ABSTRACT

We present findings of genetic information conservation between the glucocorticoid response element (GRE) DNA and the c-DNA encoding the glucocorticoid receptor (GR) DNA binding domain (DBD). The regions of nucleotide subsequence similarity to the GRE in the GR DBD occur specifically at nucleotide sequences on the ends of exons 3, 4 and 5 at their splice junction sites. These sequences encode the DNA recognition helix on exon 3, a beta strand on exon 4, and a putative alpha helix on exon 5, respectively. The nucleotide sequence of exon 5 that encodes the putative alpha helix located on the carboxyl terminus of the GR DBD shares sequence similarity with the flanking nucleotide regions of the GRE. We generated a computer model of the GR DBD using atomic coordinates derived from nuclear magnetic resonance spectroscopy (NMR) to which we attached the exon 5 encoded putative alpha helix. We docked this GR DBD structure at the 39 base pair nucleotide sequence containing the GRE binding site and flanking nucleotides which contained conserved genetic information. We observed that amino acids of the DNA recognition helix, the beta strand and the putative alpha helix are spacially aligned with trinucleotides identical to their cognate codons within the GRE and its flanking nucleotides.

INTRODUCTION

The Glucocorticoid receptor (GR) DNA binding domain (DBD) has been well characterized and consists of 150 amino acids which have been shown to be biologically active in vitro (1). We reported earlier that a nucleotide sequence within the c-DNA of the GR DBD shared a high degree of subsequence similarity with a well characterized glucocorticoid response element (GRE) from mouse mammary tumor virus (MMTV) 5' long terminal repeat. This GR DBD c-DNA subsequence encoded a predicted alpha helix structure on the carboxyl flank of the first zinc finger that we proposed as a putative DNA recognition helix (2). Recently, our prediction of the GR DNA recognition alpha helix amino acid sequence, its location within the GR DBD and its orientation toward the DNA within the GRE major groove halfsites was confirmed by NMR (3) and X-ray crystallography (4).

The genomic structure of the human GR gene was recently determined (5); the two zinc fingers of the DBD are separately encoded by two of the ten exons, 3 and 4. The DNA recognition helix encoded in exon 3 is located at the carboxyl terminus of the first zinc finger. Adjacent to the DNA recognition helix is a structure which has been determined by NMR (3) and X-ray crystallography (4) to be a beta strand. This beta strand encoded in exon 4 is located on the amino terminus of the second zinc finger at the splice junction site of exons 3 and 4. This splice site occurs at a conserved glycine residue within the steroid receptor family which connects the GR DNA binding recognition helix and beta strand structures as a bridge which joins the two zinc fingers. The carboxyl terminus of the GR DBD contains a structure which we predicted as an alpha helix and proposed to interact at the GRE (2). This putative alpha helix is encoded in exon 5 and is located adjacent to the second zinc finger at the exon 4 and 5 splice site.

We report herein that genetic information is conserved between a GRE and its flanking nucleotides within GENBANK locus MMTPRGR1 and the cDNA of GR DBD, GENBANK locus HUMGCRA, at nucleotide sequences within exons 3, 4 and 5. These GR DBD cDNA subsequences which are maximally similar to the GRE encode the GR DNA recognition helix, a beta strand and a putative alpha helix located in exons 3, 4 and 5, respectively.

MATERIALS AND METHODS

Nucleotide sequence data was taken from cited references and GENBANK, a computer database of DNA and RNA sequences. LOCAL is a program which searches for maximally similar subsequences between any two amino acid or nucleic acid sequences using a dynamic programming matrix algorithm (6). Gap weighting and mismatch values used were: unity for matches, 0.9 for mismatches and -(0.9 + 1.01 * length) for gaps. PRSTRC is protein secondary structure prediction program which uses a modified Chou-Fasman (7) algorithm. LOCAL and PRSTRC are academic software packages distributed by the Harvard Medical School Molecular Biology Computer Research Resource (MBCRR), Dana-Farber Cancer Institute, Harvard School of Public Health, 44 Binney Street JF815, Boston, Massachusetts 02115.

All computer models were created using QUANTA software running on a Silicon Graphics Inc. IRIS 4D 320-GTX graphics workstation. Quanta is a molecular modeling and display tool developed by Molecular Simulations Inc., 200 Fifth Avenue, Waltham, Massachusetts 02254 which allows the construction of molecular models of DNA sequences, point mutations of existing models and the modeling of small peptides with a selected secondary structure. The model of the GR DBD was derived from NMR coordinates (3). Atomic coordinates of the exon 5 encoded GR putative DNA binding alpha helix were computed using the SEQUENCE BUILDER module of the QUANTA program. This module allows the construction of molecular models of small peptides and folds them into a selected secondary structure. This module was also used to generate coordinates for the 39 base pair B-DNA nucleotide sequence from GENBANK locus MMTPRGR1 used in (figure 3).

RESULTS

Earlier we compared nucleotide sequences containing known hormone response elements (HREs) with cDNA sequences encoding the DBD of steroid receptor proteins (2). Using an algorithm for secondary structure prediction (7), we located putative DNA recognition alpha helices within the DBDs of members of the steroid receptor protein superfamily. At the time of our initial observations (2), others suggested that the structure for DNA interaction by the steroid hormone receptor proteins was via amino acids found within a zinc finger (9). Our predicted structure for the DBD of the steroid hormone receptor protein superfamily suggested that amino acids within a putative alpha helix adjacent to the first zinc finger may be important in specific DNA binding. Therefore, we proposed that the zinc finger may serve as a structural backbone for presentation of a putative DNA recognition alpha helix to DNA major groove halfsites.

Recently the structure for the DBD of the GR was derived from NMR and X-ray crystallography structural determinations (3-4). In these reports, the location and amino acid sequence of a GR DNA recognition helix is described. We obtained the NMR derived atomic coordinates of the GR DBD (personal communication Kaptein) (3) .We compared our predicted GR DNA recognition helix structure and its orientation relative to the GRE with the equivalent NMR derived GR DNA recognition helix. Remarkably, the amino acid sequences were 100% identical and the orientation, toward the DNA, of the sidechains of hydrophilic amino acids Lys 461, Lys 465, Arg 466, Glu 469, Gln 471 and hydrophobic amino acid Val 462 were very similar (2-4) (see figure 1).

A recent computer modeling and simulation study of the GR, estrogen receptor (ER) and progesterone receptor (PR) DBDs interacting with B-form DNA major groove halfsites of GRE and estrogen response element (ERE) via alpha helical structures (10) agrees with our original findings (2). Therefore, the NMR determination of the ER structure (11), the NMR and X- ray crystallographic structural determinatons of the GR on its corresponding GRE as well as computer simulations of ER, GR and PR interacting with DNA confirms our prediction and supports our hypothesis that conservation of genetic information is a site specific determinant for protein/DNA interaction.

In the present study, using the genomic structure of the GR gene (5) as a guide, we conducted computer based nucleotide sequence similarity searches. We compared separately the nucleotide sequence of GR DBD exon 3 ( 1319 to 1485 ), the nucleotide sequence of GR DBD exon 4 (1486 to 1602 ) and the nucleotide sequence of GR DBD exon 5 ( 1603 to 1626 ) which encodes a putative DNA binding alpha helix to MMTPRGR1 nucleotide sequence (-312 to -40). The results are shown in figure 2a. The maximally similar subsequence for exon 3 and MMTPRGR1 was found within a well characterized functional GRE of MMTPRGR1 and within the exon 3 at a region encoding the DNA recognition helix of the GR thus confirming our earlier findings (2). It is significant that this GRE sequence contains the TGTTCT recognition motif and this GRE has been reported by others to be the most critical regulatory element within the MMTV 5' long terminal repeat, for MMTV gene transcription as determined by nuclease footprinting, methylation studies and deletion mutation findings (12-15). The maximal nucleotide subsequence similarity between MMTPRGR1 and exon 4 occured at the region encoding the beta strand at the splice junction site which joins 4 to 3 as described above, but in a different GRE site of MMTPRGR1 than seen with exon 3 (data not shown); however, subsequent comparisons detected sequence similarity for the beta strand encoded region of exon 4 in the same GRE site as seen with exon 3 (figure 2b). The maximally similar subsequence between theputative alpha helix encoded region of exon 5 and the MMTPRGR1 occurred on the right flanking nucleotide sequence of the GRE (see figure 2b).

Since there are multiple codons for the majority of amino acids it is necessary to look at every possible reading frame 5' to 3' on both strands of DNA in order to locate the extent of conserved genetic information . The location of conserved genetic information between the GRE nucleotides and amino acids of the GR DNA recognition helix encoded by exon 3, the beta strand of exon 4 and the putative alpha helix of exon 5 is shown in figure 2c. Trinucleotides identical to codons for hydrophilic amino acids Lys, Arg, Ser, and Glu, as well as, hydophobic amino acids Val and Phe of the DNA recognition helix are found, primarily , in the right halfsite of the GRE. Genetic information is conserved for beta strand amino acids Gln, Asn, Tyr, Leu, and Cys within the minor groove between and extending into both major groove halfsites of the GRE. In addition, trinucleotides identical to codons for amino acids Arg, Lys, and Thr are conserved within the flanking regions of the GRE major groove halfsites. These amino acids are found within our exon 5 encoded putative DNA binding alpha helix of the GR DBD.

The recent elucidation of the genomic structure of the GR gene (5) and the structural determination of the GR DBD3 has allowed us to create a model to study separately the nucleotide sequences of the GR gene, exons 3, 4 and 5, as well as, the structural modules of the GR DBD which they encode in relationship to genetic information conservation at GRE sites. The model of the GR DBD was derived from NMR atomic coordinates of the GR DBD (personal communication, Kaptein) (5). However, critical residues following GR Arg 510 on the carboxyl flank of the GR DBD in the NMR and X-ray structural determinations were disordered, and no coordinates were reported (3-4). Since the amino acid sequence ranging from Arg 510 to Lys 517 contained our putative alpha helix of exon 5, we created an alpha helix of the exon 5 encoded amino acids ranging from 511 to 517 and attached this structure to Arg 510 in our computer model. We also created a 39 bp B-DNA computer model of the nucleotide sequence of MMTPRGR1 containing the GRE and flanking regions which showed genetic sequence similarity to the GR DBD exons 3, 4 and 5 (see figure 2). A model of the GR DBD docked at this 39 bp B-DNA sequence with areas of conserved genetic information highlighted in the protein and DNA is shown in figure 3. Remarkably, the DNA recognition helix and beta strand structures encoded by exons 3 and 4 are aligned with areas of conserved genetic information within the major groove halfsites of the GRE, see figure 3a and 3b, while the putative alpha helix structure encoded by exon 5 alignswith conserved genetic information in the flanking nucleotide regions of the GRE, figure 3c. Although not shown separately, we observed that individual codon sites within the GRE major groove halfsites and flanking regions are spaced so that they are aligned with their cognate amino acids conserved within the DNA recognition helix encoded in exon 3, and the beta strand and putative alpha helix encoded by exons 4, and 5 respectively, of the GR DBD.

DISCUSSION

We have observed that genetic information is conserved within imperfect palindrome nucleotides of the GRE major groove halfsites and flanking regions for GR DNA recognition helix amino acids (459-470) encoded by exon 3, for a beta strand structure containing amino acids (471-476) encoded by exon 4, and for a putative alpha helix containing amino acids (510-517) encoded by exon 5, respectively. We also observed that flanking the GRE major groove halfsites are nucleotide sequences with dyad symmetry, 5'TAAAACGA 3' on the right and 5'CAAAAACT 3' on the left which appear to extend the original TGTTCT imperfect palindrome. It is significant that these same flanking nucleotide sequences were determined earlier to be GR binding sites by Scheidereit et al. (13) using nuclease footprinting. The GR amino acids ranging from 510-517 of our putative alpha helix encoded by exon 5 share genetic information with the GRE flanking nucleotides. These amino acids have been reported to be partially responsible for nuclear localization of the GR protein and are related in sequence to the nuclear localization signal of the simian virus, SV-40, T-antigen (16). Therefore, our observations may indicate that in addition to nuclear localization, these GR amino acids, 510-517 , may also be important in site-specific DNA recognition and transcription initiation as well.

The origin of GR DNA site specific recognition has been reported to occur within the GRE major groove halfsite containing the TGTTCT motif. The GR DBD specifically binds to the TGTTCT major groove halfsite of the GRE as a monomer; this DNA binding reaction induces cooperative dimerization and a subsequent DNA binding interaction in the adjacent GRE major groove halfsite (17). Our findings show that the TGTTCT, or specifically its complementary ACAAGA motif, is rich in codons for the exon 3 encoded GR DNA recognition helix amino acids. (see figure 2c). Recently, van der Waals and hydrogen bonding interactions between GR DBD recognition helix amino acids and nucleotides in the GRE right major groove halfsite have been reported (4,10). Many of these observations concur withour earlier predictions (2) and include amino acids interacting at nucleotide sites which we have identified as their cognate codon/anticodon nucleotides. Therefore, conservation of genetic information offers an explanation for the binding preference reported for GR at the GRE major groove halfsite which contains the TGTTCT motif. Furthermore, MMTPRGR1 contains numerous GRE sites; however, the GRE site which has been reported to preferrentially bind GR and have the highest transcription enhancing activity (15) shares nucleotide sequence similarity with the c-DNA nucleotide subsequences of exons 3, 4 and 5 of the GR DBD which encode the DNA recognition helix, the beta strand and the putative DNA binding alpha helix, respectively, as described above, figure 2a,b, and c. It is interesting to note that the GR has been reported to bind to nucleotide regions within its own gene and down regulate its expression (18). Our findings show that the nucleotide sequence of the GR gene within exon 3 which encodes the DNA recognition helix shares a high degree of nucleotide similarity with a well characterized GRE. Consequently, this GR gene sequence may serve as a regulatory binding site for the GR protein.

The above observations as well as the findings we reported earlier of conservation of genetic information shared among prokaryotic and eukaryotic DNA regulatory proteins' DBDs and their cognate DNA binding sites (2, 19) strongly support our hypothesis of a common site-specific DNA recognition code. The basic mechanism of this recognition appears to be stereochemical complementarity between the proteins' DNA recognition alpha helix amino acids and their cognate codon/anticodon nucleotides within their specific DNA binding sites. Therefore, our findings support a stereochemical theory for the origin of the genetic code (20) based on physicochemical complementarity between amino acids and their cognate codon and/or anticodon nucleotides. Three areas of research support this theory and compliment our findings: 1) Correlations between amino acids' side chain physicochemical characteristics and the nucleotides of their cognate codons (21-23) 2) stereochemical complementarity and structural relationships between amino acids and their cognate codons and/or anticodons nucleotides (24-26) direct in-vitro binding preference for codon nucleotides by cognate amino acids (27-30).

Codon recognition has recently been observed in Tetrahymena group I self-splicing intronic RNA by arginine (29). The arginine sidechain shows stereo-selective binding for its codons AGA, CGA and AGG, which are conserved at the catalytic site in 66 group I sequences (31). These observations of specific amino acid-codon interactions are consistent with our earlier findings (2, 19) as well as those we report herein. In addition, our findings suggest that site specific DNA recognition may involve overlapping reading frames. This notion is supported in that genetic information is condensed in an octamer in the GRE right major groove halfsite which contains codon trinucleotides for six amino acids on each strand reading 5' to 3'. The sequence, 5' AAGAACAG 3', on the antisense strand has codon trinucleotides for hydrophilic amino acids Lys, Arg, Glu, Asn, Thr and Gln. All of these amino acids except Thr occur in the exon 3 encoded DNA recognition helix and exon 4 encoded beta strand. The sequence, 5'CTGTTCTT 3', on the sense strand has codon trinucleotides for hydrophobic amino acids Leu, Cys, Val, Phe, Ser and Leu. These amino acids appear in the beta strand and DNA recognition helix as described above. In addition, the GRE flanking nucleotide sequences, 5' TAAAACGA 3' on the right and 5' CAAAAACT 3' on the left, contain codon trinucleotides for Lys, Arg, and Thr of the putative alpha helix encoded by exon 5.

Our findings suggest that the DNA recognition helices and their cognate DNA binding sites may be conserved remnants of primordial structures capable of molecular recognition. The spacial alignment of amino acids of the exon 3, 4 and 5 encoded structures of the GR DBD with trinucleotides identical to their cognate codons within the GRE and its flanks suggests that these structures may have been template dependent in their evolution (i.e. peptides acting as templates for nucleotide polymerization or vice-versa) (32-33). Therefore we propose that prebiotic, template directed autocatalytic synthesis of mutually cognate peptides and polynucleotides resulted in their amplification and evolutionary conservation in contemporary prokaryotic and eukaryotic organisms as a genetic regulatory apparatus.

Our computer analyses were conducted utilizing information derived from published laboratory bench biological research (wet work) of others (1, 3-5, 9-17, 20-33) . Our observations are therefore based on precedence and offer an explanation for site specific DNA binding by DNA regulatory proteins. Our findings are consistent with our hypothesis that the origin of the genetic code and a site specific DNA recognition code have the same underlying mechanism. Furthermore, our results indicate that our hypothesis, applied to genetic sequence analysis, secondary structural prediction and molecular model building, can be used as a predictive tool for determining sites on DNA regulatory proteins which recognize cognate DNA binding sites and vice-versa.

ACKNOWLEDGEMENTS

We thank Molecular Simulations Inc. staff for software support with QUANTA, Michael Fenton of Cray Research Inc. for data reduction programs, the Minnesota Supercomputer Institute Scientific Director, Don Truhlar for support and encouragement, the Minnesota Supercomputer Center user services representatives for technical support on the CRAY-2, Temple Smith of the Harvard Medical School Molecular Biology Computing Research Resource for assistance with sequence analysis software, R. Kaptein for personal communication of GR NMR structural coordinates, and special thanks are due to Charlie Larson of Silicon Graphics Inc. for hardware support with the IRIS 4D 320-GTX workstation. We are sincerely grateful to Professor Thomas C. Spelsberg of the Department of Biochemistry and Molecular Biology, and Pamela D. Popken-Harris of the Department of Immunology, Mayo Foundation, Rochester, MN for preliminary review of the manuscript and encouragement. This work was supported in part by a research grant from the Minnesota Supercomputer Institute, Minneapolis MN. This work was also supported by a research fellowship dedicated to the memory of William Lang JR..

REFERENCES

1. Miesfeld, R., Godowski, P., Maler, B. and Yamamoto, K. Glucocorticoid receptor mutants that define a small region sufficient for enhancer activation. (1987) Science 236, 423-426.
2. Harris, L., Sullivan, M. and Hickok, D. Genetic sequences of hormone response elements share similarity with predicted alpha helices within DNA binding domains of steroid receptor proteins: A basis for site-specific recognition. (1990) Computers and Mathematics with Applications 20, 25-48.
3. Hard, T., Kellenbach, E., Boelens, R., Maler, B., Dahlman, K., Freedman, L., Carlstedt-Duke, J., Yamamoto, K., Gustafsson, J. and Kaptein, R. Sloution structure of the glucocorticoid receptor DNA-binding domain.(1990) Science 249, 157-160.
4. Luisi, B., Xu, W., Otwinowski, Z., Freedman, L., Yamamoto, K. and Sigler, P. Crystallographic analysis of the interaction of the glucocorticoid receptor with DNA. (1991) Nature (London) 352, 497-505.
5. Encio, I., Detera-Wadleigh, S. The genomic structure of the human glucocorticoid receptor. (1990) J. Biol. Chem. 266, 7182-7188.
6. Smith, T. and Waterman, M. Identification of common molecular subsequences. (1981) J. Mol. Biol. 147, 195-197.
7. Chou, P. and Fasman, G. Empirical predictions of protein conformation. (1978) Annu.Rev. Biochem. 47, 251-276.
8. Dayhoff, M. (1978) J. Mol. Biol. 48, 443-453.
9. Evans, R. and Hollenberg, S. Zinc fingers: Gilt by association. (1988) Cell 52, 1-3.
10. Kothekar, V. Transcription regulation by steroid hormones: A computer simulation study. (1992) J. Biomol. Struct. and Dynam. 10, 49-62.
11. Schwabe, J., Neuhaus, D. and Rhodes, D. Solution structure of the DNA-binding domain of the oestrogen receptor. (1990) Nature (London) 348, 458-461.
12. Payvar, F., DeFranco, D., Firestone, G., Edgar, B., Wrange, O., Okret, S., Gustafsson, and J., Yamamoto, K. Sequence-specific binding of glucocorticoid receptor to MTV DNA at sites within and upstream of the transcribed region. (1983) Cell 35, 381-392.
13. Scheidereit, C., Geisse, S., Westphal, H., and Beato, M. The glucocorticoid receptor binds to defined nucleotide sequences near the promoter of mouse mammary tumor virus. (1983) Nature (London) 304, 749-752.
14. Scheidereit, C., and Beato, M. Contacts between receptor and DNA double helix within a glucocorticoid regulatory element of mouse mammary tumor virus. (1984) Proc.Natl. Acad. Sci. USA 81, 3029-3034.
15. Buetti, E., and Kuhnel, B. Distinct sequence elements involved in glucocorticoid regulation of the mouse mammary tumor virus promoter identified by linker scanning mutagenesis. (1986) J. Mol. Biol. 190, 379-389.
16. Picard, P and Yamamoto, K. Two signals mediate hormone-dependent nuclear localization of the glucocorticoid receptor. (1987) EMBO J. 6:11, 3333-3340.
17. Tsai, S., Carlstedt-Duke, J., Weigel, N., Dahlman, K., Gustafsson, J-A., Tsai, M-J. and O'Malley, B. Molecular interactions of steroid hormone receptor with its enhancer element: Evidence for receptor dimer formation. (1988) Cell 55, 361-369.
18. Burnstein, K., Jewell,C. and Cidlowski, J. Human glucocorticoid receptor cDNA contains sequences sufficient for receptor down-regulation. (1990) J. Biol. Chem. 265, 7284-7291.
19. Harris, L., Sullivan, M. and Hickok, D. Conservation of genetic information between regulatory protein DNA binding alpha helices and their cognate operator sites: A simple code for site-specific recognition. (1990) Computers and Mathematics with Applications 20, 1-23.
20. Woese, C. Models for the evolution of codon assignments. (1969) J. Mol. Biol. 43, 235-240.
21. Jungck, J. The genetic code as a periodic table. (1978) J. Mol. Evol. 11, 211-224.
22. Pieber, M., and Toha, J. Code dependent conservation of the physico-chemical properties in amino acid substitutions. (1983) Origins Life 13, 139-146.
23. Sjostrom, M. and Wold, S. A multivariate study of the relationship between the genetic code and the physical-chemical properties of amino acids. (1985) J. Mol. Evol. 22, 272-277.
24. Hendry, L., Bransome Jr., E., Hutson, M. and Campbell, L. A newly discovered stereochemical logic in the structure of DNA suggests that the genetic code is inevitable. (1984) Perspect. Biol. Med. 27, 623-651 .
25. Lacey, J., Mullins Jr., D., and Khaled, M. The case for the anticode. (1984) Origins Life 14, 505-511.
26. Al'tshtein, A. and Efimov, A. Physicochemical basis of origin of the genetic code: Stereochemical analysis of interaction of amino acids and nucleotides on the basis of the progene hypothesis. (1988) Mol. Biol. (Moscow) 22, 1411-1429; (1988) Mol. Biol. Engl. Transl. 22, 1133-1149.
27. Saxinger, C. and Ponnamperuma, C. Interactions between amino acids and nucleotides in the prebiotic milieu. (1974) Origins Life 5, 189-200.
28. Lacey, J. and Mullins, D. Experimental studies related to the origin of the genetic code and the process of protein synthesis - a review. (1983) Origins Life 13, 3-42.
29. Yarus, M. An RNA-amino acids complex and the origin of the genetic code. (1991) New Biol. 3, 183-189.
30. Desjarlais, J. and Berg, J. Redesigning the DNA-binding specificity of a zinc finger protein: A data base-guided approach. (1992) Proteins 12, 101-104, and correction 13, 273.
31. Yarus, M. and Christian, E. Genetic code origins. (1989) Nature (London) 342, 349-350.
32. Nelsestuen, G. Amino acid-directed nucleic acid synthesis. A possible mechanism in the origin of life. (1978) Journal of Molecular Evolution 11, 109-120.
33. Nelsestuen, G. Amino acid catalyzed condensation of purines and pyrimidines with 2-deoxyribose. (1979) Biochemistry 18, 2843-2846.