J. Endocrinol. Invest. 14: 559-568,1991
Thyroid hormones in tissues from human embryos and fetuses A. Costa *, R. Arisio**, C. Benedetto***, E. Bertino****, C. Fabris****, G. Giraudi*****, L. Marozio***, V. Maula******, M. Pagliano***, O. Testori*******, and G. Zoppetti****** ******Divisione di Endocrinologia e *******Servizio di Medicina Nucleare, *Ospedale Mauriziano Umberto I, **Divisione di Anatomia Patologica e Ricerche Cliniche, Ospedale S. Anna, ***Istituto di Ostetricia e Ginecologia, ****Istituto di Discipline Pediatriche, Servizio di Neonatologia, *****Dipartimento di Chimica Analitica, Universita degli Studi di Torino, Torino, Italy concentrations of 40 and 20 times respectively from the 9th to the 12th week, when thyroid follicles oirganization takes place. In fetuses and adults T3 and T4 were measured in brain, heart, kidney, liver, lung, skeletal muscle and skin (mean concentrations: 0.86 ng/g for T 3 and 7.44 ng/g for T 4 in fetuses and neonates; 1.36 ng/g for T 3 and 12.75 ng/g for T 4 in adults). Hormones concentration increased with gestational age; the T ~ 4 ratio increased until 22-24 weeks, when the prevalent increment in T4 occurs. T 3 concentration up to 30 weeks was generally higher in tissues than in cord serum of the corresponding age. During the last month of gestation T 3 increment was faster in serum. T4 level was always predominant in serum. In conclusion, T3 and T4 have been detected in the limbs of embryos before the onset of thyroid hormone secretion. Concentrations were 1/150 and 1170, of the normal maternal blood values respectively. It is conceivable that these hormones are of maternal origin, and the question of whether such small quantities may playa role in fetal development is open.
ABSTRACT. This study was intended to quantify T3 and T 4 in various human tissues at different stages of gestation as a contribute in the evaluation of the role of thyroid hormones in fetal development, particularly before the maturation of fetal thyroid function. Moreover, for a better comprehension of the influence of thyroid hormone status in tissues, the study was extended to adults. Embryonic specimens were obtained from voluntary abortions between 6 and 12 weeks of gestation, fetal and neonatal specimens from fetuses and neonates between 15 and 36 weeks of gestation after spontaneous abortion or stillbirth, and adult specimens from men (age range: 45-65 years) after death for cardiovascular diseases. Thyroid hormones were measured by the method of Gordon and coworkers. In embryos T3 and T4 were measured in limbs, carcasses, brain and liver: considering all values measured in the period 9-12 weeks, a mean concentration of 0.11 ng/g for T 3 and 1.28 ng/g for T4 was obtained. In pooled limbs of 68 weeks T3 was barely measurable (0.01 ng/g). In the carcasses there was an increase in T 3 and T 4
INTRODUCTION
have been detected in the developing human cerebral cortex at the 11th week of gestation (3). Both T3 and its nuclear receptor have been found in the brain of a 10 week embryo (4), and T3 receptor and T4 could also be measured in a 7-week-old whole embryo (5). Serum thyroxine was identified in fetus blood after 11 weeks of gestation (6). The transfer of thyroid hormones from mother to embryo has been proved in different species, in which maternal hypothyroidism is accompanied by thyroid hormone deficiency of the conceptus before the onset of fetal thyroid activity (7-9). Such transfer in humans is still controversial. Raiti et al. (10) have reported that significant quantities of T3 cross the placenta in at least 50 per cent of the cases when high doses of T3 was administered to normal pregnant women for four to six weeks before delivery.
The data of the literature on the presence of thyroid hormones in tissues obtained from human embryos and fetuses are limited, especially in the formers. Embryonic thyroids in organ culture acquire the capacity to concentrate radioiodine and synthesize iodothyronines by 74 days of conceptional age, equivalent to 12 weeks of gestation (1). Labelled T3 and T4 have been demonstrated in thyroid glands from fetuses of 11 weeks of gestation, after the mother had undergone a 131-1 test (2). Moreover, thyroid hormones as well as 5' monodeiodinase activities
Key-words: Thyroid hormones, embryos, fetuses. Correspondence: A. Costa, Ospedale Mauriziano Umberto I, L.go Turati 62,10100 Torino, Italy. Received July
26,1990; accepted March 18, 1991.
559
Costa A., Arisio R., Benedetto C, et al.
Grumbach and Werner (11) concluded that the human placenta is capable of transferring thyroid hormones. However, their transfer from the mother is considered insufficient to support normal skeletal and brain development in case of primary thyroidal defect of the fetus. Pharoah and Connolly (12) observed that neurological endemic cretinism can be prevented only by giving iodine to the mother before conception; therefore maternal thyroid hormones, specifically T4, may have a role in the maturation of fetal nervous system before the onset of fetal thyroid function. In late gestation, the transfer of thyroid hormones is well documented, at least when fetal thyroid is impaired . Indeed infants with thyroid agenesis or with organification defects have detectable values of T3 and T4 at birth; the rapid decline of their levels suggest that they are of maternal origin (13). Based upon these assumptions, we have measured the concentrations of T3 and T4 in various human tissues at different stages of gestation in order to clarify the possible role of thyroid hormones in fetal development and maturation , particularly before the onset of fetal thyroid function . Moreover, to understand more precisely the specific influence of thyroid hormone status in tissues we have extended our study to adults.
sected from the body. Skin samples were taken from the chest or abdominal wall, the subcutaneous fat was trimmed off and the surface cleaned with a gauze moistened with phosphate buffer saline (PBS); myocardial tissue was collected from the right ventricle and cerebral tissue from the frontal lobe of the forebrain. All tissues were isolated with a semisterile technique, quickly rinsed in cold PBS (4 C) and stored in liquid nitrogen until assay. Adult specimens were obtained from 7 men (age range: 45-65 years) after death for cardiovascular diseases. Subjects with endocrine or metabolic diseases were excluded. The samples were taken at autopsy 24 hours after death . Homogenization and extraction of the tissues Human tissues were homogenized in methanol containing 1 mM PTU (6-propyl-2-thiouracil). About 1 g of tissue was added to a Polytron glass tube, containing 5 ml of methanol-PTU/g wet tissue and approximately 300 cpm of 125-I-T3 (specific activity 1200 IlCi/llg). The homogenization was carried out at 4 C by a Polytron. The homogenates were transferred to stoppered test tube, then chloroform was added in a volume double than the amount of methanol. A mixture of chloroform-methanol (2: 1) was thereafter used for the extraction as described by Gordon et al. (14), Folch et al. (15), and Morreale de Escobar et al. (8). After centrifugation of the mixture for 15 min at 2,000 rpm , the extract was removed and transferred in a glass beker. The particulate matter was extracted again by shaking with three successive portions of chloroform-methanol (2: 1), each followed by centrifugation and removal of extract. The extracts were pooled, filtered on filter paper and transferred to stoppered glass tubes for backextraction of the iodothyronines into an aequeous phase. Two ml of 0.05% CaC12 were mixed with 10 ml extract and the mixture was allowed to stand at 4 C for approximately one hour to allow separation of two clear layers. The upper phase was retained and the interphasic surface was then rinsed with a small amount of pure upper layer in a volume equal to that removed (2 ml). All the upper layers and rinses were collected in a glass baker and evaporated under nitrogen to complete dryness. The residue was dissolved in 10 ml methanol and the solution was transferred into glass tube and centrifuged. The supernatant was evaporated to dryness under vacuum at 37 C, then the residue was treated with 2.5 ml PBS (0 .1 M, pH 7.4)/g homogenized tissue. Finally, the aqueous solution was centrifuged and the supernatant was further purified by column chromatography as described by Mallol et al. (16).
MATERIALS AND METHODS Human samples Embryonic specimens were obtained from 33 voluntary abortions between 6 and 12 weeks of gestation. The samples were washed with physiologic saline and kept in glass tubes at 4 C until dissection, which was performed within 30 min from uterine aspiration . Specimens of brain, liver, limbs and carcass (body without head and limbs) were collected separately. When the weight of the sample was less than 1 g, different specimens of the same tissue were pooled together. The gestational week was calculated on the basis of the first day of the last menstruation and confirmed by ultrasounds. Patients with a history of thyroid disease were excluded from the study. The study was approved by the Ethical Committee of the Istituto di Ostetricia e Ginecologia. All patients gave informed consent. Fetal and neonatal specimens were obtained from 13 fetuses and 2 neonates, without macroscopical and histological evidence of maceration, between 15 and 36 weeks of gestation (Table 1). The samples were taken within 24 hours from spontaneous abortion or stillbirth. Specimens of brain, heart, kidney, liver, lung, skeletal muscles and skin were dis-
560
Thyroid hormones in human embryos and fetuses
Column chromatography For column chromatography the Bio-Rad resin (BioRad 1 x 2, 200-400 mesh, chloride form) was first suspended in 1.2 M acetic acid and poured into 2.5 ml plastic syringes up to 1.0 ml mark. The resin was then washed several times with distilled water until the pH was 3-4 and with acetate buffer (0.2M, pH 7.0). The resin was prevented from passing through the syringe by 2 mm layer of paper. The acetate buffer was drained just before the aqueous sample (the extract) was added. Then the column was washed with 2 ml of acetate buffer (0.2 M, pH 7) and with 4 ml of ethanol. After further washing with 2 ml of acetate buffer (0.2 M) at pH 4.0, 2 ml at pH 3.0, followed by 2 ml of 1% and 2 ml of 35% acetic acid, iodothyronines were eluted with seven or eight fractions (0.5 ml) of 70% acetic acid. The fractions of eluate containing the higher radioactivity were collected and evaporated under nitrogen. The dry sample was then counted in a gamma-counter to evaluate the overall recovery, then was dissolved in 2.5 ml PBS and used to determine the T3 and T4 concentration by RIA test.
(Sclavo) were slightly modified to achieve maximum sensitivity. Standard solutions of T3 and T4 were prepared by dissolving the authentic hormone in 0.1 M PBS (pH 7.4, 0.05% NaN3), and diluting it with the same buffer to working concentrations. To determine T3, 0.2 ml of Amerlex-M T3 antibody were reacted with 0.2 ml of 125-I-T3 tracer from the kit and 0.5 ml of standard or sample. The assay was then performed at room temperature for 3 hours. To determine T4, 0.1 ml of T4 antibody from FT4 RIA kit were reacted at 4 C overnight with 0.1 ml of 125-I-T4 tracer from the same kit and 0.5 ml standard of sample. The separation of antibody-bound from free tracer was performed on Lisophase column according to the instruction of manufacturer. The accuracy of these modified assays was proved by dilution and recovery tests on aequeous samples after chromatography. In these modified assays the least detectable doses (calculated as the dose that gives responses statistically different from zero dose at 95% confidence level) were 0.5 pg/tube for T3 and 2.0 pg/tube for T4. As the extract from about 1 g of tissue was redissolved in 2.5 ml and only 0.5 ml was assayed, the least detectable amount of T3 and T4 in tissue can be estimated as 2.5 pg/g for T3 and 10 pg/g for T4.
Radioimmunoassays of T3 and T4 To measure T3 and T4 concentrations into aqueous sample after chromatography, the Amerlex-M T3 RIA kit (Amersham) and Lisophase FT 4 RIA kit
Table 1 - Clinical data of the fetuses (gestational age: 15-36 weeks) and neonates included in this study. Sex
Gestational age (weeks)
Birth weight (g)
F
24
730
2
F
35
2,060
3
F
23
470
Spontaneous abortion (encephalomeningocele)
Case n.
Cause of death
Spontaneous abortion Diaphragmatic hernia (stillbirth)
4
F
24
300
Spontaneous abortion
5
M
15
145
Spontaneous abortion
6
F
30
1.150
7
F
25
730
Spontaneous, abortion
8
M
25
870
Acute anoxya (stillbirth)
Acute anoxya (survival 1 h)
9
M
33
1,250
10
M
25
600
Spontaneous abortion (diaphragmatic hernia)
11
M
20
400
Spontaneous abortion (myelomeningocele)
12
M
36
1,620
Acute anoxya (survival 3 h)
Acute anoxya (stillbirth)
13
F
17
180
Spontaneous abortion
14
M
18
250
Spontaneous abortion (meningocele)
15
F
21
700
Spontaneous abortion
561
Costa A., Arisio H, Benedetto C, et aJ.
RESULTS
All samples were measured in duplicate or triplicate in a single assay; when sufficient extract was available, samples were analyzed in different assays, Concentrations were calculated on seven point standard curves (with tetraparametric fitting), and corrected for overall extraction recovery. The recovery of T3 and T4 in the overall extraction procedure was dependent on the tissue. For example, T3 recovery varies from 45 ± 8% in tongue and skeletal muscle to 65 ±% in colon and epiploon and 75 ± 7 % in omentum, with a mean recovery for all tissues of 54.8 ± 22.1 %. The recovery of T4 follows a similar trend, varying from 41 ± 8% in tongue, intestine and skeletal muscle to 67 ± 6% in omentum, with a mean value on all tissues of 45.3 ± 26.4%. These figures are in agreement with the values of recovery for T3 (50-75%) and T4 (40-60%) reported in reference (8). The intra- and interassay coefficients of variation were 11.2 ± 5.5 and 35.5 ± 10.4 respectively for T3, and 9.6 ± 7.0 and 31.5 ± 11.4 for T4.
Triiodothyronine and thyroxine concentrations obtained in tissues (carcass, limbs, liver and brain) from human conceptus between 6 and 12 weeks of gestation are reported in Table 2. In pooled limbs of 6-8 weeks T3 was barely measurable (0.01 ng/g). In the carcasses there was an increase in T3 and T4 concentrations from the 9th to the 12th week of gestation of about 40 and 20 times respectively. In particular, T4 increase was statistically significant (p = 0.G1) between 9-10 and 11-12 weeks. Considering all values reported for the period 9- 12 weeks a mean concentration of approximately 0.11 ng/g T3 and 1.28 ng/g for T4 was obtained. Table 3 gives the mean values of T3 and T4 concentrations, and the ratio T3!T4' in the different tissues isolated from the fetuses and neonates during the period 15-36 weeks, and from the adults. Only the first 7 tissues listed in Table 3 (brain, heart, kidney, liver, lung, skeletal muscle, skin) were measured both in the fetuses and neonates, and in the adults. Therefore, the discussion will refer mainly to them. Considering all together the values reported for those 7 tissues the following mean concentrations were obtained: 0.86 ng/g for T3 , 7.44 ng/g for T4 and 0.20 for the ratio T3!T4 in the fetuses and neonates, and 1.36 for T3, 12.75 for T4, and 0.20 for the ratio T3!T4 in the adults. Triiodothyronine and thyroxine concentrations in
Statistics
Regression analysis between age and T3 and T4 tissue concentrations were carried out by the statistical technique known as "best fit" and were expressed as polinomial equations; the variance for each fitted regression is reported.
Table 2 - Triiodothyronine and thyroxine tissue concentration (ng/g, mean ± SO) in human conceptus from 6 to 12 weeks of gestation t Gestational age (weeks)
T3
Carcass T3(T4 T4
6-8 9
11
T3(T4
0.01
1.16 (3)
0.008
T3
Liver T4
T3(T4
T3
Brain T4 T3(T4
0.06
0.32
0.08
1.22
0.05
0.04
0.68
±
±
±
±
±
±
±
±
±
0.09
0.88 (2) 1*
0.04
0.02
0.65 (3.3)
0.17
0.02 0.13 (2,2,2,3)
0.14
/
0.030
0.56
0.19
0.01
0.58
0.03
±
±
±
±
±
±
0.001
0.67 (2) 1*
0.26
0.035
4.24
0.Q7
0.035
2.07
0.01
0.13
1.56
±
±
±
±
±
±
±
±
±
0.013
2.83 (3) 5*
0.16
0.007
0.57 2*
0.01
0.035
1.82 (3.3)
0.24
0.01
0.54 (3)
0.01
(3)
9-11 12
Limbs T4
0.022 0.009
10
T3
0.85
1.31 1*
0.66
1The number of specimens included in each pool is indicated in brackets. Samples measured individually are indicated with an asterisk.
562
0.03
1.31 (6)
0.02
0.22
Thyroid hormones in human embryos and fetuses
Table 3 - Triiodothyronine and thyroxine tissue concentrations (mean ± SO, range) in the fetuses and neonates (F&N) listed in Table 2, and in the adults (A) included in this study. Tissues (n of cases)
T3(T4
T4 (ng/g)
T3 (ng/g)
F&N; A
F&N
A
F&N
A
F&N
A
Brain (12; 4)
0.94 ± 0.60 (0.24 - 2.14)
2.07 ± 2AO (0.02 - 4.65)
1.87 ± 0.80 (0.83 - 4.13)
2.58 ± 2.25 (0.52 - 547)
0.65 ± 0.67 (0.09 - 257)
0.50 ± 0.55 (0.03 - 1.10)
Tongue (9; -)
0.94 ± 1.30 (006 - 4.28)
Heart (8;6)
0.78 ± 0.64 (0.05 - 2.16)
0.75 ± 0.64 (0.20 - 159)
5.33±2.10 (1.83 - 8.68)
441 ± 2.82 (1.19 - 7.65)
0.16±0.13 (0.007 - OAO)
0.28 ± 0.35 (0.04 - 0.96)
Lung (4; 8)
0.33 ± 0.37 (0.10 - 088)
0.76± 142 (0.07 - 4.16)
8.69 ± 7.00 (147-17.9)
10.1 ± 6.98 (1.59 - 18.5)
0.05 ± 0.05 (0.01 - 0.13)
0.D7±0.14 (0.01 - 0.13)
liver (14; 6)
1.72 ± 1.51* (0.16 - 4.59)
345 ± 1.88 (1.30 - 5.30)
11 .37 ± 9.59*** (1.53 - 343)
46.33 ± 19.3 (28.9 - 70.0)
0.21 ± 0.21 (002 - 075)
0.06 ± 0.02 (0.02 - 0.09)
4.69 ± 2.50 (1.04 - 9.25)
0.22 ± 0.24 (0.01 - 0.75)
Spleen (-; 6)
0.15 ± 0.09 (0.04 - 0.29)
4.75 ± 3.29 (0.81 - 8.34)
0.05 ± 0.08 (001 - 0.24)
Stomach and Small intestine (-; 7)
0.35 ± 0.24 (0.16 - 041)
5.66 ± 5.29 (0.81 - 8.53)
0.06 ± 0.06 (0.02 - 023)
Kidney (3; 6)
0.36 ± 0.29 (0.16 - 0.64)
1.52 ± 0.91 (0.50 - 313)
749 ± 4.36 (3.84 - 10.2
10.6 ± 4.02 (5.09 - 14.9)
0.05 ± 0.04 (0.01-010)
0.16±0.13 (003 - 043)
Skin (9;5)
0.80 ± 0.80 (003 - 2.10)
0.07 ± 0.04 (0.03 - 0.10)
9.01 ± 5.29* (1 .61 -18 .0)
2.37 ± 2.20 (0.79 - 5.56)
0.08 ± 0.08 (0 .01 - 0.25)
0.06 ± 0.05 (0.01 - 0.12)
Skeletal muscle (13; 4)
0.92 ± 0.80 (0.05 - 275)
0.39 ± 0.27 (0.07 - 0.75)
6.59 ± 3.65** (1.13 - 16.5)
1.10 ± 0.04 (1.06 - 1.15)
0.17 ± 0.16 (0.01 - 0.57)
0.23 ± 0.17 (004 - 036)
Cartilage (3; -)
108 ± 041 (0.37 - 1.35)
3.75 ± 1.02 (2 .73 - 4.95)
0.24 ± 0.11 (0.11 - 0.34)
Statistical significance vs adults: *p < 0.05; **p < 0.01; ***p < 0.005
The problem of whether thyroid hormones are necessary for intrauterine brain development is still open. We have measured T3 and T4 in embrionic tissues of 6-8 weeks when major neuronal proliferative events start (18) . The carcass concentrations of T3 and T4 increased 40 and 20 times respectively from the 9th to the 11th - 12th week. The increase in T4 was particularly fast and statistically significant (p < 0.003) between the periods 9-10 and 11-12 weeks when thyroid follicles organization takes place, and iodine concentration and thyroid hormone synthesis have been demonstrated (by 74 day) (1). The problem of whether the small quantity of T4 measured before the 11th week of gestation, that is during the precolloid phase of thyroid development, is of fetal or maternal origin is still open. It is of interest that thyroid hormones have been found in rat embryonic tissues well before the onset of fetal thyroid function (7). The faster increase in T3 in comparison with T4, and
brain and liver, and the ratio T:JT4 in the same tissues at different weeks of gestation, and in the adults are reported in Figures 1-3. Hormonal ratios between blood and tissues at different weeks of gestation and in the adults are reported in Table 4. The cord blood values used for fetuses and neonates are those found by Fisher et al . (17) at different weeks of gestation. For adults a conventional blood value of 1.50 ng/ml for T3 and 80 ng/ml for T4 was adopted . DISCUSSION T3 and T4 tissues levels The significance of the difference in T3 and T4 concentrations in various tissues is not clear since, in the cells, thyroid hormones interact with ancillary factors, which modulate their activity obscuring the role played by their absolute concentrations . Moreover, the responsiveness of the target cells is influenced by the receptor status.
563
Costa A , Arisio R , Benedetto C, et at.
Table 4 - Hormonal ratios between blood and tissues. A: ratios cord blood/tissues in fetuses (ng/ml / ng/g) B: ratios cord venous blood/tissues in adults (ng/ml / ng/g). A Fetuses age (weeks)
Cord blood T3
T4
Blood/brain T3
T4
Blood/liver
T3
T4
Blood/heart T3
T4
Blood/lung T3
T4
Blood/kidney T3
blood/m uscle
T4
T3
T4
0.05
5.18 2.80
5.00
4.20
/
/
15 (1)1
Thyroid hormones in tissues from human embryos and fetuses A. Costa *, R. Arisio**, C. Benedetto***, E. Bertino****, C. Fabris****, G. Giraudi*****, L. Marozio***, V. Maula******, M. Pagliano***, O. Testori*******, and G. Zoppetti****** ******Divisione di Endocrinologia e *******Servizio di Medicina Nucleare, *Ospedale Mauriziano Umberto I, **Divisione di Anatomia Patologica e Ricerche Cliniche, Ospedale S. Anna, ***Istituto di Ostetricia e Ginecologia, ****Istituto di Discipline Pediatriche, Servizio di Neonatologia, *****Dipartimento di Chimica Analitica, Universita degli Studi di Torino, Torino, Italy concentrations of 40 and 20 times respectively from the 9th to the 12th week, when thyroid follicles oirganization takes place. In fetuses and adults T3 and T4 were measured in brain, heart, kidney, liver, lung, skeletal muscle and skin (mean concentrations: 0.86 ng/g for T 3 and 7.44 ng/g for T 4 in fetuses and neonates; 1.36 ng/g for T 3 and 12.75 ng/g for T 4 in adults). Hormones concentration increased with gestational age; the T ~ 4 ratio increased until 22-24 weeks, when the prevalent increment in T4 occurs. T 3 concentration up to 30 weeks was generally higher in tissues than in cord serum of the corresponding age. During the last month of gestation T 3 increment was faster in serum. T4 level was always predominant in serum. In conclusion, T3 and T4 have been detected in the limbs of embryos before the onset of thyroid hormone secretion. Concentrations were 1/150 and 1170, of the normal maternal blood values respectively. It is conceivable that these hormones are of maternal origin, and the question of whether such small quantities may playa role in fetal development is open.
ABSTRACT. This study was intended to quantify T3 and T 4 in various human tissues at different stages of gestation as a contribute in the evaluation of the role of thyroid hormones in fetal development, particularly before the maturation of fetal thyroid function. Moreover, for a better comprehension of the influence of thyroid hormone status in tissues, the study was extended to adults. Embryonic specimens were obtained from voluntary abortions between 6 and 12 weeks of gestation, fetal and neonatal specimens from fetuses and neonates between 15 and 36 weeks of gestation after spontaneous abortion or stillbirth, and adult specimens from men (age range: 45-65 years) after death for cardiovascular diseases. Thyroid hormones were measured by the method of Gordon and coworkers. In embryos T3 and T4 were measured in limbs, carcasses, brain and liver: considering all values measured in the period 9-12 weeks, a mean concentration of 0.11 ng/g for T 3 and 1.28 ng/g for T4 was obtained. In pooled limbs of 68 weeks T3 was barely measurable (0.01 ng/g). In the carcasses there was an increase in T 3 and T 4
INTRODUCTION
have been detected in the developing human cerebral cortex at the 11th week of gestation (3). Both T3 and its nuclear receptor have been found in the brain of a 10 week embryo (4), and T3 receptor and T4 could also be measured in a 7-week-old whole embryo (5). Serum thyroxine was identified in fetus blood after 11 weeks of gestation (6). The transfer of thyroid hormones from mother to embryo has been proved in different species, in which maternal hypothyroidism is accompanied by thyroid hormone deficiency of the conceptus before the onset of fetal thyroid activity (7-9). Such transfer in humans is still controversial. Raiti et al. (10) have reported that significant quantities of T3 cross the placenta in at least 50 per cent of the cases when high doses of T3 was administered to normal pregnant women for four to six weeks before delivery.
The data of the literature on the presence of thyroid hormones in tissues obtained from human embryos and fetuses are limited, especially in the formers. Embryonic thyroids in organ culture acquire the capacity to concentrate radioiodine and synthesize iodothyronines by 74 days of conceptional age, equivalent to 12 weeks of gestation (1). Labelled T3 and T4 have been demonstrated in thyroid glands from fetuses of 11 weeks of gestation, after the mother had undergone a 131-1 test (2). Moreover, thyroid hormones as well as 5' monodeiodinase activities
Key-words: Thyroid hormones, embryos, fetuses. Correspondence: A. Costa, Ospedale Mauriziano Umberto I, L.go Turati 62,10100 Torino, Italy. Received July
26,1990; accepted March 18, 1991.
559
Costa A., Arisio R., Benedetto C, et al.
Grumbach and Werner (11) concluded that the human placenta is capable of transferring thyroid hormones. However, their transfer from the mother is considered insufficient to support normal skeletal and brain development in case of primary thyroidal defect of the fetus. Pharoah and Connolly (12) observed that neurological endemic cretinism can be prevented only by giving iodine to the mother before conception; therefore maternal thyroid hormones, specifically T4, may have a role in the maturation of fetal nervous system before the onset of fetal thyroid function. In late gestation, the transfer of thyroid hormones is well documented, at least when fetal thyroid is impaired . Indeed infants with thyroid agenesis or with organification defects have detectable values of T3 and T4 at birth; the rapid decline of their levels suggest that they are of maternal origin (13). Based upon these assumptions, we have measured the concentrations of T3 and T4 in various human tissues at different stages of gestation in order to clarify the possible role of thyroid hormones in fetal development and maturation , particularly before the onset of fetal thyroid function . Moreover, to understand more precisely the specific influence of thyroid hormone status in tissues we have extended our study to adults.
sected from the body. Skin samples were taken from the chest or abdominal wall, the subcutaneous fat was trimmed off and the surface cleaned with a gauze moistened with phosphate buffer saline (PBS); myocardial tissue was collected from the right ventricle and cerebral tissue from the frontal lobe of the forebrain. All tissues were isolated with a semisterile technique, quickly rinsed in cold PBS (4 C) and stored in liquid nitrogen until assay. Adult specimens were obtained from 7 men (age range: 45-65 years) after death for cardiovascular diseases. Subjects with endocrine or metabolic diseases were excluded. The samples were taken at autopsy 24 hours after death . Homogenization and extraction of the tissues Human tissues were homogenized in methanol containing 1 mM PTU (6-propyl-2-thiouracil). About 1 g of tissue was added to a Polytron glass tube, containing 5 ml of methanol-PTU/g wet tissue and approximately 300 cpm of 125-I-T3 (specific activity 1200 IlCi/llg). The homogenization was carried out at 4 C by a Polytron. The homogenates were transferred to stoppered test tube, then chloroform was added in a volume double than the amount of methanol. A mixture of chloroform-methanol (2: 1) was thereafter used for the extraction as described by Gordon et al. (14), Folch et al. (15), and Morreale de Escobar et al. (8). After centrifugation of the mixture for 15 min at 2,000 rpm , the extract was removed and transferred in a glass beker. The particulate matter was extracted again by shaking with three successive portions of chloroform-methanol (2: 1), each followed by centrifugation and removal of extract. The extracts were pooled, filtered on filter paper and transferred to stoppered glass tubes for backextraction of the iodothyronines into an aequeous phase. Two ml of 0.05% CaC12 were mixed with 10 ml extract and the mixture was allowed to stand at 4 C for approximately one hour to allow separation of two clear layers. The upper phase was retained and the interphasic surface was then rinsed with a small amount of pure upper layer in a volume equal to that removed (2 ml). All the upper layers and rinses were collected in a glass baker and evaporated under nitrogen to complete dryness. The residue was dissolved in 10 ml methanol and the solution was transferred into glass tube and centrifuged. The supernatant was evaporated to dryness under vacuum at 37 C, then the residue was treated with 2.5 ml PBS (0 .1 M, pH 7.4)/g homogenized tissue. Finally, the aqueous solution was centrifuged and the supernatant was further purified by column chromatography as described by Mallol et al. (16).
MATERIALS AND METHODS Human samples Embryonic specimens were obtained from 33 voluntary abortions between 6 and 12 weeks of gestation. The samples were washed with physiologic saline and kept in glass tubes at 4 C until dissection, which was performed within 30 min from uterine aspiration . Specimens of brain, liver, limbs and carcass (body without head and limbs) were collected separately. When the weight of the sample was less than 1 g, different specimens of the same tissue were pooled together. The gestational week was calculated on the basis of the first day of the last menstruation and confirmed by ultrasounds. Patients with a history of thyroid disease were excluded from the study. The study was approved by the Ethical Committee of the Istituto di Ostetricia e Ginecologia. All patients gave informed consent. Fetal and neonatal specimens were obtained from 13 fetuses and 2 neonates, without macroscopical and histological evidence of maceration, between 15 and 36 weeks of gestation (Table 1). The samples were taken within 24 hours from spontaneous abortion or stillbirth. Specimens of brain, heart, kidney, liver, lung, skeletal muscles and skin were dis-
560
Thyroid hormones in human embryos and fetuses
Column chromatography For column chromatography the Bio-Rad resin (BioRad 1 x 2, 200-400 mesh, chloride form) was first suspended in 1.2 M acetic acid and poured into 2.5 ml plastic syringes up to 1.0 ml mark. The resin was then washed several times with distilled water until the pH was 3-4 and with acetate buffer (0.2M, pH 7.0). The resin was prevented from passing through the syringe by 2 mm layer of paper. The acetate buffer was drained just before the aqueous sample (the extract) was added. Then the column was washed with 2 ml of acetate buffer (0.2 M, pH 7) and with 4 ml of ethanol. After further washing with 2 ml of acetate buffer (0.2 M) at pH 4.0, 2 ml at pH 3.0, followed by 2 ml of 1% and 2 ml of 35% acetic acid, iodothyronines were eluted with seven or eight fractions (0.5 ml) of 70% acetic acid. The fractions of eluate containing the higher radioactivity were collected and evaporated under nitrogen. The dry sample was then counted in a gamma-counter to evaluate the overall recovery, then was dissolved in 2.5 ml PBS and used to determine the T3 and T4 concentration by RIA test.
(Sclavo) were slightly modified to achieve maximum sensitivity. Standard solutions of T3 and T4 were prepared by dissolving the authentic hormone in 0.1 M PBS (pH 7.4, 0.05% NaN3), and diluting it with the same buffer to working concentrations. To determine T3, 0.2 ml of Amerlex-M T3 antibody were reacted with 0.2 ml of 125-I-T3 tracer from the kit and 0.5 ml of standard or sample. The assay was then performed at room temperature for 3 hours. To determine T4, 0.1 ml of T4 antibody from FT4 RIA kit were reacted at 4 C overnight with 0.1 ml of 125-I-T4 tracer from the same kit and 0.5 ml standard of sample. The separation of antibody-bound from free tracer was performed on Lisophase column according to the instruction of manufacturer. The accuracy of these modified assays was proved by dilution and recovery tests on aequeous samples after chromatography. In these modified assays the least detectable doses (calculated as the dose that gives responses statistically different from zero dose at 95% confidence level) were 0.5 pg/tube for T3 and 2.0 pg/tube for T4. As the extract from about 1 g of tissue was redissolved in 2.5 ml and only 0.5 ml was assayed, the least detectable amount of T3 and T4 in tissue can be estimated as 2.5 pg/g for T3 and 10 pg/g for T4.
Radioimmunoassays of T3 and T4 To measure T3 and T4 concentrations into aqueous sample after chromatography, the Amerlex-M T3 RIA kit (Amersham) and Lisophase FT 4 RIA kit
Table 1 - Clinical data of the fetuses (gestational age: 15-36 weeks) and neonates included in this study. Sex
Gestational age (weeks)
Birth weight (g)
F
24
730
2
F
35
2,060
3
F
23
470
Spontaneous abortion (encephalomeningocele)
Case n.
Cause of death
Spontaneous abortion Diaphragmatic hernia (stillbirth)
4
F
24
300
Spontaneous abortion
5
M
15
145
Spontaneous abortion
6
F
30
1.150
7
F
25
730
Spontaneous, abortion
8
M
25
870
Acute anoxya (stillbirth)
Acute anoxya (survival 1 h)
9
M
33
1,250
10
M
25
600
Spontaneous abortion (diaphragmatic hernia)
11
M
20
400
Spontaneous abortion (myelomeningocele)
12
M
36
1,620
Acute anoxya (survival 3 h)
Acute anoxya (stillbirth)
13
F
17
180
Spontaneous abortion
14
M
18
250
Spontaneous abortion (meningocele)
15
F
21
700
Spontaneous abortion
561
Costa A., Arisio H, Benedetto C, et aJ.
RESULTS
All samples were measured in duplicate or triplicate in a single assay; when sufficient extract was available, samples were analyzed in different assays, Concentrations were calculated on seven point standard curves (with tetraparametric fitting), and corrected for overall extraction recovery. The recovery of T3 and T4 in the overall extraction procedure was dependent on the tissue. For example, T3 recovery varies from 45 ± 8% in tongue and skeletal muscle to 65 ±% in colon and epiploon and 75 ± 7 % in omentum, with a mean recovery for all tissues of 54.8 ± 22.1 %. The recovery of T4 follows a similar trend, varying from 41 ± 8% in tongue, intestine and skeletal muscle to 67 ± 6% in omentum, with a mean value on all tissues of 45.3 ± 26.4%. These figures are in agreement with the values of recovery for T3 (50-75%) and T4 (40-60%) reported in reference (8). The intra- and interassay coefficients of variation were 11.2 ± 5.5 and 35.5 ± 10.4 respectively for T3, and 9.6 ± 7.0 and 31.5 ± 11.4 for T4.
Triiodothyronine and thyroxine concentrations obtained in tissues (carcass, limbs, liver and brain) from human conceptus between 6 and 12 weeks of gestation are reported in Table 2. In pooled limbs of 6-8 weeks T3 was barely measurable (0.01 ng/g). In the carcasses there was an increase in T3 and T4 concentrations from the 9th to the 12th week of gestation of about 40 and 20 times respectively. In particular, T4 increase was statistically significant (p = 0.G1) between 9-10 and 11-12 weeks. Considering all values reported for the period 9- 12 weeks a mean concentration of approximately 0.11 ng/g T3 and 1.28 ng/g for T4 was obtained. Table 3 gives the mean values of T3 and T4 concentrations, and the ratio T3!T4' in the different tissues isolated from the fetuses and neonates during the period 15-36 weeks, and from the adults. Only the first 7 tissues listed in Table 3 (brain, heart, kidney, liver, lung, skeletal muscle, skin) were measured both in the fetuses and neonates, and in the adults. Therefore, the discussion will refer mainly to them. Considering all together the values reported for those 7 tissues the following mean concentrations were obtained: 0.86 ng/g for T3 , 7.44 ng/g for T4 and 0.20 for the ratio T3!T4 in the fetuses and neonates, and 1.36 for T3, 12.75 for T4, and 0.20 for the ratio T3!T4 in the adults. Triiodothyronine and thyroxine concentrations in
Statistics
Regression analysis between age and T3 and T4 tissue concentrations were carried out by the statistical technique known as "best fit" and were expressed as polinomial equations; the variance for each fitted regression is reported.
Table 2 - Triiodothyronine and thyroxine tissue concentration (ng/g, mean ± SO) in human conceptus from 6 to 12 weeks of gestation t Gestational age (weeks)
T3
Carcass T3(T4 T4
6-8 9
11
T3(T4
0.01
1.16 (3)
0.008
T3
Liver T4
T3(T4
T3
Brain T4 T3(T4
0.06
0.32
0.08
1.22
0.05
0.04
0.68
±
±
±
±
±
±
±
±
±
0.09
0.88 (2) 1*
0.04
0.02
0.65 (3.3)
0.17
0.02 0.13 (2,2,2,3)
0.14
/
0.030
0.56
0.19
0.01
0.58
0.03
±
±
±
±
±
±
0.001
0.67 (2) 1*
0.26
0.035
4.24
0.Q7
0.035
2.07
0.01
0.13
1.56
±
±
±
±
±
±
±
±
±
0.013
2.83 (3) 5*
0.16
0.007
0.57 2*
0.01
0.035
1.82 (3.3)
0.24
0.01
0.54 (3)
0.01
(3)
9-11 12
Limbs T4
0.022 0.009
10
T3
0.85
1.31 1*
0.66
1The number of specimens included in each pool is indicated in brackets. Samples measured individually are indicated with an asterisk.
562
0.03
1.31 (6)
0.02
0.22
Thyroid hormones in human embryos and fetuses
Table 3 - Triiodothyronine and thyroxine tissue concentrations (mean ± SO, range) in the fetuses and neonates (F&N) listed in Table 2, and in the adults (A) included in this study. Tissues (n of cases)
T3(T4
T4 (ng/g)
T3 (ng/g)
F&N; A
F&N
A
F&N
A
F&N
A
Brain (12; 4)
0.94 ± 0.60 (0.24 - 2.14)
2.07 ± 2AO (0.02 - 4.65)
1.87 ± 0.80 (0.83 - 4.13)
2.58 ± 2.25 (0.52 - 547)
0.65 ± 0.67 (0.09 - 257)
0.50 ± 0.55 (0.03 - 1.10)
Tongue (9; -)
0.94 ± 1.30 (006 - 4.28)
Heart (8;6)
0.78 ± 0.64 (0.05 - 2.16)
0.75 ± 0.64 (0.20 - 159)
5.33±2.10 (1.83 - 8.68)
441 ± 2.82 (1.19 - 7.65)
0.16±0.13 (0.007 - OAO)
0.28 ± 0.35 (0.04 - 0.96)
Lung (4; 8)
0.33 ± 0.37 (0.10 - 088)
0.76± 142 (0.07 - 4.16)
8.69 ± 7.00 (147-17.9)
10.1 ± 6.98 (1.59 - 18.5)
0.05 ± 0.05 (0.01 - 0.13)
0.D7±0.14 (0.01 - 0.13)
liver (14; 6)
1.72 ± 1.51* (0.16 - 4.59)
345 ± 1.88 (1.30 - 5.30)
11 .37 ± 9.59*** (1.53 - 343)
46.33 ± 19.3 (28.9 - 70.0)
0.21 ± 0.21 (002 - 075)
0.06 ± 0.02 (0.02 - 0.09)
4.69 ± 2.50 (1.04 - 9.25)
0.22 ± 0.24 (0.01 - 0.75)
Spleen (-; 6)
0.15 ± 0.09 (0.04 - 0.29)
4.75 ± 3.29 (0.81 - 8.34)
0.05 ± 0.08 (001 - 0.24)
Stomach and Small intestine (-; 7)
0.35 ± 0.24 (0.16 - 041)
5.66 ± 5.29 (0.81 - 8.53)
0.06 ± 0.06 (0.02 - 023)
Kidney (3; 6)
0.36 ± 0.29 (0.16 - 0.64)
1.52 ± 0.91 (0.50 - 313)
749 ± 4.36 (3.84 - 10.2
10.6 ± 4.02 (5.09 - 14.9)
0.05 ± 0.04 (0.01-010)
0.16±0.13 (003 - 043)
Skin (9;5)
0.80 ± 0.80 (003 - 2.10)
0.07 ± 0.04 (0.03 - 0.10)
9.01 ± 5.29* (1 .61 -18 .0)
2.37 ± 2.20 (0.79 - 5.56)
0.08 ± 0.08 (0 .01 - 0.25)
0.06 ± 0.05 (0.01 - 0.12)
Skeletal muscle (13; 4)
0.92 ± 0.80 (0.05 - 275)
0.39 ± 0.27 (0.07 - 0.75)
6.59 ± 3.65** (1.13 - 16.5)
1.10 ± 0.04 (1.06 - 1.15)
0.17 ± 0.16 (0.01 - 0.57)
0.23 ± 0.17 (004 - 036)
Cartilage (3; -)
108 ± 041 (0.37 - 1.35)
3.75 ± 1.02 (2 .73 - 4.95)
0.24 ± 0.11 (0.11 - 0.34)
Statistical significance vs adults: *p < 0.05; **p < 0.01; ***p < 0.005
The problem of whether thyroid hormones are necessary for intrauterine brain development is still open. We have measured T3 and T4 in embrionic tissues of 6-8 weeks when major neuronal proliferative events start (18) . The carcass concentrations of T3 and T4 increased 40 and 20 times respectively from the 9th to the 11th - 12th week. The increase in T4 was particularly fast and statistically significant (p < 0.003) between the periods 9-10 and 11-12 weeks when thyroid follicles organization takes place, and iodine concentration and thyroid hormone synthesis have been demonstrated (by 74 day) (1). The problem of whether the small quantity of T4 measured before the 11th week of gestation, that is during the precolloid phase of thyroid development, is of fetal or maternal origin is still open. It is of interest that thyroid hormones have been found in rat embryonic tissues well before the onset of fetal thyroid function (7). The faster increase in T3 in comparison with T4, and
brain and liver, and the ratio T:JT4 in the same tissues at different weeks of gestation, and in the adults are reported in Figures 1-3. Hormonal ratios between blood and tissues at different weeks of gestation and in the adults are reported in Table 4. The cord blood values used for fetuses and neonates are those found by Fisher et al . (17) at different weeks of gestation. For adults a conventional blood value of 1.50 ng/ml for T3 and 80 ng/ml for T4 was adopted . DISCUSSION T3 and T4 tissues levels The significance of the difference in T3 and T4 concentrations in various tissues is not clear since, in the cells, thyroid hormones interact with ancillary factors, which modulate their activity obscuring the role played by their absolute concentrations . Moreover, the responsiveness of the target cells is influenced by the receptor status.
563
Costa A , Arisio R , Benedetto C, et at.
Table 4 - Hormonal ratios between blood and tissues. A: ratios cord blood/tissues in fetuses (ng/ml / ng/g) B: ratios cord venous blood/tissues in adults (ng/ml / ng/g). A Fetuses age (weeks)
Cord blood T3
T4
Blood/brain T3
T4
Blood/liver
T3
T4
Blood/heart T3
T4
Blood/lung T3
T4
Blood/kidney T3
blood/m uscle
T4
T3
T4
0.05
5.18 2.80
5.00
4.20
/
/
15 (1)1
