Team:Peking/Crosslinking

Crosslinking

Crosslinking

Background

The Necessity of Crosslinking

Compared to monomers, polymers formed by covalent crosslinking possess some special properties. For example, polymers typically offer greater mechanical strength and a larger contact area. They are more easier to handle. We thus considered crosslinking the monomers to form polymers, and take the latter as the basic functional units.


Characteristics

Firstly, the monomers need to be able to crosslink under complex environmental conditions and show high preference for the reaction with each other. Faster and more selective reactions would be highly helpful for practical applications.

Secondly, since functional proteins are to be fused to the monomers to achieve the final goals, their orthogonality must be considered. If there are no covalent interactions between monomer backbones and the attached functional elements, the target proteins could be used to functionalize a polymerized network, thus adequately taking advantage of the superiority of polymers.

What’s more, polymers are comparatively easy to recover and could thus be eco-friendly. Since the aim is to construct a brand new material, it is extremely important to consider its potential effects on the environment from the onset of the project.


Implementation of Crosslinking - the SpyTag-SpyCatcher System

Inspired by the autocatalytic formation of the isopeptide bond between a specific Lys and an Asp residue in Streptococcus pyogenes (Spy) fibronectin-binding protein FbaB, researchers split its autocatalytic domain, CnaB2, and obtained two peptides which they named SpyTag and SpyCatcher. The resulting peptides are able to form isopeptide bonds with each other spontaneously. SpyTag, SpyCatcher and their stable cross-linking products provided us with an ideal tool to manipulate the protein polymer network1.

Fig. 1. Spontaneous intermolecular amide bond formation by SpyTag. (A) Amide bond formation between Lys and Asp side chains. (B) Key residues for amide bond formation in CnaB2 shown in stick format, based on PDB 2X5P. (C) Cartoon of SpyTag construction. Streptococcus pyogenes (Spy) CnaB2 was dissected into a large N-terminal fragment (SpyCatcher, left) and a small C-terminal fragment (SpyTag, right).For sake of brevity, SpyTag and SpyCatcher shall from now on be abbreviated as A and B, respectively.

Leap Before the Practical Application - the Spy Crosslinking Network

The SpyTag / SpyCatcher System could be applied in the field via polymerization to obtain the “Spy Crosslinking Network”. The functional domains provided by SpyTag and SpyCatcher were linked via an elastin like protein (ELP), yielding the new fusion protein monomers able to undergo three-dimensional polymerization (TDP). This kind of polymerization could yield hyper-branched products and polymer network at low concentrations, and gels with a high degree of cross-linkage at high concentrations2,3.

Fig. 2. The Spy Crosslinking Network. (A) Monomers containing multiple functional groups, SpyTag or SpyCatcher, react with each other both in situ and in vitro. (B) Schematic illustration of the products formed by mixing protein precursors 3A and 2B. Their mixing at high concentration (w.t. >=7.5%) leads to the formation of a covalently cross-linked gel.



Design

Theoretical Deduction - Determining the Optimal n Count for nA-FP

We first have to find out the optimal n count of nA-FP (functional peptides), which includes both SpyTag modules and functional proteins, and will react with another monomer, such as 3B.

In order to acquire hyper-branched polymers and promote the strength and contact area of the resulting polymer network, we decided to abandon 1A-FP and 2A-FP monomers, and instead decided to increase the weight proportion of functional proteins in the monomer. We thus chose 3A-SUP(n>=3). Although 4A-SUP and 6A-SUP performed better than 3A-SUP in crosslinking since it has more SpyTag groups available to react with 3B, other experiments showed that 4/6A-SUP performed poorer than 3A-SUP in uranyl adsorption at the same concentration (see the Uranyl adsorption part). Following the analysis presented here, we defined the optimal n count as 3.

Fig. 3. Schematic diagram of triple SpyTag-SUP (3A-SUP) and triple SpyCatcher (3B).

Setting up the Scaffold - Construction of the Monomers

In order to use the SpyTag-SpyCatcher system as scaffold, we fused three SpyTag modules (A) spaced by (VPGVG)4, and combined them with an N-terminal 6xHis tag and another functional protein called the Super Uranyl-binding Protein(SUP) at the C-terminus. We also fused three SpyCatcher modules (B) spaced by (VPGVG)15, with an N-terminal 6xHis tag.

We successfully assembled the constructs SpyTag-SpyTag-SpyTag-SUP (3A-SUP), SpyTag-SpyTag-SpyTag-mSA (3A-mSA), SpyCatcher-SpyCatcher-SpyCatcher (3B), together with a number of similar fusion proteins. Since these proteins were easy to obtain either by lysing the bacteria or by bacterial secretion (See the Secretion section for further information), we were able to build a brand new biomaterial that displays multi-functionality at quite a low cost.




Methods

☆☆☆ To better elucidate the properties of the monomers, the concentration was kept at 1mg/mL, unless stated otherwise. This concentration is only conductive to oligomerization and did not induce polymer network formation. ☆☆☆




Results

Exploring the polymerization of the crosslinking polymer network is significant in view of subsequent applications. We must know what the Spy Crosslinking Network is like, and how much time the monomers and functional modules need to complete their respective reactions, in order to gain a sense of their crosslinking ability.

To answer these questions, we have analyzed the crosslinking ability of the fundamental monomers 3A, 3A-SUP, 3A-mSA and 3B in detail. They were first diluted to a certain concentration, and were made to react at equimolar concentrations of A and B. The monomers containing A modules were contacted with 3B at 25°C and pH=7.3 for 2 hours.

Fig. 4. Exploration of the polymerization ability of the 3A-SUP/3A-mSA with 3B. “3A-SUP” stands for “Triple SpyTag-SUP”, “3A-mSA” for “Triple SpyTag-mSA”, and “3B” is the abbreviation of “Triple SpyCatcher”. (A) SDS-PAGE of basic experiment, which illustrates the basic cross-linking ability of 3A-SUP/3A-mSA and 3B. Lane 1, Marker; lane 2, 3A-SUP; lane 3, 3A-mSA; lane 4, 3A-SUP+3A-mSA; lane5, 3B; lane 6, 3A-SUP+3B; lane 7, 3A-mSA+3B. (Molecular Weight: 3A-SUP, 21.4kDa; 3A-mSA, 25.4kDa; 3B, 62.0kDa.) (B) The OD value of the lane 5 of oligomers produced by the mix of 3A-SUP and 3B. Peaks illustrate the monomers and the possible products: ① 3A-SUP (21.4kDa); ② 3B (62.0kDa); ③ 1x 3A-SUP+1x 3B (83.4kDa); ④ 3x 3A-SUP+1x 3B (126.2kDa); ⑤ 2x 3A-SUP+2x 3B (166.8kDa); ⑥ 4x 3A-SUP+4x 3B (333.6kDa). (C) The OD value of the lane 7 of oligomers produced by the mix of 3A-mSA and 3B. Peaks illustrate the monomers and the possible products: ① 3A-mSA (25.4kDa); ② 3B (62kDa); ③ 1x 3A-mSA+1x 3B (87.4kDa); ④ 1x 3A-mSA+2x 3B (149.4kDa); ⑤ 2x 3A-mSA+2x 3B (174.4kDa); ⑥ 3x 3A-mSA+4x 3B (324.2kDa).

We found that new bands appeared above the band of 3B when it was mixed with monomers containing A, which demonstrated that the idea of forming functional polymer network using these monomers was indeed feasible. The products were mainly oligomers due to the ease with which such oligomers could form loops, which hinders the linkage between different monomers at such low concentrations (Fig. 4. A, B and C). Interestingly, with the restriction that A and B are supplied in equimolar concentrations and that the initial content of 3B is constant, the crosslinking abilities of these monomers at low concentrations were found to be different from each other. This was done by comparing the contents of unreacted surplus 3B.

We also drew up two tables to estimate the mass distribution of the polymer homologs produced by 3A-SUP/3A-mSA and 3B:

Table. 1. Molecular weight estimation of the spy crosslinking network of 3A-SUP and 3B.
Table. 2. Molecular weight estimation of the spy crosslinking network of 3A-mSA and 3B.

After exploring the basic character of the reactants, i.e. their ability to crosslink, we wanted to further explore the conditions that are most advantageous for polymerization. To do so, we chose to use a pH gradient, a temperature gradient, and a monomer concentration gradient.

Fig. 5. Design of the gradient experiment with variations of pH, temperature and monomer concentrations.

pH Gradient Experiment

The first part of the gradient experiment is concerned with pH changes. Given that the Spy Crosslinking Network would face an aqueous environment with nearly neutral pH in practical applications, and since the proteins have their own pH optima, we decided to use a pH gradient of 6.3/7.3/8.3. 3A-SUP and 3B monomers were first dissolved in TBS at certain pH values, after which they were contacted at pH values from the mentioned gradient at 25°C for 2 hours. Samples were taken every 30min in order to investigate the variation tendency of the monomers. Finally, SDS-PAGE was used to separate the proteins and the gels were scanned and analyzed using the software Lane 1D. Thus, the relationship between time and the content of 3B and 3A was measured. The experiments were done in triplicates to enable quantitative analysis.

Fig. 6. pH Gradient Experiment. (A) The reactive extent for 3B at different pH. Experiments were repeated three times and error bars were added. (B) The mass distribution of oligomers at different time at pH7.3 (70kDa-450kDa).

Based on the results of the SDS-PAGE analysis shown above, we clearly observed the change law of 3B and the mass distribution of polymer homologs. Ideally, a lower pH would lead to faster consumption of monomers, while the time scale should be shorter in order to show the nature of the condensation and polymerization.

Temperature Gradient Experiment

The second part of the gradient experiment is concerned with temperature. According to common environmental temperatures, three typical values, 16°C, 25°C, 37°C were introduced in this experiment. Similar to the pH Gradient Experiment, the monomers 3A-SUP and 3B were first dissolved in TBS at pH=7.3, after which they were contacted at the temperatures from the mentioned gradient. The reactions were conducted at pH=7.3 for 2 hours. The subsequent steps were the same as described for the pH gradient experiment.

Fig. 7. Temperature Gradient Experiment. (A) The reactive extent for 3B at different temperature. Experiments were repeated three times and error bars were added. (B) The mass distribution of oligomers at different time at 16℃ (70kDa-400kDa).

SDS-PAGE analysis clearly showed the variation tendency of the surplus content of 3B as a function of time, as well as that of the mass distribution of polymer homologs. According to this result, we were able to conclude that higher temperature would lead to faster consumption of monomers, whereby we did not see any changes in the weight distribution of polymer homologs at the designated time scale. We assumed that it should be easier for oligomers to form loops (intramolecular reactions) at lower concentrations, and that the time scale of sampling was too coarse to show this variation. Even though the rate of polymerization could be described by the consumption rate of functional groups, we decided to use the consumption rate of monomers to describe the overall tendency.

Concentration Gradient Experiment

The third part is concerned with optimal monomer concentrations. At low concentrations, the polymers would cross-link and not polymerize, but oligomerize. According to the basics of polymer science, the differences between these two processes could be described according to crosslinking density, which is large in the former, and rather small in the latter case. From this point of view, if we hope to increase the mass, strength and contact area of polymers, we have to mix the monomers at high concentrations. Thus, understanding the relationship between the monomer concentrations and the variation of mass distribution of polymers is quite essential.


Fig. 8. The mass distribution of oligomers of concentration gradient experiment at different concentration. (A) SDS-PAGE of concentration gradient experiment, which clearly showed the change of mass distribution according to the change of concentration. Lane 1, Marker; lane 2, 3A-SUP; lane 3, 3A-mSA; lane 4, 3A-SUP+3A-mSA; lane5, 3B; (1mg/mL) lane 6, 3A-SUP+3B; lane 7, 3A-mSA+3B; lane 8, 3A-SUP+3A-mSA+3B; (5mg/mL) lane 9, 3A-SUP+3B; lane 10, 3A-mSA+3B; lane 11, 3A-SUP+3A-mSA+3B; (10mg/mL) lane 12, 3A-SUP+3B; lane 13, 3A-SUP+3A-mSA+3B. (Molecular Weight: 3A-SUP, 21.4kDa; 3A-mSA, 25.4kDa; 3B, 62.0kDa.) The mass of samples in lanes was equal to each other. (B) The OD value of the lanes 2, 5, 6, 9, 12 of the oligomers produced by the mix of 3A-SUP and 3B. Peaks illustrate the monomers and the possible products, ① 3A-SUP (21.4kDa); ② 3B (62kDa); ③ 1x 3A-SUP+1x 3B (83.4kDa); ④ 3x 3A-SUP+1x 3B (126.2kDa); ⑤ 2x 3A-SUP+2x 3B (166.8kDa); ⑥ 4x 3A-SUP+4x 3B (333.6kDa); ⑦ oligomers larger than 750kDa (~6x 3A-sup+10x 3B or 9x 3A-SUP+9x 3B).

SDS-PAGE analysis showed that more hyper-branched products with greater molecular weights appeared with the increase of monomer concentration, which indicates that the average weight would be larger if higher concentrations of monomers were mixed together.




Conclusion

The crosslinking experiments showed that monomers containing SpyTag domains could react with triple SpyCatcher (3B) monomers, and that the crosslinking reaction may be favored at appropriately low pH (in the experiments, pH=6.3) and appropriately high temperature (in the experiments, T=25℃). Crosslinking leads to oligomerization at low monomer concentrations, and produces highly-crosslinked polymers with larger masses at high monomer concentrations. To this end, we have determined the optimal conditions for polymerization, which consisted of appropriately low pH, appropriately high temperature and comparatively higher monomer concentrations. The findings were in accordance with the universal principles of polyester condensation, as expected.



References:

[1]. Zakeri, B., et al., Peptide tag forming a rapid covalent bond to a protein, through engineering a bacterial adhesin. Proceedings of the National Academy of Sciences, 2012. 109(12): p. E690-E697.

[2]. Samuel C Reddington and Mark Howarth, Secrets of a covalent interaction for biomaterials and biotechnology: SpyTag and SpyCatcher. Current Opinion in Chemical Biology, 2015, 29:94–99

[3]. Zhang, W., et al., Controlling Macromolecular Topology with Genetically Encoded SpyTag–SpyCatcher Chemistry. Journal of the American Chemical Society, 2013. 135(37): p. 13988-13997.

[4]. Sun, F., et al., Synthesis of bioactive protein polymer network by genetically encoded SpyTag-SpyCatcher chemistry. Proceedings of the National Academy of Sciences, 2014. 111(31): p. 11269-11274.