NOVEL PEPTIDE TAG

Abstract

The present invention relates to a peptide consisting of a sequence of 5 or 6 to 30, preferably 6 to 12, most preferably 10 to 12 amino acids, wherein (a) at least ⅓, preferably at least ½ of said amino acids are amino acids having a functional group or a side chain which is positively charged at neutral pH; and (b) at least two amino acids, preferably at least three are histidines.

Claims

1. A peptide consisting of a sequence of 5 or 6 to 30, preferably 6 to 12, most preferably 10 to 12 amino acids, wherein (a) at least ⅓, preferably at least ½ of said amino acids are amino acids having a functional group or a side chain which is positively charged at neutral pH; and (b) at least two amino acids, preferably at least three are histidines.

2. The peptide of claim 1, wherein the amino acids having a functional group or a side chain which is positively charged at neutral pH are lysines, arginines or a mixture of lysine(s) and arginine(s).

3. The peptide of claim 1, wherein the peptide comprises at least one, preferably at least two and most preferably at least three alternating region(s) (S) and regions (W) each, wherein each region (S) comprises at least one amino acid having a functional group or a side chain which is positively charged at neutral pH, and each region (W) comprises at least one histidine.

4. The peptide of claim 1 wherein the at least two, preferably at least three histidines are present in a stretch or no more than 8 consecutive amino acids, preferably not more than 6 consecutive amino acids, and most preferably not more than 4 consecutive amino acids.

5. The peptide of claim 1, wherein any amino acids present in the peptide and having no functional group or side chain or having a functional group or side chain which is negatively charged at neutral pH are present in the peptide up to ⅕, preferably up to ⅒ and most preferably up to 1/20 of all amino acids of the peptide, and wherein said amino acids preferably (i) do have a functional group or a side chain which is negatively charged at neutral pH, and/or (ii) have a functional group or a side chain which does not bear a net charge at neutral pH.

6. A (poly)peptide being tagged with the peptide according to claim 1 .

7. The (poly)peptide of claim 6, wherein the peptide is genetically fused or chemically linked to the (poly)peptide.

8. The (poly)peptide of claim 6, wherein the (poly)peptide is an antibody or an antibody fragment, protein A, a hormone, an enzyme, a binding protein, a receptor, a ligand or a pharmaceutically active_(poly)peptide.

9. An inorganic surface comprising at least one peptide according to claim 1 and/or at least one (poly)peptide being tagged with the peptide, wherein the peptide and/or the (poly)peptide is directly bound to the inorganic surface.

10. The inorganic surface of claim 9, wherein the inorganic surface comprises iron oxide or silica.

11. The inorganic surface of claim 9, wherein the inorganic surface is a particle, preferably a nanoparticle or a microparticle.

12. The inorganic surface of claim 10, wherein (i) the iron oxide comprises ferrous oxide, magnetite, maghemite or a transition state between magnetite and maghemite, and/or (ii) the silica comprises silanol or siloxane.

13. A microorganism or a host cell expressing the peptide according to claim 1 and/or a (poly)peptide being tagged with the peptide .

14. A method for immobilizing the peptide according to claim 1 and/or a (poly)peptide being tagged with the peptide on an inorganic surface, the method comprising contacting the inorganic surface with the peptide, the (poly)peptide or a microorganism or host cell under conditions, wherein the microorganism or host cell expresses the peptide or the (poly)peptide, and wherein the peptide and/or the (poly)peptide directly binds to the inorganic surface via the peptide .

15. A method for purifying the peptide according to claim 1 and/or a (poly)peptide being tagged with the peptide, the method comprising (a) contacting an inorganic surface with a mixture comprising the peptide, the (poly)peptide and/or a microorganism or host cell under conditions, wherein the microorganism or host cell expresses the peptide or the (poly)peptide, and wherein the peptide and/or the (poly)peptide directly binds to the inorganic surface via the peptide, and (b) optionally eluting the peptide and/or the (poly)peptide from the inorganic surface, thereby purifying the peptide and/or the (poly)peptide from said mixture.

Description

[0108] The figures show.

[0109] FIG. 1. Adsorption isotherm of (HR).sub.4-tagged ER with 1 g L.sup.-1 magnetic nanoparticles in three different environments: 50 mM Tris pH 7.0,50 mM Tris pH 7.8 and TBS pH 7.0. Analysis via BCA. Measured as technical duplicates.

[0110] FIG. 2. Adsorption isotherm of (HR).sub.4-tagged ER with 1 g L.sup.-1 magnetic nanoparticles in three different environments: 50 mM Tris pH 7.0,50 mM Tris pH 7.8 and TBS pH 7.0.. Analysis via HPLC. Measured as technical duplicates.

[0111] FIG. 3. Competitive adsorption isotherm of (HR).sub.4-ER with GFP w/o tag and a constant E. coli lysate concentration of 2 g L.sup.-1 and 1 g L.sup.-1 IONs. 10 .Math.L of a final ION concentration of 2 g L.sup.-1 was loaded onto the gel (20 .Math.g). The ER amount on the IONs increased whereas no GFP was adsorbed. Lysate proteins are adsorbed being constant over the whole range. References (Supernatant of ION incubation (IONS)) in lane 9 and 10 show both proteins existent in the mixture.

[0112] FIG. 4. Binding of (HR).sub.4-ER to IONs. Freeze-thaw supernatant (FTS) after 3 times freeze-thaw. Incubation of FTS with different magnetic iron oxide nanoparticle (ION) concentrations. IONs were resuspended in Tris pH 7.8. MNPB: bound proteins to ION. (HR).sub.4-ER detected in MNPB fractions. MNPB normalized to 5 g L.sup.-1. The numbers in lane 2 are the sizes of the underlying band in kDa.

[0113] FIG. 5. Relative enzyme activity of immobilized (HR).sub.4-ER and ER w/o tag onto IONs. The relative enzyme activity is calculated as the adsorbed enzyme activity divided by the free enzyme activity. The activity was determined at 10 mM maleimide and different concentrations of NADH at 50 mM Tris pH 7.0. The adsorption happened at 50 mM Tris pH 7.0. Measurements performed as technical triplicates.

[0114] FIG. 6. Model for adsorption of (HR).sub.4-ER and ER w/o tag to IONs. (HR).sub.4-ER adsorbs to IONs via tag whereas ER w/o tag adsorbs unspecifically.

[0115] FIG. 7. Long-term stability of ER w/o tag and (HR).sub.4-ER in a free and immobilized state, respectively. The enzyme activity was determined at 10 mM maleimide, 0.5 mM NADH and 50 mM Tris pH 7.0 for 28 days. The enzyme activity was related to the enzyme activity of the corresponding sample at the timepoint 0. The enzyme activities were measured every 7 days as technical triplicates. The samples were stored at 4° C.

[0116] FIG. 8. Principle of IgG-binding by novel Protein A-based ligand modified, bare IONs.

[0117] FIG. 9. IgG binding studies using IONs@B8-(RH).sub.4 (A / B) and diverse control systems (C / D). Bare IONs, GFP-(RH).sub.4@IONs, without (w/o) IONs are used as references. Binding conditions (A / C): 20 mM phosphate pH 7.4, 150 mM NaCl, 1 h, 25° C. Desorption conditions (B / D): 50 mM glycine pH 2.9, 150 mM NaCl, 2 h, 25° C. All measurements were conducted as technical duplicates.

[0118] FIG. 10. Cell capture of E. coli by functionalized IONs (ION@B8-RH.sub.4@IgG) and using two different initial E. coli concentrations.

[0119] FIG. 11. Purification of GFP-(RH).sub.4 with a silica stationary phase and 50 mM TRIS pH 8 mobile phase. The purification is performed at a flow rate of 76 cm h.sup.-1. In the first 125 CV the lysate is loaded on the column. After loading the column is equilibrated till the 280 nm curve shows no protein flow through. A washing step with 1 M NaCl in 50 mM TRIS pH 8 is performed for 25 CV. The elution is performed with a 0.5 M lysine in 50 mM TRIS pH 8 buffer via a gradient over 20 CV.

[0120] FIG. 12. SDS-PAGE (12.5%) of the GFP-(RH).sub.4 Purification with silica stationary phase. GFP-(RH).sub.4 can be seen slightly over the 25 kDa ladder band in the lysate and the elution lanes. The flow-through lanes show increasing GFP-(RH).sub.4 concentrations which indicates overloading of the stationary phase. Gel evaluation with the Amersham Typhoon resulted in a purity of about 90 % for the Elution Peak (Lane 12).

[0121] FIG. 13: For each protein (green fluorescent protein (GFP), ene reductase (ER) and polymerized B binding domains of protein A (B8)) production process, a SDS-PAGE is shown containing a ladder with marker proteins indicating distinct sizes (left), the lysate (middle) and the purified protein (right) is shown (For ene reductase the ladder and the lysate are switched). All proteins contain an (RH).sub.4-tag and were loaded on a silica column with a loading buffer (TRIS 50 mM pH 8) and eluted with a similar buffer containing additional 500 mM lysine. Densitometric determination yields purities between 80-85% after just one purification step.

[0122] FIG. 14: Adsorption isotherms of tagged ene reductase to different silica species (the number in the legend indicates the pore size in Å).

[0123] FIG. 15: The “on particle” SDS-PAGE shows the proteins bound to the silica resin material after the pH shift to pH 6, 5, 4, 3 or 2 and elution with lysine (500 mM). The first and the last line represent protein ladders.

[0124] The examples illustrate the invention.

EXAMPLE 1 - MATERIAL AND METHODS

Synthesis of IONs

[0125] The iron oxide nanoparticles used for this study were synthesized by co-precipitation of Fe.sup.2+ and Fe.sup.3+ in alkaline aqueous solutions adapted to the procedure described by Roth et al. (Roth et al. 2015). Briefly, 21.2 g of FeCl.sub.3 x 6 H.sub.2O and 8.29 g of FeCl.sub.2 x 4 H.sub.2O were dissolved in 200 mL of deionized, degassed water which equals a Fe(III) : Fe(II) ratio of 1.88 : 1. This iron chloride solution was added to 1 L of a 1 M solution of NaOH prepared with deionized, degassed water stirring at 250 rpm in a reaction vessel. The reaction mixture was kept under nitrogen atmosphere at 25° C. and stirred for 30 more minutes before the nanoparticles that had formed were washed several times with degassed deionized water until the conductivity of the ION solution was below 200 .Math.S cm.sup.-1. In order to separate the particles from the washing water the mixture was placed on a NdFeB permanent magnet. FeCl.sub.3 x 6 H.sub.2O and sodium hydroxide were purchased from AppliChem GmbH, Germany in the highest purity available. FeCl.sub.2 x 4 H.sub.2O extra pure was obtained from Merck KGaA, Germany.

[0126] Silica beads have been purchased from Sigma-Aldrich (Davisil 643).

Protein Immobilization

Adsorption Isotherm

[0127] Adsorption isotherms were conducted to investigate the affinity and maximum load of pure (HR).sub.4-ER to IONs. The (HR).sub.4-ER was considered as pure if the purity was above 95% according to densitometric analysis via SDS-PAGE. Different (HR).sub.4-ER concentrations were mixed with IONs obtaining 1 g L.sup.-1 ION in every sample with different final protein concentrations. These different protein concentrations were 0, 0.025, 0.05, 0.1, 0.25, 0.5, 0.75, 1.0, 1.5 and 2 g L.sup.-1. The mixtures were incubated for 1 hour at 16° C. and 1200 rpm. The IONs were separated from the supernatant by applying a magnetic field. This supernatant was defined as MNPS (ION supernatant). The IONs were washed with the same buffer. Therefore, the same volume as before was added to the IONs and mixed by vortexing. One more time, a magnetic field was applied separating the IONs from the supernatant. That supernatant was termed as MNPW (ION wash). Finally, the IONs were resuspended in the initial volume, defining this sample as MNPB (ION bound).

[0128] The MNPS and MNPW samples were analyzed by BCA assay and by HPLC (see HPLC analysis). Additionally, the enzymatic activity of MNPS, MNPW and MNPB samples was investigated as described below. All samples were conducted as technical duplicates. Finally, the values were fitted using the Langmuir fit.

HPLC Analysis

[0129] Besides the BCA analysis, the MNPS and MNPW samples of the adsorption isotherm were analyzed via HPLC. These samples were centrifuged (17000 g, 5 min) prior analysis. 8 .Math.L of each sample were injected and analyzed on a C4 column (Aeris 3.6 .Math.m Widepore C4 200 Å 150 × 2.1 mm). A mixture of water with 20 mM TFA (Buffer A) and 100% acetronitrile with 20 mM TFA (Buffer B) was used as mobile phase. Following method was applied: Firstly, a gradient going from 40% B to 60% B and a gradient length of 10 CV was applied, followed by a 100% B step which lasted 3 CV. Finally, the column was equilibrated at 40% B for 5 CV. The samples were injected two times. The amount of (HR).sub.4-ER was calculated by a standard curve of (HR).sub.4-ER standards in the corresponding buffers.

Capturing (HR).SUB.4.-ER with IONs (Qualitative Analysis)

[0130] The binding of (HR).sub.4-ER from the lysate was detected by SDS-PAGE. For that, the E. coli cells were lysed after cultivation in shaking flasks via freeze-thaw. The IONs and the pellet of an (HR).sub.4-ER cultivation were resuspended in 50 mM Tris pH 7.8. The supernatant after freeze-thaw (FTS) was mixed 1:1 with different ION concentrations to obtain final ION concentrations of 5, 1.5 and 0.25 g L.sup.-1, respectively. The mixtures were incubated for 1 hour at 10° C. and 1000 rpm. A magnetic field was applied to separate the IONs from the supernatant. This supernatant was defined as MNPS. Washing the IONs with the same buffer, the same volume as before was added to the IONs, mixed by shortly vortexing and again a magnetic field was applied separating the IONs from the supernatant. That supernatant was termed as MNPW. Finally, the IONs were resuspended in defined volumes to obtain normalized ION concentrations (MNPB, ION bound) and to compare the binding by SDS-PAGE.

Competitive Adsorption Isotherm

[0131] The goal of the competitive adsorption isotherm was to show the adsorption of a (HR).sub.4-tagged protein in a highly competitive environment containing many other proteins and one protein with the same concentration. For this, every sample contained a constant concentration of 2 g L.sup.-1 E. coli lysate proteins and 1 g L.sup.-1 IONs. The concentrations of (HR).sub.4-ER and GFP w/o tag were added to this mixture and their concentrations were varied using 0, 0.05, 0.1, 0.25, 0.5, 0.75 and 1 g L.sup.-1. All components were prepared in 50 mM Tris pH 7.0. GFP w/o tag and (HR).sub.4-ER had a purity above 95% according to densitometry. All experiments have been performed in technical triplicates. The samples were incubated at 16° C. for 1 h at 1200 rpm. The following steps were performed analogously to the adsorption isotherm of the pure (HR).sub.4-ER. To show which proteins adsorbed to the IONs, the MNPB samples were concentrated to 4 g L.sup.-1 IONs each and mixed 1:1 with SDS loading buffer containing DTT. That mixture was boiled for 5 min at 95° C. and 10 .Math.L were loaded onto a 15% polyacrylamide gel.

Measurement of Enzymatic Activity

[0132] The enzymatic activity of ene reductase was assessed by measuring the decrease of absorption at 340 nm which correlates with the decrease of NADH which is oxidized to NAD.sup.+. Hereby, the reaction rate of the enzyme at 500 .Math.M NADH (or 200 and 800 .Math.M) and 10 mM maleimide in 50 mM Tris pH 7.0 at 30° C. was determined. The free enzyme had a final concentration of approximately 1 - 5 .Math.g mL.sup.-1 and the immobilized variant 10 - 20 .Math.g mL.sup.-1 to assure a linear decrease of the absorption. The extinction at 340 nm was determined every 40 s for 10 min. As blank served the mixtures without maleimide. Technical triplicates of each sample were measured in triplicates.

[0133] The enzyme activity was calculated as Units (turnover of 1 .Math.mol substrate per min). So, the specific enzymatic activity v.sub.x (in U mg.sup.-1) was calculated from the decreasing absorption ΔA.sub.340 by applying the Lambert-Beer law and considering the reaction volume V.sub.R (in L) and applied enzyme mass m.sub.E (in g).

[00001]$v_x=\frac{\Delta A_{340}\text{}\cdot V_R}{\Delta t\text{}\cdot {\epsilon }_{NADH}\text{}\cdot d\cdot m_E}$

[0134] The molar extinction coefficient of NADH ε.sub.NADH is 6220 L mol.sup.-1 cm.sup.-1 and the thickness d is 0.59 cm. Only a linear decrease of absorption was used for the calculation. For this case, linear means the linear decrease of absorption for at least 3 min (6 time points) showing an R.sup.2> 0.995. The “activity” of the blank was fitted over the whole 10 min since no linear fit with R.sup.2 > 0.995 was possible. The blank was subtracted from the enzyme activity of the sample.

[0135] The relative enzyme activities of the immobilized variants were calculated as the specific enzyme activity of the immobilized variant divided by the specific enzyme activity of the free variant. Regarding the long-term stability, the enzyme activity of the free ER and immobilized ER variants were measured every 7 days for 28 days. The variants were stored at 4° C. The experiments were performed as technical triplicates and each replicate measured three times. The relative enzyme activity was calculated from the specific enzyme activity at the time point 0 divided by the specific enzyme activity at the time point x.

EXAMPLE 2 - RESULTS

[0136] Ene reductase is an enzyme from the class of oxidoreductases which catalyzes the reduction of activated alkenes. This enzyme was tagged with a (HR).sub.4-tag. Since GFP tagged with this (HR).sub.4-tag adsorbed to magnetic iron oxide nanoparticles quite well, it was aimed to demonstrate its transferability to other more complex proteins. That would demonstrate the broad applicability of the tag. The tag consists of histidine and arginine showing strong interactions with iron oxide nanoparticles. Previous adsorption studies of GFP bearing different tags have been performed at Tris and a physiological pH. Thus, adsorption studies of (HR).sub.4-tagged ene reductase were made at similar conditions. The following figures show the adsorption isotherms of (HR).sub.4-ER with 1 g L.sup.-1 magnetic nanoparticles. These adsorption isotherms were analyzed via BCA and HPLC, respectively.

[0137] The adsorption isotherms analyzed via BCA and HPLC showed the same trends. For better comparison with loads from the literature, the curves were fitted with Langmuir considering the range from 0 to 0.75 g L.sup.-1 (Tris pH 7.8) or 1 g L.sup.-1 equilibrium concentration (Tris and TBS pH 7.0). The highest adsorption could be measured at 50 mM Tris pH 7.0 having a maximum load of 0.28 ± 0.008 g g.sup.-1 (BCA) or 0.383 ± 0.009 g g.sup.-1 (HPLC), respectively. At TBS pH 7.0 the load was 0.23 ± 0.042 g g.sup.-1 (BCA) or 0.243 ± 0.009 g g.sup.-1 (HPLC). The lowest load was assessed for 50 mM Tris pH 7.8 with 0.058 ± 0.006 g g.sup.-1 (BCA) and 0.114 ± 0.025 g g.sup.-1 (HPLC).

[0138] FIG. 3 depicts that with increasing GFP w/o tag and (HR).sub.4-ER concentration only a band at around 40 kDa is getting stronger and no band at 28 kDa. This means that (HR).sub.4-ER (41.02 kDa) is binding and no GFP w/o tag (28.20 kDa), although they are present in the same concentrations and they are both negatively charged (pl.sub.GFP = 5.37; pl.sub.(HR)4-ER = 6.2). Lanes 9 and 10 served as reference to show the existence of both proteins in the mixture.

[0139] These findings could be supported by another experiment. In that experiment, (HR).sub.4-ER was captured directly from an E. coli cell lysate via IONs (see FIG. 4). For this, lysate was incubated with IONs at different concentrations and a binding of (HR).sub.4-ER could be observed qualitatively whereas less lysate proteins adsorbed on the IONs.

[0140] Therefore, it is assumed that the tag is responsible for a strong interaction and outcompetes other proteins not having the tag. The experiment of the competitive adsorption isotherm and the capturing experiment together confirmed the binding because of the tag.

[0141] The strong binding and high affinity of the (HR).sub.4-tag suggest its use as immobilization tag for enzymes. One important prerequisite for enzyme immobilization is the retained activity of an enzyme despite its immobilization. The relative enzyme activity of ION-immobilized (HR).sub.4-ER and ER w/o tag were determined after mixing pure ER with IONs.

[0142] FIG. 5 depicts the relative enzyme activity as a function of NADH concentration. The relative enzyme activity increases for ER w/o tag and (HR).sub.4-ER with increasing NADH concentration. This phenomenon is presumably caused by the higher probability of a collision of NADH and enzyme when more NADH is present which in turn leads to a reaction. More importantly, ER is still active after immobilization and supports the use of IONs as carrier for immobilization purposes. A few research groups have also published an immobilization of enzymes to IONs..sup.18,21-23 Both enzymes exhibit the same relative enzyme activity, thus no advantage is seen for the tag under those conditions. Nevertheless, it seems that the immobilized ER w/o tag saturates at around 0.8 mM NADH whereas the adsorbed (HR).sub.4-ER does not. This would indicate a more “directed” adsorption of (HR).sub.4-ER to the IONs while ER w/o tag just binds to the IONs electrostatically and in a more undirected way what would limit the mass transport to the active center of the enzyme. The hypothesis is that due to adsorption via (HR).sub.4 tag the active center is believed to be more accessible for substrates than the active center of the ER w/o tag which binds non-specifically and maybe oriented its catalytic center towards the IONs. This is depicted in FIG. 6.

[0143] The activity of both ER variants and both states (free, immobilized) decreased. Interestingly, the relative enzyme activity of immobilized ER decreased faster than their free counterparts. More importantly, the ER w/o tag bound to IONs showed the fastest decrease in relative enzyme activity. The tag seems to have a stabilizing effect onto the free enzyme and the immobilized one. The immobilized (HR).sub.4-tagged ER retained 32.5 % of its initial activity whereas the immobilized ER w/o tag showed only 7.9 %. These results indicate that the ER w/o tag is less stable on the IONs. One possibility is that the untagged variant binds to the IONs nonspecifically and is distorted over time while the (HR).sub.4-ER preferentially adsorbs via tag and is less affected by particles.

[0144] Another interesting field for protein immobilization is the immobilization of protein domains for antibody purification. A prominent example for the use of this strategy is Protein A chromatography.

EXAMPLE 3 - ANTIBODIES

[0145] Antibodies are important molecules, that are used in diagnostics, sensor technology and as therapeutics inter alia in the field of hematology and oncology..sup.24,25 For their purification, a general platform process is applied. Thereby, Protein A-chromatography is the central step of this process. This step ensures capturing of the antibody, reducing the whole process volume, and resulting in purities over 95% while yields of 99% are achieved. However, during the whole antibody manufacturing, the downstream process accounts for the biggest part of the costs with the Protein A-chromatography owning the biggest share..sup.26-28

[0146] This is especially due to the rate limiting step of pore diffusion. This is a general problem of chromatography, but Protein A being a big-sized ligand (compared to e.g. ion exchange ligand) and the target molecule IgG being likewise big-sized exacerbate the problem with IgG’s low effective diffusivity. This leads to the need of high residence times and thus lower productivities..sup.29 Magnetic separation with non-porous magnetic iron oxide nanoparticles (IONs) has the potential to overcome these problems. Due to their nanoscale size, they offer high specific surface areas (~89 m.sup.2/g) without the need of pores..sup.30 For protein immobilization, usually functionalized magnetic nanoparticles e.g. by functional silanes or metal chelate materials are used. This approach differs from others: unfunctionalized IONs are used for a Protein A derivative consisting of 8 polymerized B-domains..sup.31 The ligand’s terminus is functionalized with the affinity peptide tag (RH).sub.4. FIG. 8 shows this ligand and the principle. These affinity materials could be used to efficiently bind and elute human IgG.

Methods

[0147] The polymerized B-domain protein (B8-(RH).sub.4) was placed in the plasmid pET28a and expressed in E. coli BLR(DE3). The intracellularly produced protein was released by the method described in (Kaveh-Baghbaderani et al., 2018). The supernatant obtained in this way was purified by immobilized metal ion affinity chromatography (IMAC, HiTrap FF; GE Healthcare) to a purity >95% (confirmed by SDS-PAGE). For functionalization, 40 mg/mL B8-(RH).sub.4 was incubated with 0.5 g/L IONs in 20 mM Tris pH 7.0 with 150 mM NaCl at 25° C. for 1 h while shaking. Performing the BCA-assay (Thermo Fisher Scientific GmbH) onto these particles revealed a ligand density of 19.8 mg/g IONs. Human polyclonal IgG (Gammanorm, Octapharma GmbH), that was previously purified by Protein A-chromatography (Bio-Scale SuprA, Bio-Rad GmbH) in order to remove non-binding subclasses of IgG, was used for equilibrium isotherms. The ION@B8-(RH).sub.4 were incubated with different concentrations (up to 4 g/L) human polyclonal IgG (Gammanorm, Octapharma GmbH) for 1 h at 25° C. in 20 mM PBS (pH 7.4, 150 mM NaCl) while shaking. IgG was eluted by 50 mM Glycine pH 2.9 with 150 mM NaCl for 2 h at the same incubation conditions. IgG was quantified by BCA-assay. E. coli cells were separated by IONs@B8-(RH).sub.4 that were further functionalized with bound anti E. coli antibodies (Bio-Rad, Germany) in a concentration of 25 mg per g IONs. The bacteria cells (BL21(DE3) have been labeled by GFP during expression induced by 25 .Math.M IPTG. Therefore, the gene was inserted into the pET28a vector. Varying concentrations of MNP@B8-RH4@rabbitlgG were shaken with E. coli set to an optical density 0.05 and 0.1 at 550 rpm for 30 min. The particles were resuspended and magnetically separated. The fluorescence of the supernatant was measured in a microscale thermophoresis device (Monolith NT.115, NanoTemper, Germany, fluorescence only). The E. coli removal is the ratio of the initial fluorescence to the fluorescence after incubation.

Results

[0148] FIG. 9, A and B shows adsorption isotherms of IgG and their elution dependent on the equilibrium concentration c*. This novel material results in unprecedented IgG binding capacities of up to 915 mg IgG per g IONs. Thereby, up to 55% of the bound IgG could be successfully desorbed. In order to prove the specificity of these particles, various control systems were examined for their binding and elution capabilities for antibodies (FIGS. 9, C and D): (i) non-functionalized IONs to detect unspecific binding to the iron oxide surface, (ii) IONs functionalized with a model protein (GFP), also via the affinity tag (RH).sub.4, (iii) completely without IONs to see wall adsorption effects. The binding of the latter control was related to the same (non-existent) amount of IONs for better comparability. It becomes clear that any kind of surface - be it a nanoscale particle system or vessel walls - is capable of binding antibodies. However, the pH-shift to below 3 is not sufficient to elute them again. It was thus possible to prove that the functionality of the materials presented above is specifically based on the binding to Protein A derivatives.

[0149] These ligand functionalized IONs must not only be used for purification but can also be used for cell separation and cell detection. As an example, the B8-(RH).sub.4@IONs can be bound to anti-E. coli rabbit IgG. These IONs have been used for E. coli (BL21(DE3)) separation. Since the cell serves as a model, it was intracellularly labeled by GFP during protein expression so the cell removal could be monitored by fluorescence depletion (device: Monolith NT.115, NanoTemper Technologies GmbH, Germany). FIG. 10 shows the successful cell separation by these IONs using two different starting optical densities (OD) of E. coli. Up to 41% of an E. coli suspension (OD = 0.05) could be removed. These IONs could be used for cell separation, concentration, sorting and detection. The latter could be achieved e.g. by enzyme-linked reactions or using fluorescent labelings.

EXAMPLE 4 - PURIFICATION OF TAGGED PROTEINS WITH SILICA-BASED COLUMN CHROMATOGRAPHY

Materials and Methods

[0150] TABLE-US-00001 Used Bacterial strains Strain Genotype E. coli BL21 DE3 F- ompT gal dcm Ion hsdSB(rB- mB-) λ(DE3[lacl lacUV5-T7 gene 1 ind1 sam7 in5])

TABLE-US-00002 Buffers and solutions Name Composition Washing buffer 1 (WB1) 50 mM Tris pH 8 Washing buffer 2 (WB2) 50 mM Tris 1 M NaCl pH 8 Elution buffer 1 (E1) 0.5 M Lysine 50 mM Tris pH 8 Lysogeny broth medium (LB) 1% Tryptone 0.5% Yeast Extract 0.5 % NaCl Fairbanks A (Staining solution for SDS gels) 25% Isopropanol 10% Acetic Acid 0.05% Coomassie R-250 Fairbanks D (Destaining solution) 10% Acetic Acid

[0151] The E. coli were already containing the plasmid with the (RH).sub.4-GFP Tag. A pre-culture was prepared with the cells using LB medium. A concentration of 100 .Math.g/mL of Ampicillin was added to the culture, to ensure that only the cells with the correct plasmid were growing in the medium. After the addition of the appropriate antibiotic to the medium the pre-culture was transferred to the final 1 L baffled flasks, containing 350 mL medium. The final culture volume was approximately 400 mL. These were incubated at 37° C. at 200 rpm until the OD.sub.600 reached 0.6 - 0.7. This step was followed by the induction with 1 .Math.M IPTG and the overnight incubation at 37° C.

[0152] After overnight expression, the cells were harvested via centrifugation (3200 rpm, 15 min, 8° C.). The cell pellet was re-suspended in WB 1. To ensure the stability of the (RH).sub.4-GFP and removal of remaining DNA, protease inhibitor and DNAse I was added to the re-suspended solution. The suspension was put on ice and lysed via sonication (max. amplitude 40%, 3 min - 10 s on 20 s off). The lysed cell suspension was centrifuged to remove cell particles from the lysate. The centrifugation was carried out at 7927 rpm for 35 min at 4° C.

[0153] The cleared lysate was sterile filtered using a bottle top filter (pore diameter 0.2 .Math.m) to remove remaining cell debris, DNA and other contaminants.

Chromatography

[0154] Chromatography experiment was conducted using an Aekta Purifier by GE Healthcare Life Sciences. An Omnifit Labware column (d= 0.7854 cm) was used and filled with wide pore silica slurry (Davisil 643, Sigma), resulting in an overall column volume of 1 mL. The HETP and asymmetry factor was measured using 1 M NaCl solution and measuring the peak of the conductometry. The column was loaded with 50% lysate and 50% 50 mM TRIS pH 8 at a flow rate of 1 mL min.sup.-1. Following the loading of the column, it was washed with WB1 for 5 CV until the curves of 260 nm and 280 nm reached approximately 0 mAU to ensure that no further protein was passing the column. The column was cleared of unspecifically bound proteins with WB2 for 5 CV. The higher salt concentration removes ionically bound proteins. The column was then washed for 5 CV with WB1 before the start of the elution with E1 by a gradient from 0 to 100%.

SDS-Page

[0155] A SDS-PAGE separates charged molecules in a mix according to their electrophoretic properties in a polyacrylamide gel by their differences in molecular weight in an electric field. SDS as an anionic detergent can denature proteins up to their tertiary structure and apply a negative charge in a relative proportion to the proteins’ mass. The negatively charged and linearized polypeptides will move through the gel towards the anode. Proteins that are bigger in size will move slower than smaller ones.

[0156] For the analysis of (RH).sub.4-GFP a 12.5% SDS gel was prepared as follows:

TABLE-US-00003 Conditions for SDS-gel Volume Solution 3.3 mL ddH.sub.2O 2.5 mL 40% Acrylamide 2 mL 1.5 M Tris pH 8.8 80 .Math.L 10% SDS 80 .Math.L 10% APS 8 .Math.L TEMED

[0157] The samples were incubated with 5x Laemmli buffer at 95° C. for 5 min. In addition to the samples, a protein standard (color pre-stained protein standard, NEB 11-245 kDa) was added. After the electrophoresis, the gels were stained Fairbanks A and de-stained with Fairbanks D. The bands became visible because the stain would embed itself into the alkaline sidechains of the amino acids. The evaluation of the gel was performed after scanning the gel with an Amersham Typhoon scanner (GE Healthcare, Great Britain).

[0158] The purification of GFP-HR4 with silica stationary phase is shown to be possible. During the loading, the 280 nm curve clearly shows the flow through of proteins whereas the 488 nm curve, which is specific for GFP shows no significant flow through of GFP which indicates a selective binding of GFP-(HR).sub.4. During the loading, the 488 nm curve rises slowly indicating the flow through of GFP-(RH).sub.4 which also was observed by a light green coloring of the flow through by eye. The increasing concentration of GFP-RH4 in the flow through can also be seen on the SDS gel from lane 1 to lane 6 as the band of GFP-(RH)4 is more pronounced After loading the column is equilibrated to WB1. During this equilibration, the 280 nm curve falls back to baseline level. The 488 nm curve falls significant slower back to baseline level which indicates washing out of GFP-RH4 of the column. This can also be seen on the SDS gel in lanes 7 and 9 where mostly GFP-(RH).sub.4 is seen. Following this a washing step with WB2 containing a high salt concentration (1 M NaCl) was performed to wash out proteins bound by weak ion interactions to the column. Both the 288 nm and 488 nm curves show no significant elution of any protein. After the washing step the elution with E1 buffer over 20 CV is initiated. The GFP-(RH).sub.4 elutes at a lysine concentration of about 150 mM. The eluting peak (10 mL) is collected in four fractions: The start of the elution (1 mL), the front part of the peak (1 mL), the peak part of the elution (4 mL) and the tail of the elution peak (4 mL). The elution fractions can be seen in the lanes 10 to 13 on the SDS gel. In all four lanes a clear GFP-(RH).sub.4 band can be seen. The purity of every lane was calculated with the Amersham Typhoon and resulted in purities for lane 10 to 13 of 82 %, 78 %, 86 % and 94% respectively. In all four fractions the main contaminates are two proteins with a size if about 30 kDa and 43 kDa respectively. The protein at 43 kDa is likely to be an RNA binding Protein as also found in similar approaches.

EXAMPLE 5 - AFFINITY PROTEIN PURIFICATION WITH UNDERIVATIZED SILICA COLUMNS

Summary

[0159] Example 5 shows that the novel tag system as provided herein (GFP-(RH).sub.4) can be used for affinity protein purification with underivatized silica columns. This is a significant improvement to existing separation affinity-based methods such as immobilized metal affinity columns (IMAC), where hazardous materials such as imidazole are needed for protein elution.

[0160] In Example 5 protein purities of up to 85% directly from crude cell lysates (tested with PosH-Tag fused to GFP) are achieved with one binding step, one washing step and one elution step in a silica column (75 mL). Therefore, this purity is reached with just one capture step. Other proteins, such as the enzyme enereductase and a polymerized B domain from protein A, can be purified by the same method as well. However, there is still optimization necessary and possible. We already observed higher purities of 90-95% for tagged proteins when the column is overloaded which is even closer to realistic processing of the protein purification.

Immobilization

[0161] If a strong acid such as HCl (0.1 M) is added, the bond is irreversible, and the proteins cannot be eluted with lysine anymore. However, with boiling in SDS, an elution of denatured proteins is possible.

Experimental Details

[0162] For cleared lysate purification a XK16 column (GE Healthcare, Germany) filled with Davisil 643 (Sigma, Germany) was prepared with a column volume (CV) of 75 mL. The column was equilibrated with 4 CV of equilibration buffer (50 mM Tris pH 8.0) at a flowrate of 4 mL min.sup.-1. The equilibration was followed by loading of the cleared lysate on the column. Once the lysate was loaded on the column, it was washed with 4 CV of equilibration buffer. As soon as the UV signal decreased back to its baseline, the elution process was started. Isocratic elution was performed with a 50 mM Tris pH 8.0 + 500 mM L-lysine buffer.

[0163] For the static binding capacity, the purified GFP was loaded on silicas with different pore sizes:

TABLE-US-00004 Silica Particle size [.Math.m] Pore size [Å] Specific surface area [m.sup.2 g.sup.-1] Silica gel 60 40-63 60 500 Davisil 643 35-70 150 300 SiliCycle 500 40-75 500 47

[0164] A concentration of 1 g L.sup.-1 silica per tube was used at a total volume of 1 mL. The tubes were placed in a table-top shaker at 750 rpm, 10° C. overnight. The tubes were prepared in technical triplicate. To obtain the amount of protein bound to the silica particles a supernatant analysis BCA and particle BCA was conducted.

[0165] For the immobilization experiments purified 1 g L.sup.-1 GFP-(RH).sub.4 was bound to 1 g L.sup.-1 silica (Davisil) while in 50 mM Tris pH 8.0. After 1 h incubation the supernatant was discarded and 1 mL of 50 mM Tris with a pH of 2, 3, 4, 5, or 6 was added to the tube. After 1 h incubation the supernatant was discarded again, and the silica was washed 3x with 50 mM Tris pH 8. After washing, an elution step with 50 mM Tris pH 9.0 + 500 mM L-lysine was performed. After washing again with 50 mM Tris pH 8.0 buffer the silica was loaded on a SDS gel to check for remaining protein.

Results

[0166] The transferability of the (RH).sub.4 tag to other proteins was successfully implemented. The successful purification of (HR).sub.4-ER and B8-(RH).sub.4 shows that the (RH).sub.4 tag is functional on both, the C- and N-terminus. Purities of 80-85%, obtained from densitometry of SDS PAGEs was lower, can be attributed to the fact that in these runs no overloading but upscaling was performed (FIG. 13). When overloading the column with protein of interest purities >95% can be expected. Combining this step with another chromatographic step such as an ion exchange chromatography or immobilized metal afinity chromatography is a promising outlook for upscaling the process.

[0167] The adsorption behavior is similar for all investigated silica species (FIG. 13). However, silica with an average pore sizes of 6 nm and of 50 nm demonstrate a lower binding capacity for tagged GFP. This can be explained by the small and large pore sizes, respectively. Furthermore, the Silica 500 has a smaller specific surface area. The affinity of the tag to the silica surfaces is very similar for 15 and 50 nm silica pore sizes with 0.45 and 0.32 g L.sup.-1, respectively. This verifies the high affinity of the tag to silica surfaces. The lower affinity to silica with 6 nm pores (1.1 g L.sup.-1) can be explained by mass transport limitations into the pores. This pore size is too small for protein purification of an enzyme such as ene reductase.

[0168] Aside from purification using the “novel peptide tag”, we found out that an immobilization on silica is possible, as well. If a tagged protein (e.g. GFP) is loaded on silica particles (such as chromatography resin), the tagged protein binds to the surface at typical buffer conditions (TRIS 50 mM pH 8). As we tested different possibilities for protein elution, we found out that a low pH such as pH 2 or 3 does not lead to elution of the protein. While tagged protein incubated with pH 4-6 can be removed with the elution buffer (500 mM lysine and 50 mM Tris pH 9.0), the proteins incubated at pH 2 and 3 cannot be eluted with the elution buffer. We verified the adsorption with on-particle SDS PAGE (FIG. 15). Hence, the tagged proteins seem to be separable from silica when heated up to 90° C. in an SDS containing solution.

[0169] The results shown above illustrate the possibility to immobilize proteins with the help of the novel tag as described herein on silica, in particular on underivatized silica.

REFERENCES

[0170] Berensmeier, S.; Blank-Shim, S.; Schwaminger, S.; Fraga-Garcia, P.; Wenzel, W.; Borkowska-Panek, M.; Anand, P.; Fink, K. Magnetic particle-binding peptides.

[0171] Terpe, K. Overview of tag protein fusions: From molecular and biochemical fundamentals to commercial systems. Applied microbiology and biotechnology 2003, 60, 523-533.

[0172] Colombo, M.; Carregal-Romero, S.; Casula, M. F.; Gutiérrez, L.; Morales, M. P.; Böhm, I. B.; Heverhagen, J. T.; Prosperi, D.; Parak, W. J. Biological applications of magnetic nanoparticles. Chemical Society reviews 2012, 41, 4306-4334.

[0173] Le Sage, D.; Arai, K.; Glenn, D. R.; DeVience, S. J.; Pham, L. M.; Rahn-Lee, L.; Lukin, M. D.; Yacoby, A.; Komeili, A.; Walsworth, R. L. Optical magnetic imaging of living cells. Nature 2013, 496, 486-489.

[0174] Laurent, S.; Forge, D.; Port, M.; Roch, A.; Robic, C.; Vander Elst, L.; Muller, R. N. Magnetic iron oxide nanoparticles: Synthesis, stabilization, vectorization, physicochemical characterizations, and biological applications. Chemical reviews 2008, 108, 2064-2110.

[0175] Berensmeier, S. Magnetic particles for the separation and purification of nucleic acids. Applied microbiology and biotechnology 2006, 73, 495-504.

[0176] Schwaminger, S. P.; Bauer, D.; Fraga-García, P.; Wagner, F. E.; Berensmeier, S. Oxidation of magnetite nanoparticles: Impact on surface and crystal properties. CrystEngComm 2017, 19, 246-255.

[0177] Roth, H.-C.; Schwaminger, S. P.; Schindler, M.; Wagner, F. E.; Berensmeier, S. Influencing factors in the CO-precipitation process of superparamagnetic iron oxide nano particles: A model based study. Journal of Magnetism and Magnetic Materials 2015, 377, 81-89.

[0178] Massart, R. Preparation of aqueous magnetic liquids in alkaline and acidic media. IEEE Trans. Magn. 1981, 17, 1247-1248.

[0179] Kuroda, A.; Ikeda, T.; Funabashi, H. Purification method, purification kit, and silicon oxide-binding tag for use therein.

[0180] Baneyx, F. C.; Coyle, B. L. Affinity tags and processes for purifying and immobilizing proteins using same.

[0181] Schwaminger, S. P.; Fraga-García, P.; Eigenfeld, M.; Becker, T. M.; Berensmeier, S. Magnetic Separation in Bioprocessing Beyond the Analytical Scale: From Biotechnology to the Food Industry. Frontiers in bioengineering and biotechnology 2019, 7, 233.

[0182] Block, H.; Maertens, B.; Spriestersbach, A.; Brinker, N.; Kubicek, J.; Fabis, R.; Labahn, J.; Schäfer, F. Chapter 27 Immobilized-Metal Affinity Chromatography (IMAC). Guide to Protein Purification, 2nd Edition; Methods in Enzymology; Elsevier, 2009; pp 439-473.

[0183] Hearn, M. T.; Acosta, D. Applications of novel affinity cassette methods: Use of peptide fusion handles for the purification of recombinant proteins. Journal of molecular recognition : JMR 2001, 14, 323-369.

[0184] Gaberc-Porekar, V.; Menart, V. Potential for Using Histidine Tags in Purification of Proteins at Large Scale. Chem. Eng. Technol. 2005, 28, 1306-1314.

[0185] Franzreb, M.; Siemann-Herzberg, M.; Hobley, T. J.; Thomas, O. R. T. Protein purification using magnetic adsorbent particles. Applied microbiology and biotechnology 2006, 70, 505-516.

[0186] Uzun, K.; Çevik, E.; Senel, M.; Sözeri, H.; Baykal, A.; Abasiyanik, M. F.; Toprak, M. S. Covalent immobilization of invertase on PAMAM-dendrimer modified superparamagnetic iron oxide nanoparticles. J Nanopart Res 2010, 12, 3057-3067.

[0187] Johnson, A. K.; Zawadzka, A. M.; Deobald, L. A.; Crawford, R. L.; Paszczynski, A. J. Novel method for immobilization of enzymes to magnetic nanoparticles. J Nanopart Res 2008, 10, 1009-1025.

[0188] Chang, C.-C.; Lee, S.-F.; Sun, K.-W.; Ho, C.-C.; Chen, Y.-T.; Chang, C.-H.; Kan, L.-S. Mn,Cd-metallothionein-2: A room temperature magnetic protein. Biochemical and biophysical research communications 2006, 340, 1134-1138.

[0189] Baalousha, M. Aggregation and disaggregation of iron oxide nanoparticles: Influence of particle concentration, pH and natural organic matter. The Science of the total environment 2009, 407, 2093-2101.

[0190] Schnell, F.; Kube, M.; Berensmeier, S.; Schwaminger, S. P. Magnetic Recovery of Cellulase from Cellulose Substrates with Bare Iron Oxide Nanoparticles. ChemNanoMat 2019, 5, 422-426.

[0191] Roth, H.-C.; Schwaminger, S. P.; Peng, F.; Berensmeier, S. Immobilization of Cellulase on Magnetic Nanocarriers. ChemistryOpen 2016, 5, 183-187.

[0192] Shaw, S.-Y.; Chen, Y.-J.; Ou, J.-J.; Ho, L. Preparation and characterization of Pseudomonas putida esterase immobilized on magnetic nanoparticles. Enzyme and Microbial Technology 2006, 39, 1089-1095.

[0193] Zhang, X.; Soori, G.; Dobleman, T. J.; Xiao, G. G. The application of monoclonal antibodies in cancer diagnosis. Expert review of molecular diagnostics 2014, 14, 97-106.

[0194] Saeed, A. F. U. H.; Wang, R.; Ling, S.; Wang, S. Antibody Engineering for Pursuing a Healthier Future. Frontiers in microbiology 2017, 8, 495.

[0195] Kelley, B. Very large scale monoclonal antibody purification: The case for conventional unit operations. Biotechnology progress 2007, 23, 995-1008.

[0196] Process scale purification of antibodies; Gottschalk, U., Ed., Second edition; John Wiley & Sons Inc: Hoboken, NJ, 2017.

[0197] Shukla, A. A.; Hubbard, B.; Tressel, T.; Guhan, S.; Low, D. Downstream processing of monoclonal antibodies--application of platform approaches. Journal of chromatography. B, Analytical technologies in the biomedical and life sciences 2007, 848, 28-39.

[0198] Carta, G.; Jungbauer, A. Protein chromatography: Process development and scale-up; Wiley-VCH: Weinheim, 2010.

[0199] Schwaminger, S. P.; Blank-Shim, S. A.; Scheifele, I.; Fraga-García, P.; Berensmeier, S. Peptide binding to metal oxide nanoparticles. Faraday discussions 2017, 204, 233-250.

[0200] Freiherr von Roman, M.; Berensmeier, S. Improving the binding capacities of protein A chromatographic materials by means of ligand polymerization. Journal of chromatography. A 2014, 1347, 80-86.

[0201] Sanchayita Ghose, Thomas M. McNerney, Brian Hubbard, Preparative protein purification on underivatized silica, 2004, 87(3):413-423, BIOTECHNOLOGY AND BIOENGINEERING,