Skip to content. | Skip to navigation

You are here: Home » Chemistry » Faculty » Goss » Goss Group » Assembly of Transcription Initiation Complex
Document Actions

Assembly of Transcription Initiation Complex

Overexpression of Myc protein has been linked to many human cancers. The fact that Myc/Max has been linked to oncogenic transformation makes this an important model system for cancer research and is a potential therapeutic target. It is known that Myc/Max/Mad are members of the b/HLH/z (basic, helix, loop, helix, leucine zipper) family of transcription factors. Myc/Max is a transcriptional activator and Mad/Max is a transcriptional repressor. Biological studies have revealed a number of the genes that are targets of Myc activation and Mad repression. Recently, structural data on both Myc/Max and Mad/Max heterodimers have become available. In spite of extensive biological studies, however, there is still a gap in understanding how these potential protein partners discriminate amongst one another and form the scaffolding for assembly of a biologically active transcription complex. The detailed kinetic mechanism of assembly and the equilibrium binding affinity are not known. The lack of these data is an important problem because the rate and mechanism of assembly have been shown to influence DNA site selection and order of addition of other transcription components in the Jun-Fos system. Lack of this knowledge makes it impossible to discriminate among proposed cellular assembly mechanisms, understand DNA selection, or to determine how other members of the Myc/Max/Mad network (such as Mix and Miz1) are able to exert their influence on the assembly of the biological transcription complex and influence gene regulation. Such a lack also prevents rational design of therapeutic or biological agents to regulate transcription.


Our long-term goal is to understand the assembly of transcription factor nucleic acid complexes and how the kinetics and intermediates in these processes regulate the composition of the final macromolecular assembly and lead to biological function. The objective of this research project is to determine the kinetics and equilibria for Myc/Max and Mad/Max interactions with DNA. These heterodimers form the initial interaction with DNA for assembly of the macromolecular transcription complex and are therefore the architectural scaffold for biological assembly. Our hypothesis is that formation of a functional transcription factor-DNA complex will require displacement of one or both Max monomers from the Max2-DNA complex. The few b/HLH/z proteins studied to date follow a monomer assembly pathway in vitro where two protein monomers bind DNA sequentially and form their dimerization interface while bound to DNA. It is generally thought that this model represents cellular assembly. We have formulated our quite different hypothesis based on cellular concentrations of protein and preliminary data. At cellular concentrations, it is unlikely that much monomer is present. Because Max is stably expressed at concentrations that far exceed Myc, based on our preliminary data, the steady-state species most likely is Max2-DNA. A functional model must, therefore, include dissociation of one or both of the Max monomers from DNA. Kinetic studies combined with equilibria to measure the rate-limiting steps will elucidate the relevant functional pathways and allow us to discriminate among proposed cellular models. The rationale for this proposed research is that, once it is understood how the Myc/Max and Mad/Max complexes assemble on DNA, it is expected that it will become possible to elucidate the effects of other transcription factors, such as Miz, on stability and composition of the complex and to modulate those interactions for therapeutic purposes. These studies will increase our understanding of the molecular mechanism of gene activation and identify potential sites for small molecule or genetic intervention. In the long term, these studies are expected to guide the targeting and development of novel biological and therapeutic agents against a broad spectrum of cancer and inflammatory diseases.


B.1. Helix-loop-helix Transcriptional Activators.

The transcriptional activators can be classified according to family groupings on the basis of sequence homology, oligomerization properties and three dimensional structural similarity of modular DNA binding domains (Harrison 1991; Burley SK 1994). Helix-loop-helix transcription factors, first identified by Baltimore and coworkers, were predicted to function as protein dimers and possess a modular DNA binding/dimerization motif consisting of two alpha helices (H1 and H2) separated by a loop (L) (Murre C, McCam PS, and Baltimore D 1989). Most of the helix-loop-helix proteins possess a conserved basic (b) region immediately N-terminal to H1. This region is generally required for DNA binding (Prendergast GC, Lawe D, and Ziff EB 1991; Vinson CR and Garcia KC 1992; Fisher F and Goding CR 1992). Some of the HLH family members have a conserved heptad repeat or zipper (Z) C-terminal to H2. A large number of biologically important proteins contain all three motifs: basic (b), helix-loop-helix (HLH) and leucine zipper (Z). HLH proteins can thus be considered to occur in three types: HLH, b/HLH, and b/HLH/Z. Family members are found in diverse eucaryotic systems and regulate metabolism, mediate stress responses and control cell differentiation and development (reviewed in (Sun and Baltimore 1991; Grandori, Cowley, James, and Eisenman 2000)).

Most b/HLH and b/HLH/Z transcription factors contain two independent domains: one involved in DNA binding and one for transcriptional activation. The DNA binding domains dimerize and interact with DNA targets with nanomolar binding affinity. The DNA targets, E boxes, are located in the proximal region of class II nuclear gene promoters, between 50 and 200 base pairs upstream of the transcription start site. The activation domains interact with and stabilize other elements of the transcription machinery bound to the core promoter and thereby regulate the efficiency of mRNA synthesis initiation. Some transcriptional activators may act synergistically with one another when bound to different recognition sites near the same promoter. 

B.2. Structural Basis for Dimerization and DNA Recognition.

The relative functional simplicity combined with the biological importance of the HLH family of proteins has made them the object of extensive biological and structural studies (Patikoglou G and Burley SK 1997; Grandori et al. 2000). Detailed three-dimensional structural information is available for the Max b/HLH/Z homodimer (Ferre-D'Amare AR et al. 1993; Brownlie et al. 1997), Myc-Max and Mad-Max heterodimers (Nair and Burley 2003), among others, each with their respective DNA targets. The structure of b/HLH/Z protein Max with its target DNA will be briefly discussed as a representative because there are many similarities among the structures for Max2-DNA, Myc-Max-DNA and Mad-Max-DNA. The structure of the Max b/HLH/Z domain complexed with DNA shows that the HLH motif dimerizes into a globular four-helix bundle (consisting of Helix 1 and Helix 2 from both proteins) with a well-defined hydrophobic core. The two basic regions project into the major groove of the DNA like scissors and make a number of contacts with the bases and backbone of the DNA. The other pair of alpha helices forms the leucine zipper region of the molecule. The four-helix bundle appears to confer rigidity such that the orientation of the basic regions of these proteins recognize palindromic sequences without spacing between the repeated elements. In contrast, b/Z proteins, lacking the globular core usually bind inverted repeats with a variable spacing of one or two base pairs (reviewed by (Patikoglou G and Burley SK 1997).


Structure of Max homodimer bound to DNA. This figure is based on the published crystal structure (Ferre-D'Amare AR et al. 1993) and reconstructed from the coordinates deposited in the protein database (pdb file: 1an2). The interactions between hydrophobic residues in the leucine zipper are shown on the right. The basic region is continuous with helix 1 while the leucine zipper is continuous with helix 2. The basic region contacts the DNA base pairs directly. Additional contacts are made with the loop region and DNA phosphate backbone.

Recent (Nair and Burley 2003) crystallographic structures of Myc-Max and Mad-Max with DNA reveal similar secondary structural elements. The Max homodimer structure reveals a Gln 91-Asn 92-Gln 91-Asn92 tetrad near the C-terminal end of the zipper which leads to a packing defect within the interface. In contrast, the leucine zipper of Myc-Max and Mad-Max have tighter intermolecular packing due to hydrogen bond formation and more closely resemble the coiled coils found in GCN4 homodimers (Ellenberger et al. 1994). Perhaps one of the most interesting findings from the crystallographic data is the formation of a heterodimer tetramer with two Myc-Max heterodimers, each binding a DNA E-box. These heterodimers are oriented head to tail of the leucine zipper and it has been proposed that biologically they may bind widely separated E-box sequences. However, the crystal structure reveals no specific hydrogen bond interactions to stabilize this tetramer. The interactions are hydrophobic involving alpha helices H1, H2 and the leucine zipper. We have not found evidence of tetramer formation in solution for our protein constructs; however, we will continue to explore this possibility with both cross-linked and wild-type proteins.

B.3. Helix-Loop-Helix Recognition of the E-box (5'CANNTG3').

The three regions of each protein that make contact are the first residue of H2, the loop and the basic region. There are two invariant residues in the basic regions of b/HLH and b/HLH/Z proteins capable of E-box binding. These are glutamine 32 and arginine 35 in Max. These two amino acids are implicated in recognition of the outer two base-pairs of the E-box. Histidine 28 in Max structure makes a hydrogen bond with the N7 of the outermost guanine of the E-box. Identical protein-DNA contacts were observed for the Myc-Max heterodimers to those reported for the Max homodimer and Mad-Max heterodimer. Several additional contacts were observed between residues specific to Myc and the phosphate backbone. The role of loop residues of bHLH proteins in contributing to DNA binding has been analyzed (Winston and Gottesman, 2000).

Circular dichroism spectroscopic investigations of DNA binding by the isolated b/HLH domain of MyoD (Anthony-Cahill SJ et al. 1992), the b/HLH/Z domain of TFEB (Fisher et al. 1992), full length USF (Ferre-D'Amare AR et al. 1994), as well as Myc and Max (Fieber et al. 2001) reveal that these proteins undergo

B.4. Biological Functions of Myc, Max and Mad

Myc has been extensively studied over the last almost 20 years. It is involved in a wide range of cellular functions. Myc was originally identified as the oncogenic effector of avian retroviruses inducing lymphoid tumors (Nesbit CE et al. 1999). Activation of Myc is also seen in human tumors (reviewed in (Grandori et al. 2000),(Bange J et al. 2001)) including lymphoid malignancies, lung cancer, breast cancer and colon cancer (Erisman et al. 1985). The proteins encoded by the Myc family of genes are predominantly localized in the cell nucleus and expression generally correlates with cell proliferation. Myc promotes oncogenic transformation by regulating target genes that drive cell proliferation and promote angiogenesis (Grandori et al. 2000). Myc protein has a short half-life of 20-30 min (Hann and Eisenman 1984) and is rapidly degraded. Because neither dimerization nor DNA specific binding could be readily demonstrated for Myc protein, except at very high concentrations, a search for Myc interacting proteins lead to the identification of Max. All known oncogenic functions of Myc require dimerization with Max (Amati et al. 1992-). The transcription activation function of Myc is mediated in part by recruitment of a histone acetyltransferase (McMahon SB et al. 2000). Myc interacts with other cellular factors including AP-2 (Gaubatz et al. 1995), Miz-1 (Peukert K et al. 1997) and Nmi (Bao J and Zervos AS 1996). All of these interactions depend on the bHLHz region of the Myc-Max heterodimer suggesting that this region acts to target factors to particular promoter regions and may play an architectural role in nucleoprotein assemblies.

Max is a stably expressed b/HLH/Z protein that forms homodimers and heterodimers with Myc oncoprotein. These dimers bind DNA under physiologic conditions. Max will also form heterodimers with the b/HLH/Z protein Mad (Lahoz et al. 1994; Roussel MF et al. 1996; Cerni et al. 1995). In vivo Max exists in at least three dimeric states: Max-Max, Myc-Max and Mad-Max. All of these dimers bind the sequence 5'CACGTG3' (E-box), although the relative affinities are different. Max, unlike Myc and Mad, lacks a functional activation domain and consequently acts as a transcriptional repressor when in the homodimeric form and is believed to function by sequestering the DNA targets normally recognized by b/HLH/Z activators.

Mad family proteins were all identified in expression cloning screens by their ability to bind specifically to Max (Ayer D.E. and Eisenman 1993; Zervos et al. 1993; Hurlin et al. 1995). In their biochemical behavior, Mad is much like Myc in that it homodimerizes poorly but can form specific complexes with Max. Mad-Max heterodimers recognize the E-box sequence as do Myc-Max heterodimers (Blackwell et al. 1993). Recently, specific requirements for flanking sequences have been proposed (O'Hagan RC et al. 2000). Transcription assays show that in contrast to Myc, Mad acts to repress E-box dependent expression of synthetic reporter genes (Ayer D.E. et al. 1993; Hurlin et al. 1995; Schreiber-Agus N et al. 1995). Overexpression of Mad in tissue culture and in mice interferes with cell proliferation and blocks the cooperative transformation by Myc and Ras (Grandori et al. 2000). The ability of Mad to inhibit cell proliferation is linked to its transcriptional function. Mutants of Mad that do not associate with Max or bind DNA are inactive in biological assays (Hurlin et al. 1995; Schreiber-Agus N et al. 1995; Roussel MF et al. 1996). Mad-Max heterodimers interact with mSin3, a corepressor (Ayer D.E. et al. 1995) and recruit mSin3 to the DNA promoter. This leads to condensation of chromatin structure and reduced transcription.

B.5. Mechanism of Assembly of Protein Dimer-DNA Complex

Myc-Max and Mad-Max heterodimers have unique function and DNA specificity. Dimeric DNA binding motifs, in contrast to monomeric motifs, offer a greater variety of target site recognition and functional diversity. Dimerization may also offer kinetic advantages for DNA binding. There are two proposed pathways for assembly (Kim B and Little JW 1992; Berger et al. 1998; Kohler et al. 1999). The first pathway, the monomer pathway, involves the binding of monomeric protein to DNA followed sequentially by binding of the second monomeric protein and assembly of the dimerization interface while bound to DNA. In the second pathway, the two protein monomers form a dimer first and then bind to DNA (dimer pathway). Since the two pathways form a thermodynamic cycle, the stability of the final complex will be the same regardless of the pathway followed. However, the monomer pathway, at least in some cases, offers a faster route to assembly of the complex (Kohler et al. 1999; Kohler and Schepartz 2001; Park C et al. 1996). Both of these models assume the starting point for assembly is DNA interacting with proteins. However, at cellular concentrations (Kato et al. 1992), the stably expressed species is likely Max2-DNA. While this homodimer-DNA species must assemble from DNA and Max by one of the above models, competition with Myc or Mad is likely to be more complex (See B.7 below).

Jun-Fos, Max, and ATF-2 (Kohler and Schepartz 2001; Kohler et al. 1999) have all been shown to assemble via the monomer pathway. Kinetic data have shown that assembly of the complex occurs at a rate that is inconsistent with prior formation of a protein dimer (Kohler et al. 1999). Demonstration of the formation of DNA-protein monomers further supports this mechanism (Kim B and Little JW 1992; Park C et al. 1996). However, no detailed kinetics have been reported. These studies have shown that electrostatic effects between the protein and DNA are important and accelerate the rate of complex formation (Schreiber G and Fersht AR 1996). It has also been proposed that the unfolded structure of the monomeric protein allows for a larger capture radius for the DNA (Shoemaker BA et al. 2000). Since Myc, Max and Mad have been shown to have helix 1 unfolded in the monomeric state (Fieber et al. 2001), this would allow for an increased rate of molecular recognition. These studies have also shown that the monomer-DNA intermediate is relatively unstable compared to the final complex (Kim B and Little JW 1992; Park C et al. 1996; Kohler and Schepartz 2001). This instability will allow the monomer complex to dissociate from non-productive DNA binding sites and not become kinetically trapped at the "wrong" binding site. We are determining the kinetics of the individual steps of the assembly process. By examining these initial steps we will be better able to determine how assembly is controlled and intervention points for mediating gene expression.

Reference List

  1. Amati, B., S. Dalton, M. W. Brooks, T. D. Littlewood, G. I. Evan, and H. Land. 1992-. "Transcriptional Activation by the Human C-Myc Oncoprotein in Yeast Requires Interaction With Max." Nature 359:423-26.
  2. Anthony-Cahill SJ, Benfield PA, Robert F, Wasserman ZR, Stafford WFI, Altenbach C, Hubbel WL, and DeGrado WF. 1992. "Molecular Characterization of Helix-Loop-Helix Peptides." Science 255:979-83.
  3. Ayer D.E. and R. N. Eisenman. 1993. "A Switch From Myc:Max to Mad:Max Heterocomplexes Accompanies Monocyte/Macrophage Differentiation." Genes Dev 7(0890-9369. 11):2110-2119.
  4. Ayer D.E., L. Kretzner, and R. N. Eisenman. 1993. "Mad: a Heterodimeric Partner for Max That Antagonizes Myc Transcriptional Activity." Cell 72(0092-8674. 2):211-22.
  5. Ayer D.E., Q. A. Lawrence, and R. N. Eisenman. 1995. "Mad-Max Transcriptional Repression Is Mediated by Ternary Complex Formation With Mammalian Homologs of Yeast Repressor Sin3." Cell 80(0092-8674. 5):767-76.
  6. Bange J, Zwick E, and Ullrich A. 2001. "Molecular Targets for Breast Cancer Therapy and Prevention." Nat Med 7:548-52.
  7. Bao J and Zervos AS. 1996. "Isolation and Characterization of Nmi, a Novel Partner of Myc Proteins ." Oncogene 12:2171-76.
  8. Berger, CL, L Piebelli, U Haditsch, and HR Bosshard. 1998. "Diffusion Controlled DNA Recognition by an Unfolded, Monomeric B-Zip Transcription Factor." FEBS Lett. 425:14-18.
  9. Blackwell, T. K., J. Huang, A. Ma, L. Kretzner, F. W. Alt, R. N. Eisenman, and H. Weintraub. 1993. "Binding of Myc Proteins to Canonical and Noncanonical DNA Sequences." Mol. Cell. Biol. 13(0270-7306. 9):5216-24.
  10. Brownlie, P., T. Ceska, M. Lamers, C. Romier, G. Stier, H. Teo, and D. Suck. 1997. "The Crystal Structure of an Intact Human Max-DNA Complex: New Insights into Mechanisms of Transcriptional Control." Structure 5(0969-2126. 4) 509-20.
  11. Burley SK. 1994. "DNA-Bindng Motifs JJFrom Eukaryotic Transcription Factors." Curr Opin Struct. Biol. 4:3-11.
  12. Cerni, C., K. Bousset, C. Seelos, H. Burkhardt, M. Henriksson, and B. Luscher. 1995. "Differential Effects by Mad and Max on Transformation by Cellular and Viral Oncoproteins." Oncogene 11(0950-9232. 3):587-96.
  13. Ellenberger, T., D. Fass, M. Arnaud, and S. C. Harrison. 1994. "Crystal Structure of Transcription Factor E47: E-Box Recognition by a Basic Region Helix-Loop-Helix Dimer." Genes Dev 8(0890-9369. 8):970-980.
  14. Erisman, MD, PG Rothberg, RE Diehl, CC Morese, JM Spandorfer, and SM Astrin. 1985. Mol Cell Biol 5:1969-76.
  15. Ferre-D'Amare AR, Pognonec P, Roeder RG, and Burley SK. 1994. "Structure and Function of the B/HLH/z Domain of USF." EMBO J. 13:180-189.
  16. Ferre-D'Amare AR, Prendergast GC, Ziff EB, and Burley SK. 1993. "Recognition by Max of Its Cognate DNA Through a Dimeric B/HLH/Z Domain." Nature 363(0028-0836. 6424):38-45.
  17. Fieber, W, Schneider ML., Matt T., Krautler B, Konrat R., and Bister K. 2001. "Structure, Function, and Dynamics of the Dimerization and DNA-Binding Domain of Oncogenic Transcription Factor V-Myc." J. Mol. Biol. 307(0022-2836. 5) 1395-410.
  18. Fisher, D. E., L. A. Parent, and P. A. Sharp. 1992. "Myc/Max and Other Helix-Loop-Helix/Leucine Zipper Proteins Bend DNA Toward the Minor Groove." Proc Natl Acad. Sci. USA 89(0027-8424. 24):11779-83.
  19. Fisher F and Goding CR. 1992. "Single Amino Acid Substitutions Alter Helix-Loop-Helix Protein Specificity for Bases Flanking the Core CANNTG Motif." EMBO J 11:4102-9.
  20. Gaubatz, S., A. Imhof, R. Dosch, O. Werner, P. Mitchell, R. Buettner, and M. Eilers. 1995. "Transcriptional Activation by Myc Is Under Negative Control by the Transcription Factor AP-2." EMBO J 14(0261-4189. 7):1508-19.
  21. Grandori, C., S. M. Cowley, L. P. James, and R. N. Eisenman. 2000. "The Myc/Max/Mad Network and the Transcriptional Control of Cell Behavior." Annu. Rev. Cell. Dev. Biol. 16(1081-0706):653-99.
  22. Hann, SR and RN Eisenman. 1984. "Proteins Encoded by the Human C-Myc Oncogene: Differeintial Expression in Neoplastic Cells." Mol Cell Biol 4:4286-97.
  23. Harrison, S. 1991. "A Structural Taxonomy of DNA-Binding Domains." Nature 353:715-19.
  24. Hurlin, P. J., C. Queva, P. J. Koskinen, E. Steingrimsson, D. E. Ayer, N. G. Copeland, N. A. Jenkins, and R. N. Eisenman. 1995. "Mad3 and Mad4: Novel Max-Interacting Transcriptional Repressors That Suppress C-Myc Dependent Transformation and Are Expressed During Neural and Epidermal Differentiation." EMBO J 14(0261-4189. 22):5646-59.
  25. Kato, G. J., W. M. Lee, L. L. Chen, and C. V. Dang. 1992. "Max: Functional Domains and Interaction With C-Myc." Genes Dev 6(0890-9369. 1):81-92.
  26. Kim B and Little JW. 1992. "Dimerization of a Specific DNA Binding Protein on the DNA." Science 255:203-6.
  27. Kohler, JJ, SJ Metallo, TL Schneider, and A Schepartz. 1999. "DNA Specificity Enhanced by Squential Binding of Protein Monomers." Proc Natl Acad Sci USA 96:11735-39.
  28. Kohler, JJ and A Schepartz. 2001. "Kinetic Studies of Fos-Jun-DNA Complex Formation: DNA Binding Prior to Dimerization." Biochemistry 40:130-142.
  29. Lahoz, EG, L. Xu, N. Schreiber-Agus, and RA DePinho. 1994. "Suppression of Myc, but Not Ela, Transformation Activity by Max-Associated Proteins, Mad and Mxi1." Proc. Natl. Acad. Sci. USA 91:5503-07.
  30. McMahon SB, Wood MA, and Cole MD. 2000. "The Essential Cofactor TRRAP Recruits the Histone Acetyltransferase HGCN5 to CMyc." Mol Cell Biol 20:556-62.
  31. Murre C, McCam PS, and Baltimore D. 1989. "A New DNA Binding and Dimerization Motif in Immunoglobulin Enhancer Bindng, Daughterless, MyoD and Myc Proteins." Cell 56:777-83.
  32. Nair, S. K. and S. K. Burley. 2003. "X-Ray Structures of Myc-Max and Mad-Max Recognizing DNA. Molecular Bases of Regulation by Proto-Oncogenic Transcription Factors." Cell 112(0092-8674. 2):193-205.
  33. Nesbit CE, Tersak JM, and Prochownik EV. 1999. "Myc Oncogenes and Human Neoplastic Disease." Oncogene 18:3004-19.
  34. O'Hagan RC, Schreiber-Agus N, David G, and Engelman JA. 2000. "Gene Target Recognition Among Members of the Myc Superfamily and Implications for Oncogenesis." Nat Genet 24:113-19.
  35. Park C, Campbell JL, and Goddard 3rd WA. 1996. "Can the Monomer of the Leucine Zipper Proteins Recognize the Dimer Binding Site Without Dimerization." J. Amer. Chem Soc 118:4235-39.
  36. Patikoglou G and Burley SK. 1997. "JEukaryotic Transcription Factor:DNA Complexes." Annul Rev Biophys Biomol Struct(26):287-323.
  37. Peukert K, Staller P, Schneider A, Carmichael G, Hanel F, and Eilers M. 1997. "An Alternative Pathway for Gene Regulation by Myc." EMBO J 16:5672-86.
  38. Prendergast GC, Lawe D, and Ziff EB. 1991. "Association of Myn, the Murine Homolog of Max With C-Myc Stimulates Methylation-Sensitive DNA Binding and Ras Cotransformation." Cell 65:395-407.
  39. Roussel MF, RA Ashmun, CJ Sherr, RN Eisenman, and DE Ayer. 1996. "Inhibition of Cell Proliferation by the Mad1 Transcriptonal Repressor." Mol. Cell. Biol. 16:2796-801.
  40. Schreiber-Agus N, Chin L, Chen K, Torres R, and Rao G. 1995. "An Amino Terminal Domain of Mxi1 Mediates Anti-Myc Oncogenic Activity and Interacts With a Homolog of Yeast Repressor SIN3." Cell 80:777-86.
  41. Schreiber G and Fersht AR. 1996. "Rapid, Electrostatically Assisted Association of Proteins." Nat Struc Biol 3:427-31.
  42. Shoemaker BA, Portman JJ, and Wolynes PG. 2000. Proc Natl Acad Sci USA 97:8868-73.
  43. Sun, X-H and D Baltimore. 1991. "An Inhibitory Domain of E12 Transcription Factor Prevents DNA Binding i E12 Homodimers Bu Not in E12 Heterodimers." Cell 64:459-70.
  44. Vinson CR and Garcia KC. 1992. "Molecular Model for DNA Recognition by the Family of Basic-Helix-Loop-Helix-Zipper Proteins." New Biol 4396-403.
  45. Zervos, A. S., J. Gyuris, and R. Brent. 1993. "Mxi1, a Protein That Specifically Interacts With Max to Bind Myc-Max Recognition Sites." Cell 72(0092-8674. 2):223-32.