119
Molecular and Cellular Endocrinology, 88 (1992) 119-128
0 1992 Elsevier Scientific Publishers Ireland, Ltd. 0303-7207//$05.00 MOLCEL 02834
Calbindin-D,,
gene expression during pregnancy and lactation in the rat J. Krisinger, J.L. Dann, E.B. Jeung and P.C.K. Leung
Department of Obstetrics and Gynecology, University of British Columbia, Vancouuer, B.C. V52 4H4, Canada
(Received 4 June 1992; accepted 28 June 1992)
Key words: Calbindin-D,,;
Gene expression; Pregnancy; Lactation; (Rat)
Summary Calbindin-D,, (CaBP-9k) is a calcium binding protein expressed in mammalian intestine, uterus and placenta. It is believed to be involved in transepithelial calcium transport in intestine and placenta and regulation of cytosolic calcium concentration in uterus. CaBP-9k mRNA levels were measured by Northern blot analysis in maternal duodenum, uterus, placenta and fetal/neonatal duodenum during pregnancy and lactation. In maternal duodenum a maximal increase occurred at day 15 of lactation (2.3-fold) and 20 days post-lactation levels decrease to 30.3% of non-pregnant controls. In non-pregnant uterus a lo-fold variation of CaBP-9k mRNA levels was observed between individual animals despite a uniform expression of p-actin. During pregnancy high CaBP-9k expression is found, averaging about 20% of duodenal levels, which abruptly drops below detection during early lactation. At late lactation CaBP-9k mRNA levels are again subject to great variation ranging from no expression to maximal levels found in the non-pregnant uterus. Placental CaBP-9k is maximally expressed at the end of pregnancy (day 20) reaching about 2.5% of duodenal levels. Fetal intestinal CaBP-9k mRNA was detectable in 20 Fg total RNA at day 18 of pregnancy and rose sharply in early lactation reaching about 50% of adult duodenal levels at day 20 lactation. The profound changes of uterine CaBP-9k mRNA in non-pregnant (cycling), pregnant, and lactating rats indicate a rapid hormonal regulation of gene expression, most likely involving 17P-estradiol.
Introduction Calbindin-D,, (CaBP-9k)is a cytosolic calcium binding protein 04, 9000) expressed predominantly in mammalian duodenum, placenta, and uterus (Christakos et al., 1989). It is thought to be
Correspondence to: Dr. J. Krisinger, The Research Center, Dept. of OB/GYN, 178-950 W. 28th Ave., Vancouver, B.C. VSZ 4H4, Canada. Tel. (6041875-2433; Fax 875-2496. This work was supported by grants from the British Columbia Health Research Foundation No. 7(90-2) and the Medical Research Council of Canada.
involved in transepithelial calcium transport in duodenum (Roche et al., 1986; Wasserman and Fullmer, 1986) and placenta (Mathieu et al., 1989) and to play a role in regulating intracellular calcium levels in myometrium (Mathieu et al., 1989). The regulation of the expression in duodenum by 1,25_dihydroxyvitamin D, has been well documented (Kessler and DeLuca, 1985; Brehier and Thomasset, 1990; Krisinger et al., 19901, whereas regulation of placental CaBP-9k is not understood. In uterus estrogen regulation has been demonstrated (Mathieu et al., 1989; Darwish et al., 1990; L’Horset et al., 1990) and is not af-
120
fected by vitamin D metabolites (Darwish et al., 1990). In the ovariectomized rat, uterine CaBP-9k mRNA levels fall below detection and high expression can be restored by 17P-estradiol injection (Darwish et al., 1990; L’Horset et al., 1990). Gel retardation and transfection studies have revealed an estrogen response element in the rat CaBP-9k gene, responsible for the transcriptional regulation of the uterine expression similar to other genes controlled by steroid hormones. Studies in pregnant rats have also shown gestational changes of CaBP-9k protein levels in uterus (Mathieu et al., 1989). Immunohistochemical studies found expression of CaBP-9k mainly in myometrium (Delorme et al., 1983; Warembourg et al., 1987; Bruns et al., 1988). Myometrial expression of CaBP-9k and estrogen control of its expression have led to the hypothesis that this protein participates in the control of uterine activity during gestation (Bruns et al., 1988). In particular, a role for this calcium binding protein in the development of contractions at parturition has been postulated in regard to the pivotal role of intra-myometrial calcium ions for uterine activity (Kyozuka et al., 1987; Savineau et al., 1988). In order to further understand the regulation of CaBP-9k around the time of parturition, the expression of the gene encoding CaBP-9k through late pregnancy and lactation was examined in rat uterus. In addition to the uterine regulation, changes in maternal duodenal, placental and fetal/neonatal duodenal mRNA levels were also determined. Materials
and methods
Animals
Three-month-old Sprague-Dawley (S.D.) rats were kept on a standard laboratory diet (1% calcium, 0.61% phosphorus, 4.5 II-I/g vitamin D), distilled water and a 12 h day/night cycle. For timed pregnancies, females were mated with male S.D. rats overnight and insemination was confirmed by the presence of sperm in vaginal smear the following morning (day 0 of pregnancy). Animals were sacrificed at day 16, 18, 20 of pregnancy and day 5, 10, 15, 20 of lactation and 20 days post-lactation, respectively. Maternal and fetal/neonatal weight and litter size were deter-
TABLE
1
MATERNAL AND LITTER
WEIGHT, FETAL/NEONATAL SIZE OF EACH EXPERIMENTAL Maternal weight (g)
Fetal/neonatal weight (g)
WEIGHT GROUP
Litter size
Pregnancy day 16 Pregnancy day 18 Pregnancy day 20 Lactation day 5 Lactation day 10 Lactation day 15 Lactation day 20 20 days postlactation
17
0.266 + 0.017
12.2 f 2.7
322 f 32
0.760 f 0.069
13.0 + 2.4
343 k 32
2.2
kO.12
11.7k3.8
296 + 10
9.98
* 1.33
11.3k3.2
31s+
304*11
16.8
+1.60
12.3+ 1.2
316 k 16
32.9
k3.14
7.8 f 2.5
300 & 29
41.2
k8.18
9.7 k 2.8
28Yk16
-
_
mined (Table 1). Each group consisted of six animals. The experimental protocol was approved by the University of British Columbia Animal Care Committee. Tissue and RNA preparation 8 cm of the proximal duodenum
of the adult animals were removed. The entire uterus of nonpregnant and lactating animals or 1 g of uterine tissue of pregnant animals was collected. The fetal parts of the placentae were removed and pooled to yield 1 g of total tissue. Duodena were removed from fetuses on day 18 and 20 of pregnancy, pooled from each litter for preparation of RNA. 1 g of duodenal tissue was collected from suckling pups on day 5, 10, 15, and 20 after birth. Tissues were rinsed in saline and homogenized in 4 M guanidinium isothiocyanate buffer for 30 s with a polytron. Total RNA was prepared using the guanidinium isothiocyanate/CsCl procedure (Glisin et al., 1974; Chomczynski and Sacchi, 1987). RNA concentration was determined by UV spectrophotometry. Quantification of CaBP-9k mRNA 5 pg (maternal duodenum) or 20 Fg (other
tissues) total RNA was separated
on a standard
121
denaturing gel containing formaldehyde. The samples included 0.5 pg ethidium bromide to visualize the RNA after electrophoresis. Ethidium bromide staining of the ribosomal RNA was
MATERNAL
(A)
COmrOl
P, 5
PI8
used to confirm equal loading and integrity RNA. RNA was transferred onto nylon brane and probed with a 295 nucleotide rat probe. The probe consisted of the 5’ part
DUODENUM
P20
L5
LlO
MATERNAL
CC)
L15
L20
control
PL20
PI 5
PI8
P20
of the memcDNA of the
DUODENUM
L5
LIO
L15
L20
PLZO
(8)
C
C
P16
P16
P18
P18
P20
L5
LlO
L15
L20
PL20
Fig. 1. CaBP-9k mRNA levels during pregnancy and lactation in rat duodenum. Non-pregnant animals (control), pregnant animals on days 16, 18, 20 of pregnancy (P16-20) and lactating animals on days 5, 10, 15, 20 of lactation (L5-20) and 20 days post-lactation (PL20) were sacrificed. 8 cm of the proximal duodenum were removed for preparation of total RNA. 5 pg of total RNA were electrophoresed on 1% denaturing agarose and transferred onto nylon membrane. The Northern blot was hybridized to a 295 nt rat CaBP-9k cDNA probe. After autoradiography CaBP-9k was quantified by densitometry. A: Each group consisted of six animals. Shown are the means f SEM for each group. Values are given in arbitrary units. * *, p < 0.001; *, p < 0.05 as compared to control values. B: A representative RNA sample of each group was used for Northern blot analysis. Groups are as in A. Shown are the autoradiograph and the ethidium bromide stained RNA gel. C: The same blot as in A was probed with a 1.2 kb bovine /3-actin cDNA. Results of the densitometric scanning are illustrated as in A.
122
rat CaBP-9k cDNA including coding region for 78 out of 79 amino acids (Darwish et al., 1987). DNA was labelled by random primed synthesis using [ a3* P]dCTP. For quantification of p-actin mRNA, a 1.2 kilobasepair bovine p-actin fragment (Degen et al., 1983) was labelled. After stringent hybridization (50% formamide) and washing, autoradiography was performed. Autoradiographs were scanned with a BioRad densitometer. The area under the curve value of each band was multiplied by 100 and used as arbitrary units. Where values are compared to controls, the
Results CaBP-9k mRNA levels in maternal duodenum
At day 16, 18 and 20 of gestation, CaBP-9k mRNA levels dropped slightly in comparison to non-pregnant age-matched, control rats (Fig. 1A). CC)
UTERUS
(A)
control was set as 100 units. Results were analyzed by analysis of variance. Individual groups were compared using the Student-Fisher r-test. Results are expressed as mean + SEM. Each group consisted of 5-6 animals.
UTERUS
300
Control
P 15
P18
P20
L5
LIO
L15
L20
PL20 Control
PI 6
PI8
P2O
L5
L10
L15
L20
PL20
(B)
Fig. 2. CaBP-9k mRNA levels during pregnancy and lactation in rat uterus. Groups are as in Fig. 1. 20 kg of total RNA were analyzed. A: Results of densitometric scanning of each individual sample are indicated. Due to the great variation in each group means f SEM were not calculated (N.D. = not detectable) B: Representative RNA samples of each group were used for Northern analysis. For the non-pregnant controls, rats on days 15 and 20 of lactation and 20 days post-lactation, two samples representing the highest and lowest value of each group were used. The autoradiograph and the ethidium bromide stained gel are shown. C: The same blot was hybridized with the @-actin probe. Results of densitometric scanning of each group are shown as mean+ SEM. (* p < 0.05; * *, p < 0.0001).
123
At day 18, the values decreased significantly (p < 0.05) to 52.3% of controls. During lactation, levels increased steadily to day 15 (225.7%, p < O.OOl),compared to non-pregnant controls. Interestingly, at day 20 post-lactation levels declined significantly (p < 0.001) to 30.3% of non-pregnant control levels. 5 pg total RNA was used in Northern blot analysis for densitometric scanning after autoradiography. Uniform loading of RNA was assured by measuring absorbance of 260 nm and ethidium bromide staining of the RNA samples (see Fig. 1B). When control probing using a bovine P-actin cDNA was performed, dynamic changes were observed which could not be explained by inconsistent RNA loading. Fig. 1B demonstrates a decline of p-actin in duodenum of pregnant and lactating animals except for day 15 of lactation. On day 16, 18, and 20 (t, < 0.01) of gestation, day 5 (p < 0.05) and day 10 (p < 0.02) of lactation and day 20 (p < 0.05) post-lactation, this decrease was significantly different from controls.
PLACENTA
(A)
P20
(6)
CaBP-9k mRNA levels in uterus
When CaBP-9k mRNA levels in uterus were determined, substantial variation was observed in non-pregnant control animals and in lactating animals. Fig. 2 A shows results of densitometric scanning of autoradiographs using 20 pg of total RNA. Variation in the control group reached from 16.7 to 171 arbitrary units (n = 6). Integrity of the RNA was assessed by ethidium bromide staining of the 18 and 28 S ribosomal RNAs and control probing with bovine p-actin cDNA. Throughout the experimental groups, each individual result is shown as a bar in Fig. 2A. During pregnancy, especially at day 18, levels were relatively uniform between 71-106 densitometric units (n = 5). At early lactation, CaBP-9X mRNA levels fell below detection levels in all but two animals, one at day 5 and one at day 10. On day 15 and 20 of lactation and day 20 post-lactation, the groups again showed substantial variation of CaBP-9k mRNA. While in some animals no CaBP-9k mRNA was detected (at day 15 and day 20 of lactation four animals, and at day 20 postlactation one animal), other individuals bad substantial amounts of CaBP-9k mRNA reaching up to 77 arbitrary units. Fig. 2C shows the results of
PI6
P16
PI8
PI8
P20
P20
Fig. 3. CaBP-9k mRNA in rat placenta. A: Animals on days 16, 18 and 20 of pregnancy were sacrificed and the fetal portions of the placentae removed for preparation of total RNA. 20 Fg of RNA was used for CaBP-9k quantification as described above. Means+ SEM of each group (n = 6) are shown. The level at day 18 increased significantly, ** p < 0.001, when compared to day 16. A further significant increase occurred between day 18 and 20, * p < 0.01. B: Representative RNA samples of each group were used for Northern analysis. The autoradiograph and the ethidium bromide stained gel are shown.
124
(A)
FETAIAEONATAL
DUODENUM
(B)
P18
P18
P20
P20
L5
LIO
L15
L5
LlO
L15
L20
L20
Fig. 4. CaBP-9k mRNA levels in fetal/neonatal duodenum. A: Animals were sacrificed on days 18 and 20 of pregnancy and days 5-20 of lactation (U-20) for preparation of total RNA from duodenum. CaBP-9k mRNA was quantified as above. Levels on day 10. were significantly lower compared to day 5, * * p < 0.001. Levels on day 15 were significantly when compared to levels on day 10, * p < 0.05. A further significant increase occurred between day 15 and day 20 of ** p < 0.001. B: Representative RNA samples of each group were used for Northern analysis. The autoradiograph ethidium bromide stained RNA gel are shown.
(Pig, P20) described increased lactation, and the
densitometric scanning of the same blot probed with the /3-actin probe. Values in each group were uniform and the mean f SEM were calculated. p-Actin mRNA levels increased slightly but not significantly at day 16 and day 18 of pregnancy. At day 20 of pregnancy and day 5 and day 10 of lactation, @-actin levels were decreased. On day 5 of lactation this decrease was significant (54.3% of non-pregnant controls). During lactation a steady increase occurred peaking at day 20 (244.1% of controls). On day 20 post-lactation levels again returned to near non-pregnant control levels.
UT
D
PL
ND
CaBP-9k mRNA leveki in placenta
Preliminary experiments had shown that CaBP-9k is expressed in the fetal placenta and could not be detected in the maternal portion. Fig. 3A shows the results of densitometric quantification of CaBP-9k mRNA in fetal placenta. Between day 16 and day 18 of pregnancy, levels increased 2.2-fold (p < 0.0001) and another 1.7fold between day 18 and day 20 (p < 0.01). CaBP-9k mRiVA levels in fetal/neonatal num
duode-
At day 16 of gestation, due to the size of the fetus, analysis of fetal duodenal CaBP-9k levels was not technically possible. On day 18, no CaBP-9k was detected using up to 40 pug of total RNA. As shown in Fig. 4A, trace amounts of CaBP-9k mRNA were detected in the duodena of fetuses at day 20 of pregnancy. Between day 20 of pregnancy and day 5 of lactation, a drastic increase of CaBP-9k mRNA was observed (19.2fold). Interestingly, at day 10 and day 15 of lactation, levels were found to be substantially lower than on day 5. The decrease was in both cases significant (day 5 vs. day 10, p < 0.0001; and day 5 vs. day 15, p < 0.05). At day 20 of lactation, a further significant increase of duodenal levels occurred compared to the day 5 group (p < 0.01). CaBP-9k mRNA level comparison of maternal and fetal/neonatal tissues
In order to compare the CaBP-9k mRNA levels in the tissues studied, a Northern blot experiment was performed with representative samples of maternal duodenum and uterus, placenta and
UT
D
PL
ND
Fig. 5. Comparison of maternal and fetal/neonatal CaBP-9k mRNA levels. Northern analysis was performed using 20 pg of adult uterine RNA (UT, highest value in group C in Fig. 2A), 5 pg of adult duodenal RNA (D, control group C in Fig. 1A), 20 Fg fetal placenta1 RNA (PL, 20. day pregnancy) and 20 pg neonatal duodenal RNA (ND, 20 day lactation). The autoradiograph and the ethidium bromide stained RNA gel are shown.
fetal/neonatal duodenum. Fig. 5 shows the autoradiograph of an overnight exposure of 20 pg total RNA for all samples except maternal duodenum, where only 5 pg were loaded. The uterus from a non-pregnant control animal was chosen, which showed the highest CaBP-9k level in this group (see Fig. 2A). Highest expression levels are found in the adult duodenum and in neonatal day 20 duodenum. Adult uterine CaBP-9k mRNA levels reached a maximum of approximately 25% of the duodenal levels. Within the control group of non-pregnant animals, the uterine CaBP-9k levels varied drastically by a factor of 10 (Fig. 2A). Therefore, the lowest expression level in non-pregnant uterus amounts to approximately 2.5% of that in duodenum. Placental CaBP-9k mRNA levels are low, amounting to about 2.5% of duodenal levels. Discussion
This study was carried out to gather more information about the regulation of CaBP-9k dur-
126
ing pregnancy and lactation. In particular, uterine levels were of interest due to studies published previously, demonstrating estrogen regulation of uterine CaBP-9k mRNA after ovariectomy and 17/3-estradiol injection (Darwish et al., 1990; L’Horset et al., 1990). Animals were kept on a normal rat chow providing normal amounts of vitamin D. Duodenal CaBP-9k levels were measured to monitor the response of CaBP-9k gene expression provoked by the changes in vitamin D metabolism during pregnancy and especially during lactation. Maternal duodenal CaBP-9k mRNA levels were found to increase significantly during lactation. This result was expected and the increase can be explained by elevated 1,25-dihydroxyvitamin D, levels during lactation, as a response to the high calcium demand of the lactating rat. 1,25_Dihydroxyvitamin D, levels have been shown to increase maximally at day 14 of lactation by 6-fold, coinciding with the highest recorded calcium transport activity (Halloran and DeLuca, 1980). CaBP-9k protein content was determined by Bruns et al. (1987) in lactating rat duodenum, where a 2.5-fold increase of CaBP-9k on day 7 and day 21 of lactation was found. Interestingly, the CaBP-9k mRNA levels at day 20 post-lactation were decreased to levels only 30.3% of nonpregnant controls. We have no explanation for this decrease below control levels. A possibility is the overall change of the duodenal mucosa which undergoes substantial hypertrophy during lactation (Lichtenberger and Trier, 1979) and may still be in the process of re-establishing the non-pregnant state by decreasing cell proliferation at day 20 post-lactation. Changes in /3-actin mRNA levels may support this explanation. @-Actin is found in the epithelial cells of rat intestine (Hartman et al., 1989). In addition to p-actin another unique form of actin is found in rat intestine which is distinct from other muscular and non-muscular isoforms (Sawtell et al., 1988). After day 15 of lactation, p-actin levels dropped from 108% of non-pregnant control to 75% at day 20 of lactation and 37% at day 20 post-lactation. During pregnancy, both CaBP-9k and /3-actin RNA levels were found to be lower than in non-pregnant controls. Determination of uterine CaBP-9k mRNA lev-
els demonstrated dramatic changes. When nonpregnant control uteri were analyzed, a high degree of variation was found, in some samples as much as a lo-fold difference was revealed (n = 6). These differences may be explained by changing estrogen levels during the estrous cycle (Shaikh, 1971; Butcher et al., 1974). CaBP-9k mRNA levels reached about 50% of the maximum of controls at day 16, day 18, and day 20 of pregnancy. Mathieu et al. (1989) measured myometrial CaBP-9k levels in rats and did not report a fluctuation in the protein level as observed in our experiment in the mRNA levels of non-pregnant controls. They described high CaBP-9k levels in non-pregnant animals in comparison to a steady decrease until day 15 of pregnancy, followed by a marked increase near term_ Uterine &BP-9k was not determined during lactation in the report of Mathieu et al. (1989). We detected an abrupt decrease of CaBP-9k mRNA at day 5 of lactation. Only in two out of 12 animals at day 5 (n = 6) and day 10 (n = 6) of lactation, trace levels of CaBP-9k mRNA were detectable with 20 pg total RNA. As lactation proceeded, a few animals had relatively high CaBP-9k mRNA levels on day 15 and 20, while still the majority in each group had no detectable CaBP-9k mRNA levels. There was no correlation of litter size (ranging from 8 to 13) to expression of CaBP-9k mRNA during lactation. An explanation for down-regulation of uterine CaBP-9k mRNA levels may be the low 17pestradiol levels during lactation in the rat (Labhsetwar and Watson, 1974; Smith and Neill, 1977; Taya and Greenwald, 1982). Near the end of lactation, but on varying days, the animal re-enters the estrous cycle, again possibly causing variations in estrogen levels late in lactation affecting the CaBP-9k gene. The combination of high progesterone and low estrogen levels is probably responsible for the low expression of CaBP-9k in the uterus of the lactating rat. With the approach of weaning or after removal of the suckling pups, estrogen levels increase within l-2 days (LuxLantos et al., 1990). We propose that estrogen is the main, if not sole regulator controlling uterine CaBP-9k expression during gestation and lactation. There is evidence for a sharp increase of estrogen shortly before delivery at day 22 in the rat along with elevated estradiol receptor num-
127
bers in the uterus (Cathey and Chung, 1991). The results of this study show a steady expression in late pregnancy (day 20) and a drastic down-regulation before day 5 of lactation. It remains to be determined whether there is an additional peak of CaBP-9k mRNA expression right before parturition on day 22 and when exactly the decline starts after delivery, early in lactation. L’Horset et al. (1990) reported two CaBP-9k transcripts in rat uterus after injection of 17P-estradiol. In the present study, an additional transcript was not detected in any of the animals. L’Horset et al. did not analyze a mature rat uterus without 17@estradiol treatment. The appearance of an additional CaBP-9k transcript may be caused by the injection of 17p-estradiol and the two transcripts may not be synthesized in response to endogenous estrogens in normal or pregnant rats. Equal loading of total RNA was controlled by determination of the RNA concentration (UV absorption at 260 nm) and ethidium bromide staining of the gels. When /3-actin was used as probe, changes of @-actin mRNA levels in uterus were found during pregnancy and lactation. Most interesting is a steady rise of p-actin mRNA during lactation from below non-pregnant levels to a 3-fold higher level at day 20. At day 20 post-lactation, p-actin levels have almost returned to non-pregnant control levels. Changes in /3-actin expression were expected during gestation given the enormous expansion and regression the uterus is subjected to during and after pregnancy. L’Horset et al. (1990) reported an increase in uterine p-actin mRNA following 17/?-estradiol injection into ovariectomized rats, which coincided with the induction of CaBP-9k mRNA. In the present study there was no correlation between CaBP-9k and @actin mRNA levels. Placental expression of CaBP-9k has been shown previously to be confined to the fetal part of the rat placenta (Warembourg et al., 1986). When CaBP-9k mRNA levels in fetal placenta were analyzed, a steady increase from day 16 to day 20 of gestation was found. Between day 16 and day 20 levels rose 4-fold. These data are in agreement with those reported by Bruns et al. (1981) in mouse and Mathieu et al. (1989) in rat at the protein level. High placental CaBP-9k mRNA expression at the end of gestation sup-
ports the hypothesis that the protein is involved in fetal/maternal calcium transport, which peaks at the end of pregnancy. It is unknown which factors control placental CaBP-9k expression. A possible role for vitamin D metabolites in this process is controversial. Placental CaBP-9k levels need to be analyzed in vitamin D deficient, pregnant rats to exclude any effect of vitamin D on its expression. With regard to transplacental calcium transport, Brommage and DeLuca (1984) have clearly demonstrated that vitamin D is not required for maternal/fetal calcium transport in pregnant, vitamin D-depleted rats. Placental CaBP-9k was not analyzed in this experiment. In fetal duodenum, CaBP-9k mRNA was not detectable on day 18 and only a trace level appeared on day 20 of pregnancy_ In organ culture experiments Brehier and Thomasset (1990) used fetal duodena at day 18 of pregnancy. After 10 days in culture they report trace levels of CaBP-9k mRNA and protein which are significantly induced by 1,25_dihydroxyvitamin D,. A marked increase in intestinal CaBP-9k during early development and after administration of epidermal growth factor has been demonstrated (Bruns et al., 1989). Development of the intestinal mucosa including appearance of the vitamin D receptor in the early neonatal phase is thought to be the cause of this event (Halloran et al., 1980, 1981). DeLorme et al. (1979) reported a marked increase on the last day of gestation, followed by a plateau during the first 21 days of the postnatal period. At weaning, another sharp rise was documented in this study. Similar to the plateauing CaBP-9k concentrations during the first 3 weeks of suckling, this present study revealed a decline of CaBP-9k mRNA on day 10 and day 15 postnatal compared to levels at day 5. When the experiment was repeated with six additional animals per group this result was confirmed (data not shown). Control probing with p-actin did not show any difference of p-actin expression in this group (unpublished observation). Clearly the expression of CaBP-9k during neonatal development warrants further investigation. In summary, the present results show that the highest expression of CaBP-9k occurs in adult and neonatal duodenum. Placental and uterine mRNA levels are substantially lower (Fig. 5). Of
128
particular interest is the pronounced regulation in the uterus, especially in the non-pregnant rat where a lo-fold variation occurred and the fast down-regulation revealed in early lactation. From this study we conclude that uterine CaBP-9k is subject to a tight hormonal regulation during the reproductive life of the rat. References Brehier, A. and Thomasset, M. (1990) Endocrinology 127, 580-587. Brommage, R. and DeLuca, H.F. (1984) Am. J. Physiol. 246, F526-F529. Bruns, M.E., Vollmer, S., Wallsheim, V. and Bruns, D.E. (1981) J. Biol. Chem. 256, 4649-4653. Bruns, M.E., Boass, A. and Toverud, S. (1987) Endocrinology 121, 278-283. Bruns, M.E., Overpeck, J.G., Smith, G.C., Hirsch, G.N., Mills, S.E. and Bruns, D.E. (1988) Endocrinology 122,2371-2378. Bruns, D.E., Krishnan, A.V., Feldman, D., Gray, R.W., Christakos, S., Hirsch, G.N. and Bruns, M.E. (1989) Endocrinology 125, 478-485. Butcher, R.L., Collins, W.E. and Fugo, N.W. (1974) Endocrinology 94, 1704-1708. Cathey, T.M. and Chung, K.W. (1991) Life Sci. 49, 293-298. Chomczynski, P. and Sacchi, N. (1987) Anal. Biochem. 162, 156-159. Christakos, S., Gabrielides, C. and Rhoten, W.B. (1989) Endoer. Rev. 10, 3-26. Danvish, H.M., Krisinger, J., Strom, M. and DeLuca, H.F. (1987) Proc. Natl. Acad. Sci. USA 84, 6108-6111. Darwish, H., Krisinger, J., Furlow, J.D., Smith, C., Murdoch, F.E. and DeLuca, H.F. (1990) J. Biol. Chem. 266.551-558. Degen, J., Neubauer, M., Freizner Degen, S.J., Seyfried, C.E. and Morris, D. (1983) J. Biol. Chem. 258, 12153-12162. DeLorme, A.C., Marche, P. and Garel, J.M. (1979) J. Dev. Physiol. 1, 181-194. Delorme, A.C., Danan, J.L., Acker, M.G., Ripoche, M.A. and Mathieu, H. (1983) Endocrinology 113, 1340-1347. Glisin, V.. Crkvenjakov, R. and Byus, C. (1974) Biochemistry 13, 2633-2637.
Halloran, B. and DeLuca, H. (1980a) Am. J. Physiol. 239, E64-E68. Halloran, B.P. and DeLuca, H.F. (1980b) Am. J. Physiol. 239, G473-G479. Halloran, B.P. and DeLuca, H.F. (1981) J. Biol. Chem. 256, 7338-7342. Hartman, A.L., Sawtell, N.M. and Lessard, J.L. (1989) J. Histochem. Cytochem. 37, 1225-1233. Kessler, M.A. and DeLuca, H.F. (1985) Arch. Biochem. Biophys. 236, 17-25. Krisinger, J., Strom, M., Darwish, H.M., Perlman, K., Smith, C. and DeLuca, H.F. (1990) J. Biol. Chem. 266, 1910-1913. Kyozuka, M., Crankshaw, J., Berezin, I., Collins, S. and Daniel, E. (1987) Can. J. Physiol. Pharmacol. 65, 1966-1975. Labhsetwar, A.P. and Watson, D.3. (1974) Biol. Reprod. 10, 103-110. L’Horset, F., Perret, C., Brehier, A. and Thomasset, M. (1990) Endocrinology 127, 2891-2897. Lichtenberger, L.M. and Trier, J.S. (1979) Am. J. Physiol. 237, E98-E105. Lux-Lantos, V., Tesone, M. and Libertun, C. (1990) Endocrinology 126, 680-686. Mathieu, CL., Burnett, S.H., Mills, S.E., Overpeck, J.G., Bruns, D.E. and Bruns, M.E. (1989a) Proc. Natl. Acad. Sci. USA 86, 3433-3437. Mathieu, C., Mills, S., Burnett, S., Cloney, D., Bruns, D. and Bruns, M.E. (1989b) Endocrinology 125, 2745-2750. Roche, C., Bellaton, C., Pansu, D., Miller, A. and Bronner, F. (1986) Am. J. Physiol. 251, G314-G320. Savineau, J., Mironneau, J. and Mironneau, C. (1988) Pfliig. Arch. 411, 296-303. Sawtell, N.M., Hartman, A.L. and Lessard, J.L. (1988) Cell Motil. Cytoskeleton 11, 318-325. Shaikh, A.A. (1971) Biol. Reprod. 5, 297-307. Smith, M.S. and Neill, J.D. (1977) Biol. Reprod. 17, 255-261. Taya, K. and Greenwald, G.S. (1982) Endocrinol. Jpn. 29, 453-459. Warembourg, M., Perret, C. and Thomasset, M. (1986) Endocrinology 119, 176-184. Warembourg, M., Perret, C. and Thomasset, M. (1987) Cell Tissue Res. 247, 51-57. Wasserman, R.H. and Fullmer, C.S. (1989) Adv. Exp. Med. Biol. 249, 45-65.
Molecular and Cellular Endocrinology, 88 (1992) 119-128
0 1992 Elsevier Scientific Publishers Ireland, Ltd. 0303-7207//$05.00 MOLCEL 02834
Calbindin-D,,
gene expression during pregnancy and lactation in the rat J. Krisinger, J.L. Dann, E.B. Jeung and P.C.K. Leung
Department of Obstetrics and Gynecology, University of British Columbia, Vancouuer, B.C. V52 4H4, Canada
(Received 4 June 1992; accepted 28 June 1992)
Key words: Calbindin-D,,;
Gene expression; Pregnancy; Lactation; (Rat)
Summary Calbindin-D,, (CaBP-9k) is a calcium binding protein expressed in mammalian intestine, uterus and placenta. It is believed to be involved in transepithelial calcium transport in intestine and placenta and regulation of cytosolic calcium concentration in uterus. CaBP-9k mRNA levels were measured by Northern blot analysis in maternal duodenum, uterus, placenta and fetal/neonatal duodenum during pregnancy and lactation. In maternal duodenum a maximal increase occurred at day 15 of lactation (2.3-fold) and 20 days post-lactation levels decrease to 30.3% of non-pregnant controls. In non-pregnant uterus a lo-fold variation of CaBP-9k mRNA levels was observed between individual animals despite a uniform expression of p-actin. During pregnancy high CaBP-9k expression is found, averaging about 20% of duodenal levels, which abruptly drops below detection during early lactation. At late lactation CaBP-9k mRNA levels are again subject to great variation ranging from no expression to maximal levels found in the non-pregnant uterus. Placental CaBP-9k is maximally expressed at the end of pregnancy (day 20) reaching about 2.5% of duodenal levels. Fetal intestinal CaBP-9k mRNA was detectable in 20 Fg total RNA at day 18 of pregnancy and rose sharply in early lactation reaching about 50% of adult duodenal levels at day 20 lactation. The profound changes of uterine CaBP-9k mRNA in non-pregnant (cycling), pregnant, and lactating rats indicate a rapid hormonal regulation of gene expression, most likely involving 17P-estradiol.
Introduction Calbindin-D,, (CaBP-9k)is a cytosolic calcium binding protein 04, 9000) expressed predominantly in mammalian duodenum, placenta, and uterus (Christakos et al., 1989). It is thought to be
Correspondence to: Dr. J. Krisinger, The Research Center, Dept. of OB/GYN, 178-950 W. 28th Ave., Vancouver, B.C. VSZ 4H4, Canada. Tel. (6041875-2433; Fax 875-2496. This work was supported by grants from the British Columbia Health Research Foundation No. 7(90-2) and the Medical Research Council of Canada.
involved in transepithelial calcium transport in duodenum (Roche et al., 1986; Wasserman and Fullmer, 1986) and placenta (Mathieu et al., 1989) and to play a role in regulating intracellular calcium levels in myometrium (Mathieu et al., 1989). The regulation of the expression in duodenum by 1,25_dihydroxyvitamin D, has been well documented (Kessler and DeLuca, 1985; Brehier and Thomasset, 1990; Krisinger et al., 19901, whereas regulation of placental CaBP-9k is not understood. In uterus estrogen regulation has been demonstrated (Mathieu et al., 1989; Darwish et al., 1990; L’Horset et al., 1990) and is not af-
120
fected by vitamin D metabolites (Darwish et al., 1990). In the ovariectomized rat, uterine CaBP-9k mRNA levels fall below detection and high expression can be restored by 17P-estradiol injection (Darwish et al., 1990; L’Horset et al., 1990). Gel retardation and transfection studies have revealed an estrogen response element in the rat CaBP-9k gene, responsible for the transcriptional regulation of the uterine expression similar to other genes controlled by steroid hormones. Studies in pregnant rats have also shown gestational changes of CaBP-9k protein levels in uterus (Mathieu et al., 1989). Immunohistochemical studies found expression of CaBP-9k mainly in myometrium (Delorme et al., 1983; Warembourg et al., 1987; Bruns et al., 1988). Myometrial expression of CaBP-9k and estrogen control of its expression have led to the hypothesis that this protein participates in the control of uterine activity during gestation (Bruns et al., 1988). In particular, a role for this calcium binding protein in the development of contractions at parturition has been postulated in regard to the pivotal role of intra-myometrial calcium ions for uterine activity (Kyozuka et al., 1987; Savineau et al., 1988). In order to further understand the regulation of CaBP-9k around the time of parturition, the expression of the gene encoding CaBP-9k through late pregnancy and lactation was examined in rat uterus. In addition to the uterine regulation, changes in maternal duodenal, placental and fetal/neonatal duodenal mRNA levels were also determined. Materials
and methods
Animals
Three-month-old Sprague-Dawley (S.D.) rats were kept on a standard laboratory diet (1% calcium, 0.61% phosphorus, 4.5 II-I/g vitamin D), distilled water and a 12 h day/night cycle. For timed pregnancies, females were mated with male S.D. rats overnight and insemination was confirmed by the presence of sperm in vaginal smear the following morning (day 0 of pregnancy). Animals were sacrificed at day 16, 18, 20 of pregnancy and day 5, 10, 15, 20 of lactation and 20 days post-lactation, respectively. Maternal and fetal/neonatal weight and litter size were deter-
TABLE
1
MATERNAL AND LITTER
WEIGHT, FETAL/NEONATAL SIZE OF EACH EXPERIMENTAL Maternal weight (g)
Fetal/neonatal weight (g)
WEIGHT GROUP
Litter size
Pregnancy day 16 Pregnancy day 18 Pregnancy day 20 Lactation day 5 Lactation day 10 Lactation day 15 Lactation day 20 20 days postlactation
17
0.266 + 0.017
12.2 f 2.7
322 f 32
0.760 f 0.069
13.0 + 2.4
343 k 32
2.2
kO.12
11.7k3.8
296 + 10
9.98
* 1.33
11.3k3.2
31s+
304*11
16.8
+1.60
12.3+ 1.2
316 k 16
32.9
k3.14
7.8 f 2.5
300 & 29
41.2
k8.18
9.7 k 2.8
28Yk16
-
_
mined (Table 1). Each group consisted of six animals. The experimental protocol was approved by the University of British Columbia Animal Care Committee. Tissue and RNA preparation 8 cm of the proximal duodenum
of the adult animals were removed. The entire uterus of nonpregnant and lactating animals or 1 g of uterine tissue of pregnant animals was collected. The fetal parts of the placentae were removed and pooled to yield 1 g of total tissue. Duodena were removed from fetuses on day 18 and 20 of pregnancy, pooled from each litter for preparation of RNA. 1 g of duodenal tissue was collected from suckling pups on day 5, 10, 15, and 20 after birth. Tissues were rinsed in saline and homogenized in 4 M guanidinium isothiocyanate buffer for 30 s with a polytron. Total RNA was prepared using the guanidinium isothiocyanate/CsCl procedure (Glisin et al., 1974; Chomczynski and Sacchi, 1987). RNA concentration was determined by UV spectrophotometry. Quantification of CaBP-9k mRNA 5 pg (maternal duodenum) or 20 Fg (other
tissues) total RNA was separated
on a standard
121
denaturing gel containing formaldehyde. The samples included 0.5 pg ethidium bromide to visualize the RNA after electrophoresis. Ethidium bromide staining of the ribosomal RNA was
MATERNAL
(A)
COmrOl
P, 5
PI8
used to confirm equal loading and integrity RNA. RNA was transferred onto nylon brane and probed with a 295 nucleotide rat probe. The probe consisted of the 5’ part
DUODENUM
P20
L5
LlO
MATERNAL
CC)
L15
L20
control
PL20
PI 5
PI8
P20
of the memcDNA of the
DUODENUM
L5
LIO
L15
L20
PLZO
(8)
C
C
P16
P16
P18
P18
P20
L5
LlO
L15
L20
PL20
Fig. 1. CaBP-9k mRNA levels during pregnancy and lactation in rat duodenum. Non-pregnant animals (control), pregnant animals on days 16, 18, 20 of pregnancy (P16-20) and lactating animals on days 5, 10, 15, 20 of lactation (L5-20) and 20 days post-lactation (PL20) were sacrificed. 8 cm of the proximal duodenum were removed for preparation of total RNA. 5 pg of total RNA were electrophoresed on 1% denaturing agarose and transferred onto nylon membrane. The Northern blot was hybridized to a 295 nt rat CaBP-9k cDNA probe. After autoradiography CaBP-9k was quantified by densitometry. A: Each group consisted of six animals. Shown are the means f SEM for each group. Values are given in arbitrary units. * *, p < 0.001; *, p < 0.05 as compared to control values. B: A representative RNA sample of each group was used for Northern blot analysis. Groups are as in A. Shown are the autoradiograph and the ethidium bromide stained RNA gel. C: The same blot as in A was probed with a 1.2 kb bovine /3-actin cDNA. Results of the densitometric scanning are illustrated as in A.
122
rat CaBP-9k cDNA including coding region for 78 out of 79 amino acids (Darwish et al., 1987). DNA was labelled by random primed synthesis using [ a3* P]dCTP. For quantification of p-actin mRNA, a 1.2 kilobasepair bovine p-actin fragment (Degen et al., 1983) was labelled. After stringent hybridization (50% formamide) and washing, autoradiography was performed. Autoradiographs were scanned with a BioRad densitometer. The area under the curve value of each band was multiplied by 100 and used as arbitrary units. Where values are compared to controls, the
Results CaBP-9k mRNA levels in maternal duodenum
At day 16, 18 and 20 of gestation, CaBP-9k mRNA levels dropped slightly in comparison to non-pregnant age-matched, control rats (Fig. 1A). CC)
UTERUS
(A)
control was set as 100 units. Results were analyzed by analysis of variance. Individual groups were compared using the Student-Fisher r-test. Results are expressed as mean + SEM. Each group consisted of 5-6 animals.
UTERUS
300
Control
P 15
P18
P20
L5
LIO
L15
L20
PL20 Control
PI 6
PI8
P2O
L5
L10
L15
L20
PL20
(B)
Fig. 2. CaBP-9k mRNA levels during pregnancy and lactation in rat uterus. Groups are as in Fig. 1. 20 kg of total RNA were analyzed. A: Results of densitometric scanning of each individual sample are indicated. Due to the great variation in each group means f SEM were not calculated (N.D. = not detectable) B: Representative RNA samples of each group were used for Northern analysis. For the non-pregnant controls, rats on days 15 and 20 of lactation and 20 days post-lactation, two samples representing the highest and lowest value of each group were used. The autoradiograph and the ethidium bromide stained gel are shown. C: The same blot was hybridized with the @-actin probe. Results of densitometric scanning of each group are shown as mean+ SEM. (* p < 0.05; * *, p < 0.0001).
123
At day 18, the values decreased significantly (p < 0.05) to 52.3% of controls. During lactation, levels increased steadily to day 15 (225.7%, p < O.OOl),compared to non-pregnant controls. Interestingly, at day 20 post-lactation levels declined significantly (p < 0.001) to 30.3% of non-pregnant control levels. 5 pg total RNA was used in Northern blot analysis for densitometric scanning after autoradiography. Uniform loading of RNA was assured by measuring absorbance of 260 nm and ethidium bromide staining of the RNA samples (see Fig. 1B). When control probing using a bovine P-actin cDNA was performed, dynamic changes were observed which could not be explained by inconsistent RNA loading. Fig. 1B demonstrates a decline of p-actin in duodenum of pregnant and lactating animals except for day 15 of lactation. On day 16, 18, and 20 (t, < 0.01) of gestation, day 5 (p < 0.05) and day 10 (p < 0.02) of lactation and day 20 (p < 0.05) post-lactation, this decrease was significantly different from controls.
PLACENTA
(A)
P20
(6)
CaBP-9k mRNA levels in uterus
When CaBP-9k mRNA levels in uterus were determined, substantial variation was observed in non-pregnant control animals and in lactating animals. Fig. 2 A shows results of densitometric scanning of autoradiographs using 20 pg of total RNA. Variation in the control group reached from 16.7 to 171 arbitrary units (n = 6). Integrity of the RNA was assessed by ethidium bromide staining of the 18 and 28 S ribosomal RNAs and control probing with bovine p-actin cDNA. Throughout the experimental groups, each individual result is shown as a bar in Fig. 2A. During pregnancy, especially at day 18, levels were relatively uniform between 71-106 densitometric units (n = 5). At early lactation, CaBP-9X mRNA levels fell below detection levels in all but two animals, one at day 5 and one at day 10. On day 15 and 20 of lactation and day 20 post-lactation, the groups again showed substantial variation of CaBP-9k mRNA. While in some animals no CaBP-9k mRNA was detected (at day 15 and day 20 of lactation four animals, and at day 20 postlactation one animal), other individuals bad substantial amounts of CaBP-9k mRNA reaching up to 77 arbitrary units. Fig. 2C shows the results of
PI6
P16
PI8
PI8
P20
P20
Fig. 3. CaBP-9k mRNA in rat placenta. A: Animals on days 16, 18 and 20 of pregnancy were sacrificed and the fetal portions of the placentae removed for preparation of total RNA. 20 Fg of RNA was used for CaBP-9k quantification as described above. Means+ SEM of each group (n = 6) are shown. The level at day 18 increased significantly, ** p < 0.001, when compared to day 16. A further significant increase occurred between day 18 and 20, * p < 0.01. B: Representative RNA samples of each group were used for Northern analysis. The autoradiograph and the ethidium bromide stained gel are shown.
124
(A)
FETAIAEONATAL
DUODENUM
(B)
P18
P18
P20
P20
L5
LIO
L15
L5
LlO
L15
L20
L20
Fig. 4. CaBP-9k mRNA levels in fetal/neonatal duodenum. A: Animals were sacrificed on days 18 and 20 of pregnancy and days 5-20 of lactation (U-20) for preparation of total RNA from duodenum. CaBP-9k mRNA was quantified as above. Levels on day 10. were significantly lower compared to day 5, * * p < 0.001. Levels on day 15 were significantly when compared to levels on day 10, * p < 0.05. A further significant increase occurred between day 15 and day 20 of ** p < 0.001. B: Representative RNA samples of each group were used for Northern analysis. The autoradiograph ethidium bromide stained RNA gel are shown.
(Pig, P20) described increased lactation, and the
densitometric scanning of the same blot probed with the /3-actin probe. Values in each group were uniform and the mean f SEM were calculated. p-Actin mRNA levels increased slightly but not significantly at day 16 and day 18 of pregnancy. At day 20 of pregnancy and day 5 and day 10 of lactation, @-actin levels were decreased. On day 5 of lactation this decrease was significant (54.3% of non-pregnant controls). During lactation a steady increase occurred peaking at day 20 (244.1% of controls). On day 20 post-lactation levels again returned to near non-pregnant control levels.
UT
D
PL
ND
CaBP-9k mRNA leveki in placenta
Preliminary experiments had shown that CaBP-9k is expressed in the fetal placenta and could not be detected in the maternal portion. Fig. 3A shows the results of densitometric quantification of CaBP-9k mRNA in fetal placenta. Between day 16 and day 18 of pregnancy, levels increased 2.2-fold (p < 0.0001) and another 1.7fold between day 18 and day 20 (p < 0.01). CaBP-9k mRiVA levels in fetal/neonatal num
duode-
At day 16 of gestation, due to the size of the fetus, analysis of fetal duodenal CaBP-9k levels was not technically possible. On day 18, no CaBP-9k was detected using up to 40 pug of total RNA. As shown in Fig. 4A, trace amounts of CaBP-9k mRNA were detected in the duodena of fetuses at day 20 of pregnancy. Between day 20 of pregnancy and day 5 of lactation, a drastic increase of CaBP-9k mRNA was observed (19.2fold). Interestingly, at day 10 and day 15 of lactation, levels were found to be substantially lower than on day 5. The decrease was in both cases significant (day 5 vs. day 10, p < 0.0001; and day 5 vs. day 15, p < 0.05). At day 20 of lactation, a further significant increase of duodenal levels occurred compared to the day 5 group (p < 0.01). CaBP-9k mRNA level comparison of maternal and fetal/neonatal tissues
In order to compare the CaBP-9k mRNA levels in the tissues studied, a Northern blot experiment was performed with representative samples of maternal duodenum and uterus, placenta and
UT
D
PL
ND
Fig. 5. Comparison of maternal and fetal/neonatal CaBP-9k mRNA levels. Northern analysis was performed using 20 pg of adult uterine RNA (UT, highest value in group C in Fig. 2A), 5 pg of adult duodenal RNA (D, control group C in Fig. 1A), 20 Fg fetal placenta1 RNA (PL, 20. day pregnancy) and 20 pg neonatal duodenal RNA (ND, 20 day lactation). The autoradiograph and the ethidium bromide stained RNA gel are shown.
fetal/neonatal duodenum. Fig. 5 shows the autoradiograph of an overnight exposure of 20 pg total RNA for all samples except maternal duodenum, where only 5 pg were loaded. The uterus from a non-pregnant control animal was chosen, which showed the highest CaBP-9k level in this group (see Fig. 2A). Highest expression levels are found in the adult duodenum and in neonatal day 20 duodenum. Adult uterine CaBP-9k mRNA levels reached a maximum of approximately 25% of the duodenal levels. Within the control group of non-pregnant animals, the uterine CaBP-9k levels varied drastically by a factor of 10 (Fig. 2A). Therefore, the lowest expression level in non-pregnant uterus amounts to approximately 2.5% of that in duodenum. Placental CaBP-9k mRNA levels are low, amounting to about 2.5% of duodenal levels. Discussion
This study was carried out to gather more information about the regulation of CaBP-9k dur-
126
ing pregnancy and lactation. In particular, uterine levels were of interest due to studies published previously, demonstrating estrogen regulation of uterine CaBP-9k mRNA after ovariectomy and 17/3-estradiol injection (Darwish et al., 1990; L’Horset et al., 1990). Animals were kept on a normal rat chow providing normal amounts of vitamin D. Duodenal CaBP-9k levels were measured to monitor the response of CaBP-9k gene expression provoked by the changes in vitamin D metabolism during pregnancy and especially during lactation. Maternal duodenal CaBP-9k mRNA levels were found to increase significantly during lactation. This result was expected and the increase can be explained by elevated 1,25-dihydroxyvitamin D, levels during lactation, as a response to the high calcium demand of the lactating rat. 1,25_Dihydroxyvitamin D, levels have been shown to increase maximally at day 14 of lactation by 6-fold, coinciding with the highest recorded calcium transport activity (Halloran and DeLuca, 1980). CaBP-9k protein content was determined by Bruns et al. (1987) in lactating rat duodenum, where a 2.5-fold increase of CaBP-9k on day 7 and day 21 of lactation was found. Interestingly, the CaBP-9k mRNA levels at day 20 post-lactation were decreased to levels only 30.3% of nonpregnant controls. We have no explanation for this decrease below control levels. A possibility is the overall change of the duodenal mucosa which undergoes substantial hypertrophy during lactation (Lichtenberger and Trier, 1979) and may still be in the process of re-establishing the non-pregnant state by decreasing cell proliferation at day 20 post-lactation. Changes in /3-actin mRNA levels may support this explanation. @-Actin is found in the epithelial cells of rat intestine (Hartman et al., 1989). In addition to p-actin another unique form of actin is found in rat intestine which is distinct from other muscular and non-muscular isoforms (Sawtell et al., 1988). After day 15 of lactation, p-actin levels dropped from 108% of non-pregnant control to 75% at day 20 of lactation and 37% at day 20 post-lactation. During pregnancy, both CaBP-9k and /3-actin RNA levels were found to be lower than in non-pregnant controls. Determination of uterine CaBP-9k mRNA lev-
els demonstrated dramatic changes. When nonpregnant control uteri were analyzed, a high degree of variation was found, in some samples as much as a lo-fold difference was revealed (n = 6). These differences may be explained by changing estrogen levels during the estrous cycle (Shaikh, 1971; Butcher et al., 1974). CaBP-9k mRNA levels reached about 50% of the maximum of controls at day 16, day 18, and day 20 of pregnancy. Mathieu et al. (1989) measured myometrial CaBP-9k levels in rats and did not report a fluctuation in the protein level as observed in our experiment in the mRNA levels of non-pregnant controls. They described high CaBP-9k levels in non-pregnant animals in comparison to a steady decrease until day 15 of pregnancy, followed by a marked increase near term_ Uterine &BP-9k was not determined during lactation in the report of Mathieu et al. (1989). We detected an abrupt decrease of CaBP-9k mRNA at day 5 of lactation. Only in two out of 12 animals at day 5 (n = 6) and day 10 (n = 6) of lactation, trace levels of CaBP-9k mRNA were detectable with 20 pg total RNA. As lactation proceeded, a few animals had relatively high CaBP-9k mRNA levels on day 15 and 20, while still the majority in each group had no detectable CaBP-9k mRNA levels. There was no correlation of litter size (ranging from 8 to 13) to expression of CaBP-9k mRNA during lactation. An explanation for down-regulation of uterine CaBP-9k mRNA levels may be the low 17pestradiol levels during lactation in the rat (Labhsetwar and Watson, 1974; Smith and Neill, 1977; Taya and Greenwald, 1982). Near the end of lactation, but on varying days, the animal re-enters the estrous cycle, again possibly causing variations in estrogen levels late in lactation affecting the CaBP-9k gene. The combination of high progesterone and low estrogen levels is probably responsible for the low expression of CaBP-9k in the uterus of the lactating rat. With the approach of weaning or after removal of the suckling pups, estrogen levels increase within l-2 days (LuxLantos et al., 1990). We propose that estrogen is the main, if not sole regulator controlling uterine CaBP-9k expression during gestation and lactation. There is evidence for a sharp increase of estrogen shortly before delivery at day 22 in the rat along with elevated estradiol receptor num-
127
bers in the uterus (Cathey and Chung, 1991). The results of this study show a steady expression in late pregnancy (day 20) and a drastic down-regulation before day 5 of lactation. It remains to be determined whether there is an additional peak of CaBP-9k mRNA expression right before parturition on day 22 and when exactly the decline starts after delivery, early in lactation. L’Horset et al. (1990) reported two CaBP-9k transcripts in rat uterus after injection of 17P-estradiol. In the present study, an additional transcript was not detected in any of the animals. L’Horset et al. did not analyze a mature rat uterus without 17@estradiol treatment. The appearance of an additional CaBP-9k transcript may be caused by the injection of 17p-estradiol and the two transcripts may not be synthesized in response to endogenous estrogens in normal or pregnant rats. Equal loading of total RNA was controlled by determination of the RNA concentration (UV absorption at 260 nm) and ethidium bromide staining of the gels. When /3-actin was used as probe, changes of @-actin mRNA levels in uterus were found during pregnancy and lactation. Most interesting is a steady rise of p-actin mRNA during lactation from below non-pregnant levels to a 3-fold higher level at day 20. At day 20 post-lactation, p-actin levels have almost returned to non-pregnant control levels. Changes in /3-actin expression were expected during gestation given the enormous expansion and regression the uterus is subjected to during and after pregnancy. L’Horset et al. (1990) reported an increase in uterine p-actin mRNA following 17/?-estradiol injection into ovariectomized rats, which coincided with the induction of CaBP-9k mRNA. In the present study there was no correlation between CaBP-9k and @actin mRNA levels. Placental expression of CaBP-9k has been shown previously to be confined to the fetal part of the rat placenta (Warembourg et al., 1986). When CaBP-9k mRNA levels in fetal placenta were analyzed, a steady increase from day 16 to day 20 of gestation was found. Between day 16 and day 20 levels rose 4-fold. These data are in agreement with those reported by Bruns et al. (1981) in mouse and Mathieu et al. (1989) in rat at the protein level. High placental CaBP-9k mRNA expression at the end of gestation sup-
ports the hypothesis that the protein is involved in fetal/maternal calcium transport, which peaks at the end of pregnancy. It is unknown which factors control placental CaBP-9k expression. A possible role for vitamin D metabolites in this process is controversial. Placental CaBP-9k levels need to be analyzed in vitamin D deficient, pregnant rats to exclude any effect of vitamin D on its expression. With regard to transplacental calcium transport, Brommage and DeLuca (1984) have clearly demonstrated that vitamin D is not required for maternal/fetal calcium transport in pregnant, vitamin D-depleted rats. Placental CaBP-9k was not analyzed in this experiment. In fetal duodenum, CaBP-9k mRNA was not detectable on day 18 and only a trace level appeared on day 20 of pregnancy_ In organ culture experiments Brehier and Thomasset (1990) used fetal duodena at day 18 of pregnancy. After 10 days in culture they report trace levels of CaBP-9k mRNA and protein which are significantly induced by 1,25_dihydroxyvitamin D,. A marked increase in intestinal CaBP-9k during early development and after administration of epidermal growth factor has been demonstrated (Bruns et al., 1989). Development of the intestinal mucosa including appearance of the vitamin D receptor in the early neonatal phase is thought to be the cause of this event (Halloran et al., 1980, 1981). DeLorme et al. (1979) reported a marked increase on the last day of gestation, followed by a plateau during the first 21 days of the postnatal period. At weaning, another sharp rise was documented in this study. Similar to the plateauing CaBP-9k concentrations during the first 3 weeks of suckling, this present study revealed a decline of CaBP-9k mRNA on day 10 and day 15 postnatal compared to levels at day 5. When the experiment was repeated with six additional animals per group this result was confirmed (data not shown). Control probing with p-actin did not show any difference of p-actin expression in this group (unpublished observation). Clearly the expression of CaBP-9k during neonatal development warrants further investigation. In summary, the present results show that the highest expression of CaBP-9k occurs in adult and neonatal duodenum. Placental and uterine mRNA levels are substantially lower (Fig. 5). Of
128
particular interest is the pronounced regulation in the uterus, especially in the non-pregnant rat where a lo-fold variation occurred and the fast down-regulation revealed in early lactation. From this study we conclude that uterine CaBP-9k is subject to a tight hormonal regulation during the reproductive life of the rat. References Brehier, A. and Thomasset, M. (1990) Endocrinology 127, 580-587. Brommage, R. and DeLuca, H.F. (1984) Am. J. Physiol. 246, F526-F529. Bruns, M.E., Vollmer, S., Wallsheim, V. and Bruns, D.E. (1981) J. Biol. Chem. 256, 4649-4653. Bruns, M.E., Boass, A. and Toverud, S. (1987) Endocrinology 121, 278-283. Bruns, M.E., Overpeck, J.G., Smith, G.C., Hirsch, G.N., Mills, S.E. and Bruns, D.E. (1988) Endocrinology 122,2371-2378. Bruns, D.E., Krishnan, A.V., Feldman, D., Gray, R.W., Christakos, S., Hirsch, G.N. and Bruns, M.E. (1989) Endocrinology 125, 478-485. Butcher, R.L., Collins, W.E. and Fugo, N.W. (1974) Endocrinology 94, 1704-1708. Cathey, T.M. and Chung, K.W. (1991) Life Sci. 49, 293-298. Chomczynski, P. and Sacchi, N. (1987) Anal. Biochem. 162, 156-159. Christakos, S., Gabrielides, C. and Rhoten, W.B. (1989) Endoer. Rev. 10, 3-26. Danvish, H.M., Krisinger, J., Strom, M. and DeLuca, H.F. (1987) Proc. Natl. Acad. Sci. USA 84, 6108-6111. Darwish, H., Krisinger, J., Furlow, J.D., Smith, C., Murdoch, F.E. and DeLuca, H.F. (1990) J. Biol. Chem. 266.551-558. Degen, J., Neubauer, M., Freizner Degen, S.J., Seyfried, C.E. and Morris, D. (1983) J. Biol. Chem. 258, 12153-12162. DeLorme, A.C., Marche, P. and Garel, J.M. (1979) J. Dev. Physiol. 1, 181-194. Delorme, A.C., Danan, J.L., Acker, M.G., Ripoche, M.A. and Mathieu, H. (1983) Endocrinology 113, 1340-1347. Glisin, V.. Crkvenjakov, R. and Byus, C. (1974) Biochemistry 13, 2633-2637.
Halloran, B. and DeLuca, H. (1980a) Am. J. Physiol. 239, E64-E68. Halloran, B.P. and DeLuca, H.F. (1980b) Am. J. Physiol. 239, G473-G479. Halloran, B.P. and DeLuca, H.F. (1981) J. Biol. Chem. 256, 7338-7342. Hartman, A.L., Sawtell, N.M. and Lessard, J.L. (1989) J. Histochem. Cytochem. 37, 1225-1233. Kessler, M.A. and DeLuca, H.F. (1985) Arch. Biochem. Biophys. 236, 17-25. Krisinger, J., Strom, M., Darwish, H.M., Perlman, K., Smith, C. and DeLuca, H.F. (1990) J. Biol. Chem. 266, 1910-1913. Kyozuka, M., Crankshaw, J., Berezin, I., Collins, S. and Daniel, E. (1987) Can. J. Physiol. Pharmacol. 65, 1966-1975. Labhsetwar, A.P. and Watson, D.3. (1974) Biol. Reprod. 10, 103-110. L’Horset, F., Perret, C., Brehier, A. and Thomasset, M. (1990) Endocrinology 127, 2891-2897. Lichtenberger, L.M. and Trier, J.S. (1979) Am. J. Physiol. 237, E98-E105. Lux-Lantos, V., Tesone, M. and Libertun, C. (1990) Endocrinology 126, 680-686. Mathieu, CL., Burnett, S.H., Mills, S.E., Overpeck, J.G., Bruns, D.E. and Bruns, M.E. (1989a) Proc. Natl. Acad. Sci. USA 86, 3433-3437. Mathieu, C., Mills, S., Burnett, S., Cloney, D., Bruns, D. and Bruns, M.E. (1989b) Endocrinology 125, 2745-2750. Roche, C., Bellaton, C., Pansu, D., Miller, A. and Bronner, F. (1986) Am. J. Physiol. 251, G314-G320. Savineau, J., Mironneau, J. and Mironneau, C. (1988) Pfliig. Arch. 411, 296-303. Sawtell, N.M., Hartman, A.L. and Lessard, J.L. (1988) Cell Motil. Cytoskeleton 11, 318-325. Shaikh, A.A. (1971) Biol. Reprod. 5, 297-307. Smith, M.S. and Neill, J.D. (1977) Biol. Reprod. 17, 255-261. Taya, K. and Greenwald, G.S. (1982) Endocrinol. Jpn. 29, 453-459. Warembourg, M., Perret, C. and Thomasset, M. (1986) Endocrinology 119, 176-184. Warembourg, M., Perret, C. and Thomasset, M. (1987) Cell Tissue Res. 247, 51-57. Wasserman, R.H. and Fullmer, C.S. (1989) Adv. Exp. Med. Biol. 249, 45-65.
