TPP3

TPP3 (Tomato Pistil Predominant 3 defensin) is a class II plant defensin from Lycopersicon esculentum (tomato), belonging to the Solanaceae family of cysteine-rich cationic antimicrobial peptides (CAPs). It adopts the conserved cysteine-stabilized αβ (CSαβ) motif stabilized by four disulfide bonds and forms a characteristic dimeric 'cationic grip' conformation that selectively accommodates the head group of phosphatidylinositol (4,5)-bisphosphate (PIP2) to mediate membrane permeabilization.

TPP3 (Tomato Defensin) – Scientific Classification, Structural Identity, and Procurement-Relevant Background

TPP3 (Tomato Pistil Predominant 3 defensin) is a class II plant defensin from Lycopersicon esculentum (tomato), belonging to the Solanaceae family of cysteine-rich cationic antimicrobial peptides (CAPs). It adopts the conserved cysteine-stabilized αβ (CSαβ) motif stabilized by four disulfide bonds and forms a characteristic dimeric 'cationic grip' conformation that selectively accommodates the head group of phosphatidylinositol (4,5)-bisphosphate (PIP2) to mediate membrane permeabilization [1]. TPP3 shares 63% amino acid sequence identity with the well-characterized tobacco defensin NaD1, yet exhibits a fundamentally distinct lipid binding profile [1]. Its crystal structure has been solved at 1.7 Å resolution (PDB ID: 4UJ0), providing atomic-level resolution of the dimer interface and PIP2 binding pocket [2]. Recombinant TPP3 is expressible in the Pichia pastoris yeast system, enabling scalable production for research and potential agricultural biocontrol applications [3].

• [1] Baxter AA, Richter V, Lay FT, Poon IKH, Adda CG, Veneer PK, Phan TK, Bleackley MR, Anderson MA, Kvansakul M, Hulett MD. The Tomato Defensin TPP3 Binds Phosphatidylinositol (4,5)-Bisphosphate via a Conserved Dimeric Cationic Grip Conformation To Mediate Cell Lysis. Mol Cell Biol. 2015 Jun;35(11):1964–1978. doi:10.1128/MCB.00282-15. View Source
• [2] RCSB PDB. 4UJ0: Crystal structure of the tomato defensin TPP3. Deposited 2015-04-07. doi:10.2210/pdb4UJ0/pdb. View Source
• [3] Payne JAE, Bleackley MR, Lee T-H, Shafee TMA, Poon IKH, Hulett MD, Aguilar M-I, Anderson MA. Recombinant expression and purification of the tomato defensin TPP3 and its preliminary X-ray crystallographic analysis. Acta Crystallogr Sect F Struct Biol Cryst Commun. 2012 Mar 1;68(Pt 3):366–368. doi:10.1107/S1744309112001510. View Source

Why TPP3 Cannot Be Interchanged with NaD1 or Other Class II Solanaceous Defensins – Critical Lipid Specificity Differences

Despite sharing 63% sequence identity and a conserved cationic grip dimer architecture with the tobacco defensin NaD1, TPP3 exhibits a fundamentally distinct phospholipid binding profile that renders functional substitution invalid. NaD1 binds to a broad range of membrane phosphatidylinositol phosphates (including PI(3)P, PI(4)P, PI(5)P, PI(3,5)P2, and PI(4,5)P2), whereas TPP3 is uniquely specific for PIP2, with negligible binding to any other phosphoinositide species [1]. This differential lipid specificity translates to divergent cytolytic profiles: at an equimolar concentration of 10 μM, TPP3 induces higher membrane permeabilization (69.6% PI-positive U937 cells) than NaD1 (62.3%), yet lower LDH release (35% vs. 65%), indicating a mechanistically distinct mode of membrane disruption [1]. Furthermore, key residues in the PIP2 binding pocket are not functionally equivalent: the TPP3 K6A mutation abrogates PIP2 binding and oligomerization, whereas the corresponding NaD1 K4A mutation disrupts dimer formation itself, revealing divergent structural requirements for activity [1]. Selecting NaD1, NsD7, or other in-class defensins without accounting for these quantitative functional differences will yield non-equivalent experimental outcomes.

• [1] Baxter AA, Richter V, Lay FT, Poon IKH, Adda CG, Veneer PK, Phan TK, Bleackley MR, Anderson MA, Kvansakul M, Hulett MD. The Tomato Defensin TPP3 Binds Phosphatidylinositol (4,5)-Bisphosphate via a Conserved Dimeric Cationic Grip Conformation To Mediate Cell Lysis. Mol Cell Biol. 2015 Jun;35(11):1964–1978. doi:10.1128/MCB.00282-15. View Source

TPP3 Quantitative Differentiation Evidence: Head-to-Head Comparison Data vs. NaD1 and Other Class II Defensins

TPP3 vs. NaD1 – Exclusive PIP2 Lipid Binding Specificity (Protein-Lipid Overlay Assay)

TPP3 displays an exclusive binding profile restricted to phosphatidylinositol (4,5)-bisphosphate (PIP2), as demonstrated by PIP Strip protein-lipid overlay assays. Quantitative densitometry analysis of three independent PIP Strips, normalized for PIP2 binding, revealed that TPP3 has negligible binding to all other phosphoinositides tested, including PI(3)P, PI(4)P, PI(5)P, PI(3,5)P2, and PI(3,4,5)P3 [1]. In contrast, NaD1 probed on identical PIP Strips exhibited robust binding to a broad range of phosphatidylinositol phosphates [1]. This unique specificity for PIP2 distinguishes TPP3 from all other characterized solanaceous class II defensins [2].

Evidence Dimension	Phosphoinositide binding specificity (PIP Strip densitometry)
Target Compound Data	TPP3: Specific binding to PIP2 only; negligible binding to PI(3)P, PI(4)P, PI(5)P, PI(3,5)P2, PI(3,4,5)P3
Comparator Or Baseline	NaD1: Broad binding to multiple phosphoinositide species including PI(3)P, PI(4)P, PI(5)P, PI(3,5)P2, and PI(4,5)P2
Quantified Difference	Qualitatively distinct: TPP3 exhibits exclusive PIP2 specificity vs. NaD1's promiscuous phosphoinositide binding
Conditions	In vitro PIP Strip and Membrane Lipid Strip overlay assays; 1 μg/mL defensin; densitometry normalized to PIP2 signal; three independent experiments [1]

Why This Matters

For researchers requiring selective PIP2 targeting without confounding off-target lipid interactions, TPP3 provides a uniquely specific tool that NaD1 cannot substitute.

• [1] Baxter AA, Richter V, Lay FT, Poon IKH, Adda CG, Veneer PK, Phan TK, Bleackley MR, Anderson MA, Kvansakul M, Hulett MD. The Tomato Defensin TPP3 Binds Phosphatidylinositol (4,5)-Bisphosphate via a Conserved Dimeric Cationic Grip Conformation To Mediate Cell Lysis. Mol Cell Biol. 2015 Jun;35(11):1964–1978. doi:10.1128/MCB.00282-15. (Figure 1B–D). View Source
• [2] Poon IKH, Baxter AA, Lay FT, Mills GD, Adda CG, Payne JAE, Phan TK, Ryan GF, White JA, Veneer PK, van der Weerden NL, Anderson MA, Kvansakul M, Hulett MD. The lure of the lipids: how defensins exploit membrane phospholipids to induce cytolysis in target cells. Cell Death Dis. 2017 Mar 30;8(3):e2712. doi:10.1038/cddis.2017.69. View Source

TPP3 vs. NaD1 – Differential Membrane Permeabilization Efficacy in U937 Tumor Cells (PI Uptake Assay)

In a direct head-to-head comparison using flow cytometry-based propidium iodide (PI) uptake assays on human myeloid leukemia U937 cells, TPP3 at 10 μM induced 69.6% PI-positive cells (membrane permeabilization) versus 62.3% for NaD1 at the identical 10 μM concentration after 30 min incubation at 37°C [1]. TPP3 exhibited a clear concentration-dependent response, with PI-positive cell percentages decreasing at lower concentrations (5, 2.5, and 1.25 μM) [1]. The superior permeabilization efficacy of TPP3 is consistent with its distinct PIP2-specific binding mechanism, which may enable more efficient higher-order oligomerization at the target membrane [1].

Evidence Dimension	Membrane permeabilization (% PI-positive U937 cells)
Target Compound Data	TPP3 at 10 μM: 69.6% PI-positive cells (SEM from 3 independent experiments)
Comparator Or Baseline	NaD1 at 10 μM: 62.3% PI-positive cells (positive control); PBS: baseline (negligible PI uptake)
Quantified Difference	TPP3 achieves 7.3 percentage points higher membrane permeabilization than NaD1 at equimolar 10 μM concentration
Conditions	U937 human myeloid leukemia cells; 30 min incubation at 37°C; flow cytometry-based PI uptake; 10, 5, 2.5, and 1.25 μM TPP3 tested; NaD1 at 10 μM as positive control [1]

Why This Matters

For applications requiring maximal tumor cell membrane disruption at a given defensin concentration, TPP3 provides measurably superior potency compared to NaD1, enabling lower dosing requirements.

• [1] Baxter AA, Richter V, Lay FT, Poon IKH, Adda CG, Veneer PK, Phan TK, Bleackley MR, Anderson MA, Kvansakul M, Hulett MD. The Tomato Defensin TPP3 Binds Phosphatidylinositol (4,5)-Bisphosphate via a Conserved Dimeric Cationic Grip Conformation To Mediate Cell Lysis. Mol Cell Biol. 2015 Jun;35(11):1964–1978. doi:10.1128/MCB.00282-15. (Figure 4A). View Source

TPP3 vs. NaD1 – Distinct Cytotoxicity Profile Revealed by Differential LDH Release in U937 Cells

Despite inducing greater membrane permeabilization to small molecules (PI uptake), TPP3 at 10 μM triggered only 35% LDH release from U937 cells, compared with 65% LDH release induced by NaD1 at 10 μM [1]. This nearly twofold lower release of the large (≤140 kDa) cytoplasmic enzyme lactate dehydrogenase suggests that TPP3 creates membrane lesions with fundamentally different size characteristics than NaD1 [1]. This functional divergence—higher small-molecule permeability yet lower macromolecular leakage—is consistent with TPP3's exclusive PIP2-dependent oligomerization forming structurally distinct membrane pores or defects compared to NaD1's broad-lipid-mediated oligomeric assemblies [1].

Evidence Dimension	Cytotoxicity (% LDH release from U937 cells)
Target Compound Data	TPP3 at 10 μM: 35% cytotoxicity (LDH release)
Comparator Or Baseline	NaD1 at 10 μM: 65% cytotoxicity (LDH release); cell lysis buffer: 100% lysis control
Quantified Difference	NaD1 induces 1.86-fold higher LDH release than TPP3 at equimolar 10 μM concentration, despite TPP3 inducing higher PI uptake
Conditions	U937 human myeloid leukemia cells; LDH cytotoxicity assay kit II (Abcam); 10, 5, and 2.5 μM TPP3 tested; SEM from 3 independent experiments [1]

Why This Matters

The divergent PI uptake vs. LDH release profiles mean TPP3 and NaD1 are not functionally interchangeable for studies of membrane lesion architecture; selection of TPP3 enables investigation of PIP2-specific, size-restricted membrane permeabilization.

• [1] Baxter AA, Richter V, Lay FT, Poon IKH, Adda CG, Veneer PK, Phan TK, Bleackley MR, Anderson MA, Kvansakul M, Hulett MD. The Tomato Defensin TPP3 Binds Phosphatidylinositol (4,5)-Bisphosphate via a Conserved Dimeric Cationic Grip Conformation To Mediate Cell Lysis. Mol Cell Biol. 2015 Jun;35(11):1964–1978. doi:10.1128/MCB.00282-15. (Figure 4A and 4C). View Source

TPP3 vs. NaD1 – Structural Divergence at Atomic Resolution in the Cationic Grip Dimer

The crystal structure of TPP3 was determined at 1.7 Å resolution (PDB ID: 4UJ0), revealing a dimeric cationic grip conformation that superimposes with the NaD1-PIP2 dimer with a root mean square deviation (RMSD) of only 1.2 Å over 93 Cα atoms (Z-score 17.1 by Dali-Lite analysis) [1]. Despite this overall structural conservation, critical residue-level differences exist in the lipid binding pocket: TPP3 residues K6, H35, and K42 correspond to NaD1 residues K4, H33, and R40, respectively [1]. Crucially, the TPP3 K6A mutation abrogates PIP2 binding and higher-order oligomerization without disrupting dimer formation, whereas the equivalent NaD1 K4A mutation disrupts dimerization itself [1][2]. This demonstrates that while the dimer architecture is conserved, the structural determinants of PIP2 binding and oligomerization are mechanistically distinct between TPP3 and NaD1 [1].

Evidence Dimension	Tertiary/quaternary structure comparison (X-ray crystallography)
Target Compound Data	TPP3: 1.7 Å resolution structure; dimer in cationic grip; K6 required for PIP2 binding/oligomerization but NOT dimer formation
Comparator Or Baseline	NaD1: Previously solved structure (PDB 4AAZ); dimer in cationic grip; K4 required for BOTH dimer formation and PIP2 binding
Quantified Difference	RMSD 1.2 Å over 93 Cα atoms (Z-score 17.1); TPP3 K6A: dimer intact but PIP2 binding abolished vs. NaD1 K4A: dimer disrupted
Conditions	X-ray crystallography; TPP3 crystals obtained using Pichia pastoris-expressed recombinant protein; molecular replacement using NaD1 monomer (PDB 4AAZ) as search model; refinement to Rwork 18.66%, Rfree 20.79% [1]

Why This Matters

For structure-guided engineering of defensin variants with tailored lipid specificity or oligomerization properties, TPP3 provides a distinct structural template with non-redundant residue-function relationships compared to NaD1.

• [1] Baxter AA, Richter V, Lay FT, Poon IKH, Adda CG, Veneer PK, Phan TK, Bleackley MR, Anderson MA, Kvansakul M, Hulett MD. The Tomato Defensin TPP3 Binds Phosphatidylinositol (4,5)-Bisphosphate via a Conserved Dimeric Cationic Grip Conformation To Mediate Cell Lysis. Mol Cell Biol. 2015 Jun;35(11):1964–1978. doi:10.1128/MCB.00282-15. (Figures 1A, 2, 6). View Source
• [2] Lay FT, Mills GD, Poon IKH, Cowieson NP, Kirby N, Baxter AA, van der Weerden NL, Dogovski C, Perugini MA, Anderson MA, Kvansakul M, Hulett MD. Dimerization of Plant Defensin NaD1 Enhances Its Antifungal Activity. J Biol Chem. 2012 Jun 8;287(24):19961–19972. doi:10.1074/jbc.M111.331132. View Source

TPP3 Antifungal Activity Is PIP2-Dependent – K6A Mutant Shows ~3-Fold Reduction Against Fusarium graminearum

The functional requirement of PIP2 binding for TPP3's antifungal activity was demonstrated through site-directed mutagenesis of the critical lipid-binding residue K6. The TPP3(K6A) mutant, which retains dimer formation but loses PIP2 binding and higher-order oligomerization, exhibited an approximately 3-fold reduction in the ability to inhibit hyphal growth of the filamentous fungus Fusarium graminearum compared with wild-type TPP3 [1]. This contrasts with the antifungal mechanism of other class II defensins such as NaD1, where the K4A mutation (equivalent position) disrupts the dimer itself rather than specifically abrogating lipid binding [1][2]. The PIP2-dependent antifungal activity confirms that TPP3's mechanism is inseparable from its unique lipid specificity, and that mutations differentially affect TPP3 vs. NaD1 antifungal function.

Evidence Dimension	Antifungal hyphal growth inhibition (Fusarium graminearum)
Target Compound Data	Wild-type TPP3: Full antifungal activity; TPP3(K6A) mutant: ~3-fold reduction in hyphal growth inhibition
Comparator Or Baseline	NaD1(K4A) mutant: Loss of dimer formation and antifungal activity (distinct mutational phenotype at equivalent residue position)
Quantified Difference	Approximately 3-fold reduction in antifungal activity for TPP3(K6A) vs. wild-type TPP3; distinct mutational phenotypes at equivalent residues between TPP3 and NaD1
Conditions	Fusarium graminearum hyphal growth inhibition assay; comparison of wild-type TPP3 vs. TPP3(K6A) point mutant; data from Figure 7 [1]

Why This Matters

For agricultural biocontrol applications targeting Fusarium head blight and other crop diseases, TPP3's PIP2-dependent antifungal mechanism provides a defined, genetically tractable mode of action that is distinct from NaD1 and other class II defensins.

• [1] Baxter AA, Richter V, Lay FT, Poon IKH, Adda CG, Veneer PK, Phan TK, Bleackley MR, Anderson MA, Kvansakul M, Hulett MD. The Tomato Defensin TPP3 Binds Phosphatidylinositol (4,5)-Bisphosphate via a Conserved Dimeric Cationic Grip Conformation To Mediate Cell Lysis. Mol Cell Biol. 2015 Jun;35(11):1964–1978. doi:10.1128/MCB.00282-15. (Figures 6, 7). View Source
• [2] Lay FT, Mills GD, Poon IKH, Cowieson NP, Kirby N, Baxter AA, van der Weerden NL, Dogovski C, Perugini MA, Anderson MA, Kvansakul M, Hulett MD. Dimerization of Plant Defensin NaD1 Enhances Its Antifungal Activity. J Biol Chem. 2012 Jun 8;287(24):19961–19972. doi:10.1074/jbc.M111.331132. View Source

TPP3 Membrane Permeabilization Is Functionally Dependent on Inner Leaflet PIP2 Accessibility – 2-Fold Kinetic Delay with PIP2 Sequestration

The functional dependence of TPP3 on accessible plasma membrane PIP2 was demonstrated in live-cell confocal imaging experiments using HeLa cells. Cells overexpressing GFP-PH(PLCδ1), a high-affinity PIP2 biosensor that sequesters inner leaflet PIP2, exhibited an average time to membrane permeabilization of 1,210 seconds following TPP3 addition—a 2.1-fold delay compared with 570 seconds for free-GFP-expressing control cells [1]. Furthermore, pre-treatment of U937 cells with neomycin (a PIP2-sequestering aminoglycoside) produced concentration-dependent inhibition of TPP3-mediated membrane permeabilization, whereas the amphipathic peptide LL-37 (which acts via a PIP2-independent pore-forming mechanism) was unaffected by neomycin pre-treatment [1]. This demonstrates that TPP3's cytolytic activity is specifically and obligately dependent on PIP2 availability at the target membrane inner leaflet.

Evidence Dimension	Kinetics of membrane permeabilization with/without PIP2 sequestration
Target Compound Data	TPP3 with GFP-PH (PIP2 sequestered): 1,210 s to permeabilization; TPP3 with free GFP (PIP2 available): 570 s to permeabilization
Comparator Or Baseline	Control: LL-37 (PIP2-independent pore-forming peptide) – no inhibition by neomycin pre-treatment; NaD1 activity also inhibited by neomycin-mediated PIP2 sequestration [1]
Quantified Difference	2.1-fold delay in TPP3-mediated membrane permeabilization when PIP2 is sequestered by GFP-PH(PLCδ1) overexpression
Conditions	HeLa cells transfected with GFP-PH(PLCδ1) or free GFP; CLSM live-cell imaging over 30 min (1,800 s) time course; 10 μM TPP3; neomycin (0–20 mM) pre-treatment in U937 cells with flow cytometry PI uptake readout [1]

Why This Matters

For studies investigating PIP2-dependent membrane permeabilization mechanisms or targeting PIP2-enriched cancer cells, TPP3 provides a validated, PIP2-dependent tool with defined kinetic parameters that allow quantitative assessment of target membrane PIP2 status.

• [1] Baxter AA, Richter V, Lay FT, Poon IKH, Adda CG, Veneer PK, Phan TK, Bleackley MR, Anderson MA, Kvansakul M, Hulett MD. The Tomato Defensin TPP3 Binds Phosphatidylinositol (4,5)-Bisphosphate via a Conserved Dimeric Cationic Grip Conformation To Mediate Cell Lysis. Mol Cell Biol. 2015 Jun;35(11):1964–1978. doi:10.1128/MCB.00282-15. (Figure 5C–E). View Source

Optimal Application Scenarios for TPP3 Based on Quantitatively Demonstrated Differentiation

PIP2-Specific Membrane Targeting in Fundamental Lipid Biology Research

TPP3 is the defensin of choice when experimental protocols demand exclusive, high-specificity targeting of phosphatidylinositol (4,5)-bisphosphate (PIP2) without cross-reactivity with other phosphoinositide species. As demonstrated by PIP Strip densitometry, TPP3 exhibits negligible binding to PI(3)P, PI(4)P, PI(5)P, PI(3,5)P2, or PI(3,4,5)P3, unlike NaD1 which binds broadly across the phosphoinositide panel [1]. This exclusivity makes TPP3 a superior molecular probe for dissecting PIP2-specific signaling pathways, membrane trafficking events, or phospholipid distribution studies where confounding off-target lipid interactions must be stringently excluded. The availability of a high-resolution crystal structure (1.7 Å, PDB 4UJ0) further enables rational design of labeled or modified TPP3 variants for imaging applications [2].

PIP2-Dependent Antifungal Biocontrol Agent Development for Fusarium Head Blight and Related Crop Diseases

For agricultural biotechnology programs developing protein-based biofungicides against Fusarium graminearum and related phytopathogens, TPP3 offers a genetically validated, PIP2-dependent antifungal mechanism. The ~3-fold reduction in hyphal growth inhibition observed with the TPP3(K6A) mutant confirms that PIP2 binding is mechanistically coupled to antifungal activity [1]. This defined structure-activity relationship supports rational protein engineering strategies—such as enhancing PIP2 binding affinity or oligomerization efficiency—that are not equally applicable to NaD1 given its distinct mutational phenotype at the equivalent residue position (K4A disrupts dimerization rather than specifically abrogating lipid binding) [1][3]. Furthermore, recombinant expression in the Pichia pastoris system enables scalable, cost-effective production for field application development [4].

Tumor Cell Membrane Permeabilization Studies Requiring Defined Lesion Architecture

In cancer research contexts where the mechanism of membrane lesion formation must be controlled experimentally, TPP3 provides a distinct functional profile: high small-molecule permeability (69.6% PI uptake at 10 μM) combined with relatively restricted macromolecular leakage (35% LDH release at 10 μM) [1]. This contrasts sharply with NaD1, which generates lower PI uptake (62.3%) but substantially higher LDH release (65%) at the same concentration [1]. This differential indicates that TPP3 forms membrane lesions with fundamentally different pore dimensions or structural characteristics than NaD1, making TPP3 the preferred candidate for studies specifically investigating PIP2-restricted, oligomerization-dependent membrane permeabilization with controlled molecular weight cut-off properties.

Structure-Guided Defensin Engineering Using a Validated Dimer Template with Non-Redundant Mutational Phenotypes

For protein engineering programs aiming to develop defensin variants with customized lipid specificity, oligomerization properties, or cytotoxicity profiles, TPP3 provides a structurally validated engineering scaffold at 1.7 Å resolution (PDB 4UJ0) [2]. The demonstrated functional divergence between TPP3 residue K6 (required for PIP2 binding but dispensable for dimer integrity) and the equivalent NaD1 residue K4 (required for dimerization itself) means that TPP3 and NaD1 offer non-redundant mutational landscapes for engineering [1]. Engineering efforts targeting enhanced antifungal potency, altered lipid specificity, or reduced mammalian cell toxicity can exploit TPP3's unique structural features that are not accessible using NaD1 as a starting template.

• [1] Baxter AA, Richter V, Lay FT, Poon IKH, Adda CG, Veneer PK, Phan TK, Bleackley MR, Anderson MA, Kvansakul M, Hulett MD. The Tomato Defensin TPP3 Binds Phosphatidylinositol (4,5)-Bisphosphate via a Conserved Dimeric Cationic Grip Conformation To Mediate Cell Lysis. Mol Cell Biol. 2015 Jun;35(11):1964–1978. doi:10.1128/MCB.00282-15. View Source
• [2] RCSB PDB. 4UJ0: Crystal structure of the tomato defensin TPP3. Deposited 2015-04-07. doi:10.2210/pdb4UJ0/pdb. View Source
• [3] Lay FT, Mills GD, Poon IKH, Cowieson NP, Kirby N, Baxter AA, van der Weerden NL, Dogovski C, Perugini MA, Anderson MA, Kvansakul M, Hulett MD. Dimerization of Plant Defensin NaD1 Enhances Its Antifungal Activity. J Biol Chem. 2012 Jun 8;287(24):19961–19972. doi:10.1074/jbc.M111.331132. View Source
• [4] Payne JAE, Bleackley MR, Lee T-H, Shafee TMA, Poon IKH, Hulett MD, Aguilar M-I, Anderson MA. Recombinant expression and purification of the tomato defensin TPP3 and its preliminary X-ray crystallographic analysis. Acta Crystallogr Sect F Struct Biol Cryst Commun. 2012 Mar 1;68(Pt 3):366–368. doi:10.1107/S1744309112001510. View Source