Recombinant expression and purification of AF1q and its interaction with T-cell Factor 7

Protein Expression and Purification 165 (2020) 105499

Contents lists available at ScienceDirect

Protein Expression and Purification journal homepage: www.elsevier.com/locate/yprep

Recombinant expression and purification of AF1q and its interaction with Tcell Factor 7

T

Nazimuddin Khan, Jino Park, William L. Dean, Robert D. Gray, William Tse, Donghan Lee∗, T. Michael Sabo∗∗ Department of Medicine, James Graham Brown Cancer Center, University of Louisville, 505 S. Hancock St., Louisville, KY, 40202, USA

ARTICLE INFO

ABSTRACT

Keywords: Human AF1q T-cell Factor 7 (TCF7) High-mobility group domain (HMG) NMR spectroscopy

The protein ALL1 fused from chromosome 1q (AF1q) is overexpressed in a variety of cancers and acts to activate several signaling pathways that lead to oncogenesis. For example, AF1q has been shown to interact with T-cell Factor 7 (TCF7; also known as TCF1) from the Wnt/β-catenin pathway resulting in the transcriptional activation of the CD44 and the enhancement of breast cancer metastasis. Despite the importance of AF1q in facilitating oncogenesis and metastasis, the structural and biophysical properties of AF1q remain largely unexplored due to the absence of a viable method for producing recombinant protein. Here, we report the overexpression of AF1q in E. coli as a fusion to a N-terminal His6-tag, which forms inclusion bodies (IBs) during expression. The AF1q protein was purified from IBs under denaturing conditions by immobilized metal affinity chromatography followed by a successful one-step dialysis refolding. Refolded AF1q was further purified to homogeneity by gel filtration chromatography resulting in an overall yield of 35 mg/L culture. Our nuclear magnetic resonance (NMR) and analytical ultracentrifugation (AUC) measurements reveal AF1q interacts with TCF7, specifically with TCF7's high-mobility group (HMG) domain (residues 154–237), which is, to our knowledge, the first biophysical characterization of the AF1q and TCF7 interaction.

1. Introduction ALL1 fused from chromosome 1q (AF1q; gene MLLT11; Uniprot Q13015; 90 amino acids; 10 kDa) was first identified as part of an overexpressed fusion gene related to the development of acute myeloid leukemia [1,2]. Since then, the overexpression of AF1q has been associated with many different types of cancers [3–11] and is considered an oncogenic and poor prognostic factor [4,5,10]. The AF1q protein exerts its oncogenic effects by activating the following signaling pathways: Wnt/β-catenin, protein kinase B/phosphatidylinositol (PI3K/AKT), platelet-derived growth factor receptor (PDGFR)/signal transducer and activator of transcription 3 (STAT3) [8,12,13]. Interestingly, AF1q stimulates drug and radiation-induced apotosis [14–16], indicating that AF1q possesses both pro- and anti-oncogenic functions, similar to the proteins Ras and Myc [17].To date, little is known concerning AF1q′s primary biological function except that it can promote T-cell differentiation in the thymus [18] and it may play a role in neurodevelopment [19,20].

Despite AF1q′s involvement in the development and progression of cancer, a detailed biophysical and structural analysis is still lacking due to the absence of a viable method for producing recombinant protein. For example, interaction of AF1q with T-cell Factor 7 (TCF7; also known as TCF1) from the Wnt/β-catenin pathway is proposed to support breast cancer metastasis by causing the transcriptional activation of the CD44 gene and other downstream targets [12]. However, the evidence for the AF1q-TCF7 interaction is supported only by co-immunoprecipitation and an electrophoretic mobility shift assay [12,21]. Here, we report a method for the recombinant expression and purification of AF1q, which possesses secondary structural elements that are primarily α-helical. In addition, we demonstrate that AF1q interacts with TCF7, specifically TCF7's HMG domain (residues 154–237), using nuclear magnetic resonance (NMR) spectroscopy and sedimentation velocity analytical ultracentrifugation (SV-AUC).

Abbreviations: TCF7, T-cell Factor 7; HMG, high-mobility group domain; NMR, nuclear magnetic resonance spectroscopy; HSQC, heteronuclear single quantum coherence; IBs, inclusion bodies ∗ Corresponding author. ∗∗ Corresponding author. E-mail addresses: [email protected] (D. Lee), [email protected] (T.M. Sabo). https://doi.org/10.1016/j.pep.2019.105499 Received 15 July 2019; Received in revised form 12 September 2019; Accepted 17 September 2019 Available online 18 September 2019 1046-5928/ © 2019 Published by Elsevier Inc.

Protein Expression and Purification 165 (2020) 105499

N. Khan, et al.

2. Materials and methods 2.1. Molecular cloning The 270 bp open reading frame (ORF) of the synthetic and codonoptimized DNA of human AF1q for expression in E. coli was obtained in the standard cloning vector pUC57 from NeoScientific, USA. It was excised from pUC57 by NdeI and BamHIHF restriction enzymes and cloned into a modified version of the E. coli expression plasmid pET-28a (Novagen) under the T7 promoter control. The pET-28a vector was modified by incorporating a duplex of the following two single stranded DNAs using NCOI & BamHI cloning sites: strand-1, 5′-CATGGGCAGCA GCCATCATCATCATCATCACAGCAGCGGCGAAAACCTGTATTTTCAGG GCCATATGCTCGAG-3′ and strand-2, 5′-GATCCTCGAGCATATGGCCCT GAAAATACAGGTTTTCGCCGCTGCTGTGATGATGATGATGATGGCTGC TGCC-3′. This modification introduced a N-terminal His6-tag followed by a TEV protease cleavage site (ENLYFQ/G) for tag removal instead of the thrombin cleavage site in the original plasmid (Fig. S1). Similarly, the 435 bp ORF of the human TCF7 (Uniprot P36402-6, residues 125–269) [12], was cloned into pET14bΔThrSUMO [22] plasmid having an N-terminal His6SUMO-tag. In addition, we also truncated TCF7 to the high-mobility group (HMG) domain (residues 154–237) plus an initiator methionine residue by PCR amplification using the following two oligonucleotides: 5′-GGGAATTCCATATGATC AAGAAGCCCCTCAATG-3′ and 5′-CGCGGATCCTTACTTTTCCCTCGA CCG-3’. It was cloned into the pET-28His7MBP plasmid having an Nterminal His7MBP-tag followed by a TEV cleavage site. This plasmid was constructed by introducing a His7MBP-tag into the pET-28a vector using the following primers: 5′-CTAGTCTAGAAATAATTTTGTTTAACT TTAAGAAGGAGATATACC ATGGGCAGCAGCCACCACCATCATCATCA CCATACCGGGAAAATTGAAGAAGGTAAACTGG-3′ and 5′-GGGAATTC CATATGGCCCTGAAAATACAGGTTTTCACCAGAGCCTGAGCTCGAATT AGTCTGCGCGTC-3’ (Fig. S1). All positive clones were confirmed by restriction enzyme digestion and DNA sequence analysis (DNA Core Facility, University of Louisville).

Fig. 1. Overexpression of AF1q in E. coli Rosetta (DE3)PLysS cells. The expression level was tested on a 20% SDS-PAGE gel. Lanes: M, Marker proteins (Gold Biotech); 1, His6AF1q (12.6 kDa) after induction with 0.4 mM IPTG for 4 h at 37 °C; 2, Cell pellet from non-induced cultures; 3, Supernatant after cell lysis; 4, inclusion bodies solubilized in the denaturing buffer (50 mM phosphate, pH 8.0, 300 mM NaCl, 8 M urea, and 1 mM TCEP).

For affinity purification, the XK16/20 (GE Healthcare; bed volume 31 ml) empty column was packed with 15 ml of the Ni-NTA Superflow resin (Qiagen; binding capacity 50 mg/ml resin; max pressure 140 psi) and attached to a ÄKTAprime plus system (GE Healthcare). It was equilibrated with 60 ml of the wash buffer (50 mM phosphate, pH 8.0, 300 mM NaCl, 5 mM imidazole, 8 M urea, and 1 mM TCEP) at a flow rate of 1.5 ml/min. The supernatant containing His6AF1q was loaded on the column. It was washed with 60–120 ml wash buffer and the bound His6AF1q was eluted from the column with a linear gradient of 0–500 mM imidazole in the wash buffer in a total volume of 300 ml at a flow rate of 1.5 ml/min. Fractions were collected and their purity was tested by a 20% SDS-PAGE gel. Before the refolding step, the His6AF1q containing fractions were pooled and diluted to a final concentration of 0.5 mg/ml in the wash buffer. The His6AF1q was refolded by dialyzing it in a 3.5 kDa MWCO dialysis bag (Spectra/Por) against a refolding buffer (50 mM phosphate, pH 8.0, 150 mM NaCl, and 1 mM TCEP) overnight at 4 °C with constant stirring. The His6-tag was cleaved during a second overnight dialysis step at 4 °C against a fresh refolding buffer by adding TEV protease to the sample in a 1:50 M ratio of TEV protease versus sample. Once the His6-tag cleavage was confirmed by a 20% SDS-PAGE gel, the sample was loaded on the Ni-NTA column that was pre-equilibrated with the native wash buffer (50 mM phosphate, pH 8.0, 300 mM NaCl, 5 mM imidazole, and 1 mM TCEP) and the flow through having the pure AF1q was collected. To reduce the AF1q sample size for gel filtration, it was concentrated by AmiconUltra-15 centrifugal filters (3,000 MWCO) with centrifugation at 4,000×g and 4 °C. The AF1q sample was loaded on the Superdex 75 column (Hi Load 16/600, sample volume up to 5 ml, GE Healthcare) attached to an ÄKTA pure system (GE Healthcare) that was pre-equilibrated with 400 ml of 25 mM phosphate buffer of pH 7.4 containing 100 mM NaCl, 3 mM NaN3, and 1 mM TCEP. The AF1q peaks were collected and tested by Western blot analysis. The highly pure AF1q sample was concentrated by AmiconUltra-15 centrifugal filters (3,000 MWCO) with centrifugation at 4,000×g and 4 °C. To analyze the size and molecular weights of the AF1q peaks, a molecular weight marker proteins mixture (BioRad's Gel Filtration Standard, Catalog # 1511901) was loaded on the Superdex 75 column under purification conditions as mentioned above for AF1q. Likewise, for the overexpression and purification of unlabeled TCF7 (125–269) and 15N-labeled TCF7-HMG (154–237), the protocols mentioned in Refs. [22,23] were used, respectively, with the only modification being

2.2. Expression and purification of AF1q AF1q in the pET-28a plasmid was used to transform E. coli Rosetta (DE3)pLysS cells. A starter culture was prepared by inoculating 20 ml of LB media having kanamycin and chloramphenicol antibiotics with a single colony of the AF1q construct and the starter culture was incubated overnight at 37 °C with rapid shaking at 250 rpm. For the expression of unlabeled AF1q, the starter culture was added to 980 ml LB containing the required antibiotics and incubated at 37 °C with rapid shaking at 250 rpm until the OD600 reached 0.65. For the induction of AF1q expression, 0.4 mM IPTG was added, and the culture was incubated at 37 °C for an additional 4 h with constant shaking. The cells were harvested by centrifugation at 5,000×g for 20 min at 4 °C, then stored at −20 °C. The expression level was monitored with a 20% SDSPAGE gel. For the overexpression of 15N-labeled AF1q in M9 medium containing 1 g/L of 15NH4Cl, the protocol stated in Ref. [22] was used. For chromatographic purification of AF1q, 5.3 g of cell pellet was resuspended in 40 ml lysis buffer (50 mM phosphate, pH 8.0, 300 mM NaCl, 0.44 mM EDTA, 2 mM TCEP, 1 mM PMSF, and protease inhibitors cocktail) in a 50 ml falcon tube, and sonicated for 5 min on ice with an amplitude of 50% and a repeated pulse rate of 5 s pulse ON and 30 s OFF using a Branson 450 Digital Sonifier. The lysate was cleared by centrifugation at 18000×g for 30 min at 4 °C. To analyze the solubility of His6AF1q, the supernatant and pellet were loaded on a 20% SDSPAGE gel that indicated AF1q forms IBs. The IBs were washed three times with the lysis buffer containing 1% Triton X-100 to remove residual membranes and contaminants. The IBs were solubilized in 40 mL of the denaturing buffer (50 mM phosphate, pH 8.0, 300 mM NaCl, 8 M urea, and 1 mM TCEP) while stirring for 30–60 min at 4 °C and the supernatant containing His6AF1q was saved. 2

Protein Expression and Purification 165 (2020) 105499

N. Khan, et al.

Fig. 2. Affinity purification of AF1q using a Ni-NTA column. (A) His6AF1q was eluted as a single peak after applying the following linear gradient: 0–500 mM imidazole in the wash buffer in a total volume of 300 ml at a flow rate of 1.5 ml/min using an ÄKTAprime plus system (GE Healthcare). The column was washed with 60–120 ml (4–8 column volumes) wash buffer before elution of His6AF1q. (B) Purity of the affinity peak was analyzed by a 20% SDS-PAGE gel. Lanes: M, Marker proteins (Gold Biotech); 1–6, different fractions from the affinity peak of His6AF1q.

2.5. Analytical ultracentrifugation

Table 1 Summary of the recombinant AF1q purification. Step

Total Protein (mg)c

AF1q (mg)d

Purity (%)f

Yield (%)

Cell lysatea Washed IBs Solubilized material First Ni-NTA column (pooled peak) Second Ni-NTA columnb (Flow through) Superdex 75 column

390 166 135 95

137 111 96 92

35 67 71 97

100 81 70 67

44.4

44

99

30

35

(30 + 5)e

100

(22 + 4)e

Sedimentation velocity experiments were carried out in a Beckman Coulter ProteomeLab XL-A analytical ultracentrifuge (Beckman Coulter Inc., Brea, CA) at 20 °C and 40,000 rpm in standard 2-sector cells. Buffer density was determined on a Mettler/Paar Calculating Density Meter DMA 55A at 20 °C, and viscosity was measured using an Anton Parr AMVn Automated Microviscometer at 20 °C. Data were analyzed with the program Sedfit (free software: www.analyticalultracentrifugation. com) using the continuous c(s) distribution model. The partial specific volume of AF1q, TCF7 (125–269) and TCF7-HMG (154–237) were calculated from the amino acid composition using the Protparam tool in ExPASy (free software: web.expasy.org). Experimental sedimentation coefficients were corrected to s20,w using the corrections based on the measured density and viscosity.

a

From 5.3 g of wet weight E. coli cell pellet (from 1 L culture). Second Ni-NTA column purification step, after refolding and tag cleavage (flow through collected). c Protein concentration determined by Bradford assay using BSA as a standard protein. d Determined from total protein concentration and purity. e Gel filtration peak I plus peak II. f Purity determined by densitometric assessment of SDS-PAGE and Western Blot. b

2.6. NMR spectroscopy Four different 2D [1H,15N] HSQC measurements were carried out on a Bruker Avance Neo 600 MHz instrument equipped with a nitrogencooled Prodigy TCI cryoprobe: 1.) 1 mM 15N-labeled AF1q; 2.) 1 mM 15 N-labeled AF1q and 1 mM unlabeled TCF7 (125–269); 3.) 70 μM 15Nlabeled TCF7-HMG (154–237); and 4.) 70 μM 15N-labeled TCF7-HMG (154–237) and 70 μM unlabeled AF1q. The spectra for samples 1 and 2 were recorded at 298 K with 1024 and 128 complex points in the direct (t2) and indirect (t1) dimensions, respectively, with 8 and 4 scans per t1 increment for samples 1 and 2, respectively. The t1,max and t2,max were 89.1 ms and 125 ms, respectively. The spectra for samples 3 and 4 were recorded at 277 K with 1024 and 100 complex points in the t2 and t1 dimensions, respectively, with 16 scans per t1 increment. The t1,max and t2,max were 69.6 ms and 125 ms, respectively. Samples 1 and 2 were in 25 mM phosphate, pH 7.4, 100 mM NaCl, and 3 mM NaN3 with 10% D2O for the lock signal and sample 3 and 4 were in the same buffer with pH 5.5. All data were processed using NMRPipe [26] and analyzed with CARA [27].

that the His7MBP-tag was cleaved by TEV-protease. 2.3. Western-blot analysis For SDS-PAGE, 20 μg of the recombinant AF1q samples were mixed with equal volume of 2X-Laemmli buffer and heated at 95 °C for 5 min. They were resolved on a 20% SDS-PAGE after running at 100–150 V for 1 h and blotted onto a nitrocellulose membrane. The blot was probed with different antibodies including anti-AF1q monoclonal antibody from Abcam (Cat# ab109016). An enhanced chemiluminescence system (Denville) was used for developing blot. 2.4. Circular dichroism measurements of AF1q The two AF1q samples were analyzed by circular dichroism (CD) using a Jasco J-810 spectropolarimeter (Jasco, Inc, Easton, MD). Concentrations of AF1q peak I and peak II stock solutions were estimated from the A280 determined in 6 M Gdn-HCl and the amino acid composition of AF1q as described by Grimsley and Pace [24]. The protein samples for CD analysis were diluted into 50 mM sodium phosphate, pH 7.4 to give a final concentration of 4–5 μM (in terms of the monomeric sequence). CD spectra were recorded at room temperature (∼24 °C) in a quartz cuvette of 0.1 cm pathlength (Starna Cells, Atascadero, CA) in the range from 190 to 320 nm with a 1-nm slit width, a scan rate of 50 nm/min and 8-s integration time. The CD spectrum of the buffer collected under identical conditions was subtracted from that of AF1q samples. An average of five consecutive, baseline corrected CD spectra were analyzed in terms of secondary structure using the web server CAPITO–CD Analysis and Plotting Tool (http:// capito.nmr.leibniz-fli.de/index.php) [25].

3. Results and discussion 3.1. Expression and purification of AF1q The AF1q [pET-28a] construct transformed into Rosetta(DE3)pLysS cells displayed high levels of expression in both LB and 15N-labeled M9 media after inducing with 0.4 mM IPTG for 4 h at 37 °C, however, AF1q forms inclusion bodies (IBs) (Fig. 1). Interestingly, most recombinant proteins expressed in E. coli form IBs [28,29], which can be advantageous due to higher levels of purity [30] and protection from the deleterious effects of soluble proteases from E. coli [31]. Previous studies have shown that refolding proteins in IBs to their native form can be achieved by utilizing high concentrations of chaotropic agents, such as urea, for solubilizing the IBs followed by gradual dialysis to remove the chaotropic agents [28,32–34]. Similarly, we isolated the AF1q IBs, solubilized the IBs in a 3

Protein Expression and Purification 165 (2020) 105499

N. Khan, et al.

Fig. 3. Gel filtration purification, molecular weight analysis, immunodetection and sedimentation velocity analytical ultracentrifugation (SV-AUC) of AF1q. (A) AF1q was eluted as a major peak I at 70 ml retention volume from a Superdex 75 (16/600, GE Healthcare) column at a flow rate of 0.7 ml/min in the buffer 25 mM phosphate buffer, pH 8.0, 300 mM NaCl, and 1 mM TCEP using ÄKTA pure system (GE Healthcare). A minor peak II eluted at 95 ml. (B) The relative molecular masses of the AF1q peaks I and II were determined to be 199.5 KDa and 14.5 kDa respectively, with the calibration curve for standard proteins eluted on a Superdex 75 16/ 600 GL column. (C) Western blot analysis of peak I and peak II using anti-AF1q monoclonal antibodies. SV-AUC measurement of peak I (D) and peak II (E) at 280 nm. 23% of peak I runs as a monomer at 8.7 kDa and the rest forms oligomeric species.

Fig. 4. Fitted and experimental circular dichroism spectra for (A) peak I (5.7 μM) and (B) peak II (4 μM), both in 50 mM sodium phosphate, pH 7.4, using the secondary structural elements shown in Table 2.

denaturing buffer containing 8 M urea, and purified the IBs by affinity chromatography using Ni-NTA column under denaturing conditions. The His6AF1q (MW = 12.6 kDa) eluted as a single peak at 166 mM imidazole after applying a linear imidazole gradient (Fig. 2A) with a purity of 97% (Fig. 2B). Table 1 summarizes the amounts and percent yields of AF1q after each step of the purification process. In the next step, we refolded AF1q by overnight dialysis [35,36], with a gradual decrease in the urea concentration until urea was essentially removed

Table 2 Predicted and calculated secondary structures of AF1q. Protein

% helix

% β-strand

% irregular

AF1q predicted (Chou-Fasman) AF1q-peak I-from CD AF1q-peak II-from CD

29 18 7

17 16 14

54 69 74

4

Protein Expression and Purification 165 (2020) 105499

N. Khan, et al.

Fig. 5. NMR measurements reveal AF1q interacts with TCF7 (125–269). 2D [1H,15N] HSQC NMR spectra of (A) 1 mM 15N-labeled AF1q peak II and (B) 1 mM labeled AF1q peak II with 1 mM unlabeled TCF7 (125–269). A strip of the detected tryptophan peaks is shown in (B).

from the buffer. During a second overnight dialysis step at 4 °C against a fresh refolding buffer, the His6-tag was cleaved by TEV-protease [37] and the mixture was passed again through a Ni-NTA column to collect AF1q (MW = 10 kDa) in the flow through (data not shown). We further purified AF1q to homogeneity after passing through a Superdex 75 column that resulted into two peaks (Fig. 3A) with measured molecular weights of 199.5 kDa for peak I and 14.5 kDa for peak II, indicating high oligomeric and monomeric forms of AF1q, respectively (Fig. 3B), with high purity as analyzed by an immunoblot using anti-AF1q antibodies (Fig. 3C). Both AF1q samples appeared to be folded as denoted by analysis of the CD spectra shown in Fig. 4A and B. Table 2 summarizes the theoretical distribution of secondary structures calculated from the amino acid sequence of AF1q using the method of Chou and Fasman [38]. Also shown in Fig. 4A and B is the best fitting distribution of secondary structures estimated from the CD spectra using the program CAPITO [25], indicating that both peaks I and II form primarily irregular structures. Finally, sedimentation velocity analytical ultracentrifugation (SV-AUC) analysis of peak I indicated the presence of a significant amount of aggregation, while peak II behaved as a monomeric species, supporting the gel filtration results (Fig. 3D and E). The final yield for the highly pure AF1q samples (unlabeled and 15N-labeled) was 30 mg and 5 mg protein/liter E. coli culture for peak I and peak II, respectively.

15

N-

appeared coupled with a substantial decrease in the line-broadening of the signals (Fig. 5B). Though AF1q is clearly no longer exchanging between higher molecular weight species in the presence of TCF7 (125–269), the wide-ranging variation in the line-widths of the peaks suggest the presence of chemical exchange between multiple different species and/or conformational sub-states. Regardless, this spectrum indicates that AF1q and TCF7 (125–269) interact, which is also supported by our SV-AUC measurements (Fig. S3) and previously shown by co-immunoprecipitation and an electrophoretic mobility shift assay [12,21]. TCF7, similar to the lymphoid enhancer-binding factor 1 (LEF-1, PDB: 2LEF), is an HMG transcription factor that forms the TCF7/LEF-1/ β-Catenin complex in the Wnt signaling pathway to activate multiple downstream targets [12,39]. We were curious as to whether TCF7 might be interacting with AF1q through its HMG domain. Therefore, we truncated TCF7 to just the HMG domain (TCF7-HMG (154–237); Fig. S2B) and analyzed its interactions with AF1q by measuring 2D [1H,15N] HSQC NMR spectra of the 15N-labeled TCF7-HMG (154–237) (70 μM) with and without unlabeled AF1q peak I (70 μM) (Fig. S4). In the presence of AF1q peak I, we observed a significant decrease in the peak intensities for the backbone amides of the TCF7-HMG (154–237), namely a decrease that is almost an order of magnitude larger than the noise of the measurement, which indicates that the peak intensity differences are significantly larger than the uncertainty in the measurements. This decrease in peak intensity coupled with precipitation of the proteins upon addition of AF1q was also noted by SV-AUC experiments (Fig. S5), where addition of AF1q depletes the amount of free TCF7HMG (154–237) being measured during the SV-AUC experiment through precipitation. These results show that AF1q targets TCF7's HMG domain (154–237) supporting the previously proposed mechanism of enhanced transcriptional rates, especially of CD44, due to AF1q binding to the TCF7/LEF-1/β-catenin complex [12].

3.2. AF1q interacts with TCF7 as shown by nuclear magnetic resonance spectroscopy (NMR) and SV-AUC The 2D [1H,15N] HSQC NMR spectrum of 1 mM 15N-labeled AF1q (peak II) is depicted in Fig. 5A. Only a few broad peaks for the backbone amide chemical shifts can be seen in the spectrum. The spectrum indicates that AF1q peak II also forms oligomeric species when concentrated to 1 mM, as shown with AF1q peak I by gel filtration chromatography and AUC, with a significant amount of chemical exchange leading to linebroadening. Interestingly, when 1 mM 15N-labeled AF1q peak II was mixed with the unlabeled 1 mM TCF7 (125–269) (pI = 10) that we expressed and purified to homogeneity (Fig. S2A), all the expected peaks

4. Conclusion We have successfully purified human AF1q after overexpression in E. coli. Our protocol produces a high yield of pure AF1q in only two 5

Protein Expression and Purification 165 (2020) 105499

N. Khan, et al.

chromatographic purification steps and a simple refolding dialysis step. Our results indicate that AF1q forms higher order oligomeric species in the absence of a binding partner. For the first time, our NMR and SVAUC measurements clearly demonstrate that AF1q and TCF7 interact, with TCF7 (125–269) interactions disrupting AF1q oligomerization and binding of TCF7 HMG (154–237) leading to precipitation of the complex. With this robust overexpression and purification system in hand, our future studies will focus on investigating with more detail the structural basis for the interaction of AF1q and TCF7.

[14] N.N. Co, W.P. Tsang, T.W.L. Wong, H.H. Cheung, T.Y. Tsang, S.K. Kong, T.T. Kwok, Oncogene AF1q enhances doxorubicin-induced apoptosis through BAD-mediated mitochondrial apoptotic pathway, Mol. Cancer Ther. 7 (2008) 3160–3168, https:// doi.org/10.1158/1535-7163.MCT-08-0416. [15] P. Tiberio, E. Cavadini, M. Callari, M.G. Daidone, V. Appierto, AF1q: a novel mediator of basal and 4-HPR-induced apoptosis in ovarian cancer cells, PLoS One 7 (2012), https://doi.org/10.1371/journal.pone.0039968 e39968. [16] N.N. Co, W.P. Tsang, T.Y. Tsang, C.L.A. Yeung, P.L. Yau, S.K. Kong, T.T. Kwok, AF1q enhancement of gamma irradiation-induced apoptosis by up-regulation of BAD expression via NF-kappa B in human squamous carcinoma A431 cells, Oncol. Rep. 24 (2010) 547–554, https://doi.org/10.3892/or_00000891. [17] J. Shortt, R.W. Johnstone, Oncogenes in cell survival and cell death, Cold Spring Harb. Perspect. Biol. 4 (2012) a009829, https://doi.org/10.1101/cshperspect.a009829. [18] A. Parcelier, N. Maharzi, M. Delord, M. Robledo-Sarmiento, E. Nelson, H. Belakhdar-Mekid, M. Pla, K. Kuranda, V. Parietti, M. Goodhardt, N. Legrand, I.D. Bernstein, J.C. Gluckman, F. Sigaux, B. Canque, AF1q/MLLT11 regulates the emergence of human prothymocytes through cooperative interaction with the Notch signaling pathway, Blood 118 (2011) 1784–1796, https://doi.org/10.1182/ blood-2011-01-333179. [19] H.J. Lin, K.M. Shaffer, Z.R. Sun, G. Jay, W.W. He, W. Ma, AF1q, a differentially expressed gene during neuronal differentiation, transforms HEK cells into neuronlike cells, Mol. Brain Res. 131 (2004) 126–130, https://doi.org/10.1016/j. molbrainres.2004.07.022. [20] Y. Hu, Q. Sun, C. Zhang, Q. Sha, X. Sun, RE1 silencing transcription factor (REST) negatively regulates ALL1-fused from chromosome 1q (AF1q) gene transcription, BMC Mol. Biol. 16 (2015) 15, https://doi.org/10.1186/s12867-015-0043-7. [21] H. Zhang, R. Ren, J. Du, T. Sun, P. Wang, P. Kang, AF1q contributes to adriamycininduced podocyte injury by activating Wnt/beta-catenin signaling, Kidney Blood Press. Res. 42 (2017) 794–803, https://doi.org/10.1159/000484329. [22] N. Khan, D. Ban, P. Trigo-Mourino, M.G. Carneiro, M. Konrad, D. Lee, T.M. Sabo, 1 H, 13C and 15N resonance assignment of human guanylate kinase, Biomol. NMR Assign. 12 (2018) 11–14, https://doi.org/10.1007/s12104-017-9771-6. [23] R. Jain, N. Khan, A. Menzel, I. Rajkovic, M. Konrad, S. Techert, Insights into open/ closed conformations of the catalytically active human guanylate kinase as investigated by small-angle X-ray scattering, Eur. Biophys. J. 45 (2016) 81–89, https://doi.org/10.1007/s00249-015-1079-9. [24] G.R. Grimsley, C.N. Pace, Spectrophotometric determination of protein concentration (Chapter 3), Curr. Protein Pept. Sci. (2004), https://doi.org/10.1002/ 0471140864.ps0301s33 Unit 3.1–3.1.9. [25] C. Wiedemann, P. Bellstedt, M. Görlach, CAPITO–a web server-based analysis and plotting tool for circular dichroism data, Bioinformatics 29 (2013) 1750–1757, https://doi.org/10.1093/bioinformatics/btt278. [26] F. Delaglio, S. Grzesiek, G.W. Vuister, G. Zhu, J. Pfeifer, A. Bax, NMRPipe: a multidimensional spectral processing system based on UNIX pipes, J. Biomol. NMR 6 (1995) 277–293, https://doi.org/10.1007/BF00197809. [27] R.L. Keller, Optimizing the Process of Nuclear Magnetic Resonance Analysis and Computer Aided Resonance Assignment, Ph.D. thesis Institute of Molecular Biology and Biophysics, ETH, Zürich, 2004. [28] Z. Yang, L. Zhang, Y. Zhang, T. Zhang, Y. Feng, X. Lu, W. Lan, J. Wang, H. Wu, C. Cao, X. Wang, Highly efficient production of soluble proteins from insoluble inclusion bodies by a two-step-denaturing and refolding method, PLoS One 6 (2011), https://doi.org/10.1371/journal.pone.0022981 e22981. [29] I. Palmer, P.T. Wingfield, Preparation and extraction of insoluble (inclusion-body) proteins from Escherichia coli, Curr. Protein Pept. Sci. 6 (2012), https://doi.org/10. 1002/0471140864.ps0603s70 6.3.1–6.3.20. [30] A. Ramón, M. Señorale-Pose, M. Marín, ... , Inclusion bodies: not that bad, Front. Microbiol. 5 (2014) 56, https://doi.org/10.3389/fmicb.2014.00056. [31] A.K. Upadhyay, A. Murmu, A. Singh, A.K. Panda, Kinetics of inclusion body formation and its correlation with the characteristics of protein aggregates in Escherichia coli, PLoS One 7 (2012), https://doi.org/10.1371/journal.pone.0033951 e33951. [32] S.M. Singh, A.K. Panda, Solubilization and refolding of bacterial inclusion body proteins, J. Biosci. Bioeng. 99 (2005) 303–310, https://doi.org/10.1263/jbb.99. 303. [33] L.F. Vallejo, U. Rinas, Strategies for the recovery of active proteins through refolding of bacterial inclusion body proteins, Microb. Cell Factories 3 (2004) 11, https://doi.org/10.1186/1475-2859-3-11. [34] R.R. Burgess, Refolding solubilized inclusion body proteins, Methods Enzymol. 463 (2009) 259–282, https://doi.org/10.1016/S0076-6879(09)63017-2. [35] C.A. Santos, L.L. Beloti, M.A.S. Toledo, A. Crucello, M.T.P. Favaro, J.S. Mendes, A.S. Santiago, A.R. Azzoni, A.P. Souza, A novel protein refolding protocol for the solubilization and purification of recombinant peptidoglycan-associated lipoprotein from Xylella fastidiosa overexpressed in Escherichia coli, Protein Expr. Purif. 82 (2012) 284–289, https://doi.org/10.1016/j.pep.2012.01.010. [36] K. Tsumoto, D. Ejima, I. Kumagai, T. Arakawa, Practical considerations in refolding proteins from inclusion bodies, Protein Expr. Purif. 28 (2003) 1–8. [37] R.B. Kapust, J. Tözsér, T.D. Copeland, D.S. Waugh, The P1' specificity of tobacco etch virus protease, Biochem. Biophys. Res. Commun. 294 (2002) 949–955, https:// doi.org/10.1016/S0006-291X(02)00574-0. [38] P.Y. Chou, G.D. Fasman, Empirical predictions of protein conformation, Annu. Rev. Biophys. Biomol. Struct. 47 (1978) 251–276, https://doi.org/10.1146/annurev.bi. 47.070178.001343. [39] J.J. Love, X. Li, D.A. Case, K. Giese, R. Grosschedl, P.E. Wright, Structural basis for DNA bending by the architectural transcription factor LEF-1, Nature 376 (1995) 791–795, https://doi.org/10.1038/376791a0.

Acknowledgements We would like to thank David Ban, Mark Vincent Carreon dela Cerna, William Holmes and Vilius Stribinskis for their contributions toward the development of the AF1q expression and purification protocol. This research was supported by the James Graham Brown Cancer Center. Appendix A. Supplementary data Supplementary data to this article can be found online at https:// doi.org/10.1016/j.pep.2019.105499. References [1] W. Tse, W. Zhu, H.S. Chen, A. Cohen, A novel gene, AF1q, fused to MLL in t(1;11) (q21;q23), is specifically expressed in leukemic and immature hematopoietic cells, Blood 85 (1995) 650–656. [2] M. Busson-Le Coniat, F. Salomon-Nguyen, J. Hillion, O.A. Bernard, R. Berger, MLLAF1q fusion resulting from t(1;11) in acute leukemia, Leukemia 13 (1999) 302–306. [3] W. Li, M. Ji, F. Lu, Y. Pang, X. Dong, J. Zhang, L. Peng, J. Ye, S. Zang, D. Ma, C. Ji, Novel AF1q/MLLT11 favorably affects imatinib resistance and cell survival in chronic myeloid leukemia, Cell Death Dis. 9 (2018) 855, https://doi.org/10.1038/ s41419-018-0900-7. [4] W. Tse, S. Meshinchi, T.A. Alonzo, D.L. Stirewalt, R.B. Gerbing, W.G. Woods, F.R. Appelbaum, J.P. Radich, Elevated expression of the AF1q gene, an MLL fusion partner, is an independent adverse prognostic factor in pediatric acute myeloid leukemia, Blood 104 (2004) 3058–3063, https://doi.org/10.1182/blood-2003-124347. [5] C.J. Strunk, U. Platzbecker, C. Thiede, M. Schaich, T. Illmer, Z. Kang, P. Leahy, C. Li, X. Xie, M.J. Laughlin, H.M. Lazarus, S.L. Gerson, K.D. Bunting, G. Ehninger, W. Tse, Elevated AF1q expression is a poor prognostic marker for adult acute myeloid leukemia patients with normal cytogenetics, Am. J. Hematol. 84 (2009) 308–309, https://doi.org/10.1002/ajh.21396. [6] X.-Z. Chang, D.-Q. Li, Y.-F. Hou, J. Wu, J.-S. Lu, G.-H. Di, W. Jin, Z.-L. Ou, Z.Z. Shen, Z.-M. Shao, Identification of the functional role of AF1Q in the progression of breast cancer, Breast Canc. Res. Treat. 111 (2008) 65–78, https://doi.org/10. 1007/s10549-007-9761-y. [7] H. Jin, W. Sun, Y. Zhang, H. Yan, H. Liufu, S. Wang, C. Chen, J. Gu, X. Hua, L. Zhou, G. Jiang, D. Rao, Q. Xie, H. Huang, C. Huang, MicroRNA-411 downregulation enhances tumor growth by upregulating MLLT11 expression in human bladder cancer, Mol. Ther. Nucleic Acids 11 (2018) 312–322, https://doi.org/10.1016/j.omtn. 2018.03.003. [8] J. Hu, G. Li, L. Liu, Y. Wang, X. Li, J. Gong, AF1q mediates tumor progression in colorectal cancer by regulating AKT signaling, Int. J. Mol. Sci. 18 (2017) 987, https://doi.org/10.3390/ijms18050987. [9] P. Tiberio, L. Lozneanu, V. Angeloni, E. Cavadini, P. Pinciroli, M. Callari, M.L. Carcangiu, D. Lorusso, F. Raspagliesi, V. Pala, M.G. Daidone, V. Appierto, Involvement of AF1q/MLLT11 in the progression of ovarian cancer, Oncotarget 8 (2017) 23246–23264, https://doi.org/10.18632/oncotarget.15574. [10] W. Tse, H.J. Deeg, D. Stirewalt, F.R. Appelbaum, J. Radich, T. Gooley, Increased AF1q gene expression in high-risk myelodysplastic syndrome, Br. J. Haematol. 128 (2005) 218–220, https://doi.org/10.1111/j.1365-2141.05306.x. [11] C. Jacques, O. Baris, D. Prunier-Mirebeau, F. Savagner, P. Rodien, V. Rohmer, B. Franc, S. Guyetant, Y. Malthiery, P. Reynier, Two-step differential expression analysis reveals a new set of genes involved in thyroid oncocytic tumors, J. Clin. Endocrinol. Metab. 90 (2005) 2314–2320, https://doi.org/10.1210/jc.2004-1337. [12] J. Park, M. Schlederer, M. Schreiber, R. Ice, O. Merkel, M. Bilban, S. Hofbauer, S. Kim, J. Addison, J. Zou, C. Ji, S.T. Bunting, Z. Wang, M. Shoham, G. Huang, Z. Bago-Horvath, L.F. Gibson, Y. Rojanasakul, S. Remick, A. Ivanov, E. Pugacheva, K.D. Bunting, R. Moriggl, L. Kenner, W. Tse, AF1q is a novel TCF7 co-factor which activates CD44 and promotes breast cancer metastasis, Oncotarget 6 (2015) 20697–20710, https://doi.org/10.18632/oncotarget.4136. [13] J. Park, S. Kim, J. Joh, S.C. Remick, D.M. Miller, J. Yan, Z. Kanaan, J.-H. Chao, M.M. Krem, S.K. Basu, S. Hagiwara, L. Kenner, R. Moriggl, K.D. Bunting, W. Tse, MLLT11/AF1q boosts oncogenic STAT3 activity through Src-PDGFR tyrosine kinase signaling, Oncotarget 7 (2016) 43960–43973, https://doi.org/10.18632/ oncotarget.9759.

6