An analysis of protein domain linkers: their classification and role in protein folding

Recent advances in protein engineering have come from creating multi-functional chimeric proteins containing modules from various proteins. These modules are typically joined via an oligopeptide linker, the correct design of which is crucial for the desired function of the chimeric protein. Here we analyse the properties of naturally occurring inter-domain linkers with the aim to design linkers for domain fusion. Two main types of linker were identified; helical and non-helical. Helical linkers are thought to act as rigid spacers separating two domains. Non-helical linkers are rich in prolines, which also leads to structural rigidity and isolation of the linker from the attached domains. This means that both linker types are likely to act as a scaffold to prevent unfavourable interactions between folding domains. Based on these results we have constructed a linker database intended for the rational design of linkers for domain fusion, which can be accessed via the Internet at http://mathbio.nimr.mrc.ac.uk.     

Armadillo: domain boundary prediction by amino acid composition

The identification and annotation of protein domains provides a critical step in the accurate determination of molecular function. Both computational and experimental methods of protein structure determination may be deterred by large multi-domain proteins or flexible linker regions. Knowledge of domains and their boundaries may reduce the experimental cost of protein structure determination by allowing researchers to work on a set of smaller and possibly more successful alternatives. Current domain prediction methods often rely on sequence similarity to conserved domains and as such are poorly suited to detect domain structure in poorly conserved or orphan proteins. We present here a simple computational method to identify protein domain linkers and their boundaries from sequence information alone. Our domain predictor, Armadillo (http://armadillo.blueprint.org), uses any amino acid index to convert a protein sequence to a smoothed numeric profile from which domains and domain boundaries may be predicted. We derived an amino acid index called the domain linker propensity index (DLI) from the amino acid composition of domain linkers using a non-redundant structure dataset. The index indicates that Pro and Gly show a propensity for linker residues while small hydrophobic residues do not. Armadillo predicts domain linker boundaries from Z-score distributions and obtains 35% sensitivity with DLI in a two-domain, single-linker dataset (within +/-20 residues from linker). The combination of DLI and an entropy-based amino acid index increases the overall Armadillo sensitivity to 56% for two domain proteins. Moreover, Armadillo achieves 37% sensitivity for multi-domain proteins, surpassing most other prediction methods. Armadillo provides a simple, but effective method by which prediction of domain boundaries can be obtained with reasonable sensitivity. Armadillo should prove to be a valuable tool for rapidly delineating protein domains in poorly conserved proteins or those with no sequence neighbors. As a first-line predictor, domain meta-predictors could yield improved results with Armadillo predictions.     

Effect of backbone cyclization on protein folding stability: chain entropies of both the unfolded and the folded states are restricted

Circular versions of a large number of proteins have been designed by connecting the N and C termini via peptide linkers. A motivation for these designs is the assumed enhancement in folding stability, because backbone cyclization reduces the chain entropy of the unfolded state. Here, it is recognized that backbone cyclization also reduces the chain entropy of a flexible peptide linker in the folded state. Specifically, the end-to-end distance of the linker is restricted to fluctuations around the average displacement between the N and C termini of the folded protein. The balance of the chain-entropy reductions in the folded and unfolded states is used to predict the change in the unfolding free energy, deltadeltaG(cycl), by backbone cyclization. Predicted values of deltadeltaG(cycl) are in quantitative agreement with results of a careful study on cyclizing the 34 residue PIN1 WW domain by linkers with two to seen residues. The experimental results of an optimal linker length l=4 and a maximum stabilization of 1.7 kcal/mol are reproduced. Calculations of deltadeltaG(cycl) for a broad selection of circular proteins suggest that the stabilizing effect of backbone cyclization is modest, reflecting entropy reductions in both the unfolded and the folded states.     

Design of the linkers which effectively separate domains of a bifunctional fusion protein

With the aim of separating the domains of a bifunctional fusion protein, the ability of several lengths of helix-forming peptides to separate two weakly interacting beta-can domains was compared with that of flexible linkers or of a three alpha-helices bundle domain. We introduced helix-forming peptide linkers A(EAAAK)nA (n = 2-5) between two green fluorescent protein variants, EBFP and EGFP, and investigated their spectral properties. The fluorescence resonance energy transfer from EBFP to EGFP decreased as the length of the linkers increased. The circular dichroism spectra analysis suggested that the linkers form an alpha-helix and the alpha-helical contents increased as the length of the linkers increased. The results clearly suggested the ability of the helical linkers to control the distance and reduce the interference between the domains. This 'linker engineering' may open a way to the rational design of linkers which maximize the multiple functions of fusion proteins or de novo multi-domain proteins.     

Linkers in the structural biology of protein–protein interactions

Linkers or spacers are short amino acid sequences created in nature to separate multiple domains in a single protein. Most of them are rigid and function to prohibit unwanted interactions between the discrete domains. However, Gly‐rich linkers are flexible, connecting various domains in a single protein without interfering with the function of each domain. The advent of recombinant DNA technology made it possible to fuse two interacting partners with the introduction of artificial linkers. Often, independent proteins may not exist as stable or structured proteins until they interact with their binding partner, following which they gain stability and the essential structural elements. Gly‐rich linkers have been proven useful for these types of unstable interactions, particularly where the interaction is weak and transient, by creating a covalent link between the proteins to form a stable protein–protein complex. Gly‐rich linkers are also employed to form stable covalently linked dimers, and to connect two independent domains that create a ligand‐binding site or recognition sequence. The lengths of linkers vary from 2 to 31 amino acids, optimized for each condition so that the linker does not impose any constraints on the conformation or interactions of the linked partners. Various structures of covalently linked protein complexes have been described using X‐ray crystallography, nuclear magnetic resonance and cryo‐electron microscopy techniques. In this review, we evaluate several structural studies where linkers have been used to improve protein quality, to produce stable protein–protein complexes, and to obtain protein dimers.     

Quantitative understanding of the energy transfer between fluorescent proteins connected via flexible peptide linkers

The fusion of different protein domains via peptide linkers is a powerful, modular approach to obtain proteins with new functions. A detailed understanding of the conformational behavior of peptide linkers is important for applications such as fluorescence resonance energy transfer (FRET)-based sensor proteins and multidomain proteins involved in multivalent interactions. To investigate the conformational behavior of flexible glycine- and serine-containing peptide linkers, we constructed a series of fusion proteins of enhanced cyan and yellow fluorescent proteins (ECFP-linker-EYFP) in which the linker length was systematically varied by incorporating between 1 and 9 GGSGGS repeats. As expected, both steady-state and time-resolved fluorescence measurements showed a decrease in energy transfer with increasing linker length. The amount of energy transfer observed in these fusion proteins can be quantitatively understood by simple models that describe the flexible linker as a worm-like chain with a persistence length of 4.5 A or a Gaussian chain with a characteristic ratio of 2.3. The implications of our results for understanding the properties of FRET-based sensors and other fusion proteins with Gly/Ser linkers are discussed.     

重组单链抗体-低分子质量尿激酶融合蛋白在 CHO 细胞中的抗降解研究

抗人纤维蛋白单链抗体-低分子质量尿激酶(Ⅱn-UK)融合蛋白,兼有单链抗体对纤维蛋白的亲和性和尿激酶的溶栓活性,有望开发成为新型导向溶栓药物.但基于通用连接肽(G4S)3的Ⅱn-linker-UK融合蛋白在CHO细胞中表达时出现明显的降解.为了解决此问题,利用分子生物学方法,对Hn-UK融合蛋白进行了分子改造,包括置换连接肽,改变两个半分子(moiety)的相对位置,以及对连接肽附近明确的蛋白酶位点进行突变等方法,并分别研究了改造后的11种Ⅱn-1inker-UK或UK-linker-Ⅱn突变体在CHO细胞中分泌性表达时的稳定性,最终筛选到一种抗降解的突变体.     

Characterization and prediction of linker sequences of multi-domain proteins by a neural network

In this paper, we describe a neural network analysis of sequences connecting two protein domains (domain linkers). The neural network was trained to distinguish between domain linker sequences and non-linker sequences, using a SCOP-defined domain library. The analysis indicated that a significant difference existed between domain linkers and non-linker regions, including intra-domain loop regions. Moreover, the resulting Hinton diagram showed a position-dependent amino acid preference of the domain linker sequences, and implied their non-random nature. We then applied the neural network to predict domain linkers in multi-domain protein sequences. As the result of a Jack-knife test, 58% of the predicted regions matched actual linker regions (specificity), and 36% of the SCOP-derived domain linkers were predicted (sensitivity). This prediction efficiency is superior to simpler methods derived from secondary structure prediction that assume that long loop regions are putative domain linkers. Altogether, these results suggest that domain linkers possess local characteristics different from those of loop regions.     

Bifunctional enhancement of a β-glucanase-xylanase fusion enzyme by optimization of peptide linkers

The flexible peptides (GGGGS)n (n < or = 3), the alpha-helical peptides (EAAAK)n (n < or = 3) and two other peptides were used as linkers to construct bifunctional fusions of beta-glucanase (Glu) and xylanase (Xyl) for improved catalytic efficiencies of both moieties. Eight Glu-Xyl fusion enzymes constructed with different linkers were all expressed as the proteins of ca. 46 kDa in Escherichia coli BL21 and displayed the activities of both beta-glucanase and xylanase. Compared to all the characterized fusions with the parental enzymes, the catalytic efficiencies of the Glu and Xyl moieties were equivalent to 304-426% and 82-143% of the parental ones, respectively. The peptide linker (GGGGS)(2) resulted in the best fusion, whose catalytic efficiency had a net increase of 326% for the Glu and of 43% for the Xyl. The two moieties of a fusion with the linker (EAAAK)(3) also showed net increases of 262 and 31% in catalytic efficiency. Our results highlight, for the first time, the enhanced bifunctional activities of the Glu-Xyl fusion enzyme by optimizing the peptide linkers to separate the two moieties at a reasonable distance for beneficial interaction.     

Probing the primary structural determinants of streptokinase inter-domain linkers by site-specific substitution and deletion mutagenesis

The bacterial protein streptokinase (SK) contains three independently folded domains (α, β and γ), interconnected by two flexible linkers with noticeable sequence homology. To investigate their primary structure requirements, the linkers were swapped amongst themselves i.e. linker 1 (between α and β domains) was swapped with linker 2 (between β and γ domains) and vice versa. The resultant construct exhibited very low activity essentially due to an enhanced proteolytic susceptibility. However, a SK mutant with two linker 1 sequences, which was proteolytically as stable as WT-rSK retained about 10% of the plasminogen activator activity of rSK When the native sequence of each linker was substituted with 9 consecutive glycine sequences, in case of the linker 1 substitution mutant substantial activity was seen to survive, whereas the linker 2 mutant lost nearly all its activity. The optimal length of linkers was then studied through deletion mutagenesis experiments, which showed that deletion beyond three residues in either of the linkers resulted in virtually complete loss of activator activity. The effect of length of the linkers was then also examined by insertion of extraneous pentapeptide sequences having a propensity for adopting either an extended conformation or a relatively rigid conformation. The insertion of poly-Pro sequences into native linker 2 sequence caused up to 10-fold reduction in activity, whereas its effect in linker 1 was relatively minor. Interestingly, most of the linker mutants could form stable 1:1 complexes with human plasminogen. Taken together, these observations suggest that (i) the functioning of the inter-domain linkers of SK requires a critical minimal length, (ii) linker 1 is relatively more tolerant to insertions and sequence alterations, and appears to function primarily as a covalent connector between the α and β domains, and (iii) the native linker 2 sequence is virtually indispensable for the activity of SK probably because of structural and/or flexibility requirements in SK action during catalysis.     

Structure of the Intermediate Filament-Binding Region of Desmoplakin

Desmoplakin (DP) is a cytoskeletal linker protein that connects the desmosomal cadherin/plakoglobin/plakophilin complex to intermediate filaments (IFs). The C-terminal region of DP (DPCT) mediates IF binding, and contains three plakin repeat domains (PRDs), termed PRD-A, PRD-B and PRD-C. Previous crystal structures of PRDs B and C revealed that each is formed by 4.5 copies of a plakin repeat (PR) and has a conserved positively charged groove on its surface. Although PRDs A and B are linked by just four amino acids, B and C are separated by a 154 residue flexible linker, which has hindered crystallographic analysis of the full DPCT. Here we present the crystal structure of a DPCT fragment spanning PRDs A and B, and elucidate the overall architecture of DPCT by small angle X-ray scattering (SAXS) analysis. The structure of PRD-A is similar to that of PRD-B, and the two domains are arranged in a quasi-linear arrangement, and separated by a 4 amino acid linker. Analysis of the B-C linker region using secondary structure prediction and the crystal structure of a homologous linker from the cytolinker periplakin suggests that the N-terminal ~100 amino acids of the linker form two PR-like motifs. SAXS analysis of DPCT indicates an elongated but non-linear shape with R_g = 51.5 Å and D_max = 178 Å. These data provide the first structural insights into an IF binding protein containing multiple PRDs and provide a foundation for studying the molecular basis of DP-IF interactions.     

A study on the effects of linker flexibility on acid phosphatase PhoC-GFP fusion protein using a novel linker library

Fusion strategy has been widely used to construct artificial multifunction proteins. The flexibility or rigidity of linkers between two fused partners is an important parameter that affects the function of fusion proteins. By combining the flexible unit GGGGS (F) and rigid unit EAAAK (R), ten linkers consisting of five elementary units that cover the fully rigid RRRRR linker to the fully flexible FFFFF linker were used to construct acid phosphatase-green fluorescence protein fusion protein (PhoC-GFP). By varying the linker flexibility in PhoC-GFPs, the relative specific activity of phosphotransferase and phosphatase varied from ∼19.0% to 100% and ∼9.35% to 100%, respectively. There exists an optimal linker capable of achieving the highest phosphotransferase/phosphatase activity and GFP fluorescence intensity. We found that the highest activities were achieved neither with the rigid RRRRR linker nor with the flexible FFFFF linker, but with the FFFRR linker. Linker flexibility could adjust the activity ratio between phosphotransferase and phosphatase and varied between ∼30% to 100%. PhoC-GFP with FRRRR linker achieved the highest relative specific phosphotransferase activity/relative specific phosphatase activity (T/P) value. Our results show that applying a linker library with controllable flexibility to the fusion proteins will be an efficient way to adjust the function of fusion enzymes.     

Spatial orientation and dynamics of the U1A proteins in the U1A-UTR complex

The human U1A protein contains three distinct domains: the N-terminal RBD1 (amino acids 1-101), the C-terminal RBD2 (amino acids 195-282), and the linker region (amino acids 102-194). The RBD1 domains of two U1A proteins bind specifically to two internal loops in the 3' untranslated region (3' UTR) of its own pre-mRNA. Tryptophan fluorescence and fluorescence resonance energy transfer data show that the two RBD2 domains do not interact with any regions of the UTR complex and display an overall tumbling that is uncorrelated from the core of the complex (formed by RBD1-UTR), indicating that the linker regions of the two U1A proteins remain flexible. The two RBD2 domains are separated by an apparent distance greater than 74 A in the UTR complex. The linker region adjacent to the RBD1 domain (103-ERDRKREKRKPKSQETP-119) is supposedly involved in protein-protein interactions (12). A single cysteine, introduced at position 101 or 121 of the U1A protein, was used as a specific attachment site for the fluorophore pair IAEDANS [N'-iodoacetyl-N'-(1-sulfo-5-n-naphthyl)ethylenediamine]/DABMI [4-(dimethylamino)-phenylazophenyl-4'-maleimide]. In the U1A-UTR complex (2:1), the dyes at the 101 position are separated by = approximately 51 A, while the dyes at the 121 position are at an apparent distance = approximately 58 A. The 101-121 crossed distance on adjacent U1A proteins averages to = 55 A. These results suggest that the amino acid sequence 101-121 of the two U1A proteins in the complex are held in proximity to each other in a compact conformation.     

Structure-function analysis of a series of novel GIP analogues containing different helical length linkers

Glucose-dependent insulinotropic polypeptide (GIP1-42) is a potent glucose-lowering intestinal peptide hormone. The equipotent GIP1-30NH2 was structurally modified by linking N- and C-terminal fragments with several different linkers. Substitution of the middle region of GIP by a flexible aminohexanoic linker resulted in greatly reduced binding affinity and reduction or complete loss of bioactivity. Connection of the bioactive domains GIP1-14 and GIP19-30NH2 by EKEK or AAAA linkers resulted in peptide agonists with approximately 3-4-fold increased bioactivity as compared to GIP1-30NH2. Conformational analysis by CD spectroscopy of GIP fragments and analogues suggests a helical region in the C-terminal (19-30) portion of GIP. It was demonstrated that stabilization of this C-terminal helical region by the introduction of helical linkers favored binding and activation of the GIP receptor. Our results suggest an important contribution of a direct interaction of the first 14 amino acids with the GIP receptor, an appropriate relative orientation of N- and C-terminal parts of GIP, and the presence of helical linkers to be essential for bioactivity.     

Intermodular Linker Flexibility Revealed from Crystal Structures of Adjacent Cellulosomal Cohesins of Acetivibrio cellulolyticus

Cellulosome complexes comprise an intercalated set of multimodular dockerin-containing enzymatic subunits connected to cohesin-containing nonenzymatic subunits called scaffoldins. The adjoining modules in each cellulosomal subunit are interconnected by a variety of linker segments of different lengths and composition. The exact role of the cellulosomal linkers has yet to be described, although it is assumed that they contribute to the architecture and action of the cellulosome by providing the protein subunits with flexibility and by providing spacers between the enzymatic modules that could enhance interactions with the cellulose substrate. Here we present four crystal structures of Acetivibrio cellulolyticus cellulosomal type II cohesins with linker extensions. Two of the structures represent two different crystal forms (trigonal and orthorhombic) of the same N-terminal cohesin module (CohB1) together with its full (6-residue) native C-terminal linker, derived from scaffoldin B. The other two structures belong to the adjacent (second) cohesin module (CohB2), each of which was crystallized with the same 6-residue linker segment, but now positioned at the N-terminus and with either a truncated (5-residue) or a full-length (45-residue) C-terminal linker, respectively. Comparison between the two CohB1 structures revealed significant differences in the conformation of their equivalent C-terminal linker segment. In one crystal form a helical conformation was observed, as opposed to an extended conformation in the other. The CohB2 structures also displayed diverse conformations in their linker segments. In these structures, different linker conformations were observed in the individual molecules within the asymmetric unit of each structure. This conformational diversity implies that the linkers may adopt alternative conformations in their natural environment, consistent with varying environmental conditions. The findings suggest that linkers can play an important role in the assembly, dynamics and function of the cellulosomal components.     

Quantitative account of the enhanced affinity of two linked scFvs specific for different epitopes on the same antigen

Protein and other antigens typically have a number of different epitopes. This presents an opportunity for designing high-affinity antibodies by connecting via a flexible peptide linker two antibody fragments recognizing non-overlapping epitopes on the same antigen. The same strategy was employed in natural and designed DNA-binding proteins. According to a previous theory, the linking enhances the antigen-binding affinity over those of the individual antibody fragments (with association constants K(A) and K(B)) by p(d(0))K(B) or p(d(0))K(A), where p(d(0))=(3/4pil(p)bL)(3/2)exp(-3d(0)(2)/4l(p)bL)(1-5l(p)/4bL+ cdots, three dots, centered ) is the probability density for the end-to-end vector of the flexible linker with L residues to have a distance d(0). The predicted affinity enhancement is found to be actually approached by a bi-specific antibody against hen egg lysozyme consisting of scFv fragments of D1.3 and HyHEL-10. The wide applicability of the theory is demonstrated by diverse examples of protein-protein interactions constrained by flexible linkers.     

NMR studies on domain diffusion and alignment in modular GB1 repeats

Modular proteins contain individual domains that are often connected by flexible, unstructured linkers. Using a model system based on the GB1 domain, we constructed tandem repeat proteins and investigated the rotational diffusion and long-range angular ordering behavior of individual domains by measuring NMR relaxation parameters and residual dipolar couplings. Although they display almost identical protein-solvent interfaces, each domain exhibits distinct rotational diffusion and alignment properties. The diffusion tensor anisotropy of the N-terminal domain (NTD) is D_‖/D_⊥ = 1.5-1.6, similar to that of single-GB1 domains (D_‖/D_⊥ = 1.6-1.7), whereas the value for the C-terminal domain (CTD) is D_‖/D_⊥ = 2.0-2.2. In addition, the two domains have different rotational correlation times. These effects are observed for linkers of three to 24 residues, irrespective of linker length. The NTD and CTD also differ in their degree of magnetic alignment, even with a flexible linker of 18 residues, exhibiting D_a values of 7.7 Hz and 9.7 Hz, respectively. Our results suggest that diffusion differences and long-range influences may persist in modular protein systems, even for systems that have highly flexible linkers and exhibit no domain-domain or domain-linker interactions.     

Protein Linkers Provide Limits on the Domain Interactions in the ABC Importer GlnPQ and Determine the Rate of Transport

GlnPQ is an ATP-binding cassette importer with a unique domain organization and intricate transport behavior. The protein has two extracytoplamic substrate-binding domains (SBDs) per membrane subunit, each with different specificity for amino acids and different spacing to the translocator domain. We determined the effect of the length and structure of the linkers, which connect the SBDs to each other and to the membrane-embedded translocator domain, on the transport by GlnPQ. We reveal that varying the linker length impacts transport in a dual manner that depends on the conformational dynamics of the SBD. Varying the linker length not only changes the time for the SBD to find the translocator (docking) but also changes the probability to release the substrate again, thus altering the transport efficiency. On the basis of the experimental data and mathematical modeling, we calculate the docking efficiency as function of linker length and lifetime of the closed conformation. Importantly, not only linker length but also features in the sequence are important for efficient delivery of substrate from SBD to the translocator. We show that the linkers provide a platform for SBD docking and are not merely flexible structures.     

Kainate-binding proteins are rendered functional ion channels upon transplantation of two short pore-flanking domains from a kainate receptor

Kainate-binding proteins belong to an elusive class of putative ionotropic glutamate receptors that to date have not been shown to form functional ion channels in heterologous expression systems, despite binding glutamatergic agonists with high affinity. To test the hypothesis that inefficient or interrupted signal transduction from the ligand-binding site via linker domains to the ion pore (gating) might be responsible for this apparent lack of function, we transplanted the short homologous linker sequences from the fully functional rat kainate receptor GluR6 into frog kainate-binding protein. We were able to generate chimeric receptors that are functional in the Xenopus oocyte expression system and in human embryonic kidney 293 cells. The linker domains A and B in particular appear to be crucial for gating, because a functional kainate-binding protein was observed when at least parts of both linkers were derived from GluR6. We speculate that to enable signal transduction from the ligand-binding site to the ion pore of the frog kainate-binding protein, the linker structure of the protein has to undergo an essential conformational alteration, possibly mediated by an as yet unknown subunit or modulatory protein.