A small surface hydrophobic stripe in the coiled-coil domain of type I keratins mediates tetramer stability.
Through molecular modeling and site-directed mutagenesis, we document a hitherto unnoticed hydrophobic stripe exposed at the surface of coiled-coil keratin heterodimers that contributes to the extraordinary stability of heterotetramers.The inability of K16 to form urea-stable tetramers in vitro correlates with an increase in its turnover rate in vivo.The data presented support a specific conformation for the assembly competent IF tetramer, provide a molecular basis for their differential stability in vitro, and point to the physiological relevance associated with this property in vivo.
Affiliation: Department of Biological Chemistry, Johns Hopkins University School of Medicine, Baltimore, MD 21205, USA.
Intermediate filaments (IFs) are fibrous polymers encoded by a large family of differentially expressed genes that provide crucial structural support in the cytoplasm and nucleus in higher eukaryotes. The mechanisms involved in bringing together approximately 16 elongated coiled-coil dimers to form an IF are poorly defined. Available evidence suggests that tetramer subunits play a key role during IF assembly and regulation. Through molecular modeling and site-directed mutagenesis, we document a hitherto unnoticed hydrophobic stripe exposed at the surface of coiled-coil keratin heterodimers that contributes to the extraordinary stability of heterotetramers. The inability of K16 to form urea-stable tetramers in vitro correlates with an increase in its turnover rate in vivo. The data presented support a specific conformation for the assembly competent IF tetramer, provide a molecular basis for their differential stability in vitro, and point to the physiological relevance associated with this property in vivo.
- Hydrophobic and Hydrophilic Interactions*
- Amino Acid Sequence
- Amino Acid Substitution
- Cells, Cultured
- Chromatography, Ion Exchange
- Consensus Sequence
- Cross-Linking Reagents/pharmacology
- Electrophoresis, Agar Gel
- Electrophoresis, Polyacrylamide Gel
- Escherichia coli/genetics
- Models, Molecular
- Molecular Sequence Data
- Mutagenesis, Site-Directed
- Precipitin Tests
- Protein Conformation
- Protein Structure, Tertiary
- Recombinant Proteins/analysis/isolation & purification/metabolism
- Sequence Homology, Amino Acid
© Copyright Policy
fig1: Mouse and human keratin 16 form unstable tetramers with type II K6. (a) Various purified type I keratins (human K14, mouse K17, human K16, mouse K16) were individually mixed with the mouse type II keratin K6β in a 55:45 M ratio, applied to a Mono Q anion-exchange chromatography column, and eluted with a gradient of guanidine-HCl. Fractions were analyzed by SDS-PAGE. Monomeric type II keratins elute first, followed by monomeric type I keratins and heterodimers of type I and II keratins, and finally heterotetramers of type I and II keratins (Wawersik et al., 1997). White lines indicate that intervening lanes have been spliced out. (b) Type I–type II heterotypic complexes (see panel a) were chemically cross-linked with BS3. Cross-linked products (4 μg proteins) were resolved on a 4–16% gradient SDS-PAGE and stained with Coomassie blue. Although individual keratins do not cross-link under these conditions (not depicted; Coulombe and Fuchs, 1990), the type I/type II mixes cross-link as oligomers. The migration standards are indicated at left, and oligomeric complex positions of tetramer (T), dimer (D) type II (II M) and type I (I M) monomers are indicated on the right. Dotted boxes indicate region quantitated in c. The presence of multiple bands for each of the cross-linked heterodimer and heterotetramer products corresponds to distinct intramolecular cross-links. For each combination tested, antibodies to types I and II keratins react with all of these bands (not depicted), reflecting their heterotypic character. K14, K17, and K16 have comparable numbers of lysine residues (23, 22, and 19, respectively), such that side chain availability likely did not influence cross-linking outcome. There is excellent concordance between the chromatography and chemical cross-linking assays. White lines indicate that intervening lanes have been spliced out. (c) Densitometry was performed on the cross-linked products in b and in replicate experiments. The intensities of monomer, dimer, and tetramer in each lane were summed to 100%, and the percentages of tetramers (black box) and dimers (gray box) were graphed (mean ± SEM). (d) Cultured mouse primary keratinocytes were lysed with 8 M or 6 M urea, mixed with Coomassie G250 dye and separated on a blue native gel (5–13% acrylamide without SDS). Western blotting was performed after transfer to membrane and revealed monomer, dimer, and tetramer complexes of the various native keratins. The approximate migration of various recombinant proteins is indicated to the right; however, these marks should not be considered an exact molecular mass. Although a general calibration curve can be calculated for blue native gels, the apparent mass of a specific complex can vary up to 20% due to different solubilization conditions and post-translational modifications (Schagger, 2001). This explanation may also account for the difference in migration of the dimer band in 8 M versus 6 M urea. Alternatively, there may be a keratin binding protein present in 6 M urea that is dissociated in 8 M urea. Note that two different exposures are shown for K6. The darker exposure (K6′) shows that K6 is present in both monomer and heterodimer bands in 8 M urea.
We compared the ability of mK16 to form stable heterotetramers in vitro to that of hK16 and related type I keratins, hK14 and mK17. Purified types I and II keratins were mixed in the presence of 6.5 M urea and low salt and subjected to an anion exchange chromatography assay that resolves monomers, heterodimers, and heterotetramers (Wawersik et al., 1997). Whereas hK14/mK6 and mK17/mK6 elute largely as heterotetramers, both hK16/mK6 and mK16/mK6 elute primarily as monomers and heterodimers (Fig. 1 a). Complexes containing mouse or human K6 paralogues and K17 display identical properties in this assay (not depicted), establishing that behavior in this assay is not dictated by the species of origin. Purified heterotypic complexes were subjected to cross-linking of lysine side chains with Bis(sulfosuccinimidyl)suberate (BS3), followed by electrophoretic separation of products via SDS-PAGE. Cross-linked hK14/mK6 and mK17/mK6 migrated predominantly as tetramers (>40% tetramers, <35% dimers), whereas hK16- and mK16-containing complexes migrated predominantly as dimers (>50%) with a small amount of tetramers (<20%; Fig. 1, b and c). K14, K17, and K16 have comparable numbers of lysine residues (23, 22, and 19, respectively), such that side chain availability likely did not influence cross-linking outcome.