Vol. June
177,
No.
2, 1991
BIOCHEMICAL
AND
BIOPHYSICAL
RESEARCH
COMMUNICATIONS
Pages 790-796
14. 1991
A PIT-l
BINDING
SITE 3' TO THE TRANSCRIPTION START SITE TRANSCRIPTION ELONGATION --IN VITRO Kelly
P. Smith
and Z. Dave Sharp1
Department University
of Cellular and Structural of Texas Health Science 7703 Floyd Curl Drive San Antonio, Texas 78284-7762
Received
April
17,
INHIBITS
Biology Center
1991
The transcription factor Pit-l can autoregulate its own expression by interacting with binding sites associated with the Pit-l gene. One of these binding sites is located within the transcribed region of the gene and appears to be responsible for negatively regulating Pit-l transcription. Using unique constructions in an in vitro transcription system, we show here that binding of Pit-l to its recognition element can impede the progress of transcription elongation through the occupied site. We propose that this is a viable mechanism for Pit-l autoregulation. 0 1991Academrc mess, Inc. Enhanced cell-specific transcription of the prolactin and growth hormone genes is controlled by a pituitary-specific transcription factor referred to as Pit-l (l-3). Pit-l is a member of a family of structurally related POU-domain transcription factors which appear to be involved in development (4-6). The regulatory regions of the rat prolactin gene has at least 7 Pit-l binding sites and the growth hormone gene at least two sites (1). A single proximal Pit-l binding site is sufficient to stimulate cell-specific transcription in vitro Analysis of the 5' regulatory region of the Pit-l gene (7,8). has identified two Pit-l binding sites (9). One site is within a DNAase I footprinted region from -38 to -70. This is very nearly the same position as the proximal Pit-l binding site of the rat prolactin gene (-36 to -63). The second binding site is located from +4 to +34, within the transcribed region of the Pit-l gene. 1To whom correspondence
should be addressed at current
of Biotechnology, Center for Molecular Texas Health Science Center, 7703 Floyd Texas 78284-6250. 0006-291X/91 $1.50 Cq)Fright 0 1991 by Academic Pwss. Inc. All rights oj reprodrrction in cm?jiwm reserved.
790
address: Institute Medicine, University of Curl Drive, San Antonio,
Vol.
177,
No.
2, 1991
BIOCHEMICAL
AND
BIOPHYSICAL
RESEARCH
COMMUNICATIONS
These two binding sites appear to be central to the autoregulation of the Pit-l gene since binding of Pit-l to the 5t site stimulates transcription while binding to the 3' site inhibits transcription (9). It is unclear whether binding of Pit-l to the 3' site inhibits transcription by hindering transcription initiation or by blocking elongation through the occupied binding site. This study employs a series of prolactin promoter constructs in an in vitro transcription system to further investigate the transcriptional regulatory role of a Pit-l binding site 3' to the cap site.
MATERIALS
AND METHODS
Template Preparation. To produce the templates used in this study, two prolactin Bal 31 deletion mutants containing 36 and 66 bp of 5' flanking DNA were used. The 36 bp template possesses a TATA box and cap site only (basal promoter). The 66 bp template has a basal promoter as well as a single proximal Pit-l binding site (-36 to -63) (Fig la). Each of these templates has a =I site inserted 34 bp downstream of the cap site. Into these sites were inserted annealed oligonucleotide DNA fragments possessing either an intact Pit-l binding site (homologous to the proximal Pit-l binding site footprinted region in the rat prolactin gene) or a DNA sequence slightly altered such that Pit-l does not bind (10) *
Structure of prolactin test templates for in vitro (a) The drawings indicate the positions of Pit-l binding sites and altered non-binding sites relative to the Boxes indicate entire footprinted transcription start site. Templates were produced as outlined in Materials region of DNA. (b) Positions of the 5' and 3' prolactin and Methods. oligonucleotides used for primer extension analysis relative to the Pit-l binding site (or altered site) located at +50. Fiqure
1.
transcription.
791
Vol.
177,
No.
2, 1991
BIOCHEMICAL
AND
BIOPHYSICAL
RESEARCH
COMMUNICATIONS
Normal Pit-l binding site: AATTCGATTATATATATATTCATGAAGGTGTCGAATT Altered site (no binding): AGCTTGCCTGATTATATATATCTTCATGAAGGTGTCT This procedure shown in Fig.
resulted la. Their
in the production of the structures were verified
4 templates by sequencing.
All templates were assayed by in vitro Assay method. transcription with GHB nuclear extracts prepared by a method In our protocol, potassium modified from Dignam (11,12). glutamate replaced potassium chloride in the extract buffers. In each assay, 200 ng of test template and 75 ng of control template pHTX-B, a plasmid containing three copies of the adenovirus major late promoter, were incubated in 15 ~1 of GH3 nuclear extract. The transcripts produced by in vitro transcription were assayed by primer extension (7) using [32P]-labelled oligonucleotide primers homologous to specific sequences within the RNA products from the prolactin and major late promoter templates. Two primers were used for the prolactin RNA assays. The first (or 5') primer is homologous to a 25 nucleotide region of RNA located from +20 to +45 and produces a 45 nucleotide extension product (Fig lb). The second (or 3') primer recognizes a region from +98 to +127 and results in a 127 nucleotide product (Fig lb). As shown in Fig lb, the 5' primer recognizes a sequence upstream of the normal or altered Pit-l binding sites at +50 which were cloned into each template. It is designed to measure the total level of transcription initiation for each template. The 3' primer is homologous to a region downstream of the inserted binding sites and is designed to measure the level of transcription which proceeds through an occupied Pit-l binding site versus transcription through an unoccupied sequence. Two adenovirus major late promoter primers which produced a 54 nucleotide primer extension product were used to measure control template transcription. A 30 nt MLP oligomer was used along with the 3' prolactin primer and a shorter 25 nt MLP oligomer was used with the 5' prolactin primer. The shorter primer was used so that the labelled primer would not interfere with the 45 nt prolactin primer extension product when assayed on an acrylamide gel (Fig 2). Bands corresponding to the expected size product from the primer extension analyses were quantified using a Betascope (Betagen). Each template product was then standardized versus the averaged levels of major late promoter control template transcription and transcription values were expressed as percentages of the level of transcription obtained from template 66. This allowed values from more than one experiment to be combined for statistical analysis (ANOVA, Student-Newman-Keuls test). RESULTS The autoradiograph of a typical experiment (Fig 2) and the graph shown in Fig 3 indicates the relative level of transcription of each template as measured from the 5' and 3' primers. A 4 to 5 fold increase in transcription activity was evident when the proximal Pit-l binding site was present (template 66) as opposed to the transcription level with the 792
Vol.
177,
No.
2, 1991
BIOCHEMICAL
AND
BIOPHYSICAL
3’Primer
RESEARCH
COMMUNICATIONS
SPrimer
prl
127ntm ma
MLP 54 nt
Fiqure 2. Gel analysis of in vitro transcription products assayed by primer extension (described in Methods). [ 32P]labelled products were run on an 8% polyacrylamide gel. Bands corresponding to prolactin (prl) and major late promoter (MLP) template products were quantified using a Betascope (Betagen).
basal
promoter
(template
36).
These results
are consistent
with
other studies (13). Comparison of transcription from template vs template 36-3' as measured from the 5' primer indicated a statistically significant ( Q = 0.01) 2 fold increase in transcription. located
Thus,
3'
to
the
it
appears
that
transcription
start
a Pit-l
site
binding
36
site
can positively
influence transcription. However, an increase in transcription proximal Pit-l was not detected in comparing template 66 (single binding
site)
sites). binding
Our interpretation site may be acting
facilitate the only cap site.
with
template
66-3'
(proximal
of these to attract
and 3'
binding
results is that the 3' transcription factors
to
the establishment of an initiation complex when it is transcription factor binding site in the vicinity of the this activity is rendered insignificant when However,
a properly positioned 5' Pit-l binding site is present and actively enhancing the construction and/or stability of the transcription initiation complex. 793
Vol.
177,
No.
BIOCHEMICAL
2, 1991
AND
BIOPHYSICAL
RESEARCH
COMMUNICATIONS
120
r-
100 80 60
36-3'
NT) El
5’ Primer
3’ Primer
Fiqure 3. Graph indicating the relative transcription activity of each template (expressed as a percentage of activity from template 66) as measured from primers 5' (upstream) and 3' (downstream) of the Pit-l binding site or altered site located 4-50. The graph represents two separate experiments.
Comparisons
of
the
levels
of
transcription
measured
at
from
the
5' and 3' primers in the 36-3' and 66-3' templates showed that a significantly smaller proportion of the RNA produced in the in vitro transcriptions was actually long enough to be measured by primer extension from the 3' (more distal) primer. This difference was not due to random transcription termination since the transcription measured with either primer from the 36 template (which has no 3' Pit-l binding site) was not significantly different and transcription measured from the 66 template with the 3' and 5' primers are assigned equivalent values. Therefore, these results must be due to the occupancy of Pit-l at the 3' binding sites in the 36-3' and 66-3' templates which acts to impede the progress of transcription elongation through the binding site. Thus, Pit-l bound to a recognition element that is 3# to the transcription start site can act to negatively regulate transcription in vitro. DISCUSSION Autoregulation appears to be a widely used means of controlling levels of transcription factors within the cell. A variety of Drosonhila homeodomain proteins, which are structurally related to the POU-domain transcription factors, have been shown to autoregulate their own expression (14-16). There are also reports of Ultrabithorax proteins activating their own expression by binding to downstream Ubx binding sites (17). However, the functional downstream binding sites in Ubx were 794
Vol.
177,
No.
2, 1991
BIOCHEMICAL
AND
BIOPHYSICAL
RESEARCH
COMMUNICATIONS
further downstream (+218 to +306) than the 3# Pit-l binding site in the Pit-l gene (+4 to +34), and resulted in transcription activation rather than repression. Interestingly, the deletion of two more proximal sites (+41 to +88) resulted in an increase in transcription of Ubx. Nevertheless, the presence of a single transcription factor binding site downstream of the transcription start site which functions to negatively autoregulate gene expression is thusfar unique to Pit-I. By placing a Pit-l binding site 3' to the transcription start site in the rat prolactin gene, we were able to gain some insight as to how this repression works in vitro. Primer extension analysis of transcription products using primers homologous to regions 5' and 3' to the Pit-l binding site showed that a binding site occupied by a Pit-I protein can hinder the progression of transcription elongation through the site. The RNA polymerase III transcription factor TFIIIA binds to an internal control region within 5s RNA gene of Xenopus and acts as an assembly factor for the Pol III transcription complex (18). TFIIIA allows the passage of the transcription complex through the binding site while remaining bound to the DNA. This is apparently accomplished by the ability of the zinc-finger DNA binding region of TFIIIA to lift up from the DNA and allow the complex to pass (19). Although these systems are quite different, this functional dissimilarity between TFIIIA and Pit-l may point to a fundamental difference in the mechanism of DNA binding by zinc-finger proteins and POU-domain proteins. Since the Pit-l binding site we inserted into the transcribed region of the prolactin gene is further downstream than the Pit-l binding site in the Pit-l gene (+50 to +77 vs +4 to +34), we cannot rule out the possibility that binding to the site in the Pit-l gene inhibits transcription by sterically interfering with the formation of the initiation complex. However, we feel confident that this inhibition of elongation is genuine and is at least site represses Pitpart of the mechanism by which the 3' binding 1 transcription. This study has also shown that the presence of a single Pit1 binding site downstream from the cap site can have a moderate We feel that this stimulatory effect on transcription. stimulation is due to Pit-l attracting general transcription factors and thereby increasing the local concentration of these This factors, enhancing the assembly of the initiation complex. 795
Vol.
177,
No.
2, 1991
BIOCHEMICAL
AND
BIOPHYSICAL
RESEARCH
COMMUNICATIONS
effect is probably not seen in the templates with two binding sites because the transcription activation by a 5' Pit-l binding site in close proximity to the TATA box is much more efficient at stimulating transcription. The results of this study have increased our understanding of the mechanisms of transcription activation and autoregulation by Pit-l. ACKNOWLEDGMENTS The authors would like to thank Sharon Helsel and Kazi This work was supported by Public for technical assistance. health Service Grant # DK38546 from the National Institutes Health to Z.D.S.
Begum of
REFERENCES 1.
2. 3. 4. 5.
6. 7. 8. 9. 10. Il. 12. 13. 14. 15. 16. 17. 18. 19.
V.R., Elsholtz, H., Lu, L.I.-W., and Nelson, C., Albert, 239, 1400-1405. Rosenfeld, M.G. (1988) Science Mangalam, H.J., Albert, V-R., Ingraham, H.A., Kapiloff, M., Wilson, L., Nelson, C., Elsholtz, H., and Rosenfeld, M.G. (1989) Genes DeV. 3, 946-958. J.W., Smith, K-P., and Sharp, Z.D. (1991) Cao, Z, Poser, Submitted G. and Finney, M. (1991) Cell 64, 475-478. Ruvkun, Ingraham, H.A., Albert, V.R., Chen, R., Crenshaw, E.B.,III, Elsholtz, H.P., He, X., Kapiloff, M.S., Mangalam, H.J., Swanson, L.W., Treaty, M.N., and Rosenfeld, M.G. (1990) Annu. Rev. Physiol. 52, 773-791. He, X., Treaty, M.N., Simmons, D.M., Ingraham, H.A., Swanson, L.W., and Rosenfeld, M.G. (1989) Nature 340, 35-42. Cao, Z., Barron, E.A., and Sharp, Z.D. (1988) Mol. Cell. Biol. 8, 5432-5438. Barron, E.A., Cao, Z., Schneider, B.G., Kraig, E., Carrillo, and Sharp, Z.D. (1989) Mol. Cell. Biol. 9, 817-819. A.J., Chen, R., Ingraham, H.A., Treaty, M.N., Albert, V.R., Wilson, L., and Rosenfeld, M.G. (1990) Nature 346, 583-586. Sharp, Z.D., Helsel, S., Cao, Z., Barron, E.A., and Sanchez, Y. (1989) Nucleic Acids Res. 17, 2705-2722. Dignam, J.D., Lebovitz, R-M., and Roeder, R.G. (1983) Nut. Acids Res. 11, 1475-1489. Cao, Z., Barron, E.A., Carrillo, A.J., and Sharp, Z.D. (1987) Mol. Cell. Biol. 7, 3402-3408. Smith, K.P., Barron, E.A., and Sharp, Z.D. (1991) Submitted Harding, K., Hoey, T., Warrior, R., and Levine, M. (1989) EMBO J. 8, 1205-1212. Biggin, M.D. and Tjian, R. (1988) Cell 53, 699-711. Kuziora, M.A. and McGinnis, W. (1988) Cell 55, 477-485. Krasnow, M.A., Saffman, E.E., Kornfeld, K., and Hogness, D.S. (1989) Cell 57, 1031-1043. Kassavetis, G.A., Braun, B.R., Nguyen, L.H., and Geiduschek, E.P. (1990) Cell 60, 235-245. Miller, J., McLachlan, A.D., and Klug, A. (1985) EMBO J. 4, 1609-1614.
796
177,
No.
2, 1991
BIOCHEMICAL
AND
BIOPHYSICAL
RESEARCH
COMMUNICATIONS
Pages 790-796
14. 1991
A PIT-l
BINDING
SITE 3' TO THE TRANSCRIPTION START SITE TRANSCRIPTION ELONGATION --IN VITRO Kelly
P. Smith
and Z. Dave Sharp1
Department University
of Cellular and Structural of Texas Health Science 7703 Floyd Curl Drive San Antonio, Texas 78284-7762
Received
April
17,
INHIBITS
Biology Center
1991
The transcription factor Pit-l can autoregulate its own expression by interacting with binding sites associated with the Pit-l gene. One of these binding sites is located within the transcribed region of the gene and appears to be responsible for negatively regulating Pit-l transcription. Using unique constructions in an in vitro transcription system, we show here that binding of Pit-l to its recognition element can impede the progress of transcription elongation through the occupied site. We propose that this is a viable mechanism for Pit-l autoregulation. 0 1991Academrc mess, Inc. Enhanced cell-specific transcription of the prolactin and growth hormone genes is controlled by a pituitary-specific transcription factor referred to as Pit-l (l-3). Pit-l is a member of a family of structurally related POU-domain transcription factors which appear to be involved in development (4-6). The regulatory regions of the rat prolactin gene has at least 7 Pit-l binding sites and the growth hormone gene at least two sites (1). A single proximal Pit-l binding site is sufficient to stimulate cell-specific transcription in vitro Analysis of the 5' regulatory region of the Pit-l gene (7,8). has identified two Pit-l binding sites (9). One site is within a DNAase I footprinted region from -38 to -70. This is very nearly the same position as the proximal Pit-l binding site of the rat prolactin gene (-36 to -63). The second binding site is located from +4 to +34, within the transcribed region of the Pit-l gene. 1To whom correspondence
should be addressed at current
of Biotechnology, Center for Molecular Texas Health Science Center, 7703 Floyd Texas 78284-6250. 0006-291X/91 $1.50 Cq)Fright 0 1991 by Academic Pwss. Inc. All rights oj reprodrrction in cm?jiwm reserved.
790
address: Institute Medicine, University of Curl Drive, San Antonio,
Vol.
177,
No.
2, 1991
BIOCHEMICAL
AND
BIOPHYSICAL
RESEARCH
COMMUNICATIONS
These two binding sites appear to be central to the autoregulation of the Pit-l gene since binding of Pit-l to the 5t site stimulates transcription while binding to the 3' site inhibits transcription (9). It is unclear whether binding of Pit-l to the 3' site inhibits transcription by hindering transcription initiation or by blocking elongation through the occupied binding site. This study employs a series of prolactin promoter constructs in an in vitro transcription system to further investigate the transcriptional regulatory role of a Pit-l binding site 3' to the cap site.
MATERIALS
AND METHODS
Template Preparation. To produce the templates used in this study, two prolactin Bal 31 deletion mutants containing 36 and 66 bp of 5' flanking DNA were used. The 36 bp template possesses a TATA box and cap site only (basal promoter). The 66 bp template has a basal promoter as well as a single proximal Pit-l binding site (-36 to -63) (Fig la). Each of these templates has a =I site inserted 34 bp downstream of the cap site. Into these sites were inserted annealed oligonucleotide DNA fragments possessing either an intact Pit-l binding site (homologous to the proximal Pit-l binding site footprinted region in the rat prolactin gene) or a DNA sequence slightly altered such that Pit-l does not bind (10) *
Structure of prolactin test templates for in vitro (a) The drawings indicate the positions of Pit-l binding sites and altered non-binding sites relative to the Boxes indicate entire footprinted transcription start site. Templates were produced as outlined in Materials region of DNA. (b) Positions of the 5' and 3' prolactin and Methods. oligonucleotides used for primer extension analysis relative to the Pit-l binding site (or altered site) located at +50. Fiqure
1.
transcription.
791
Vol.
177,
No.
2, 1991
BIOCHEMICAL
AND
BIOPHYSICAL
RESEARCH
COMMUNICATIONS
Normal Pit-l binding site: AATTCGATTATATATATATTCATGAAGGTGTCGAATT Altered site (no binding): AGCTTGCCTGATTATATATATCTTCATGAAGGTGTCT This procedure shown in Fig.
resulted la. Their
in the production of the structures were verified
4 templates by sequencing.
All templates were assayed by in vitro Assay method. transcription with GHB nuclear extracts prepared by a method In our protocol, potassium modified from Dignam (11,12). glutamate replaced potassium chloride in the extract buffers. In each assay, 200 ng of test template and 75 ng of control template pHTX-B, a plasmid containing three copies of the adenovirus major late promoter, were incubated in 15 ~1 of GH3 nuclear extract. The transcripts produced by in vitro transcription were assayed by primer extension (7) using [32P]-labelled oligonucleotide primers homologous to specific sequences within the RNA products from the prolactin and major late promoter templates. Two primers were used for the prolactin RNA assays. The first (or 5') primer is homologous to a 25 nucleotide region of RNA located from +20 to +45 and produces a 45 nucleotide extension product (Fig lb). The second (or 3') primer recognizes a region from +98 to +127 and results in a 127 nucleotide product (Fig lb). As shown in Fig lb, the 5' primer recognizes a sequence upstream of the normal or altered Pit-l binding sites at +50 which were cloned into each template. It is designed to measure the total level of transcription initiation for each template. The 3' primer is homologous to a region downstream of the inserted binding sites and is designed to measure the level of transcription which proceeds through an occupied Pit-l binding site versus transcription through an unoccupied sequence. Two adenovirus major late promoter primers which produced a 54 nucleotide primer extension product were used to measure control template transcription. A 30 nt MLP oligomer was used along with the 3' prolactin primer and a shorter 25 nt MLP oligomer was used with the 5' prolactin primer. The shorter primer was used so that the labelled primer would not interfere with the 45 nt prolactin primer extension product when assayed on an acrylamide gel (Fig 2). Bands corresponding to the expected size product from the primer extension analyses were quantified using a Betascope (Betagen). Each template product was then standardized versus the averaged levels of major late promoter control template transcription and transcription values were expressed as percentages of the level of transcription obtained from template 66. This allowed values from more than one experiment to be combined for statistical analysis (ANOVA, Student-Newman-Keuls test). RESULTS The autoradiograph of a typical experiment (Fig 2) and the graph shown in Fig 3 indicates the relative level of transcription of each template as measured from the 5' and 3' primers. A 4 to 5 fold increase in transcription activity was evident when the proximal Pit-l binding site was present (template 66) as opposed to the transcription level with the 792
Vol.
177,
No.
2, 1991
BIOCHEMICAL
AND
BIOPHYSICAL
3’Primer
RESEARCH
COMMUNICATIONS
SPrimer
prl
127ntm ma
MLP 54 nt
Fiqure 2. Gel analysis of in vitro transcription products assayed by primer extension (described in Methods). [ 32P]labelled products were run on an 8% polyacrylamide gel. Bands corresponding to prolactin (prl) and major late promoter (MLP) template products were quantified using a Betascope (Betagen).
basal
promoter
(template
36).
These results
are consistent
with
other studies (13). Comparison of transcription from template vs template 36-3' as measured from the 5' primer indicated a statistically significant ( Q = 0.01) 2 fold increase in transcription. located
Thus,
3'
to
the
it
appears
that
transcription
start
a Pit-l
site
binding
36
site
can positively
influence transcription. However, an increase in transcription proximal Pit-l was not detected in comparing template 66 (single binding
site)
sites). binding
Our interpretation site may be acting
facilitate the only cap site.
with
template
66-3'
(proximal
of these to attract
and 3'
binding
results is that the 3' transcription factors
to
the establishment of an initiation complex when it is transcription factor binding site in the vicinity of the this activity is rendered insignificant when However,
a properly positioned 5' Pit-l binding site is present and actively enhancing the construction and/or stability of the transcription initiation complex. 793
Vol.
177,
No.
BIOCHEMICAL
2, 1991
AND
BIOPHYSICAL
RESEARCH
COMMUNICATIONS
120
r-
100 80 60
36-3'
NT) El
5’ Primer
3’ Primer
Fiqure 3. Graph indicating the relative transcription activity of each template (expressed as a percentage of activity from template 66) as measured from primers 5' (upstream) and 3' (downstream) of the Pit-l binding site or altered site located 4-50. The graph represents two separate experiments.
Comparisons
of
the
levels
of
transcription
measured
at
from
the
5' and 3' primers in the 36-3' and 66-3' templates showed that a significantly smaller proportion of the RNA produced in the in vitro transcriptions was actually long enough to be measured by primer extension from the 3' (more distal) primer. This difference was not due to random transcription termination since the transcription measured with either primer from the 36 template (which has no 3' Pit-l binding site) was not significantly different and transcription measured from the 66 template with the 3' and 5' primers are assigned equivalent values. Therefore, these results must be due to the occupancy of Pit-l at the 3' binding sites in the 36-3' and 66-3' templates which acts to impede the progress of transcription elongation through the binding site. Thus, Pit-l bound to a recognition element that is 3# to the transcription start site can act to negatively regulate transcription in vitro. DISCUSSION Autoregulation appears to be a widely used means of controlling levels of transcription factors within the cell. A variety of Drosonhila homeodomain proteins, which are structurally related to the POU-domain transcription factors, have been shown to autoregulate their own expression (14-16). There are also reports of Ultrabithorax proteins activating their own expression by binding to downstream Ubx binding sites (17). However, the functional downstream binding sites in Ubx were 794
Vol.
177,
No.
2, 1991
BIOCHEMICAL
AND
BIOPHYSICAL
RESEARCH
COMMUNICATIONS
further downstream (+218 to +306) than the 3# Pit-l binding site in the Pit-l gene (+4 to +34), and resulted in transcription activation rather than repression. Interestingly, the deletion of two more proximal sites (+41 to +88) resulted in an increase in transcription of Ubx. Nevertheless, the presence of a single transcription factor binding site downstream of the transcription start site which functions to negatively autoregulate gene expression is thusfar unique to Pit-I. By placing a Pit-l binding site 3' to the transcription start site in the rat prolactin gene, we were able to gain some insight as to how this repression works in vitro. Primer extension analysis of transcription products using primers homologous to regions 5' and 3' to the Pit-l binding site showed that a binding site occupied by a Pit-I protein can hinder the progression of transcription elongation through the site. The RNA polymerase III transcription factor TFIIIA binds to an internal control region within 5s RNA gene of Xenopus and acts as an assembly factor for the Pol III transcription complex (18). TFIIIA allows the passage of the transcription complex through the binding site while remaining bound to the DNA. This is apparently accomplished by the ability of the zinc-finger DNA binding region of TFIIIA to lift up from the DNA and allow the complex to pass (19). Although these systems are quite different, this functional dissimilarity between TFIIIA and Pit-l may point to a fundamental difference in the mechanism of DNA binding by zinc-finger proteins and POU-domain proteins. Since the Pit-l binding site we inserted into the transcribed region of the prolactin gene is further downstream than the Pit-l binding site in the Pit-l gene (+50 to +77 vs +4 to +34), we cannot rule out the possibility that binding to the site in the Pit-l gene inhibits transcription by sterically interfering with the formation of the initiation complex. However, we feel confident that this inhibition of elongation is genuine and is at least site represses Pitpart of the mechanism by which the 3' binding 1 transcription. This study has also shown that the presence of a single Pit1 binding site downstream from the cap site can have a moderate We feel that this stimulatory effect on transcription. stimulation is due to Pit-l attracting general transcription factors and thereby increasing the local concentration of these This factors, enhancing the assembly of the initiation complex. 795
Vol.
177,
No.
2, 1991
BIOCHEMICAL
AND
BIOPHYSICAL
RESEARCH
COMMUNICATIONS
effect is probably not seen in the templates with two binding sites because the transcription activation by a 5' Pit-l binding site in close proximity to the TATA box is much more efficient at stimulating transcription. The results of this study have increased our understanding of the mechanisms of transcription activation and autoregulation by Pit-l. ACKNOWLEDGMENTS The authors would like to thank Sharon Helsel and Kazi This work was supported by Public for technical assistance. health Service Grant # DK38546 from the National Institutes Health to Z.D.S.
Begum of
REFERENCES 1.
2. 3. 4. 5.
6. 7. 8. 9. 10. Il. 12. 13. 14. 15. 16. 17. 18. 19.
V.R., Elsholtz, H., Lu, L.I.-W., and Nelson, C., Albert, 239, 1400-1405. Rosenfeld, M.G. (1988) Science Mangalam, H.J., Albert, V-R., Ingraham, H.A., Kapiloff, M., Wilson, L., Nelson, C., Elsholtz, H., and Rosenfeld, M.G. (1989) Genes DeV. 3, 946-958. J.W., Smith, K-P., and Sharp, Z.D. (1991) Cao, Z, Poser, Submitted G. and Finney, M. (1991) Cell 64, 475-478. Ruvkun, Ingraham, H.A., Albert, V.R., Chen, R., Crenshaw, E.B.,III, Elsholtz, H.P., He, X., Kapiloff, M.S., Mangalam, H.J., Swanson, L.W., Treaty, M.N., and Rosenfeld, M.G. (1990) Annu. Rev. Physiol. 52, 773-791. He, X., Treaty, M.N., Simmons, D.M., Ingraham, H.A., Swanson, L.W., and Rosenfeld, M.G. (1989) Nature 340, 35-42. Cao, Z., Barron, E.A., and Sharp, Z.D. (1988) Mol. Cell. Biol. 8, 5432-5438. Barron, E.A., Cao, Z., Schneider, B.G., Kraig, E., Carrillo, and Sharp, Z.D. (1989) Mol. Cell. Biol. 9, 817-819. A.J., Chen, R., Ingraham, H.A., Treaty, M.N., Albert, V.R., Wilson, L., and Rosenfeld, M.G. (1990) Nature 346, 583-586. Sharp, Z.D., Helsel, S., Cao, Z., Barron, E.A., and Sanchez, Y. (1989) Nucleic Acids Res. 17, 2705-2722. Dignam, J.D., Lebovitz, R-M., and Roeder, R.G. (1983) Nut. Acids Res. 11, 1475-1489. Cao, Z., Barron, E.A., Carrillo, A.J., and Sharp, Z.D. (1987) Mol. Cell. Biol. 7, 3402-3408. Smith, K.P., Barron, E.A., and Sharp, Z.D. (1991) Submitted Harding, K., Hoey, T., Warrior, R., and Levine, M. (1989) EMBO J. 8, 1205-1212. Biggin, M.D. and Tjian, R. (1988) Cell 53, 699-711. Kuziora, M.A. and McGinnis, W. (1988) Cell 55, 477-485. Krasnow, M.A., Saffman, E.E., Kornfeld, K., and Hogness, D.S. (1989) Cell 57, 1031-1043. Kassavetis, G.A., Braun, B.R., Nguyen, L.H., and Geiduschek, E.P. (1990) Cell 60, 235-245. Miller, J., McLachlan, A.D., and Klug, A. (1985) EMBO J. 4, 1609-1614.
796
