Oligomerization properties of GCN4 leucine zipper e and g position mutants

Thermodynamic cycle analysis of mutant coiled coils indicated that a charged residue at a g position at residue i can interact favorably or unfavorably with a charged residue at the e' position at i+5.

“Thermodynamic cycle analysis of mutant coiled coils showed that charged residues at a g position at residue i can interact favorably or unfavorably with a charged residue at the e' position at i + 5 (Krylov et al., 1994; Zhou et al., 1994a, 1994b; Kohn et al., 1995), consistent with the orientation of ion pairs observed in the crystal structures.”

• cites Krylov et al., 1994 sourceSupported
• cites Zhou et al., 1994aNot matched

  Couldn’t be matched to a paper in Crossref, so it can’t be checked.

• cites Kohn et al., 1995 sourcePartly supported

  The Conclusions report that an i-to-i′+5 Glu–Glu interaction destabilized the mutant coiled coil, supporting the unfavorable-interaction part of the claim. They explicitly say that double-mutant analysis was not necessary for this study; the passages provided do not show a favorable interaction demonstrated by Kohn et al. (LLM, conf=0.88)

  Experimental — accuracy not yet measured

Altering the amino acids at the e and g positions can be enough to change a leucine zipper's dimerization specificity.

“Changing the amino acids at the e and g positions can be sufficient to change the dimerization specificity of a leucine zipper (Schuermann et al., 1991; O'Shea et al., 1992, 1993; Vinson et al., 1993; John et al., 1994; Zhou et al., 1994a, 1994b).”

• cites Schuermann et al., 1991 sourcePartly supported

  In “Determinants of preferential dimer formation” (page unavailable), the paper concludes that substitutions at position g change binding specificity, but says this is only probable for position e. The claim presents the effect of changing residues at e and g as established, whereas the evidence for e is qualified. (LLM, conf=0.83)

  Experimental — accuracy not yet measured
• cites O'Shea et al., 1992 sourceNot checked — no full text of the source
• cites Vinson et al., 1993 sourceSupported

About 90% of the GCN4 leucine zipper e/g mutants in the pool do not form stable homooligomers under the lower-expression screen.

“About 90% of the mutants do not form stable homooligomers.”

• Also stated in: § Identification of homooligomeric mutants · p. 2 · plain

  “Among 111 independent subclones screened for phage immunity, only about 10% of the mutants formed enough functional homooligomers at this lower expression level to confer immunity to λ (data not shown).”

Among 25 mutants tested in both reporters, every mutant that expressed lacZ from λ112OsP at 10% or less of the unrepressed control and could not grow on 150 μg/mL Cm in the λ112OsPs strain showed a repression pattern consistent with higher-order oligomer formation.

“Among 25 mutants tested in both reporters, every mutant that expressed lacZ from λ112OsP at 10% or less of the unrepressed control and was unable to grow on 150 μg/mL Cm in the λ112OsPs strain showed a pattern of repression consistent with formation of higher-order oligomers.”

• AssumptionRests on an assumption: The oligomerization behavior of the λ repressor-leucine zipper fusion proteins reflects the oligomerization properties of the leucine zipper domain.

Sedimentation velocity experiments indicate that the n01 fusion protein consists of a complex mixture of different oligomeric species.

“Sedimentation velocity experiments indicate that the n01 fusion protein consists of a complex mixture of different oligomeric species (data not shown).”

• AssumptionRests on an assumption: The oligomerization behavior of the λ repressor-leucine zipper fusion proteins reflects the oligomerization properties of the leucine zipper domain.

Sedimentation of fusion proteins containing the wild-type GCN4 leucine zipper is consistent with a mixture of monomer-dimer and monomer-tetramer equilibria with dissociation constants of about 3.6 × 10-6 M and 2 × 10−15 M3, respectively.

“Sedimentation of fusion proteins containing the wild-type GCN4 leucine zipper is consistent with a mixture of monomer-dimer and monomer-tetramer equilibria with dissociation constants of about 3.6 × 10-6 M and 2 × 10−15 M3, respectively (Fig. 2A).”

• AssumptionRests on an assumption: The oligomerization behavior of the λ repressor-leucine zipper fusion proteins reflects the oligomerization properties of the leucine zipper domain.

The distributions of net ion-pair change for interacting and noninteracting mutant pairs overlap, even when one-way titration pairs are excluded.

“Even if we exclude those pairs where one-way titration occurs, the distributions for interacting and noninteracting pairs of mutants are overlapping.”

No clear correlation was observed between the number of repulsive, attractive, or net ion pairs and the ability to form heterodimers.

“In no case did we observe a clear correlation between the number of repulsive, attractive, or the net ion pairs and the ability to form heterodimers (not shown).”

The three consensus sequences are not matched by any of the 45 nonfunctional mutants, the 13 Type D mutants, or the 67 natural leucine zippers.

“None of the 45 nonfunctional mutants (not shown; Hu et al., 1993), or the 13 Type D mutants (Table 2), or the 67 naturally occurring leucine zippers (Hurst, 1994) match any of the three consensus sequences.”

• cites Hu et al., 1993Not matched

  Couldn’t be matched to a paper in Crossref, so it can’t be checked.

• cites Hurst, 1994Not checked — no full text of the source

The remaining 13 mutants repressed cat and lacZ from λ112OPs less than 10-fold.

“The remaining 13 mutants repressed cat and lacZ from λ112OPs less than 10-fold.”

• AssumptionRests on an assumption: The oligomerization behavior of the λ repressor-leucine zipper fusion proteins reflects the oligomerization properties of the leucine zipper domain.

The Type III mutant n01 fusion protein formed a variety of higher-order oligomers depending on solution conditions.

“The Type I mutant behaved as a tetramer under all tested conditions, whereas the Type III mutant formed a variety of higher-order oligomers, depending on the solution conditions.”

• Also stated in: § Higher-order oligomers form in vitro · p. 3 · hedged

  “By contrast, the class III mutant n01 behaved as if it formed higher-order aggregates.”

• AssumptionRests on an assumption: The oligomerization behavior of the λ repressor-leucine zipper fusion proteins reflects the oligomerization properties of the leucine zipper domain.

Putative repulsive ion pairs are absent among the homodimeric mutants.

“Putative repulsive ion pairs are not found among the homodimeric mutants.”

Stable homodimers make up less than 3% of the mutant pool.

“Stable homodimers comprise less than 3% of the pool; several g-e'positions in these mutants could form attractive ion pairs.”

The pattern of repressor activities for the Type D mutants was similar to the dimeric controls.

“The pattern of repressor activities for these mutants (Type D) was similar to the dimeric controls (Table 1).”

• AssumptionRests on an assumption: The oligomerization behavior of the λ repressor-leucine zipper fusion proteins reflects the oligomerization properties of the leucine zipper domain.

The n01 peaks elute at apparent molecular weights on the order of 140-200 kDa.

“The peaks elute at apparent molecular weights on the order of 140–200 kDa.”

• AssumptionRests on an assumption: The oligomerization behavior of the λ repressor-leucine zipper fusion proteins reflects the oligomerization properties of the leucine zipper domain.

For one class I mutant and one class III mutant, higher-order oligomers were demonstrated in vitro.

“For one class I mutant and one class III mutant, we have been able to show formation of higher-order oligomers in vitro.”

A single substitution at e1 away from the Type I consensus changes s32 (Type I) into s38 (Type D).

“For example, a single substitution at the e1 position away from the Type I consensus converts s32 (Type I) to s38 (Type D).”

The Type I mutant s09 fusion protein behaved predominantly as a tetramer under the tested in vitro conditions.

“The Type I mutant behaved as a tetramer under all tested conditions, whereas the Type III mutant formed a variety of higher-order oligomers, depending on the solution conditions.”

• Also stated in: § Higher-order oligomers form in vitro · p. 3 · hedged

  “Crosslinking and size-exclusion chromatography experiments were consistent with the s09 fusion proteins being predominantly tetrameric, whereas analysis of sedimentation velocity data also suggested that the protein is predominantly tetrameric with small but significant amounts of monomers and higher-order oligomers (data not shown).”

• AssumptionRests on an assumption: The oligomerization behavior of the λ repressor-leucine zipper fusion proteins reflects the oligomerization properties of the leucine zipper domain.

Among 55 functional mutants, 42 were classified as forming higher-order oligomers.

“Using these tests, 42 of 55 functional mutants were classified as forming higher-order oligomers.”

The net change in ionic interactions upon heterodimer formation for both interacting and noninteracting pairs ranges from 0 to -16.

“If we calculate the net change in ion pair formation based on 1mol of each homodimer giving rise to 2 mol of heterodimers, the net change in ionic interactions upon heterodimer formation for both interacting and noninteracting pairs ranges from 0 to -16 (Fig. 5).”

Among Type D mutants, only one repulsive g-e' ionic interaction is found, between g8E and e13E of mutant s38.

“Among the Type D mutants (Table 3), only one repulsive g-e' ionic interaction, between g8E and e13E of mutant s38, is found.”

Several g-e' positions in the homodimeric mutants could form attractive ion pairs.

“Stable homodimers comprise less than 3% of the pool; several g-e'positions in these mutants could form attractive ion pairs.”

Several other pairs of mutants with different oligomerization properties differ by only one or two amino acid changes.

“In several other examples, mutants with different oligomerization properties differ by only one or two amino acid changes.”

Homodimers of wild-type GCN4 and all 13 Type D mutants contain at least one attractive g-e' and g'-e ionic interaction (KE or EK).

“Homodimers of the wt GCN4 leucine zipper and the 13 Type D mutants all have at least one pair of attractive g-e' and g'-e ionic interactions (KE or EK).”

Approximately 8% of the mutant sequences in the GCN4 leucine zipper e/g library have phenotypes consistent with forming higher-order oligomers larger than dimers.

“Surprisingly, approximately 8% of the mutant sequences have phenotypes consistent with the formation of higher-order (>dimer) oligomers, which can be classified into three types based on sequence features.”

At 200 mM KCl, the n01 fusion protein elutes as a broad peak with an apparent molecular weight of about 400 kDa, corresponding to 20 or more subunits in the complex.

“At 200 mM KCl, the n01 fusion protein elutes as a broad peak with an apparent molecular weight of about 400 kDa, corresponding to 20 or more subunits in the complex (Fig. 3A).”

• AssumptionRests on an assumption: The oligomerization behavior of the λ repressor-leucine zipper fusion proteins reflects the oligomerization properties of the leucine zipper domain.

The effect predicted by charge complementarity is not observed.

“In fact, this effect is not observed.”

All combinations of mutants n34, s03, s20, s24, and s56 titrate each other.

“All combinations of the mutants n34, s03, s20, s24, and s56 titrate each other.”

• AssumptionRests on an assumption: Sensitivity to phage infection indicates that enough heterodimers form to deplete active homodimers below the threshold required for immunity.

All eight Type D mutants tested have dimerization specificities different from that of wild-type GCN4.

“The 81 pairwise interactions in Figure 4 show that all eight of the Type D mutants tested have dimerization specificities that are different from that of wt GCN4.”

• AssumptionRests on an assumption: Sensitivity to phage infection indicates that enough heterodimers form to deplete active homodimers below the threshold required for immunity.

The wild-type GCN4 leucine zipper fails to titrate or be titrated by any of the mutants tested.

“In contrast, the wt GCN4 leucine zipper fails to titrate or be titrated by any of the mutants tested.”

• AssumptionRests on an assumption: Sensitivity to phage infection indicates that enough heterodimers form to deplete active homodimers below the threshold required for immunity.

Among 64 pairwise combinations involving only mutants, only 13 retain immunity to λ.

“Among 64 pairwise combinations involving only mutants, only 13 retain immunity to λ.”

• AssumptionRests on an assumption: Sensitivity to phage infection indicates that enough heterodimers form to deplete active homodimers below the threshold required for immunity.

One-way titration is observed in the following pairs: n34 + s26, s24 + s26, s24 + s67, s56 + s26, s64 + n34, s64 + s03, s64 + s20, s64 + s24, s64 + s26, s64 + s67, and s67 + s03.

“This is observed in several of the pairs: n34 + s26, s24 + s26, s24 + s67, s56 + s26, s64 + n34, s64 + s03, s64 + s20, s64 + s24, s64 + s26, s64 + s67, and s67 + s03.”

• AssumptionRests on an assumption: Sensitivity to phage infection indicates that enough heterodimers form to deplete active homodimers below the threshold required for immunity.

Each mutant has a unique pattern of interactions with the other four mutant leucine zippers.

“each mutant has a unique pattern of interactions with the other four mutant leucine zippers.”

• AssumptionRests on an assumption: Sensitivity to phage infection indicates that enough heterodimers form to deplete active homodimers below the threshold required for immunity.

None of the eight Type D mutants can titrate or be titrated by GCN4.

“Specifically, none of the eight Type D mutants can titrate or be titrated by GCN4”

• AssumptionRests on an assumption: Sensitivity to phage infection indicates that enough heterodimers form to deplete active homodimers below the threshold required for immunity.

Under the conditions used to study Type D mutants, cI+ repressor fusions to mutants that form higher-order oligomers were not titrated by dominant negative fusions to the same mutant.

“Under the conditions used to study the Type D mutants, cI+ repressor fusions to mutants that form higher-order oligomers were not titrated by dominant negative fusions to the same leucine zipper mutant.”

• AssumptionRests on an assumption: Sensitivity to phage infection indicates that enough heterodimers form to deplete active homodimers below the threshold required for immunity.

For any pair of leucine zipper mutants, the dimerization specificity assay results fall into one of three patterns.

“For any pair of leucine zipper mutants, the results of the dimerization specificity assays fall into one of three patterns.”

• AssumptionRests on an assumption: Sensitivity to phage infection indicates that enough heterodimers form to deplete active homodimers below the threshold required for immunity.

Mutants s56 and s64 do not titrate each other.

“mutants s56 and s64 do not titrate each other.”

• AssumptionRests on an assumption: Sensitivity to phage infection indicates that enough heterodimers form to deplete active homodimers below the threshold required for immunity.

All combinations of mutants s34, s03, s20, s24, and s56 are sensitive to phage.

“Although all combinations of mutants s34, s03, s20, s24, and s56 are sensitive to phage”

• AssumptionRests on an assumption: Sensitivity to phage infection indicates that enough heterodimers form to deplete active homodimers below the threshold required for immunity.

The data set contains no examples of pairs in which heterodimer formation leads to a net increase in attractive ion pairs.

“our data set includes no examples of pairs where the formation of heterodimers leads to a net increase in the number of attractive ion pairs”

Mutant s26 appears relatively unstable despite having the potential to form two K-E ionic pairs.

“Mutant s26seems to be relatively unstable despite having the potential to form two pairs of K-E ionic interactions.”

The broad tail of the n01 peak extends beyond the elution position expected for tetramers and dimers, suggesting dissociation of the aggregates as the protein is diluted through the column.

“Note, however, that the broad tail of the peak extends beyond the elution position expected for tetramers and dimers, suggesting dissociation of the aggregates as the protein is diluted through the column.”

• AssumptionRests on an assumption: The oligomerization behavior of the λ repressor-leucine zipper fusion proteins reflects the oligomerization properties of the leucine zipper domain.

The direction of one-way titration often reflects the relative stabilities of homodimers and heterodimers in a pair of leucine zippers.

“However, the direction of”

• cites Zeng et al., 1997Not matched

  Couldn’t be matched to a paper in Crossref, so it can’t be checked.

These other factors are significant enough to prevent prediction of the dimerization specificities of the mutants.

“Whatever these other factors are, they are significant enough to prevent us from predicting the dimerization specificities of the mutants.”

Other factors involving the e and g positions can also play a significant role in determining dimerization specificity.

“However, our results indicate that other factors involving the e and g positions can also play a significant role in determining dimerization specificity.”

Changes of one or two amino acids at e and g positions can be correlated with differences in oligomerization properties between mutants.

“The significance of the consensus residues is demonstrated by those mutants where one or two amino acid changes at the e and g positions can be correlated with a difference in the oligomerization properties of a pair of mutants.”

The fusion to the tetrameric GCN4-aLdI mutant represses lacZ expression from λ112OsPs to about one-sixth the level expressed from λXZ970.

“By contrast, the fusion to GCN4-aLdI, a tetrameric mutant of GCN4 (Harbury et al., 1993), represses lacZ expression from λ112OsPs to one-sixth the level expressed from λXZ970.”

• cites Harbury et al., 1993 sourceNot checked — no full text of the source

• AssumptionRests on an assumption: The oligomerization behavior of the λ repressor-leucine zipper fusion proteins reflects the oligomerization properties of the leucine zipper domain.

In cited cases, the pH dependence of homodimer and heterodimer formation suggests that charge-charge interactions help determine specificity.

“In these cases, the pH dependence of homodimer and heterodimer formation suggests that charge-charge interactions play a role in determining specificity.”

• cites O'Shea et al., 1992, 1993Not matched

  Couldn’t be matched to a paper in Crossref, so it can’t be checked.

• cites Graddis et al., 1993 sourceNot checked — no full text of the source
• cites Zhou et al., 1994aNot matched

  Couldn’t be matched to a paper in Crossref, so it can’t be checked.

In the earlier study of this mutant pool, about 73% of mutants were immune when fusion proteins were expressed from the uninduced lacUV5 promoter.

“In a previous study (Hu et al., 1993) of this mutant pool, the fusion proteins were expressed from the uninduced lacUV5 promoter. Under these conditions, about 73% of the mutants were immune.”

• cites Hu et al., 1993Not matched

  Couldn’t be matched to a paper in Crossref, so it can’t be checked.

At the lower expression level, some weakly dimerizing proteins may not score as immune.

“Note that, at this expression level, some of the proteins that dimerize weakly may not score as immune; cells expressing control fusions with leucine zippers from GCN4 and Jun are immune to λ, whereas cells expressing a C/EBP fusion are sensitive (Zeng et al., 1997).”

• cites Zeng et al., 1997Not matched

  Couldn’t be matched to a paper in Crossref, so it can’t be checked.

In the mutant pool, each of eight e and g positions is occupied by all possible combinations of alanine, glutamic acid, lysine, or threonine.

“In this pool, each of eight e and g positions (Fig. 1) are occupied by all possible combinations of alanine (A), glutamic acid (E), lysine (K), or threonine (T).”

• cites Hu et al., 1993Not matched

  Couldn’t be matched to a paper in Crossref, so it can’t be checked.

In the authors' implementation of the dominant negative scheme, titration is observed whenever the same homodimeric leucine zipper is used in both fusion proteins.

“In our implementation of this scheme, titration is observed whenever the same homodimeric leucine zipper is used in both fusion proteins (Zeng et al., 1997).”

• cites Zeng et al., 1997Not matched

  Couldn’t be matched to a paper in Crossref, so it can’t be checked.

s64 behaves as if it is more stable than any other mutant it interacts with.

“Note that, although s64, which can form two pairs of favorable g-e'and g'-e ionic interactions, behaves as if it is more stable than any other mutant it interacts with”

In crystal structures of GCN4 homodimers and Jun-Fos heterodimers, g-e' intersubunit salt bridges are seen between oppositely charged residues at positions gi and e/+5.

“In the crystal structures of GCN4 homodimers and Jun-Fos heterodimers, g-e' intersubunit salt bridges are visible between oppositely charged amino acids in the gi and e/+5 positions (O'Shea et al., 1991; Ellenberger et al., 1992; Konig & Richmond, 1993; Glover & Harrison, 1995).”

• cites O'Shea et al., 1991Not matched

  Couldn’t be matched to a paper in Crossref, so it can’t be checked.

• cites Ellenberger et al., 1992 sourceNot checked — no full text of the source
• cites Konig & Richmond, 1993 sourceNot checked — no full text of the source

Factors beyond the number of favorable g-e ionic interactions must contribute comparable favorable or unfavorable energies to dimer stabilization.

“Other factors beyond the number of favorable interactions between ionic groups in a pair of g and e positions must contribute comparable favorable or unfavorable energies to stabilizingthe dimers.”

Common features among Type D mutants may indicate important determinants of the parallel dimeric structure of the wild-type leucine zipper.

“Although this set of sequences is small, common features among these mutants may reflect important determinants of the parallel dimeric structure of the wild-type leucine zipper.”

When a cI+ fusion and a dominant negative fusion are expressed in the same cell, the cell's immunity is determined by whether the two leucine zippers interact.

“When a cI+ fusion and a dominant negative fusion are expressed in the same cell, the immunity of the cell is determined by the interaction or the lack thereof between the two leucine zippers.”

Significantly increased repression of λ112OsPs compared to λXZ970 indicates that a fusion protein forms higher-order oligomers either in solution or after DNA binding.

“Thus, significantly increased repression of λ112OsPs compared to λXZ970 indicates that a fusion protein forms higher-order oligomers either in solution or after DNA binding.”

n34 and s56 behave as if they are more stable than s26.

“n34 and s56, which have only one favorable pair of ionic interactions, behave asif they are more stable than s26, which should be able to form two pairs of ionic interactions.”

Sedimentation of the fusion protein containing the control aLdI tetrameric leucine zipper is well described by a monomer-tetramer equilibrium, with a small amount of aggregation to higher-order complexes.

“Sedimentation of the fusion protein containing the control aLdI tetrameric leucine zipper is well described by a monomer-tetramer equilibrium, with a smalì amount of aggregation to higher-order complexes (Fig. 2B).”

• AssumptionRests on an assumption: The oligomerization behavior of the λ repressor-leucine zipper fusion proteins reflects the oligomerization properties of the leucine zipper domain.

The QL44, IS84 double mutant is a better dominant negative inhibitor of repressor activity than QL44 alone.

“The QL44, IS84 double mutant is a better dominant negative inhibitor of repressor activity than QL44 alone and was therefore used in these studies (unpubl.).”

A higher molar ratio of dominant negative to cI+ protein would be needed to titrate enough cI+ into heterotetramers with only one cI+ DNA binding domain.

“In this case, a higher molar ratio of dominant negative to cI+ protein would be needed to titrate enough of the latter into heterotetramers with only one cI+ DNA binding domain.”

• AssumptionRests on an assumption: Sensitivity to phage infection indicates that enough heterodimers form to deplete active homodimers below the threshold required for immunity.

These four sequences can form any combination of heterodimers.

“suggesting that these four sequences can form any combination of heterodimers”

• AssumptionRests on an assumption: Sensitivity to phage infection indicates that enough heterodimers form to deplete active homodimers below the threshold required for immunity.

The most striking difference between GCN4 and the mutants is the presence of leucines at e2 and g4.

“the most striking difference between GCN4 and the mutants is the presence of leucines at e2 and g4.”

Only a few combinations of sequences in the mutant set destabilize heterodimers enough to prevent detection by the assay.

“This suggests that only a few combinations of sequences in our mutant set destabilize heterodimers enough to prevent their detection by this assay.”

• AssumptionRests on an assumption: Sensitivity to phage infection indicates that enough heterodimers form to deplete active homodimers below the threshold required for immunity.

Each mutant can be regarded as having an overlapping but distinct dimerization specificity.

“Thus, each mutant can be thought of as having an overlapping but distinct dimerization specificity.”

• AssumptionRests on an assumption: Sensitivity to phage infection indicates that enough heterodimers form to deplete active homodimers below the threshold required for immunity.

If one-way titration is taken as evidence for relative stabilities, the data in Figure 4 allow a partial ranking of some mutants.

“If one-way titration is taken as evidence for the relative stabilities of pairs of leucine zippers, then the data in Figure 4 allows a partial ranking of some of the mutants.”

Stabilizing interactions between a and a positions, which are the same in homodimers and heterodimers, presumably provide the driving force for many heterodimer combinations to form.

“Stabilizing interactions between the a and a positions, which are the same in homodimers and heterodimers, presumably provide the driving force to allow the many different combinations of heterodimers to form.”

• AssumptionRests on an assumption: Sensitivity to phage infection indicates that enough heterodimers form to deplete active homodimers below the threshold required for immunity.

The genetic test requires that tested mutants form fusion proteins with strong enough dimerization to be detected as functional repressors.

“because our genetic test currently requires that the mutants we tested be able to form fusion proteins with strong enough dimerization to be detected as functional repressors”

QL44 has been shown to act as a dominant negative inhibitor in leucine zipper fusions.

“QL44 has also been shown to act as a dominant negative inhibitor in leucine zipper fusions (Zeng et al., 1997).”

• cites Zeng et al., 1997Not matched

  Couldn’t be matched to a paper in Crossref, so it can’t be checked.

The relative stabilities of Type D mutants have not been measured directly.

“The relative stabilities of the Type D mutants have not been measured directly.”