MAP Tag: A Novel Tagging System for Protein Purification and Detection

Protein purification is an essential procedure in fields such as biochemistry, molecular biology, and biophysics. Acquiring target proteins with high quality and purity is still difficult, although several tag systems have been established for protein purification. Affinity tag systems are excellent because they possess high affinity and specificity for acquiring the target proteins. Nevertheless, further affinity tag systems are needed to compensate for several disadvantages of the presently available affinity tag systems. Herein, we developed a novel affinity tag system designated as the MAP tag system. This system is composed of a rat anti-mouse podoplanin monoclonal antibody (clone PMab-1) and MAP tag (GDGMVPPGIEDK) derived from the platelet aggregation-stimulating domain of mouse podoplanin. PMab-1 possesses high affinity and specificity for the MAP tag, and the PMab-1/MAP tag complex dissociates in the presence of the epitope peptide, indicating that the MAP tag system is suitable for protein purification. We successfully purified several proteins, including a nuclear protein, soluble proteins, and a membrane protein using the MAP tag system. The MAP tag system is very useful not only for protein purification but also in protein detection systems such as western blot and flow cytometric analyses. Taken together, these findings indicate that the MAP tag system could be a powerful tool for protein purification and detection.


Introduction
A ffinity tag systems are highly useful for protein purification, detection, and analysis. Affinity tag systems are classified broadly as ''peptide tags'' and ''protein tags.'' For example, GST tag (1) and MBP tag (2) are protein tags. While these tags are useful for soluble fraction proteins, they are large (over 25 kDa) and thus can affect target protein function, requiring their removal before protein analysis. In contrast, peptide tags are less likely to affect the structure and function of target proteins because they are small (typically 1-2 kDa). Thus, removal of the tag portion is not always required before protein analysis. The most appropriate peptide tag system can be chosen from a variety of options, including the His tag, (3) FLAG tag, (4) TARGET tag, (5) PA tag, (6,7) AGIA tag, (8) and many others. (9,10) The fusion peptides are compatible with virtually all expression host cells, including yeast, insect, and mammalian cells, and Escherichia coli and can be expressed extracellularly or intracellularly.
An excellent affinity tag system should have high affinity and high specificity. However, not all peptide-based tag sys-tems meet these criteria. For example, the purification of oligohistidine-tagged proteins using metal chelate affinity resin often results in the co-purification of metal-binding proteins present in the starting material, necessitating further purification steps. (11) Generally, epitope tag systems that utilize peptide tags and anti-peptide monoclonal antibodies (mAbs) are highly specific. However, we often encounter nonspecific binding of mAbs to endogenous proteins in certain cell types, (12,13) even when using the most popular tag system, FLAG tag/anti-FLAG M2 antibody. (14) Importantly, the most suitable tag systems must be chosen based on the target protein, expression host, and many other variables. Therefore, the development of further affinity tag systems is needed to overcome the disadvantages of available affinity tag systems.
We previously established a useful rat mAb (clone PMab-1) against a 14-residue peptide in the platelet aggregation-stimulating (PLAG) domain of mouse podoplanin. (15) Podoplanin is a type I transmembrane protein that is highly expressed in malignant cancer cells and is implicated in tumor-induced platelet aggregation. (16) In our another study, we developed the PA tag system with high affinity and specificity. (6) The PA tag is derived from the human podoplanin PLAG domain. Three tandem repeats of the PLAG domain are conserved in podoplanin orthologs of the rat, hamster, dog, cow, human, and mouse. (17) Additionally, PMab-1 possesses high affinity and specificity toward mouse podoplanin. (18) Therefore, it was predicted that PMab-1 would have characteristics suitable for an anti-tag antibody. Here, we report the development of a novel affinity tag system comprising PMab-1 and its epitope peptide MAP tag.

Binding kinetics analysis using surface plasmon resonance
Rat anti-MAP mAb PMab-1 (IgG 2a , k) was described previously. (15) PMab-1 was immobilized on CM5 sensor chips using amino coupling chemistry according to the method provided by the manufacturer. The MAP-tagged ATRXepi proteins were diluted in PBST (phosphate-buffered saline, pH 7.4, containing 0.005% Tween 20) and injected at a flow rate of 30 mL/min. The ATRXepi was diluted in PBST (12.5, 25, 50, 100, and 200 nM) and passed over the biosensor chip. The binding was monitored for 60 seconds, followed by dissociation in running buffer for 120 seconds using the Single-cycle kinetics method. Binding curves were analyzed using BIAcore X100 Evaluation Software (GE Healthcare, Piscataway, NJ) with curve fitting using a 1:1 binding model.

Protein expression and purification
PMab-1 was coupled to CNBr-activated Sepharose 4B (GE Healthcare) according to the protocol provided by the manufacturer, routinely yielding coupling levels of *2 mg IgG/mL gel. Recombinant ATRXepi (PA-ATRXepi-RAP-MAP) was expressed in transformed E. coli BL21 (DE3) cultured for *3 hours at 37°C in the presence of 1 mM IPTG, and the bacterial pellet from a 100 mL culture was suspended in 10 mL of Tris-buffered saline (TBS; pH 7.5). Bacteria were lysed by sonication, and the soluble fraction was obtained by centrifugation at 15,000 g for 15 minutes. The lysates were passed through a 0.45 mm filter to remove any trace amounts 294 FUJII ET AL.
of insoluble materials. The cleared lysates were mixed with PMab-1-Sepharose (1 mL bed volume) and incubated at 4°C for 3 hours under gentle agitation. The beads were then transferred to a column and washed with 20 mL of TBS. The bound protein was eluted with TBS containing 0.1 mg/mL epitope peptide (GDGMVPPGIEDKIT). The elution was conducted at room temperature in a stepwise manner (1 mL · 10), with a dissociation time of 5 minutes allowed for each elution step. For the purification of soluble EGFR ec and HER2 ec secreted from mammalian cells, LN229/EGFR ec , and LN229/ HER2 ec were plated in 225 cm 2 flask (Thermo Fisher Scientific, Inc.), and 1 L of culture supernatant was harvested. Cleared supernatant was passed through PMab-1-Sepharose (4 mL bed volume), and the same process was repeated three times. The beads were then washed with 80 mL of TBS and eluted with the same solution containing 0.1 mg/mL epitope peptide in a stepwise manner (4 mL · 10) as described above. For the purification of EGFR from mammalian cell membranes, LN229/EGFR was plated onto hundred 10-cm culture dishes. Cells were detached from the plates using PBS containing 1 mM EDTA. The pelleted cells were washed using PBS and solubilized by adding 10 mL of PBS containing 1% (w/v) Triton X-100 and 0.05 mg/mL aprotinin. The solubilized cell lysates were incubated at 4°C for 30 minutes and further centrifuged 15 minutes at 13,000 g at 4°C. The pelleted cells were solubilized in 5 mL of the same solution, and the process was repeated twice. The cleared lysates were mixed with PMab-1-Sepharose (1 mL bed volume) and incubated at 4°C for 3 hours under gentle agitation. The beads were then transferred to a column and washed with 20 mL of PBS. The bound protein was eluted with PBS containing 0.1 mg/mL epitope peptide in a stepwise manner (1 mL · 10) as described above.
Ten peptide-elution fractions from the column chromatography were subjected to 5%-20% SDS-PAGE under reducing conditions and stained with Coomassie Brilliant Blue (CBB) or Oriole fluorescent protein stain (Bio-Rad Laboratories Inc., Berkeley, CA).

Specificity analysis of PMab-1/MAP tag interaction
We next evaluated the specificity of PMab-1 for the MAP tag. As shown in Figure 1B, PMab-1 produced a strong single band corresponding to the expressed CD133 with MAP tag (*120 kDa) in western blot analysis of the cell lysate. PMab-1 did not show any reactive bands in CHO-K1 cells, indicating that PMab-1 is very specific against the MAP tag. The specificity was also analyzed using flow cytometry (Fig. 1C). PMab-1 reacted with the hPDPNdN55 plus Nterminal MAP tag (extracellular domain), which is expressed in HEK293T, COS-7, and CHO-K1 cells (Fig. 1C, middle panel) and did not react with the corresponding untransfected cells (Fig. 1C, upper panel). LpMab-17, a mouse anti-human podoplanin mAb, recognized all transfected cell lines (Fig. 1C, lower panel).

Application to the protein purification system
We attempted to purify the four proteins with MAP tag using the MAP tag system. ATRXepi (*25 kDa) was expressed in E. coli and purified from the soluble fraction of the bacterial lysate (Fig. 2). This protein eluted in fractions 2-10, with virtually no contamination. The soluble ectodomain fragments of EGFR (EGFR ec ) and HER2 (HER2 ec ) were expressed in LN229 and purified from the culture supernatant (Figs. 3A and 3B). Because EGFR ec and HER2 ec are highly glycosylated, they electrophoresed more slowly than predicted by the theoretical molecular weights. These proteins were eluted in elution fractions 2-10, with virtually no contamination. EGFR (*170 kDa) was expressed in LN229 and was purified from the membrane fraction of LN229 (Fig. 4). This protein eluted in fractions 2-10, and the single bands were detected in elution fractions 4-10 (Fig. 4A). It was also confirmed that the strong bands in elution fractions were derived from EGFR in western blot analysis using an anti-RAP tag mAb (clone: PMab-2) (Fig. 4B). All MAP-tagged proteins were captured onto PMab-1-Sepharose and eluted from the resin with a solution containing 0.1 mg/mL free epitope peptide (GDGMVPPGIEDKIT). Despite the high affinity between PMab-1 and MAP tag (Fig. 1A), these antigens were eluted by the peptide. We performed the elution at room temperature in a stepwise manner, with a dissociation time of 5 minutes allowed for each elution step. These eluted proteins showed high purity (Figs. 2-4).

Discussion
An excellent affinity tag system requires high affinity and specificity. PMab-1, a rat anti-mouse podoplanin mAb, was produced by immunization with the synthetic peptide ''GDGMVPPGIEDKIT''. (15) PMab-1 has high affinity and specificity toward mouse podoplanin as described previously. (18) Therefore, we investigated whether PMab-1 maintains high affinity and specificity toward the MAP tag ''GDGMVPPGIEDK'' that was attached to several proteins.
First, we analyzed the binding affinity between PMab-1 and the MAP tag. Curve-fitting analysis showed high-affinity  1-10) from the column chromatography were subjected to 5%-20% SDS-PAGE under reducing conditions and transferred to a membrane. The membrane was immunoblotted with 1 mg/mL of PMab-2 (anti-RAP tag) for 30 minutes and incubated with peroxidase-conjugated secondary antibody specific for mouse IgG. PBS, phosphate-buffered saline; SS, signal sequence.
PMab-1/MAP tag interaction (K D , 3.7 · 10 -9 M) (Fig. 1A). These data do not differ from those of previous studies that analyzed PMab-1/mouse podoplanin interaction using flow cytometry (K D , 3.2 · 10 -9 M). (18) Next, we evaluated the specificity of PMab-1 against MAP tag. As shown in Figures 1B and 1C, PMab-1 has high specificity toward MAP tag in Western blot analysis and flow cytometry. Although HEK293T cells are reported to express human podoplanin, (23) PMab-1 did not react with untransfected HEK293T cells because PMab-1 does not cross-react with human podoplanin. In fact, the important amino acids for PMab-1/ MAP tag interaction are the first aspartic acid and the methionine in the MAP tag (data not shown); human podoplanin does not have this sequence. These results indicate that PMab-1 has a high affinity and specificity toward not only mouse podoplanin but also the MAP tag. Therefore, we presumed that the PMab-1/MAP tag system would be applicable to the protein purification system.
We attempted to purify MAP-tagged ATRXepi, EGFR ec , HER2 ec , and EGFR using the MAP tag system (Figs. 2-4). These purified proteins showed high purity and were ready for use in the downstream experiments without further purification. Purifying membrane proteins is generally difficult because membrane proteins require detergent to make them soluble in an aqueous solution. In contrast, the MAP tag system purified membrane proteins such as EGFR in the presence of detergent (Fig. 4). These results indicate that the MAP tag system is a powerful protein purification tool that does not depend on the host or the type of protein.
In conclusion, we successfully developed a novel affinity tagging system designated ''MAP tag'' that employs a unique mAb PMab-1 against mouse podoplanin. Owing to its high affinity and specificity toward linear dodecapeptide antigen, PMab-1 can be used to detect and purify tagged proteins in various assay formats. In the near future, we will compare the MAP tag to several other affinity tag systems against many targets and determine the types of applications most suitable for the MAP tag system.