizing fraction. The propriety of using the terms ‘follicle stimulator’ and ‘luteinizer’ for hormones eventually produced by the hypophyses of both sexes may well be questioned. On the other hand, a revision of the nomenclature of the gonadotropic hormones seems unwarranted at the present time. In this paper the conventional names will be used simply because in the work here described the hypophyses of the experimental animals were tested for both follicle stimulating and luteinizing capacities. s
In a recent publication (’35) the present writer has reported that the female hypophysis can be transformed to the male type by testis transplants in the newborn. The following paper contains a report of experiments which were planned to analyze the process of sexual diﬂerentiation of the hypophysis of the rat after birth and the possibilities of its modiﬁcation by castration and gonad transplantation.
In the course of these investigations some observations have been made concerning operative technique and on the value of diiferent body regions for the site of gonadal im plants which seem suﬁiciently important to be presented also in this paper.
It is a pleasure to express my appreciation to Prof. Emil Witschi for suggesting this investigation and for his continued interest throughout the work.
MATERIAL AND METHODS
Both albino and hooded strains of rats were used in these experiments. Whenever possible host and donor were litter mates. In operating upon the newborn rats ether was at ﬁrst used as an anesthetic, but the mortality was very high. A simple and highly satisfactory procedure has been worked out which takes advantage of the known fact that homolothermism is not fully established in the newborn rat. The animals are placed in a small thick glass dish in the freezing chamber of an electric refrigerator until they become immotile. This usually requires 30 to 45 minutes. The glass must be dry or the skin Will freeze. The animals are then sorted as to sex and are placed on ice, each in an individual compartment of the ice cube tray at room temperature. A second tray with its cover serves as an operating board. This stays sufﬁciently cool to keep the animal immotile during any type of operation. Alcohol (95 per cent) is used to sterilize the operating board and the area of incision. For best results, the operation should be performed as quickly as possible and the animals transferred to a warm place. When they have become completely active, they are readily accepted again by the mother. A similar procedure has been published by Wiesner (’35).
Figs. 1to3 Drawings of newborn rats, showing positions of transplants and of the several incisions. G, graft; J, external jugular vein; K, kidney; 0, ovary; T, testis; 1, incision for transplantation into the neck; 2, incision for transplantation into the liver; 3, incision for ovariotomy; 4, incision for castration. X 1. Figure 1, lateral view. Figure 2, dorsal view. Figure 3, ventral view.
Incisions are made, as shown in ﬁgures 1 to 3, with a pair of ﬁne invertebrate dissecting scissors and are kept as small as possible. The incision through the body Wall is made slightly lateral to that through the skin as this facilitates healing without suture. If the incision is at a point of great strain, a little collodion may be applied. The testis is pushed into the area of the incision by slight pressure on the abdomen. After a little practice, ovariotomy is readily accomplished as a blind operation, thus greatly reducing the size of the incision. The ovary is grasped With a pair of ﬁne forceps andis pulled to the surface where it is snipped off with the small scissors. There is little or no bleeding due to cessation of cardiac function (Taylor and Wiesner, ’32). The Wounds heal readily and require no further attention.
In these experiments 1126 animals were operated at birth of Which 435 lived to be used for further tests, usually in the adult stage. This gives a mortality of 62 per cent as against 12 per cent in normal litters. The mortality varied from 76 per cent in the ether anesthesia group to 6 per cent in the cold group with ovaries or testes transplanted into the neck region. In addition, sixty adult females were used in implantation experiments.
Ovary transplants into the anterior eye chamber of the adult rat Were made according to Goodman’s technique (’34). The grafts (both ovarian and testicular) and the ovaries, oviduct, and uterus of the host were sectioned at 10 p and were stained with Delaﬁeld’s hematoxylin and Congo red. The hypophyses were prepared according to the technique of Reese and McQueen-Wiﬂiams (’32).
Daily smear records Were kept on all experimental females, and observations on the ovarian grafts in the eye were made daily. The development of all types of grafts was followed by histological studies at Weekly intervals.
The oestrin used in injection experiments was extracted from pregnancy urine and was standardized in rat units by the usual procedure. The luteinizer was a puriﬁed product
prepared from dried anterior pituitary powder of sheep by the method of Fevold et al. (’33).2
- The luteinizer was prepared by the author in the zoiilogical laboratory of the
University of Wisconsin under the supervision of Prof. F. L. Hisaw and Dr. H. L. Fevold. The author wishes to express his gratitude for the courtesy extended by these two investigators.
This investigation consists of eighteen experiments as outlined in table 1. Male castmtes
Males that had been castrated at birth were submitted to different tests in parabiosis experiments (Witschi and Levine,
TABLE 1 Survey of types of operations and Izypophyseal reaction. in the positive cases
EXPEBL Type or ornaarron uggw Ab b srisaas. ' out pu erty
I 11 3‘ Castration Ovary in eye F & L*
II 10 C3‘ l Castration, testes in neck Ovary in eye F
III 28 5‘ I Castration, ovaries in neck Ovary in eye F & L* IV 10 9 I Castration Ovary in eye F & L
V 14 9 Castration, ovaries in neck | . . . . . . . . . . . . F & L
VI 16 9 Castration, testes in neck Ovary in eye F** VII 50 3‘ Ovaries in body cavity . . . . . . . . . . . . F
VIII 22 6‘ Ovaries in neck l Ovary in eye F
IX 39 9 Testis under skin . . . . . . . . . . . . F 85 L
X 117 9 Testes in body cavity . . . . . . . . . . . . F”
XI 69 SB Testes in neck . . . . . . . . . . . . F**
XII 12 9 . . . . . . . . . . . . . . . . . . . . . Adult testis in neck F & L
XIII 12 9 . . . .’., . . . . . . . . . . . . . . . . . Testes of newborn in F & L
neck XIV 12 S? . . . . . . . . . . . . . . . . . . . . . Testes in kidney F & L XV 12 S? _ . . . . . . . . . . . .. . . . . . . . . . Tetes in liver F & L XVI 12 S2 . . . . . . . . . . . . . . . . . . . . . Teste dorsal to ovary F & L XVII 27 S? Testes in liver . . . . . . . . . . . . . . . . . . . F & L XVIII 22 9 , Testes in neck Injection of hormones F** Total 495
(F) follicle stimulator. (L) luteinizer. * Experimentally caused maintenance of L potentiality. ** Experimental uppressicn of L potentiality.
’34). The normal female co-twin immediately or a month after puberty goes into continuous oestrus indicating that the hypophysis of the early castrate has a high output of follicle stimulating hormone like that of an animal castrated at puberty or later. The question arises whether these cas200 CARROLL A. PFEIFFER
trate hypophyses are identical in all other respects. Furthermore, we wish to know whether the hypophysis of newborn males can still differentiate in either the male or the female direction. To test these factors the following experiments with males castrated at birth were performed.
In experiment I, eleven males were castrated at birth. The development of the secondary sex characters under this condition is described in detail by Wiesner (’35). At 90 days of age the ovary of a 15- to 18-day-old rat was grafted into the anterior chamber of the right eye. The grafts took in every case. Corpora lutea formed within 2 weeks. This is very interesting since Goodman (’34) has shown that only follicle development is observed if the host is a male castrated when mature. This experiment proves that in normal male development the testes actually suppress the factor for the production of luteinizer. It is highly interesting to note that the hypophysis of the early castrate retains the capacity to produce luteinizing hormone, considering the above stated fact that it produces high amounts of the follicle stimulator like that of an adult castrate.
Experiment II consisted of ten males operated at birth. Both testes were removed from the scrotal region and were transplanted to the region of the right jugular vein. The grafts took in all cases. When recovered after 6 to 8 months some resembled cryptorchid testes with spermatogenesis proceeding to the spermatocyte stage. Others showed different degrees of reduction, sterility, and fatty degeneration. The prostates and seminal vesicles were intermediate between the normal and the castrate type. After puberty the typical fat deposits of the castrate were laid down. The hypophysis showed castrate cells in all cases.
At about 150 days of age the ovaries of 15-day-old rats were transplanted into the anterior eye chamber. They took quite readily but showed only follicular development. Even after 3 months there was no indication of corpus luteum formation. Comparing this with the previous experiment we realize that the testis graft is able to diﬁerentiate the hypophysis in the male direction just as well as the normal testis.
In experiment III, twenty-eight males were castrated and received ovarian implants in the neck region at birth. The host and donor were litter mates. The ovarian grafts took in 90 per cent of the cases. Follicular development was observed at 35 days, and evidently considerable oestrin was produced as the short segment of uterus attached to the graft was distended with ﬂuid (ﬁg. 13). In most cases the ovary matured at approximately 45 days. This was determined by the presence of corpora lutea in the grafts removed between 45 and 60 days after implantation. In some of the animals autopsied at 8 months mainly cystic follicles were found in the graft. Many showed the beginning of castrate changes. Castrate cells were found in the hypophyses of the older animals but not in the younger ones with normal appearing grafts. In all cases where the ovarian graft took, whether in the neck or in the eye, the seminal vesicles were considerably enlarged. This was probably due to the stimulation of the smooth muscle by oestrin, as pointed out by Moore et al. ( ’34), though the possibility of the production of male sex hormone by the ovarian medulla may also be considered.
The ovaries of 23-day-old females were transplanted into the eye in ten of the animals at 60 days of age. These grafts showed characteristic cyclic changes in follicular growth and corpus luteum formation and were similar in every way to those in the female.
Females castrated at birth
In experiment IV, ten females were castrated at birth. They developed normally up to puberty, but the uterus remained infantile. As in control females of our colony the vagina opened so that smear records could be started by the sixtieth day. In general characters, such as body size, these animals resembled the normal female up to puberty, but later they began to develop the somatic features of the castrate.
At 70 to 75 days of age the ovaries of 15-day-old rats were transplanted into the anterior eye chamber of ten of these castrates. There was a 100 per cent take, as evidenced by the vaginal smears which started after 6 to 8 days. In the ﬁrst or second cycle these animals showed a prolonged corniﬁed cell stage but came to dioestrus after 7 to 10 days and then ran fairly normal cycles. This cyclic change in the vagina was correlated with a cyclic production of corpora lutea in the ovarian graft in the eye which represents the same condition as reported by Goodman (’3-4) for adult castrated females with ovarian grafts in the eye.
In experiment V, fourteen females were operated at birth. The ovaries were removed from their normal position and were transplanted into the neck region in close proximity to the jugular vein. The grafts took in all but one, as evidenced later by the vaginal smears. These ovaries matured earlier than they would have in the normal position. In the case of the good grafts oestrous cycles started onabout the fortyﬁfth day. Controls do not begin until the sixtieth or seventieth day. Three animals in this experiment ran only a few cycles and soon developed the somatic features of the castrate. At autopsy the ovarian tissue had completely degenerated. The other eleven varied considerably as to their vaginal smears but ran fairly normal cycles for several months. This period was followed by constant oestrus and ﬁnally by anoestrus in most instances (ﬁg. 4). In all cases the smear record could be correlated with the condition of the graft. Corpora lutea were found during the normal cyclic changes in the vagina and cystic follicles during the constant oestrous phase. Where the graft became non-functional, the anoestrous condition began. Ovarian grafts were not made into the eyes of animals of this group.
The above conditions are quite similar to those observed in ovaries of newborn rats grafted into the kidney of adult castrated females (Pfeiffer, ’34). Most of the hypophyses show castrate cells. However, if the animal is killed early, before continuous oestrus is established, the hypophysis presents the normal histological picture. The presence of castrate cells shows that the graft is not in equilibrium with the hypophysis. The oestrin output for a time is high enough to produce corniﬁed cells in tlie Vagina but not enough to maintain the normal hypophysis.
castrate cells appear
a a - - - - - g -n--Q-an--a
follicles and corpora lutea only cystic follicles graft dagenerates O 10 20 30 40 50 60 70 80 90 100 110 180 130 3.40 150 3.60 3.70 180190
Fig.4 Oestral condition correlated with histology of the hypophysis and of the ovarian graft. Constant oestrus may continue for a longer time as indicated by the broken line.
In experiment VI, sixteen females were castrated at birth and the testes of litter mate males were transplanted into the neck region in close proximity to the jugular Vein. The grafts took in all but three cases, as evidenced by masculiniza— tion. In four cases the clitoris was deﬁnitely enlarged into a penis-like organ about te11 times the size (length) of the normal clitoris (compare 5 and 6 with ﬁgs. 7 and 8).
Figs.5to8 Ca.mera lueida. drawings of genital system of female rats. BV, bulbo-vestibular gland; C, clitoris; O, ovary; U, uterus; V, vagina; V0, Vaginal oriﬁce. X 1. Figure. 5, normal female. Figure 6, castrate female. Figure 7, female castrate with testis graft. Figure 8, female castrate with testis and ovarian grafts. Note the hypertrophy of the clitoris in ﬁgures 7 a.nd 8.
This response was ﬁrst described by Sand (’18 and ’19). In two of the cases the clitoris was so hypertrophied that it extended beyond the dermal sac, but it did not completely form a penis. In most of the otl1er cases the appearance was that of the young female except that the dermal sac had a slit—like opening and the clitoris was enlarged. In all cases the vagina opened at the same time as in controls and castrated females.
Wlien the animals were 40 to 70 days old, ovaries from 12-day-old rats were implanted into the anterior eye chamber. The grafts took in all cases and corniﬁed cells appeared i11 the vaginal smear Within 6 days. There was a tendency to run constant oestrus in this group which was directly correlated with the tendency for the secondary sex characters to be transformed to the male type. Thus, the four animals in which the clitoris was highly hypertrophied ran the characteristic constant oestrous type of smears. No corpora lutea appeared i11 these ovarian grafts. The three which were not masculinized showed the typical corpus luteum formation of the female and ran true, though slightly prolonged, cycles. The remainder were intermediate (irregular) in regard to both corpus luteum formation and cyclic changes in the vagina.
This correlation of masculinization of the secondary sex characters and the behavior of the ovary allows the assumption that the hypophysis is differentiated according to the gonad present in the early postnatal period. The constant oestrus of the vagina shows that the interaction between gonads and hypophysis is non-cyclic in the male regardless of Whether an ovary or a testis is present.
Normal males grafted at birth
In experiment VII, ﬁfty males were grafted at birth with the ovaries of litter mate females. The ovaries were placed in the body cavity and usually became attached to the body wall at the Wound area or to the mesentery. The grafts were recovered in only 45 per cent of the cases in which autopsy SEXUAL DIFFERENCES or THE HYPOPHYSES 205
was at about 1 year of age. Takes would have been found more frequently had the grafts been removed at an earlier time. About 70 per cent of the twenty—eight autopsied before puberty contained grafts.
In experiment VIII, twenty-two males had the ovaries of litter mate females grafted into the neck region at birth. The grafts were placed in close proximity to the jugular vein and were recovered from this position in approximately 80 per cent of the cases.
Fig. 9 Drawing of normal ovary at puberty. O, ovary; OC, ovarian capsule; OD, oviduct; U, uterus. X 2.
Figs. 10 to 12 Drawings of ovarian grafts in male hosts. Legend as in ﬁgure 9. X 2. Figure 10, 1-week-old graft. Figure 11, 2-week-old graft. Figure 12, 3-Weck—o1d graft.
Fig. 13 Drawing of 4-week-old graft in castrate ma.1e. Legend as in figure 9. X 2.
Fig. 14 Drawing of 4-month-old ovarian graft in a castrate female host. Legend as in ﬁgure 9. X 2.
Before puberty in both experiments the ovaries showed a fairly normal development (ﬁgs. 10 to 12, 21 and 22). There was considerable follicular development but no corpora lutea were ever found in any ovarian graft in these two experiments. The grafts did not develop much after puberty and underwent degeneration. The hosts appeared normal and their reproductive organs were not impaired in any Way. Many of the males with established ovarian grafts were used in the breeding colony with good success. The prostate and seminal vesicles were normal, and the testes of all were normal in size and histological structure. The histology of the hypophysis was normal. Ovarian grafts were not made into the eye cavity.
Since ovarian grafts in males show only follicular development, it is indicated that the ovarian graft in the presence of tlie l1ost testes does not have any inﬂuence upon the differentiation of the hypophysis.
Female rats with gonad grafts
In experiment IX, thirty-nine females received one testis graft under the skin of the dorsal body wall. The host and donor were newborn litter mates. Transplanted material was found in 65 per cent of the cases. However, upon sectioning, only 45 per cent showed implants of recognizable testicular structure. Fifteen animals were autopsied before puberty to determine the development of the graft. Biopsy was performed on the remaining‘ animals when they were between 6 and 8 months of age. The graft, if present, was removed at this time, as well as the right ovary and a section of one uterus. Complete smear records were kept until autopsy 5 to 6 months later.
Of the fourteen animals that had testicular grafts at the time of biopsy, thirteen were running normal cycles. The remaining animal, however, was in constant oestrus from
the time smears were established until about 2 months after
the removal of the graft, when numerous leucocytes appeared among the corniﬁed cells. The ovaries of the constant oestrous animal showed tumorous growth. It remains doubtful whether the grafted testes had anything to do with the constant oestrous condition in this case. However, the portion of the ovary not affected by the tumor showed only follicular development. The testis graft showed a few sterile seminiferous tubules a11d a small amount of epididymis. The grafts as a whole showed abundant infiltration by fat and connective tissue, with a few sterile seminiferous tubules, but in some cases they contained primary spermatocytes. The epididymis showed a much greater persistence than the seminiferous tubules.
It was thought that if the testes were placed within the body cavity, there might be less inﬁltration by connective tissue and fat. Therefore, in experiment X, 117 females were grafted with testes in the abdominal cavity. To demonstrate a possible quantitative effect, forty-eight animals received only one testis and the remaining sixty-nine each received two testis transplants. Host and donor were newborn litter mates. Forty—ﬁve animals were autopsied before puberty. Complete smear records "were kept on the remaining eightyseven. The grafts were recovered in 69 per cent of the cases in both groups. The percentage was slightly higher in those autopsied before puberty.
Fifty-ﬁve animals had grafts at biopsy (7 to 12 months of age) when the graft and the right ovary were removed. Seventeen grafts were found under the skin at the site of the incision; thirty-one were attached to the body wall, six to the intestine, and one to the kidney. Of these ﬁfty-ﬁve animals, thirty-one ran normal cycles, nine ran constant oestrus, and ﬁfteen showed a condition of ‘prevalent dioestrus.’ This prevalent dioestrus consists of extended dioestral periods of 11 to 15 days separated by short phases of oestrus (corniﬁed cells for 2 days). Such a condition frequently persisted for 3 or more months. Biopsy showed ovaries with large corpora lutea (ﬁgs. 25 and 26). However, the uterus did not produce deciduomata by the thread reaction of Long and Evans (’22). Animals with prevalent dioestrus may after some time change to normal or near-normal cycles which is correlated with the normal appearance of the ovary (ﬁg. 24). Others transform to the ‘constant oestrus’ type. The histological picture shows clearly that during prevalent dioestrus the ovary . is under the inﬂuence of large quantities of both follicle stimulating and luteinizing hormones. Obviously both hormone-producing centers of the hypophysis are stimulated. The constant oestrous condition becomes established if the production of luteinizer is discontinued. It consists of exclusively follicular development of the ovary (ﬁg. 27) and oestrous condition of the vagina. The uterus is not distended as at the height of the normal oestrous cycle. The constant oestrous condition, once established, is permanent and is not affected by the removal of the graft.
There was a slight increase in endocrine disturbance in the group receiving two testis transplants. This was due to the fact that the chances for a well-developed graft were doubled by the increase in the number of testes transplanted.
In this experiment some of the grafts were attached to the intestine. These were the largest and healthiest observed in the entire investigation (ﬁg. 19), yet none of. them had any apparent effect upon the endocrine system. The grafts resemble the typical cryptorchid testis of the rat; spermatogenesis proceeds to the spermatocyte stage, at which time the spermatocytes move to the lumen of the seminiferous tubules where they degenerate. This degeneration takes place here earlier than in. the grafts in the neck (ﬁgs. 18 and 20), probably because of the higher temperature. There is considerably more space between the seminiferous tubules than in grafts in the neck region (compare ﬁgs. 18 and 19), but there is not much diﬁerence in the actual amount of interstitial elements (compare with the normal testis, ﬁg. 15).
The greater development of the grafts attached to the intestine must be attributed to their rich blood supply. The fact that the grafts have no influence upon the endocrine system of any of the hosts suggests that the hormones are
I destroyed by the liver. This makes it imperative that a con dition be found where there is room for growth and ease of establishing a good blood supply which will drain directly to the heart. The region of the jugular veins appears to fulﬁll these conditions.
In experiment XI, sixty-nine females were used. Two testes were placed under the loose skin of the neck near the jugular veins. The donor and the host were newborn litter mates. Eleven of these were either autopsied or biopsied before puberty to determine the development of the grafted testes. When the remaining ﬁfty-eight animals had reached sexual maturity, six of them selected at random from the group running normal cycles were placed with males. All became pregnant and gave birth to normal litters. At autopsy, at the time their litters were weaned, ﬁve showed well developed grafts. Of the remaining ﬁfty-two animals, ﬁve were used in injection experiments (p. 212), and forty-seven were biopsied when 7 to 9 months old; forty-six of these ﬁfty-two animals showed grafts attached near the jugular veins. Including the pregnant group above, this gives a total of 79 per cent takes.
Of the forty-six females carrying testis grafts, eighteen ran constant oestrous cycles, six ran very irregular cycles, while the remaining twenty-two were normal. The onset of constant oestrus varied from puberty to 3 months afterward. In the experimental animals, as well as in the controls, the oriﬁce of the vagina opened within the normal range given by Long and Evans ( ’22). In two cases constant oestrus began 5 to 10 days before the removal of the graft and the right ovary. Two animals in which the grafts had been recovered at 35 days of age started constant oestrus at the time of sexual maturity.
In all cases where continuous oestrus is once deﬁnitely established it continues after the graft is removed and until senility impairs the endocrine function of the ovary. The host ovary shows the same characters as an ovary grafted into a male. This is conﬁrmatory proof that even in the presence of an ovary the testis is able to inﬂuence the differentiation of the hypophysis toward maleness.
It is thus apparent that there is a variation in the time at which the inﬂuence of the testis graft is superimposed upon the endocrine function of the hypophysis and ovary. Therefore, it seemed necessary to test the possibility that the adult testis grafted at puberty might disturb the endocrine function of the ovary in spite of negative reports published by Takewaki (’33). In experiment XII, twelve adult females were grafted with either one-fourth or one-half of the adult testis in the region of the jugular vein. About 50 per cent retained some grafted material for 711- months, of which only one-half showed sterile tubules still present. The grafts were not as well deﬁned as were those where the testis of the newborn was grafted at birth. Eleven of the animals ran normal cycles, and one ran irregular cycles, having prolonged corniﬁed cell stages interspersed between normal cycles. It is doubtful if this condition was due to the effect of the graft. The animal had a respiratory infection of which it died. The ovary was quite small and showed only follicular development at the time of death. s
The immature testis develops so much better than the grafted adult testis that in experiment XIII, twelve females were grafted at puberty with testes of the newborn. The testes were transplanted into the neck region near the jugular vein. The grafts took in ten of the cases and varied from almost completely resorbed to fairly well-developed grafts. Ten animals ran normal cycles, while two ran almost constant oestrus but only after running normal cycles for almost 6 months. Of these two the graft in one had been completely resorbed at autopsy 9 months after implantation.
A series of experiments was made testing the various grafting sites in the adult rat. In each experiment twelve females at puberty received testes transplants from the newborn. They were all autopsied 8 months after transplantation. In experiment XIV the grafts were placed in the kidney and recovered in 50 per cent of the cases at autopsy. In experiment XV implantation was into the liver with only 33 per cent of the grafts recovered at autopsy; none were well developed. In experiment XVI the testes were placed dorsal to the ovaries. As many as 83 per cent were recovered, but the condition of the grafts was inferior to those transplanted into the body cavity of the newborn. .
There was no disturbance of the oestral cycles or of the reproductive functions in any of these cases, nor was there any alteration in the histology of the genital system.
Experiment XVII was designed to prove whether the liver really prevented the grafted testes from exerting an inﬂuence upon the hypophysis and through it the host ovary. Testis transplants were made into the liver of twenty-seven females. The hosts and donors were both newborn litter mates. They were autopsied between 7 and 10 months of age. Twenty-one showed implants at the grafting site, but upon sectioning only eight showed tissue of recognizable testicular structure. The oestrous cycles and reproductive functions were undisturbed in all cases.
Previous experiments (Pfeiifer, ’35) showed that the constant oestrous animal responds to the injection of luteinizer by formation of numerous corpora lutea in the ovary and the dioestrous condition of the vagina. However, constant oestrus reappears after a few days. Oestrin causes a suppression of follicle development, and consequently anoestrus, but no corpora lutea formation.
Experiment XVIII was made to further investigate the above conditions. Twenty-two animals were grafted with two testes along the jugular vein. Host and donor were newborn. As many as ﬁfteen of this group (68 per cent) established constant oestrus. This high percentage was due entirely to improved technique. The grafts were not removed, and these animals were used for injection of luteinizer and oestrin. Only animals that continuously showed corniﬁed smears were used, so that any break in the smear record could be attributed to the injection. The smear records were always checked by biopsy observations on the ovary and uterus. Since it was desirable to ﬁnd whether the constant oestrous condition could be altered or normal oestrus established after various periods of injection, the ovaries were allowed to remain intact if their condition could be determined by macroscopic observation. If not histological studies were made.
All injections of luteinizer caused formation of corpora lutea within 3 days (table 2). Injection of 0.5 mg. induced luteinization of only a few of the follicles, while the 1 and 2 mg. injections gave numerous corpora lutea in each ovary. After the second day the animals went into dioestrus (table 2), though corniﬁed cells reappeared in the smears 2 to 6 days later, and all animals resumed constant oestrus. The experiment was repeated 30 days from the ﬁrst injection with the same results except that the smallest dose did not appear to be as effective as it Was the ﬁrst time. Injections of 2 mg. of luteinizer (single dose) Were made on four animals every 5 days for 1 month. Irregular cycles were established which were correlated roughly with the injections. After the cessation of injections two of the animals returned to constant oestrus, one died from respiratory infection about 3 Weeks after the last injection, and the fourth continued to run irregular cycles for over 3 Weeks. The ovaries in this case con TABLE 2 Injection of Zutetnteer and oestrta into constant oestrous females with testis grafts DAILY DURATION NU3‘:'ER I INJEPTIONS TOTAL $3 15333 EFFECT ON .F(ﬁ§:§R:;'F or INTERANIMALS L.H. Oestrin INJECTED INJEOTED OVABIES WITHIN RUPTION ' mg} RU. - IN DAYS 2 l 0.5 _ . . . 2 mg. - 4 -Corpora lutea. 3 days‘ I 3 3 1.0 . . . 4 mg. 4 Corpora lutea. 3 days‘ 4 2 2.0 . . . 8 mg. 4 Corpora lutea 3 days‘ 4 2 5 20 R.U._ 4 No change 4 days’ 1 3 . . . 10 40 R.U. 4 Reduction 4 days’ 2 3 . . . 100 400 R.U. 4 Reduction 4 days’ I 4
‘After ﬁrst injection.
’ After last injection.
3 Thirty-ﬁve milligrams are equivalent to 1 gm. acetone dried anterior pituitary powder.
tained almost the normal distribution of corpora lutea. It is not certain Whether this indicates a partial recovery of the luteinizing function by the hypophysis.
In order to test Whether the injection of luteinizer produced normal ovulation or whether the follicles luteinized Without ovulation, ﬁve animals (from experiment XI) which had been in constant oestrus for 4 to 5 months received 4 mg. of luteinizer in a single injection. Autopsies were made at 12, 25, 30, 44 and 45 hours after injection. Ovulation occurred in all animals about 10 hours from the time of injection. The number of follicles that ovulated varied from four to six for each ovary. This is the normal number for the rat (Long and Evans, ’22). Follicles with fragmented ova did not respond to the luteinizer. A few apparently normal mature follicles luteinized without ovulation. It would seem then that the experimental constant oestrous condition in these rats is similar to that normally occurring in the rabbit during the breeding season. Obviously the ovary requires a small amount of luteinizer to produce normal ovulation. The uterus was distended and showed the typical vascularization of the late prooestrum. It is interesting to note that the round nucleated cells of stage 1 always appeared in the vaginal smear 24 hours after the injection. The remainder of the cycle was abbreviated so that the animals autopsied at 44 and 45 hours had already attained the dioestrous condition. 4
Oestrin was injected as shown in table 2. The 5 R.U. caused only an inﬁltration of leucocytes in the smear on the fourth day, and practically no morphological changes occurred in the ovary. The 10 R.U. and the 100 R.U. injections produced the anoestralcondition in the vagina, 4 days after the last injection; but no corpora lutea were formed. However, the follicular development was impaired. The ovary (ﬁg. 28) showed large sterile areas and only small follicles. The constant oestrus was again established within 2 to 5 days. Injection of the 100 R.U. group was repeated after 30 days with identical results. Injections of 100 R.U. were made every 5 days to this group for 1 month. In all three animals each injection was followed by anoestrus on the fourth day, but after the cessation of the direct effect of the last injection the animals returned to constant oestrus.
Histotogy of the testis grafts
The histology of the testis grafted under various conditions in the rat is thoroughly discussed by Moore (’26), and only a few notes as to the general conditions when the grafts are of testes of the newborn are discussed here. If good vascularity is established, the grafts develop normally almost up to maturity. Figures 16 to 18 are photomicrographs of grafts recovered 2 weeks, 3 Weeks, and 6 months after implantation (compare these with the normal 2-weeks-old testis, ﬁg. 15).
In all cases spermatogenesis never proceeds further than the spermatocyte stage. Under the best conditions development proceeds to the spermatocyte stage and then disintegration follows as has been described by Moore (’26) for grafts of immature testes. Even under the skin in the neck region spermatogenesis does not proceed any farther. The main factor in development of the graft is its vascular supply. The best grafts are recovered from the intestine where little invasion of connective tissue or fat is encountered. The grafts take well under the skin or on the body wall at the point of incision but are readily invaded by connective tissue and fat, often leaving only the tunica albuginea and a very few sterile tubules. .
The testes start to develop well in the liver and kidney but usually become inﬁltrated with fat and degenerate. However, this is much more prevalent in the liver than in the kidney. Degeneration begins with the germ cells, then follow seminal tubules and epididymis, and ﬁnally the unorganized stroma.
The disturbance of the endocrine system of the female is not correlated absolutely with the histology of the graft, as much variation exists. However, the disturbed function is never observed unless an active graft has been present. The testis grafts seem to take equally well in females, and castrated males and females.
The histology and physiology of ovarian grafts has been thoroughly discussed by such workers as Sand (’26), Lipschiitz ( ’27), Smelser (’33), Goodman (’34) and others. The ovarian graft develops fairly normally up to puberty if a good blood supply is established. In the male there is only follicular development unless the host is castrated at birth, in which case numerous corpora lutea are formed (ﬁg. 23). The endocrine function of the graft is established early, in either the castrate male or female, so that portions of uteri, the oviducts, and the ovarian capsules are often greatly disSEXUAL DIFFERENCES or THE HYPOPHYSES 215
tended with fluid at 3 weeks of age (ﬁg. 12). This is presumably due to the effect of the oestrin. The greatest distension occurs in females castrated at birth (compare ﬁgs. 9 and 14). These grafts are able to maintain cyclic function for a time, but they usually become cystic and produce a subsequent constant oestrous reaction of the vagina. The graft brings the host female to sexual maturity 2 to 3 Weeks earlier than if in the normal position.
In the animals running constant oestrus the host ovaries contain numerous follicles in all stages of development up to maturity (ﬁg. 27). Although the ovaries of all these animals were sectioned, no corpora lutea were found, and no indication of ovulation was ever observed unless luteinizer was injected. It is easy to ﬁnd all stages of development up to the ﬁrst maturation division. After this stage all that one ﬁnds is fragmentation of the ovum (ﬁg. 29), in some cases resembling the early cleavage stages of normal development. The follicles degenerate Without any indication of either thecal or granulosa luteinization. The oviducts and uteri resemble the quiescent state of dioestrus. The uterus has the simple columnar epithelium and slit»like lumen of the dioestrous stage and never becomes distended with ﬂuid as in the normal oestrus or the constant oestrus induced in parabioses experiments. The ovary during prevalent dioestrus contains numerous corpora lutea similar to those during pregnancy (ﬁgs. 25 and 26).
Sand, in 1918, was the ﬁrst to point out that in the guinea pig, ovarian grafts exhibit extensive corpora lutea formation in the female but only follicular development and atresia in the male. Pettinari (’25) is the only one to report corpora lutea in an ovarian graft in the male guinea pig. Sand (’18), Moore (’19), Wang, Richter and Guttmacher (’25) show that corpora lutea may be found, although rarely, in the male rat. Sand states that in ‘ovary testis’ corpora lutea form in both the immature and the adult male rat. Moore (’19) reports a single case of corpora lutea formation in a male castrated at 30 days of age and receiving an ovarian transplant from a litter mate female. The graft showed corpora lutea at autopsy 200 days later. Judging from his illustration, the corpora lutea seem to be undersized and not quite normal. Most of the rats used by Wang, Richter and G‘rutten— macher were also castrated at 30 days of age and were implanted at about the ﬁftieth day of life. Here again the corpora lutea. shown in the photomicrographs do not appear typical. If these grafted ovaries develop corpora lutea of some kind, then we must conclude that at 30 days of age the hypophysis is not completely differentiated and still retains the capacity for luteinizer production.
Smelser ( ’33) makes the general statement that corpora lutea are formed in ovarian grafts in the female but not in the male. This is further borne out by Goodman (’34) with ovarian transplants in the anterior eye chamber of the adult. Here the graft is under constant observation, and any corpora lutea formed can be easily detected. Gonadotropic hormone from pregnancy urine luteinizes these ovarian grafts (Goodman, ’34). Therefore, we can conclude that in the normal male or the male castrated after puberty, the hypophysis does not produce the luteinizing hormone. Experiments described in the present paper prove that this particularity of the male hypophysis is acquired in response to the presence of active testes in any animal, whether of male or female genetical constitution.
Hisaw et al. ( ’34) and Wolfe (’35) came to the conclusion that injection of appropriate amounts of oestrin causes the release of luteinizer by the female hypophysis. The exact mechanism of luteinizer release is not deﬁnitely known. Moore and Price (’32) report that oestrin suppresses the hypophysis of males as well as of females. Hisaw et al. (’34) have shown that it is the follicular fraction that is suppressed. Witschi and Levine (’34) show that the release of luteinizer (and probably even its production) is inhibited by relatively high concentrations of follicle stimulator, even in the presence of large amounts of oestrin. It is plausible then to assume that in the female hypophysis the luteinizer is held in check by high concentrations of the follicle stimulator and that it can be released only after the follicle stimulator has been suppressed by large amounts of oestrin. As demonstrated by experiments presented in the foregoing description, the injection of oestrin suppresses the output of follicle stimulating hormone. However, it does not bring about a release of luteinizing hormone. This indicates again that the hypophysis is no longer of the female type.
Smelser (’33) expresses the opinion that the sexual di morphism in the guinea pig hypophysis is a sex limited
genetic character. On the contrary, the author in a previous paper (’35) furnished evidence to prove that in the rat the sex type of the hypophysis is not genetically determined but that its diﬂerentiation is dependent upon the gonad. The present investigation conﬁrms this view and shows that the hypophysis at birth is bipotential and still undiﬁerentiated. The possible explanation why Smelser was led to the opposite conclusion is that in his cases the hypophysis had already been determined by the gonad; the experiment would have to be started at an earlier age (probably at a foetal stage in the guinea pig).
Ovaries under the inﬂuence of the musculinized hypophysis are never maximally stimulated nor does ovulation occur. This is signiﬁcant in view of the ﬁndings of Evans and Simpson (’29) and Clark (’35 b) which show that the hypophysis of the adult male is 1.6 to 2 times as potent as that of the female. Hisaw et al. (’34- and ’35) contend that ovulation depends on the sudden release of luteinizer and that the ﬁnal stage of follicular development in the normal cycle represents an ‘augmentation’ or ‘synergic’ effect of the two gonadotropic substances. The fact that the ovaries in the experimental animals used in the present study fail to produce fully mature follicles that ovulate and do not stimulate the uterus to a full oestrous condition may be considered, therefore, as another consequence of the absence of luteinizing hormone. Such an interpretation is strongly supported experimentally by the fact (p. 212) that ovulation can be induced at any time by the injection of an appropriate dose of luteinizer.
Hellbaum (’35) assaying acetone dried pituitary powder, demonstrates that young mares produce large quantities of luteinizer, while stallions produce only follicle stimulator. The colt and young gelding are intermediate. He ﬁnds that ‘old’ (110 longer breeding?) mares and older geldings also produce only follicle stimulator. He believes that there are
both sex and age factors in this differentiation toward pure follicle stimulator production. Our experiments suggest that horses that produce only follicle stimulator will fall into two classes: those with a masculine hypophysis, incapable of producing luteinizer, and those with a feminine or neutral hypophysis which fail to produce luteinizer merely because of lack of oestrin. Hellbaum’s tests relate only to the actual function but not to the potential capacities of the different classes of pituitaries.
SUMMARY AND CONCLUSIONS
1. A technique for anesthetizing and operating newborn rats is described.
2. The physiological activity of the hypophysis, as measured by the eﬂect upon the gonads, is investigated under eighteen experimental conditions in 495 successfully operated animals.
3. The hypophysis in the rat at birth is bipotential and capable of being differentiated as either male or female, depending upon whether an ovary or as testis is present.
4. The hypophysis of the male, whether normal or castrated and with its testes reimplanted ectopically, releases only folli cle stimulator. The hypophysis of the female releases both follicle stimulator and luteinizer.
5. In both males and females castrated at birth the hypophysis remains bipotential until puberty. If an ovary is transplanted into the eye, such a hypophysis will secrete both follicle stimulator and luteinizer in cyclic order. Corpora lutea form normally in the graft.
6. Testis transplants in the newborn, castrated female have a masculinizing eﬁfect on the clitoris and cause differentiation of the hypophysis to the male type in 81 per cent of the cases (implants in neck region).
7. Testis transplants into the newborn female control the differentiation of the hypophysis, completely suppressing the luteinizer production in 68 per cent of the cases (implants in neck region).
8. Ovarian transplants into newborn normal males remain without inﬂuence on the sex type of the hypophysis.
9. The gonads of immature and newborn rats take more readily than those of the adult. The ovaries form follicles which become cystic or luteinize, depending upon the physiological state of the hypophysis. In testis grafts spermatogenesis proceeds to the spermatocyte stage. Testis grafts can develop to the same condition regardless of the age of the host and donor.
10. Various transplantation sites were studied. Testes in the neck of the female give as high as 68 per cent reversal of the sex type of the hypophysis. The effect of implants in the body region is much less, and where the blood drains from the graft through the hepatic portal system no effect is shown. This lends evidence to the theory that the sex hormones are destroyed in the liver.
11. The ovary under the inﬂuence of a masculinized hypophysis produces only follicles. The ova mature and then undergo fragmentation. However, they never ovulate. The oestrin level is such that it causes the constant oestrous condition of the vagina, but it is not high enough to cause the full oestral changes in the uterus.
12. The ovaries of the constant oestrous group are normal in their response to luteinizing hormone and oestrin. Appro220 CARROLL A. PFEIFFER
priate doses of luteinizer bring about full oestrus with ovulation and corpus luteum formation, while oestrin injections merely suppress follicle development without luteinization.
13. The sex type of the hypophysis after the age of puberty
is relatively stable and attempts to change it experimentally A
have failed almost entirely.
14. These experiments prove that the sex diﬁerence in the hypophysis is not genetic but is secondary and dependent upon the presence of differentiated sex glands.
CLARK, H. M. 1935a A prepubertal reversal of the sex differences in the gonadotropic hormone content of the pituitary gland of the rat. Anat. Rec., vol. 61, pp. 175-192. 1935b A sex diﬁerence in the change in potency of the anterior hypophysis following bilateral castration in newborn rats. Anat. Rec., vol. 61, pp. 193-202. ELLISON, E. T., M. CAMPBELL AND J. M. WOLFE 1932 Comparison of the capacity of the anterior lobe tissue of the hypophysis of male and female rats to induce ovulation. Anat. Rec., vol. 52 (suppl.), p. 54. EVANS, H. M., AND M. E. SIMPSON -1929 A sex difference in the hormone content of the anterior pituitary of the rat. Am. J. Physiol., vol. 89, pp. 375-378. Fnvonn, H. L., F. L. HIsAvv, A. HELLBAUM AND R. Hnnrrz 1933 Sex hormones of the anterior lobe of the hypophysis. Am. J. Physiol., vol. 104, pp. 710-723. FOSTER, A. M., AND F. L. HISAW 1935 Experimental ovulation and resulting pseudopregnancy in anoestrus cats. Anat. Rec., vol. 62, pp. 75-94. GOODMAN, LEROY 1934 Observations on transplanted immature ovaries in the eyes of adult male and female rats. Anat. Rec., vol. 59, p. 223-252. HELLBAUM, A. A. 1935 The gonad stimulating activity of pituitary glands from horses of various ages and sexual activities. Anat. Rec., vol. 61 (suppl.), p. 23. . HISAW, F. L., H. L. FEvoLn, M. A. FOSTER AND A. A. HnI.L.BAUM 1934 A physiological explanation of oestrus. Anat. Rec., vol. 60 (suppl.): p. 52. _ Llrscﬁiitrz, A. 1927 On some fundamental laws of ovarian dynamics. Biol. Rev., vol. 2, pp. 263-280. Lone, J. A., AND H. M. EVANS 1922 The oestrus cycle in the rat and associated phenomenon. Mem. Univ. California, vol. 6. MCQUEEN’-WILLIAMS, M. 1934 A sex diﬂerence in cytology of the anterior hypophysis of the rat. Anat. Rec., vol. 58 (suppl.), p. 78." Moons, O. R. -1919 On the physiological properties of the gonads as controllers of somatic and psychical characteristics. I. The rat. J. Exp. Zo6l., vol. 28, pp. 137-160. '
Moons}, C. R. 1926 On the properties of the gonads as controllers of somatic and psychical characteristics. IX. Testis graft reactions in different environments (rat). Am. J. Anat., vol. 37, pp. 351-416.
MOORE, C. R., AND D. PRICE 1932 Gonad hormone functions, and the reciprocal inﬂuence between gonads and hypophysis with its bearing on the problem of sex-hormone antagonism. Am. J. Anat., vol. 50, pp. 13-71.
Moonn, C. R., G. F. Smmorrs, L. J. WELLS, M. ZALESKY AND W. O. NELSON 1934 On the control of reproductive activity in an annual-breeding mammal (Citellus tridecemlineatus). Anat. Rec., vol. 60, pp. 279-290.
PETIINARI, V. 1925 Feminisation et hyperfeminiation par greﬁe ovarienne. Compt. Bend. Soc. Biol., T. 92, pp. 1228-1229.
PFEIFFER, O. A. 1934 Functional capacities of ovaries of new-born after transplantation into adult ovariotomized rats. Proc. Soc. Exp. Biol. and Med., vol. 31, pp. 479-481.
1935 Origin of functional differences between male and female hypophyses. Proc. Soc. Exp. Biol. and Med., vol. 32, pp. 603-605.
REESE, J. D., AND M. MCQUEEN-WILLIAMS 1932 Prevention of ‘castration cells’ in the anterior pituitary of the male rat by administration of the male sex hormone. Am. J. Physiol., vol. 101, pp. 239-245.
SAND, K. 1918 Studier over Ksnskarakterer. Copenhagen.
1919 Experiments on the internal secretion of the sexual glands, especially on experimental hermaphroditism. J. Physiol., vol. 53, pp. 257-263. 1926 Transplantation der Keimdriisen bei Wirbeltieren. Abh. Handb. Normal. u. Patholog. Physiol., Bd. 14/1 (Fortplanz. Entvvickl. Wochst.), S. 251-292.
SMELSER, G. K. 1933 The response of guinea pig mammary glands to sex hormones and ovarian grafts and its bearing on the problem of sex hormone antagonism. Physiol. Zoii1., vol. 6, pp. 396-449.
TAKEWAKI, K. 1933 Inﬂuence of transplantation of testicular tissue on oestrus
cycles of female albino rats. J. Fae. Sci. Imp. Univ. Tokyo, Sec. IV, vol. 3, pp. 145-151.
TAYLOR, H., AND B. P. WIESNER 1932 Experiments on temperature coeﬂicient of heart activity. J. Physiol., vol. 75, p. 36P.
WANG, G. H., C. P. RICHTER AND A. F. GUTTMACHER 1925 Activity studies on male castrated rats with ovarian transplants, and correlation of the activity with the histology of the grafts. Am. J. Physiol., vol. 73, pp. 581-599.
WIESNER, B. P. 1935 The post-natal development of the genital organs in the albino rat. J. Obstet. and Gynaecol. British Empire, vol. 41, p. 867 and vol. 42, p. 8.
WITSGHI, E., AND W. T. LEVINE 1934 Oestrus in hypophysectomized rats parabiotically connected with eastrates. Proc. Soc. Exp. Biol. and Med., vol. 32, pp. 101-107.
WOLFE, J. M. 1935 Reaction of ovaries of mature female rats to injections of oestrin. Proc. Soc. Exp. Biol. and Med., vol. 32, pp. 757-759.
WOLFE, J. M., AND E. T. ELLISON 1934 Morphological diﬁerences in the anterior pituitaries of male and female rats. Anat. Rec., vol. 58 (suppl.), p. 78.
PLATE 1 EXPLANATION or Houses
15 Photomicrograph of the normal 2-week-old testis. X 15.
16 Photomicrograph of a testis graft implanted at birth and recovered after 2 weeks. )( 15.
17 Photomicrograph of a testis transplanted at birth and recovered after 3 weeks, from the neck. of the castrate male. Note the normal appearance of both the testis and the epididymis. The seminiferous tubules are further developed than in the normal 3- to 4-week-old testis. )( 15.
18 Photomicrograph of a good testis graft in the neck of a female, recovered after 6 months’ implantation. A typical cryptorchid testis. X 15.
19 Photomicrograph of a testis graft attached to the intestine. Note the larger size and the greater space between the seminiferous tubules without actual increase in the interstitial cells. X 15.
20 Same graft as in ﬁgure 18. Photomicrograph of a seminiferous tubule, showing spermatocytes falling into the lumen and disintegrating. )< 400.
PLATE 2 EXPLANATION or rreoarzs
21 Cross section through a 1-week-old ovarian graft in a male host. 0, ovary; OD, oviduct; U, uterus. X 15.
22 Photomicrograph of a section through a 3- to 4-week-old ovarian gra.ft from the neck of a male host. Note the normal appearance. X 15.
23 Photomierograph of a cross section of an ovarian graft 3 months after transplantation into the neck of a castrate male. Host was castrated at birth and ovary transplanted at the same time. Note the amount of luteinization. X 15.
24 Pliotomicrograph of a cross section of the ovary of a. normal female. X 15.
25 Photomierograph of a cross section of the ovary of a. female in prevalent dioestrus; early dioestral phase. X 15.
26 Photomierograph of a cross section of the ovary from a female after about 3 months of prevalent dioestrus; late dioestral phase. X 15.
27 Photomicrograph of a section through the ovary of a fema.le running constant oestrus due to a testis implant. X 15.
28 Same type of animal as in ﬁgure 27 except that it had received 100 R.U. of oestrin every 5 days for a month, which treatment stopped the constant oestrous condition. X 15.
29 Photomicrograph of a fragmenting ovum. X 750.