~Gopyright 1999 by Humana Press, Inc. All rights of any nature whatsoever reserved. 1085-9195/99/31/001--049/$14.00

Regulation of Lactoferrin Gene Expression by Estrogen and Epidermal Growth Factor Molecular Mechanism

Christina T Teng Gene Regulation Group, Laboratory of Reproductive and Developmental Toxicology, National Institute of Environmental Health Sciences, National Institutes of Health, Research Triangle Park, North Carolina 27709, E-maih [email protected] ABSTRACT Lactoferrin (LF) is a member of the transferrin gene family. Its expression in the mouse uterus is regulated by estrogen and epidermal growth factor (EGF). The author et al. cloned the LF gene promoter/enhancer region, and demonstrated that multihormone signaling pathways are involved in modulating LF gene activity. Three short but complex modules, within 400 bp from the transcription initiation site of the mouse LF gene, contain the response elements that are responsible for estrogen, retinoic acid, mitogen, and growth factor stimulation. These elements have been identified and characterized, using reporter constructs transiently transfected into human endometrial carcinoma RL95-2 cells. The author et al. used molecular approaches, such as deletion, insertion, and site-directed mutagenesis, to determine the relationship between the response elements, and to fine-map the crucial nucleotides within them. This article reviews the characterization of the estrogen and EGF response elements of the mouse LF gene promoter.

Index Entries: Lactoferrin gene promoter; estrogen; epidermal growth factor (EGF); response elements.

nisms, Fe metabolism, bacteria killing, growth promotion, and modulating i m m u n e response (4-7). The bacteriostatic effect and Fe metabolism roles of LF have been related to the highaffinity Fe-binding property of the protein (8-10). Other functional implications, for instance, modulation of the i m m u n e response and host defense system, acting as growth factor, and killing bacteria with peptide fragment of the protein, may involve multiple mechanisms that are unrelated to Fe-binding. These

INTRODUCTION Lactoferrin (LF), a nonheme iron (Fe)-binding glycoprotein, is highly expressed in various tissues, such as the mammalian gland, mucosal surface, exocrine gland, neutrophilic leucocytes, uterus, and so on, and is present in milk, tears, saliva, and other biological fluids (1-3). The protein has been implicated in a wide variety of biological functions, including a potential role in the host defense mechaCell Biochemistry and Biophysics

49

Volume 31, 1999

50 biological properties of the LF have been recently reviewed by a number of laboratories, either in general (1), or with special emphasis on the structure-function relationship (4,5) and the mechanisms of bacteria killing (6-8). This laboratory is intrigued by the potential role for LF in reproduction (2,11-13) and by the complex mechanisms of regulating the LF gene in the reproductive tissue (14-22). Therefore, this review emphasizes information on LF distribution in reproductive tissue and the regulation of LF gene expression in the uterine cells. We have previously demonstrated that LF is a major protein in uterine epithelial cells (ECs) and the uterine secretory fluid during adult pseudopregnancy or early pregnancy, in estrogen-treated immature mice, and in the estrus stage of the cycling animal (2,14,15,23). Although there is no information on other animal models, estrogen does play a role in LF expression in the human endometrium (24,25) and the vaginal epithelium (26). The presence of LF in the human reproductive tract is not limited to females; the protein has also been localized in the prostate and seminal vesicle of males (27), and is a major coat protein of human sperm (28,29). These observations demonstrated that LF is present in the environment during fertilization and in early pregnancy, presumably during the preimplantation stage. The function of estrogen-stimulated LF in the uterus is not well understood, nor is it known whether androgen regulates LF expression in man. However, the secretion of LF in the mouse uterus during the normal estrus cycle, and during early pregnancy, correlates well with the secretion of immunoglobulins (Igs) (24,30). LF could play a role in immuno/inflammatory response, by interacting with macrophages and monocytes present in the uterine stroma at these stages (31-34). Therefore, LF may have a previously unknown role in reproduction. Recently, LF was found to be synthesized in the preimplantation mouse embryo (13), with a timing corresponding to the synthesis of uterine LF (2,11). Later, in the developing fetus, LF expression is restricted to hematoCell Biochemistry and Biophysics

Teng poietic and glandular ECs. These findings further enhance the possibility that LF is essential during fertilization and implantation of the mouse embryo. Whether this is also true for other species needs to be investigated. Although the presence of LF protein in various tissues is under intense investigation, the regulation of its expression, and the molecular mechanisms that are involved, remain unclear. In the mouse uterus, LF gene has been shown to be regulated by estrogen and epidermal growth factor (EGF) (14,15,35). In tissue culture, the LF promoter can be activated by changes in the cell shape and the actin cytoskeleton (36), or by treatment with retinoic acid (37), EGF (20,21), and estrogens (14-18,38). These studies demonstrate that LF mRNA synthesis is responsive to multiple signaling pathways, and can be regulated either directly or indirectly by many effectors. The author proposes to dissect the regions of the LF promoter that respond to the different signaling pathways that influence LF expression. This article reviews recent findings that further characterize the EGF and estrogen response elements of the LF gene.

LF EXPRESSION IN MOUSE UTERUS IS CORRELATED WITH SERUM ESTROGEN LEVEL LF is detected in the mouse uterus at the proestrus and estrus stages, but not at the diestrus stage of the estrus cycle. As indicated in Fig. 1, positive immunostaining (the dark stain), with affinity-purified rabbit antimouse LF IgG, is found in the uterine ECs of proestrus and estrus. A detailed analysis of the LF protein and mRNA expression, during the mouse estrus cycle, was previously reported (23,39). In these studies, LF mRNA and protein levels in the mouse uterus were compared to the estrogen level in the plasma, and a similar pattern of fluctuation was found. In the proestrus stage, circulating estrogen rose to its highest level, and the LF immunostaining appeared in the luminal epithelium, with a punctate staining pattern in the cytoplasm. Volume 31, 1999

LF Regulation by Estrogen and EGF

Proestrus

51 expression in mouse uterus follows closely the change of serum estrogen level. This correlation during the estrus cycle suggests that the LF gene in mouse uterus is sensitive to the physiological level of estrogen, and gain in vivo relevance of the author's earlier studies of LF gene activation by exogenous estrogens

(2,14,15).

Estrus

Diestrus Fig. 1. LF gene expression in mouse uterus during the estrus cycle. Adult female CD-1 mice were obtained from Charles River Laboratories. To determine the stages of the estrus cycle, vaginal smears were analyzed before the mice were killed, and the stage of the cycle was confirmed by histological analysis of the entire reproductive tract (23). Tissues were fixed in cold Bouin's fixative, and processed for histology and immunohistochemistry. The immunostaining was carried out by using a 1000-fold-diluted affinitypurified rabbit antimouse LF IgG and the Vectastain ABC-AP staining kit. Dark staining in the luminal and glandular epithelium of the proestrus and estrus stages indicates immunoreactivity to the anti-LF IgG. When entering estrus, the estrogen level started to fall, and LF became visible in all luminal and glandular ECs with a homogenous staining pattern. No LF was detected in the uterine epithelium at either metestrus or diestrus stages, as the serum estrogen quickly declined and progesterone level went up. The pattern of immunostaining was similar to the pattern observed in in situ hybridization with LF cDNA (23,39). Therefore, LF gene Cell Biochemistry and Biophysics

In addition to estrogens, LF can be upregulated by EGF in the uterus of a mouse whose ovary, adrenal gland, and hypothalamus have been removed (35). The catecholestrogen, 4-hydroxyestradiol-1713 (4-OH-E2), and chlordecone (kepone), an environmental estrogen, both can upregulate the uterine expression of LF in estrogen receptor (ER)-0~-deficient mice (40), which suggests that both ERa-dependent and -independent signaling pathways are regulating the LF gene in the mouse uterus. The following subheadings analyze some of the response elements of the mouse LF gene that mediate the estrogen and EGF-stimulated gene activity.

ESTROGEN RESPONSE ELEMENT IS LOCATED AT - 3 5 1 / - 3 2 7 OF LF GENE The author et al. cloned portions of the 5'flanking sequence of the LF gene into a chloramphenicol acetyltransferase (CAT)-reporter construct, and tested for estrogen responsiveness in a transiently transfected human endometrial carcinoma (HEC) RL95-2 cell line (RL95). The RL95 cells are easy to transfect, but the cells grow slowly. Another HEC cell line, HEC-1B, was used for some of the experiments, but the majority of the work was conducted with RL95. Experimental results from several publications (16-19,41,42) on estrogen responsiveness, are summarized in Fig. 2. The construct that contained a 600-bp promoter fragment (0.6mLF-CAT) responded to estrogen stimulation (Fig. 2A, at fragment 0.6; compare (DES)-treated and control samples); in contrast, other promoter fragments tested (1.7, 0.9, 0.3, and 0.2) were not estrogen-responsive. Within the region of -600 and -300, the author Volume 31, 1999

52

Teng

A

i 100-

[] DES [] Control

,~ 500 .~_ a-

&

0 -1.7

0.9 m

0.6

0.30.2 m

m

~

B coup - 3 51

AAGTGTCACAGGTCAAGGTAACCC

-32 7

ERE

Fig. 2 Identification of the estrogen response region of the mouse LF gene. The RL95 cells were grown in a 1:1 mixture of Dulbecco's minimal essential medium: Ham's F12, supplemented with 10% fetal bovine serum (FBS), 5 ~tg/mL bovine insulin, and 100 U / m L penicillin/streptomycin under 5% CO2. Transient transfections were performed by the calcium phosphate method with Cellphect transfection kit (Pharmacia), and the CAT assays were previously described (18). The cells were co-transfected with vector alone, or with 5 ~tg/well reporter plasmid and 0.5 ~tg/well HEO expression plasmid. After transfection, the cells were cultured in 10% charcoalstripped FBS, with or without treatment of 10-8 M DES for 24 h. All experiments were repeated at least 3x with duplicated samples. (A) Various fragments of the 5'-flanking region of the LF gene were ligated into the CAT-reporter construct, and the estrogen responsiveness was tested. The thick bars indicate the regions protected in DNase I footprinting analysis with nuclear extract from the RL95 cells (16). (B) Sequence of the ERE overlaps with a COUP-TF binding element. Adapted from refs. 16-19,41,42. found an imperfect estrogen response element (ERE) that overlapped with a COUP-TF-binding element (COUP) (Fig. 2B). This region was one of the areas (solid bar under the CAT construct in Fig. 2A) that were protected by RL95 nuclear extract from DNase I digestion in a footprinting analysis. This complex sequence, COUP-ERE, was further analyzed for its ability to bind ER and COUP-TF by electromobility shift assay (EMSA), and for its ability to confer ER-mediated transactivation in transiently transfected RL95 cells (18,19,41,42). These studies demonstrated that the 15-bp imperfect ERE was a functional response element, and the results are summarized in Fig. 3. In the presence of estrogen, a single ERE caused Cell Biochemistry and Biophysics

a 40-fold increase in CAT-reporter activity, regardless of the orientation (Fig. 3A, psvmCAT5 and 6); two EREs stimulated the CATreporter activity to its maximum level in the assay system (psv-mCAT 7). The COUP element alone showed some enhancer activity in the presence of estrogen (psv-mCAT 10 and 11), but it was low, compared to the strength of the E R E . Multiple COUP elements responded well to estrogen stimulation (psvmCAT 12 to 14). The mechanism could be promiscuous binding of ER to variously spaced direct repeats of the AGGTCA motifs (43,44). When they were adjacent to one another, the COUP element enhanced the strength of ERE (psv-mCAT 15-18). Volume 31, 1999

LF Regulation by Estrogen and EGF

53 I

r.~

~

:::I ~ 0 ~

I

I,~

~'~

~ .,~-~

03

0

~

I

r---~ t

~

0 ~.a..~ u tm..-~ ~ ~

~

I

~

~J ~

~

f:~.-' ;=It'%

~ ~

~ 0'-~.,~ ~" . . . ~ m

~

U

0

I

~-~

,".~ ,.--~

r

~ ~

,..

~.~

~.~

I=i

~

~. ~

~

~

~

I

~

~--~

;...~ l,m l

.~

~

e~

0

r

I~ ~ 0

r~

~_~.

b,.

O o u

0

r~

~L

o

'~'u e ~ ~ ~ N ~ ~ ~ ' ~

n

o 0

& ~

E

o.~.~

u

~

f= ~

" ~J U bO .,~

I~

0 ~

~

TI

0

~

~

0

0

,4.-, ~

I~1~

~

.i.~

0-~'~

~ m

~ "~

~

~.

~,~

~

~

~

~

__

uew __

0

El

~

o 8,

(%) esuodseu

4-) ~ ,~ e4 tn

~

~ N

l;;~ r ,-~ ~

~ ~'~ ~ 9 ~ ~-.i ~ N ~ =~ ~ ~,~ ~ ~ ~ ~o~

~T~

-

~.~

~

~

~

~.~ ~ . . ~

0"~

~'C:I

tU ~

~|

o

o~ ~ ~ ~ r..T4 , . , . ~

F~9

~

~"

~-

~-

~

E.

~,

",~

~ ~ ~o

~'~

o ~< ~ 9 ~

q "--

+1 r

+1

+1

+1

+1

+1

+1

+t

+l

+1

+1

+1

44

+1

+1

04

04

~

~

d

(5

ci

+I

+I

+I

(5

d

d

d

o

d

o

d

d

d

d

+I

+I

o o +1

d

d

"

'

'

+I

+I

+I

+I

+I

+I

-H

~

+I

04

~

~

I~

~

0

~

0

0

9

d

~

d

(:5

,..:

d

,.:

,-:

0

t , e-

o 0

m{

0 im el"

~ ~ ~ ~, ~ Cell Biochemistry and Biophysics

~ ~ ~ ~ Volume 31, 1999

54

Teng

COUP-TF0 t~ ~1' inactive ER

activated ER

COUP'~~~~ll I)~ ~., Fig. 4.

Proposed model for the molecular mechanism of estrogen action in mouse uterus.

In the natural context of the mouse LF gene, COUP element overlaps with the ERE (Fig. 2B). It is reasonable to predict that such organization will create competition for binding to the overlapping sequence between COUP-TF and ER. As expected, COUP-TF was found to interfere with ER binding to the mouse LF COUP-ERE, and subsequently represses ER-mediated response (19). These points were illustrated in Fig. 3B,C. Mutation in the COUP element outside the ERE overlap (by site-directed mutagenesis) blocked COUP-TF binding and increased estrogen-stimulated CAT-reporter activity to 50% above the wild-type element (Fig. 3B, m3). Mutation at the overlapping Gs hampered ER and COUP-TF binding, and the estrogen response decreased to 20% of the wild-type (Fig. 3B, ml and m2). The repressing role of the COUP-TF in the ER-mediated response was studied by transfecting increasing amounts of a COUP-TF expression plasmid into the RL95 cell; the relationship between amounts of COUP-TF and estrogen responsiveness was then evaluated (Fig. 3C). Co-transfection of 4 ~tg COUP-TF and 1 ~tg estrogen receptor expression plasmids into Cell Biochemistry and Biophysics

the RL95 cells blocked nearly all the estrogenstimulated activi~. The result strongly implied that COUP-TF acts as a repressor to modulate estrogen responsiveness of the mouse LF gene in the uterus. In fact, COUP-TF was found to repress transactivation by a number of steroid receptors and transcription factors (45). Because COUP-TF can bind variously spaced repeated AGGTCA motifs, depending on the organization of the steroid receptor response element in the gene, the COUP-TF can function as both repressor and activator (46). We found that the mouse uterus contains a high level of COUP-TF relative to the level of the ER (18). Based on the natural sequence of COUP-ERE in the mouse LF gene, the author proposes that COUP-TF is the major occupant of the COUP-ERE in mouse uterus, until the ER is activated by estrogen. The half-life of COUP-TF binding to its DNA element is 3-5 min; ligand-bound ER and the ERE interaction are much more stable, with half-lives of 30 rain (19). Therefore, the activated ER is able to replace the COUP-TF binding at the COUP-ERE, and to activate transcription of the LF gene. The proposed model is presented in Fig. 4. Volume 31, 1999

LF Regulation by Estrogen and EGF

A

55

GC-Rich Region of the Mouse Lactoferrin Promoter

-103 -92 -81 -62 0.1mLF G G C A A T G G G G C T G G A A G G C A G G C C T A T T G G G C A A T A G G G T G G G G C C A G C C C GC-I GC-II -42 -22 +I GGTGAGGTCACCCAGCACAGATAAAGGGCCCCGGGGAGAGGGCAGAAGCCAGG GC-III

B

Fold of Stimulation 0.1 mLF-CAT

1.00 5.28 2.53 4.06 5.62 16.83 17.11 7.03 10.12

FSK TPA EGF TGFa FSK + EGF FSK + TGFa TPA + EGF TPA + TGFa

C

Construct

Relative CAT

Fold

0.1mLF-CAT +PKC~ +PKC7 +PCDM8 +CEV36

7.54 17.83 34.05 10.06 55.45

1.00 2.43 4.66 1.37 7.59

• • • + •

0.91 2.28 1.14 1.37 8.43

Fig. 5. Identification of the mitogen and growth factor response regions of the mouse LF gene. The 0.1mLF-CAT reporter constructs were transfected into RL95 cells, and the CAT activity was measured. (A) GC-rich regions of the mouse LF promoter. (B) Effect of treatment with mitogens: 25 ~tM FSK, 100 n g / m L TPA, and growth factors: 100 ng/mL EGF, 10 ng/mL TGFa, alone or in combination on the activities of 0.1mLF-CAT reporter constructs transiently transfected into RL95 cells. The results are the average of three experiments. (C) Effect of overexpression of PKC (PKCct), PKC (PKC7), and PKA (CEV 36) catalytic subunits on the activities of transiently transfected 0.1mLF-CAT reporter constructs. Adapted from refs. 20,21,52.

MITOGEN AND GROWTH FACTOR RESPONSE ELEMENTS OF MOUSE LF GENE ARE LOCATED IN A GC-RICH REGION OF PROMOTER In many EGF response genes, such as the prolactin (47), transin (48), gastrin (49), tyrosine hydroxylase (50), and transferrin receptor genes (51), germinal-center (GC)-rich sequences, present in the promoter regions, confer EGF responsiveness. The mouse LF gene contains Cell Biochemistr~ and Biophysics

a GC-rich region located between -103 and +1, immediately upstream of the start codon (41,42,52). This region can be further subdivided into three parts (Fig. 5A), termed GC-I (-103 to -81), GC-II, (-76 to -40), and GC-III (-27 to -10). To test whether the mitogen and growth factor response elements involve these GC-rich regions, we cloned the -103/+1 sequence into a CAT-reporter gene (0.1mLFCAT), and examined its responses to forskolin (FSK), 12-0-tetradecanoylphorbol-13-acetate (TPA), EGF, and transforming growth factor (TGF)-a using transiently transfected RL95 Volume 31, 1999

56 cells (20,21,52). The results from these publications were summarized in Fig. 5. The 0.1 mLF-CAT was capable of responding to mitogens and growth factors (Fig. 5B); FSK had stronger effect (5.2-fold) than the TPA (2.5fold); and EGF (4-fold) and TGF-0~ (5.6-fold) were equally effective in stimulating the GCrich promoter of the mouse LF gene. In addition, a synergistic effect was observed when both mitogen and growth factor were present. TPA was a weak mitogen in these experiments, possibly as a result of both the response element sequence and the cell type. To examine the signaling pathways involved in the stimulation of the LF gene by mitogen or growth factor, the author et al. transfected the expression plasmids of the the protein kinase C (PKC00, its catalytic subunit (PKC7), or the protein kinase A (PKA) catalytic subunit (CEV 36) in the RL95 cells. Overexpressing these expression plasmids in the transfected RL95 cells mimicked the effects of exogenously added EGF and FSK (Fig. 5C). The control plasmid, pCDM8, had no effect on the 0.1mLF-CAT activity. These experiments demonstrated that both PKC and PKA pathways are involved in the regulation of LF gene expression, acting through the proximal 103 bp of the promoter.

CHARACTERIZATION OF EGF AND MITOGEN RESPONSE ELEMENTS OF MOUSE LF GENE Several transcription-factor-binding elements reside in the GC-rich region of the LF 5'-flanking sequence (Figs. 5A and 6A). A CAAT/GT box exists in both GC-I and GC-II, and an AP1/CREB-binding element (CRE) is found in GC-II. To map the location of EGF and mitogen response elements in the GC-rich region, a set of deletion mutants was constructed (20,21). The diagram of the deletion mutants is presented in Fig. 6A, and the summarized results in Fig. 6B. When GC-I and GC-III regions were either partially (D1 and D5) or completely (D2) deleted, the FSK and EGF response was retained at a slightly lowCell Biochemistry and Biophysics

Teng

A

GC-I

GC-III

GC-II

CAAT/GT

CAAT/GT

CRE

TATA

+1

wt -103 D1

-92 -82

D2

-62

D3

-42

D4

-22

D5 -54

D7

-36 -35

D8

B

Fold of Stimulation

wt D1 D2 D3 D4 D5

A A n A A A

-103 to -103 to -103 to -103 to -103 to -22 to

+1 -92 -82 -62 -42 +1

FSK

EGF

4.5 3.3 3.0 2.4 1.7 --

4.4 3.5 3.9 1.0 1.3 3.3

Fig. 6. Mapping the EGF and FSK response elements of the mouse LF gene. Cell transfection, CAT assay, and FSK, EGF treatment are described in Figs. 2 and 5. (A) Diagram of the deletion mutants of 0.1mLF-CAT. (B) FSK and EGF responses of the deletion mutants. Adapted from refs. 20 and 21. ered level (Fig. 6B). However, partial deletion of the GC-II region (D3) attenuated the EGF response, and reduced the FSK response. The sequence that was deleted in D3 consists of the proximal CAAT/GT box sequence. To block the FSK response, the CRE sequence must be deleted (D4). Thus, within the GC-II region, there is a functional distinction between FSK and EGF. Although GC-I also contains a distal CAAT/GT box, apparently this region does not play a critical role in mediating LF gene responsiveness to EGF. A degenerate GT box may render it nonfunctional. To fine-map the nucleotides (nts) in GC-II critical for the binding of transcription factors and for conferring EGF and FSK responsiveness, a series of mutants was generated (20,21). The location of mutated nts, and the summarized results from these studies, were presented in Fig. 7. An EMSA study demonstrated that CRE can bind both AP1 and CREB, whereas the proximal CAAT/GT box binds an unVolume 31, 1999

LF Regulation by Estrogen and EGF

57

Con~ruct

Fold of Stimulation

-75

-40 CAAT/GT

wt 5 6 7 8 19 i0 ii 12 15

CRE

FSK

f I i J GGGCAATAGGGTGGGGCCAGCCCGGTGAGGTCACCC .................... TAA ............. ....................... CTA ............................ CT ...... .............................. CTT--....................... CTA .... CTT . . . . ---A-C .................................... ........ AT ................................ ............ TTT ........................... ........ CGTGCACT ..........................

TPA

F+T

5.6 4.0 4.5 2.0 1.3 1.2 4.5 1.5 3.8 2.0 . . . . .

EGF

10.7 8.2 6.0 5.9 1.6 6.2 5.0 5.6 5.6 5.0 . . . . 1.5 2.4 2.2 1.9

TGF~ 8.5 5.8 6.2 6.3 5.9 -----

Fig. 7. Identification of the nts in the proximal CAAT/GT box and CRE of the mouse LF gene, which are crucial to mitogen or growth factor responses. Cell transfection, CAT assay, and mitogen growth factor treatment are described in Figs. 2 and 5. F+T, FSK plus TPA. Adapted from refs. 20 and 21. Fold of EGF stimulation

Construct wt 16 17 22 20 23 24 25 19

-73 GCAATAGGGTGGGGCCAG GCAATAGGGTGGGGCCAG GCAATAGGGTGGGGCCAG GCAATAGGGTGGGGCCAG GCAATAGGGTGGGGCCAG GCAATAGGGTGGGGCCAG GCAATAGGGTGGGGCCAG GCAATAGGGTGGGGCCAG CRE mut

IS4 IS10 IS15 IS20 IS24 IS30 IS42

-41 CCCGGTGAGGTCACC CCCGGTGAGGTCACC CCCGGTGAGGTCACC CCCGGTGAGGTCACC CCCGGTGAGGTCACC CCCGGTGAGGTCACC CCCGGTGAGGTCACC CCCGGTGAGGTCACC

8.2 1.8 1.9 2.6 3.0 3.0 2.7 2.5 1.4

Fig. 8. Distance constraints of the proximal CAAT/GT box and CRE required for EGF response. Random nts were inserted between the two elements. Cell transfection, CAT assay, and EGF treatment are described in Figs. 2 and 5. Adapted from refs. 21 and 52. known protein (20). The most important nts in the CRE region required for AP1 and CREB binding were GGT, and mutation of these nts (m6) made the reporter constructs insensitive to stimulation by both FSK and TPA (Fig. 7, m6). Other nts in the CRE were also important, because reduced FSK and TPA responses were observed with specific mutations within the CRE (m5, m7, and m8). The CAAT/GT box was found to be essential, both for protein binding and for EGF-stimulated activity (m10-15). Therefore, the author et al. named this proximal CAAT/GT box EGFRE in previous publications (20,21,52). Recently, the protein that binds to this element was identified as the basic transcription element binding (BTEB2) protein (52). It will be interesting to determine the relationship between BTEB2 and AP1/CREB. Cell Biochemistry and Biophysics

The potential function of the proximal CAAT/GT box and CRE as enhancers was tested with heterologous promoters (both thymidine kinase and SV40). CRE was shown to function as an enhancer element, but the proximal CAAT/GT box cannot (21). To explore the relationship between the proximal CAAT/GT box and CRE, the author et al. mutated the CRE, or altered the spacing between the elements, by inserting different numbers of nts and examined EGF responsiveness in transiently transfected RL95 cells (21,52). The results from these studies are summarized in Fig. 8. The EGF response decreased dramatically when the CRE was destroyed (m19), or when 4 or 10 bp were inserted between the two elements. The results suggest that distance constraints exist for the location of the proximal CAAT/GT Volume 31, 1999

58 box and the CRE. Protein-protein interaction may be an integral part of the EGF response. Insertion of more than 10 nt between the two elements did not abolish the EGF response completely: After EGF treatment, CAT-reporter activity was consistently 2-3-fold above the basal level. Introduction of many nts between the elements may inadvertently create other transcription factor recognition sites, which may contribute to the observed EGFstimulated activity.

BASIC PROMOTER AND COMPLEX RESPONSE MODULES OF MOUSE LF GENE Because of the close interrelationship between the proximal CAAT/GT box and CRE, we examined the relationship of the CRE and the noncanonical TATA box, and determined the minimal promoter of the mouse LF gene (21). The basal promoter activities from previous publications (20,21,52), are summarized in Fig. 9. Removing the GC-I sequence, which contains the distal CAAT/GT box (see map in Fig. 6A), doubled the basal promoter activity (D1 and D2, 200 and 280%, respectively); removing the GC-III (D5, 129%), and both distal and proximal CAAT/GT boxes (D3, 136%), caused the promoter activity to increase slightly. These results strongly implied the presence of repressor sequences in these regions. The promoter activity was attenuated by deletion of either CRE or the noncanonical TATA (D4, D7, D8), and by the CRE mutation (m19). When the CRE was moved 70 bp from the ATAAA (m21), basal promoter activity was drastically reduced. Therefore, the CRE and ATAAA depend on each other to produce a minimum promoter activity for the mouse LF gene. Based on the results of the above studies, we determined the basic promoter, and analyzed two multisignal response modules, A and B, within 400 bp of the 5'-flanking region of the mouse LF gene. These modules contain composite response elements that confer responsiveness to a variety of steroid hormones and growth factors. Figure 10 shows Cell Biochemistry and Biophysics

Teng Basal CAT Activity % wt D1 D2 D3 D4 D5 D7 D8 m19 m21

A A

-54 to -36 -35 to +1 CRE mut 70bp apart

100.0 200.0 280.0 136.9 18.6 129.0 14.6 12.8 20.0 33.3

Fig. 9. Data indicating that CRE and ATAAA constitute the basic promoter of the mouse LF gene. Cell transfection and CAT assay are described in Fig. 2. See Fig. 6 for identity of mutant D1-D8 and Fig. 7 for mutant 19. the locations of the response elements (A), and their actual sequences (B). During the course of these studies, we discovered that both the estrogen- and EGFresponse element are part of the complex multihormone response modules of the mouse LF gene. The short but complex regions of the mouse LF gene promoter, involved in steroid hormone and mitogen responsiveness of the gene, can be divided into three modules: A, B, and the basic promoter regions (Fig. 10). Module B, which contains the ERE, is composed of four steroid receptor half-site motifs, AGGTCA, and two extended half-site motifs, TCAAGGTCATC. In addition to the classical palindromic ERE (inverted repeat [IR] with 3-bp space, IR3), these half-site motifs can be arranged as direct repeat (DR) with 0, 1, and 7 spaces (DR0, DR1, and DR7) and IR10. Based on the organization of the half-site motifs, module B favors ER binding, as well as binding of homodimer and heterodimers of retinoid X receptor (RXR) with retinoic acid receptor (RAR), COUP, and peroxisome proliferator activated receptor (PPAR) (53). Recently, a retinoic acid response element (RARE), in the 5'-flanking region of the LF gene promoter, was mapped to the same location as the ERE (37). A band-shift assay demonstrated that the ERE-RARE bound strongly with RXR homodimers and RXR-RAR heterodimers. As in the response to estrogen, it responds to 9-cisretinoic acid (9-cis-RA) in a cell-type-specific Volume 31, 1999

LF Regulation by Estrogen and EGF

59

A

50bp I

B

A

BP

Half Site

CAAT ChAT

+1

-400 EXON

PU

1

PU

Extended Half Site

~IAPI/CREB

Extended

B

--

COUp

-350

Half

Site

ERE

-320

~TCACAGTG~CCAGT~CCATTG(~GTGTTTATA RARE

),

m

Half

-i

)m

Site

B

4 EGFRE

l -106

CAAT

GT

CAAT

! GT

-56

CTCCC~TT~CCCGACC~TCCGTCCGGAT~CCCTT~_~CCCAeCCC~GTC ..... P-distal

A~

.................

--I

k-proximal

A~ ....

I. . . .

A

CRE

! -55

CpS

l A~I/CREB

TATA

epG

-16

GGGCC~CTCCA~GGTCGTGTqTATT~CCCGGGGCCCCT I

BP

i ............

Fig. 10. Modular organization of the mouse LF promoter DNA elements. (A) Diagram of the 400-bp 5'-flanking region of the LF gene. (B) DNA sequence of modules B, A, and basic promoter. manner. Similarly, the 9-cis-RA-stimulated activity could be blocked by excess COUP-TF expressed in the cells. Other natural gene promoters, such as oxytocin (54) and the laminin B1 gene (55), which contain multiple DR motifs with varying space between the repeats, are capable of responding to ,multiple hormones through their DR motifs. While studying the ERE of the human LF gene, we discovered that an extended steroid Cell Biochemistr~ and Biophysics

receptor half-site near the ERE, TCAAGGTCATC, modulates the ER-mediated response in RL95 cells (22). Through cloning and sequencing, the author et al. determined that the binding protein for this extended half-site is the estrogen receptor-related orphan receptor ix1 (ERR 0d) (22,56). Module B of the mouse LF gene promoter consists of two such extended half-sites, which have the potential to bind ERR 0d. Therefore, members of the Volume 31, 1999

60 steroid receptor superfamily (57), which bind extended steroid half-sites as monomers, could also interact with the mouse LF module B region. Together, these studies suggest the existence of promiscuous transcriptional regulation through a common response element and possible receptor crosstalk; in addition, these studies demonstrate that the unique organization of the multihormone response module of a gene confers cell-type and tissue-specific responsiveness to various steroid hormones. Module A consists of two CAAT/GT box motifs, designated as distal A and proximal A (named EGFRE in previous publications: 20,21,52). Transcriptional activation of the mouse LF gene by EGF requires the proximal A element, because mutation or deletion of this element abolishes the EGF response. Although the distal A element seems not to be involved in the EGF response in these studies, this element certainly influences the basal promoter activity (Fig. 9). The sequences of the distal and proximal A motifs are similar, and the reason why only one of the two is involved in the EGF response is not clear. It is possible that the GT box in the distal A is degenerate and unable to interact with the transcription factor; alternatively, the nts 5' to the CAAT, which are missing in the 0.1mLFCAT construct, may be important for protein binding, and for stimulation by EGF. Recently, the author et al. identified the protein that binds the GT box in the proximal A motif as BTEB2 (52). Whether BTEB2 also binds the distal A motif should be examined. BTEB2 is a newly discovered subgroup of zinc finger proteins in the early growth response (EGR) transcription factor gene family (58) that exhibits specificity for GC-rich DNA binding. Although little is known of the specific role of BTEB2, members of the EGR-1 and EGR-2 were found to play a deterministic role in governing the development of hematopoietic, macrophage, and monocytic cells (59,60). Therefore, regulation of LF gene expression by BTEB2 may imply such a biological role. It will also be interesting to determine what

Cell Biochemistry and Biophysics

Teng other GC-box-binding proteins, such as Spl family members, are involved in regulation of LF gene expression through module A. To achieve the EGF response in the RL95 cells, the basic promoter of the LF gene is essential, because the proximal A element, previously named EGFRE, is inactive with heterologous promoters (21). The CRE motif in the basic promoter module, which can bind both AP1 and CREB (20), is crucial for EGF action, and is also required for response to FSK and phorbol ester. Binding of AP1 and CREB to the CRE seems mutually exclusive in the band-shift assay (20). However, this finding does not preclude the possibility that members of both transcription factor families interact with each other. AP1 is a family of transcription factors whose members are dimers and heterodimers of the Jun and Fos oncogene products (61,62). Although most members of the Jun and Fos gene families are immediate-early response genes, several Fosrelated proteins display an extended duration of expression following stimulation (63). Thus, it is possible that one of these factors could participate in regulating the LF gene promoter through CRE in a delayed response fashion. Furthermore, some members of the CREB family, such as ATF-2, ATF-3, and ATF-4, can heterodimerize with Fos and Jun (64,65). Thus, transcription factors that are regulated by the PKA and PKC pathways can interact with each other in the nucleus, at the level of DNA binding. There are numerous examples of crosstalk between signaling pathways that involve protein phosphorylation. In addition, phosphorylation of the transcription factors by EGF and FSK at the serine or tyrosine residues could directly affect the protein-DNA or proteinprotein interaction at the promoter region. The author et al. found that the EGF-induced protein-DNA complex indeed contains phosphorylated protein, because phosphatase treatment disrupts the complex (21). Thus the protein-protein and protein-DNA interactions involving the mouse LF module A must be complicated, possibly reflecting the sum of

Volume 31, 1999

LF Regulation by Estrogen and EGF interactions among host proteins at multiple regions. Certainly, they can not be understood by band-shift assays alone. Transient transfection experiments are valuable tools to understand gene transcription with relatively small promoter and DNA elements. However, in the native environment, regulatory elements that are involved in specific expression may be several hundreds or thousands of nts apart. Module A and B of the mouse LF gene are 230 bp away from each other; possibly they are brought into close physical proximity by a nucleosome during chromatin assembly. Therefore, detailed analysis of the response elements, and the nuclear proteins that interact with them, define not only the LF promoter response to a specific signal, but will help explain the crosstalk between the various signaling pathways, which, together, produce the unique pattern of tissue-specific LF expression.

CONCLUSION Expression of the LF gene in the reproductive tract is regulated by hormones and mitogens. To understand how estrogen and EGF regulate the LF gene in uterine cells at the molecular level, the author et al. characterized the ERE (16-19), and identified an EGF response element in the promoter of the gene (20,21,52). Thus progress was made in understanding the hormone and growth factor regulation of LF expression at the gene level. Nontheless, the studies were limited to mouse and h u m a n LF gene promoters. There is little information on the LF gene promoter of other species (66), and no other animal model is available for studying biological properties of the LF. Because of the species differences in the milk content of LF (1), the regulating mechanism of LF gene expression in a tissue may vary greatly between species. Developing animal and cell models, to study the biological roles and the regulation of LF gene expression, will broaden knowledge in this field.

Cell Biochemistry and Biophysics

61

ACKNOWLEDGMENTS The contributions and suggestions of Youhua Liu, Nengyu Yang Huiping Shi, Tatsuya Sueyoshi, and Gregg Richards are gratefully acknowledged. The author thanks Loretta Moore for editing the paper.

REFERENCES 1. Levay, E F. and Viljoen, M. (1995) Lactoferrin: a general review. Haematologica 80, 252-267. 2. Teng, C. T., Pentecost, B. T., Chen, Y. H., Newbold, R. R., Eddy, E. M., and McLachlan, J. A. (1989) Lactotransferrin gene expression in the mouse uterus and mammary gland. Endocrinology 124, 992-999. 3. Shigeta, H., Newbold, R. R., McLachlan, J. A., and Teng, C. T. (1996) Estrogenic effect on the expression of estrogen receptor, COUP-TF, and lactoferrin mRNA in developing mouse tissues. Mol. Reprod. Dev. 45, 21-30. 4. Lonnerdal, B. and Iyer, S. (1995) Lactoferrin molecular structure and biological function. Ann. Rev. Nutr. 15, 93-110. 5. Nuijens, J. H., van Berkel, P. H. C., and Schanbacher, F. L. (1996) Structure and biological action of lactoferrin. J. Mam. Gland Biol. Neopl. 1, 285-295. 6. Ellison, R. T., 3rd. (1994) The effects of lactoferrin on gram-negative bacteria. Adv. Exp. Med. Biol. 357, 71-90. 7. Tomita, M., Takase, M., Bellamy, W., and Shimamura, S. (1994) Active peptide of lactoferrin. Acta Paediatr. Jpn. 36, 585-91. 8. Naidu, A. S., and Arnold, R. R. (1997) Influence of lactoferrin on host-microbe interactions, in Lactoferrin: Interactions and Biological Functions (Hutchens, T. W. and Lonnerdal B., eds.), Humana, Totowa, NJ, pp. 259-275. 9. Baker, E. N., Anderson, B. E, Baker, H. M., Haridas, M., Jameson, G. B. Norris, G. E., Rumball, S. V., and Smith C. A. (1991) Structure, function and flexibility of human lactoferrin. Int. J. Biol. Macromol. 13, 122-129. 10. Baker, E. N., Anderson, B. F., Baker, H. M., Day, C. L., Haridas, M., Norris, G. E., et al. (1994) Three-dimensional structure of lactoferrin in various functional states. Adv. Exp. Med. Biol. 357, 1-12.

Volume 31, 1999

62 11. McMaster, M. T., Teng, C. T., Dey, S. K., and Andrews, G. K. (1992) Lactoferrin in the mouse uterus: analyses of the preimplantation period and regulation by ovarian steroids. Mol. Endocrinol. 6, 101-111. 12. Ward, P. P., Mendoza-Meneses, M., MulacJericevic, B., Cunningham, G. A., SaucedoCardenas, O., Teng, C. T., and Conneely, O. M. (1999) Restricted spatiotemporal expression of lactoferrin during murine embryonic development. Endocrinol. 140, 1852--1860. 13. Ward, P. P., Mendoza, M., Saucedo-Cardenas, O., Teng, C. T., and Conneely, O. M. (1998) Restricted spatiotemporal expression of lactoferrin during murine embryogenesis, in Advances in Lactoferrin Research (Spik, G., Legrand, D., Mazurier, J., Perrauden, J.-P., and Pierce, A., eds.), Plenum, New York, pp. 91-100. 14. Teng, C. T., Walker, M. P., Bhattacharyya, S. N., Klapper, D. G., DiAugustine, R. P., and McLachlan, J. A. (1986) Purification and properties of an estrogen-stimulated mouse uterine glycoprotein (approx. 70 kDa). Biochem. J. 240, 413--422. 15. Pentecost, B. T. and Teng, C. T. (1987) Lactotransferrin is the major estrogen inducible protein of mouse uterine secretions. J. Biol. Chem. 262, 10134-10139. 16. Liu, Y. H. and Teng, C. T. (1991) Characterization of estrogen-responsive mouse lactoferrin promoter. J. Biol. Chem. 266, 21880-21885. 17. Teng, C. T., Liu, Y. H., Yang, N. Y., Walmer, D., and Panella, T. (1992) Differential molecular mechanism of the estrogen action that regulates lactoferrin gene in human and mouse. Mol. Endocrinol. 6, 1969-1981. 18. Liu, Y. H. and Teng, C. T. (1992) Estrogen response module of the mouse lactoferrin gene contains overlapping chicken ovalbumin up-stream promoter transcription factor and estrogen receptor-binding elements. Mol. Endocrinol. 6, 355-364. 19. Liu, Y. H., Yang, N. Y., and Teng, C. T. (1993) COUP-TF acts as a competitive repressor for estrogen receptor-mediated activation of the mouse lactoferrin gene. Mol. Cell. Biol. 13, 1836-1846. 20. Shi, H. P. and Teng, C. T. (1994) Characterization of a mitogen-response unit in the mouse lactoferrin gene promoter. J. Biol. Chem. 269, 12,973-12,980. 21. Shi, H. P. and Teng, C. T. (1996) Promoterspecific activation of mouse lactoferrin gene CellBiochemistryand Biophysics

Teng

22.

23.

24.

25.

26.

27.

28.

29.

30.

31.

by epidermal growth factor involves two adjacent regulatory elements. Mol. Endocrinol. 10, 732-741. Yang, N. Y., Shigeta, H., Shi, H. P., and Teng, C. T. (1996) Estrogen related receptor, hERR1, modulates estrogen receptor mediated response of human lactoferrin gene promoter. J. Biol. Chem. 271, 5795-5804. Newbold, R. R., Teng, C. T., Beckman, W. C., Jr., Jefferson, W. N., Hanson, R. B., Miller, J. V., and McLachlan, J. A. (1992) Fluctuations of lactoferrin protein and messenger ribonucleic acid in the reproductive tract of the mouse during the estrous cycle. Biol. Reprod. 47, 903-915. Tourville, D. R., Ogra, S. S., Lippes, J., and Tomasi, T. B., Jr. (1970) The human female reproductive tract: Immunohistological localization of ~/A, ~G, ~/M secretory "piece," and lactoferrin. Am. J. Obstet. Gynecol. 108, 1102-1108. Kelver, M. E., Kaul, A., Nowicki, B., Findley, W. E., Hutchens, T. W., and Nagamani, M. (1996) Estrogen regulation of lactoferrin expression in human endometrium. Am. J. Reprod. Immunol. 36, 243-247. Cohen, M. S., Britigan, B. E., French, M. and Bean, K. (1987) Preliminary observations on lactoferrin secretion in human vaginal mucus: variation during the menstrual cycle, evidence of hormonal regulation, and implications for infection with Neisseria gonorrhoeae. Am. J. Obstet. Gynecol. 157, 1122-1125. Wichmann, L., Vaalasti, A., Vaalasti, T., Tuohimaa, P. (1989) Localization of lactoferrin in the male reproductive tract. Int. J. Androl. 12, 179-186. Hekman, A. and Rumke, P. (1969) The antigens of human seminal plasma (with special reference to lactoferrin as a spermatozoacoating antigen), in Protides Biological Fluids Colloquium, vol. 16 (Peeters, H., ed.), Pergamon, Oxford, pp. 549-552. Goodman, A. S. and Young, L. G. (1981) Immunological identification of lactoferrin as a shared antigen on radioiodinated human sperm surface and in radioiodinated human seminal plasma. ]. Reprod. Immunol. 3, 99-108. Rachman, E, Casimiri, V., Psychoyos, A., and Bernard, O. (1983) Immunoglobulins in the mouse uterus during the oestrous cycle. J. Reprod. Fert. 69, 17-21. Tachi, C., Tachi, S., Knyszynski, A., and Linder, H. R. (1981) Possible involvement of macrophages in embryo-maternal relationVolume 31, 1999

L F Regulation by Estrogen and E G F

ships during embryo implantation in the rat.

63 43. Kato, S., Sasaki, H., Suzawa, M., Masushige, S.,

J. Exp. Zool. 217, 81-92. 32. Birgens, H. S. (1991) The interaction of lactoferrin with human monocytes. Dan. Med. Bull.

38, 244-252. 33. McMaster, M. T., Newton, R. C., Dey, S. K.,

34.

35.

36.

37.

38.

39.

40.

41.

42.

and Andrews, G. K. (1992) Activation and distribution of inflammatory cells in the mouse uterus during the preimplantation period. J. Immunol. 148, 1699-1705. McMaster, M. T., Dey, S. K., and Andrews, G. K. (1993) Association of monocytes and neutrophils with early events of blastocyst implantation in mice. J. Reprod. Fertil. 99, 561-569. Nelson, K. G., Takahashi, T., Bossert, N. U, Walmer, D. K., and McLachlan, J. A. (1991) Epidermal growth factor replaces estrogen in the stimulation of female genital-tract growth and differentiation. Proc. Natl. Acad. Sci. USA 88, 21-25. Close, M. J., Howlett, A. R., Roskelley, C. D., Desprez, P. U., Bailey, N., Rowning, B., et al. (1997) Lactoferrin expression in mammary epithelial cells is mediated by changes in cell shape and actin cytoskeleton. J. Cell Sci. 110, 2861-2871. Lee, M. O., Liu, Y., and Zhang, X. K. (1995) A retinoic acid response element that overlaps an estrogen response element mediates multihormonal sensitivity in transcriptional activation of the lactoferrin gene. Mol. Cell. Biol. 15, 4194-4207. Teng, C. T. and Yang, N. Y. (1997) Estrogen regulation of human lactoferrin gene activity, in Lactoferrin: Interactions and Biological Functions (Hutchens, T. W. and Lonnerdal, B. eds.), Humana, Totawa, NJ, pp. 135-153. Walmer, D. K., Wrona, M. A., Hughes, C. L., and Nelson, K. G. (1992) Lactoferrin expression in the mouse reproductive tract during the natural estrous cycle: correlation with circulating estradiol and progesterone. Endocrinology 131, 1458-1466. Das, S. K., Taylor, J. A., Korach, K. S., Paria, B. C., Dey, S. K., and Lubahn, D. B. (1997) Estrogenic responses in estrogen receptor-c~ deficient mice reveal a distinct estrogen signaling pathway. Proc. Natl. Acad. Sci. USA 94, 12786-12791. Teng, C. T. (1994) Lactoferrin gene promoter in human and mouse. Adv. Exp. Med. Biol. 357, 183-196. Teng, C. T. (1995) Mouse lactoferrin gene: A marker for estrogen and epidermal growth factor. Environ. Health Perspect. 103 (suppl 7), 17-20.

CellBiochemistryand Biophysics

44.

45.

46.

47.

48.

49.

50.

51.

52.

53.

Tora, L., Chambon, P., and Gronemeyer, H. (1995) Widely spaced, directly repeated PuGGTCA elements act as promiscuous enhancers for different classes of nuclear receptors. Mol. Cell. Biol. 15, 5858-5867. Aumais, J. P., Lee, H. S., DeGannes, C., Horsford, J., and White, J. H. (1996) Function of directly repeated half-sites as response elements for steroid hormone receptors. J. Biol. Chem. 271, 12,568-12,577. Kliewer, S. A., Umesono, K., Heyman, R. A., Mangelsdorf, D. J., Dyck, J. A., and Evans, R. M. (1992) Retinoid X receptor-COUP-TF interactions modulate retinoic acid signaling. Proc. Natl. Acad. Sci. USA 89, 1448-1452. Tsai, S. Y. and Tsai, M. J. (1997) Chick ovalbumin upstream promoter-transcription factors (COUPTFs): coming of age. Endocr. Rev. 18, 229-240. Elsholtz, H. P., Mangalam, H. J., Potter, E., Albert, V. R., Supowit, S., Evans, R. M., Rosenfeld, M. G. (1986) Two different cis-active elements transfer the transcriptional effects of both EGF and phorbal esters. Science 234, 1552-1557. Matrisian, M. L., Leroy, P., Ruhlmann, C., Gesnel, M. C., and Breathnach, R. (1986) Isolation of the oncogene and epidermal growth factor-induced transin gene: complex control in rat fibroblasts. Mol. Cell. Biol. 6, 1679-1686. Marchant, J. L., Demedick, B., and Brand, S. J. (1991) A GC-rich element confers epidermal growth factor responsiveness to transcription from the gastrin promoter. Mol. Cell. Biol. 11, 2686-2696. Lewis, E. J. and Chikaraishi, D. M. (1987) Regulated expression of the tyrosine hydroxylase gene by epidermal growth factor. Mol. Cell. Biol. 7, 3332-3336. Ouyang, Q., Bommakanti, M., Miskimins, W. K. (1993) A mitogen-responsive promoter region that is synergistically activated through multiple signalling pathways. Mol. Cell. Biol. 13, 1796-1804. Teng, C. T., Shi, H. P., Yang, N. Y., and Shigeta, H. (1998) Mouse lactoferrin gene: promoterspecific regulation by EGF and cDNA cloning of the EGF-response-element binding protein, in Advances in Lactoferrin Research (Spik, G., Legrand, D., Mazurier, J., Perrauden, J.-P., and Pierce, A., eds.), Plenum, New York, pp. 65-78. Glass, C. K. (1994) Differential recognition of target genes by nuclear receptor monomer, dimers, and heterodimers. Endocr. Rev. 15, 391--407. Volume 31, 1999

64

Teng

54. Richard, S., and Zingg, H. H. (1990) The

55.

56.

57.

58.

59.

60.

human oxytocin gene promoter is regulated by estrogens. J. Biol. Chem. 265, 6098-6103. Vasios, G., Mader, S., Gold, J. D., Leid, M., Lutz, Y., Gaub, M. P., Chambon, P., and Gudas, L. (1991) The late retinoic acid induction of laminin B1 gene transcription involves RAR binding to the responsive element. EMBO J. 10, 1149-1158. Shi, H. E, Shigeta, H., Yang, N. Y., Fu, K. S., O'Brian, G., and Teng, C. T. (1997) Human estrogen receptor-like 1 (ESRL1) gene: genomic organization, chromosomal localization, and promoter characterization. Genomics 44, 52-60. Mangelsdorf, D. J., Thummel, C., Beato, M., Herrlich, P., Schutz, G., Umesono, K., et al. (1995) The nuclear receptor superfamily: the second decade. Cell 83, 835-839. Liu, C., Calogero, A., Ragona, G., Adamson, E., and Mercola, D. (1996) EGR-1, the reluctant suppression factor. Crit. Rev. Oncogenesis 7, 101-125. Lee, S. L., Tourtellotte, L. C., Wesselschmidt, R. L., and Milbrandt, J. (1995) Growth and differentiation proceeds normally in cells deficient in the immediate early gene NGFI-A. J. Biol. Chem. 270, 9971-9977. Krishnaraju, K., Nguyen, H. Q., Leibermann, D. A., and Hoffman, B. (1995) The zinc finger

CellBiochemistr9 and Biophysics

61. 62.

63.

64.

65.

66.

transcription factor Egr-1 potentiates macrophage differentiation of hematopoietic cells. Mol. Cell. Biol. 15, 5499-5507. Curran, T. and Franza, B. T., Jr. (1988) Fos and Jun, the AP1 connection. Cell 55, 395-397. Angel, P., and Karin, M. (1991) The role of Jun, Fos and AP1 complex in cell-proliferati0n and transformation. Biochim. Biophys. Acta 1072, 129-157. Franza, B. R., Jr., Sambucetti, L. C., Cohen, D. R., and Curren, T. (1987) Analysis of fos protein complexes and fos-related antigens by high-resolution two-dimensional gel electrophoresis. Oncogene 1, 213-221. Hai, T., and Curran, T. (1991) Cross-family dimerization of transcription factors Fos/ Jun and ATF/CREB alters DNA binding specificity. Proc. Natl. Acad. Sci. USA 88, 3720-3724. Masquiller, D. and Sassone-Corsi, P. (1992) Transcriptional cross-talk: nuclear factors CREM and CREB bind to AP-1 sites and inhibit activation by Jun. J. Biol. Chem. 267, 22,460-22,466. Seyfert, H. M., Tuckoricz, A., Interthal, H., Koczan, D., and Hobom, G. (1994) Structure of the bovine lactoferrin-encoding gene and its promoter. Gene 143, 265-269.

Volume 31, 1999