Viral mutations enhance the Max binding properties of the vMyc b-HLH-LZ domain

Abstract

Interaction with Max via the helix-loop-helix/leucine zipper (HLH-LZ) domain is essential for Myc to function as a transcription factor. Myc is commonly upregulated in tumours, however, its activity can also be potentiated by virally derived mutations. vMyc, derived from the virus, MC29 gag-Myc, differs from its cellular counterpart by five amino acids. The N-terminal mutation stabilizes the protein, however, the significance of the other mutations is not known. We now show that vMyc can sustain longer deletions in the LZ domain than cMyc before complete loss in transforming activity, implicating the viral mutations in contributing to Myc:Max complex formation. We confirmed this both in vitro and in vivo, with loss of Max binding correlating with a loss in the biological activity of Myc. A specific viral mutation, isoleucine383>leucine (I383>L) in helix 2 of the HLH domain, extends the LZ domain from four to five heptad repeats. Significantly, introduction of I383>L into a Myc mutant that is defective for Max binding substantially restored its ability to complex with Max in vitro and in vivo. We therefore propose that this virally derived mutation is functional by significantly contributing to establishing a more hydrophobic interface between the LZs of Myc and Max.

Figure 1

Sequence alignment of the LZ domains of different transcription factors. The amino acids that correspond to position 7 in a heptad repeat are boxed. The leucine repeats are indicated below, with leucine 1 (L₁) and leucine 5 (L₅) being the most N-terminal and C-terminal of the heptad repeats, respectively. Sequences were taken from the following accession numbers: human cMyc (NM_002467), murine cMyc (NM_010849), feline cMyc (M22727), avian cMyc (J00889), avian MC29 vMyc (VO1173), avian MH2 vMyc (K02082), human Nmyc (NM_005378), murine Nmyc (NM_008709), human Max (NM_002382), human cJun (NM_002228), human cFos (BC004490) and human C/EBP (NM_005194).

Figure 2

LZ mutants of MC29 vMyc and avian cMyc. (A) Schematic representation showing location of MC29 vMyc mutations. Five mutations are contained within MC29 vMyc (11). These are found in the transactivation domain (T₆₁>M), adjacent to the basic region (B) (serine₃₂₅>leucine—S₃₂₅>L), within helix 1 (serine₃₅₀>arginine—S₃₅₀>R) and helix 2 (I₃₈₃>L) of the HLH domain, and within the LZ domain (lysine₄₀₇>arginine—K₄₀₇>R). LZ deletion mutants of MC29 vMyc and avian cMyc are shown. The leucine repeats (L₁–L₅) within the LZs of MC29 vMyc and avian cMyc are also indicated. Leucine L_4a is shown offset to indicate its internal location (position 3) within the most C-terminal heptad repeat. Premature translation termination codons (indicated by an asterisk) were introduced into MC29 v-myc and avian c-myc (18) by site-directed mutagenesis. The Δ7 and Δ14 mutants truncate specifically at L₅ and L₄ within the heptad, respectively, whilst the Δ10 truncation removes L₅ and L_4a. (B) Retrovirally-expressed cMyc, vMyc and their respective LZ mutants in CEFs were detected by western blot of total cell lysates using an anti-Myc antibody (upper panel). Equivalence of loading was shown by Max9, the main isoform of Max in CEF (17,26), which was detected using an anti-Max antibody (21) (lower panel). Vector control is shown by a dash.

Figure 3

vMyc can sustain longer deletions in the LZ than cMyc before loss of biological activity (A) Cell morphology of CEF expressing cMyc, vMyc and their respective LZ mutants are shown. (a) cMyc, (b) cMycΔ7, (c) cMycΔ10, (d) cMycΔ14, (e) vMyc, (f) vMycΔ7, (g) vMycΔ10, (h) vMycΔ14 and (i) vector control. (B) Growth rate of CEF expressing cMyc, vMyc and their respective LZ mutants. Growth rate was measured by cumulative cell counts over 4 days. These data are representative of at least two different experiments. (C) The ability of retrovirally-expressed cMyc, vMyc and their respective LZ mutants to induce anchorage-independent growth was determined by plating infected CEF into soft agar. Colony counts were taken after 14 days. Vector control is shown by a dash. These data are representative of at least two different experiments.

Figure 4

Complex formation between Max/Max9 and the LZ mutants of v- and c-Myc in vitro. cMyc, vMyc, LZ mutants and Max/Max9 were produced in rabbit reticulocyte lysate and Myc/Max complex formation determined by immunoprecipitation of Max using an anti-myc specific antibody. [³⁵S]-labelled proteins were resolved by SDS–PAGE and visualized by fluorography. Central lane was Max/Max9 alone.

Figure 5

Complex formation between Max/Max9 and the LZ mutants of v- and c-Myc in yeast. Two different experimental systems based on sensitive yeast two hybrid assays were used to measure complex formation in vivo. (A) Max/Max9-dependent dimerization and DNA binding were determined in vivo. β-Galactosidase activity was quantitatively measured as a result of transcriptional activation resulting from the co-transfection of Pho4-cMyc, Pho4-vMyc or their respective LZ mutants into yeast, together with the PHO5 UAS-CYC-LacZ reporter. This assay has been shown previously to a reliable assay for Myc/Max complex formation (28). (B) Dimerization which is independent of Myc/Max DNA binding was measured by co-transfecting Pho4-cMyc, Pho4-vMyc or their respective LZ mutants into yeast, along with the reporter, pGV256-lex-OP (23) and pRS315-lexA-Max9. β-Galactosidase activity was then determined.

Figure 6

I₃₈₃>L mutation in cMycΔ10 partially restores interaction with Max/Max9 in vitro and in vivo. (A) Schematic representation of cMyc, cMycΔ10 and cMycI>LΔ10 mutants. I₃₈₃>L mutation was introduced into cMycΔ10, which does not bind to Max (18). (B) Complex formation between Max/Max9 and cMyc, cMycΔ10, and cMyc I>LΔ10 co-translated in vitro was determined by immunoprecipitation of Max/Max9 using an anti-Myc antibody, and [³⁵S]-labelled proteins resolved by SDS–PAGE and visualized by fluorography. (C) Complex formation between Max/Max9 and Pho4-cMyc, Pho4cMycΔ10 and Pho4-cMycI>LΔ10 was measured in vivo by co-transfecting the appropriate plasmids into yeast. β-Galactosidase activity was then determined.