Solution Nuclear Magnetic Resonance Structure of Membrane-Integral Diacylglycerol Kinase

See allHide authors and affiliations

Science  26 Jun 2009:
Vol. 324, Issue 5935, pp. 1726-1729
DOI: 10.1126/science.1171716

Opening the Portico

Escherichia coli diacylglycerol kinase (DAGK) represents a family of integral membrane phosphotransferases that function in prokaryotic-specific metabolic pathways. Van Horn et al. (p. 1726) determined the structure of the 40-kilodalton functional homotrimer of E. coli DAGK by solution nuclear magnetic resonance spectroscopy. Each monomer comprises three transmembrane helices. The third transmembrane helix from each subunit is domain-swapped to pack against the first and second transmembrane helices from an adjacent subunit. These three helices frame a portico-like membrane-submerged cavity that contains residues critical for activity in close proximity to residues critical for folding. The structure provides insight into the determinants of lipid substrate specificity and phosphotransferase activity.


Escherichia coli diacylglycerol kinase (DAGK) represents a family of integral membrane enzymes that is unrelated to all other phosphotransferases. We have determined the three-dimensional structure of the DAGK homotrimer with the use of solution nuclear magnetic resonance. The third transmembrane helix from each subunit is domain-swapped with the first and second transmembrane segments from an adjacent subunit. Each of DAGK’s three active sites resembles a portico. The cornice of the portico appears to be the determinant of DAGK’s lipid substrate specificity and overhangs the site of phosphoryl transfer near the water-membrane interface. Mutations to cysteine that caused severe misfolding were located in or near the active site, indicating a high degree of overlap between sites responsible for folding and for catalysis.

Escherichia coli diacylglycerol kinase (DAGK) is encoded by the dgkA gene and catalyzes the direct phosphorylation of diacylglycerol (DAG) by Mg(II)–adenosine triphosphate (MgATP) to form phosphatidic acid as part of the membrane-derived oligosaccharide (MDO) cycle (13). In Gram-positive organisms, the dgkA homolog encodes an undecaprenol kinase, indicating a role in oligosaccharide assembly or in related signaling pathways (4). The DAGK homolog in Streptococcus mutans is known to be a virulence factor for smooth-surface dental caries (5). DAGK was among the first integral membrane enzymes to be solubilized, purified, and mechanistically characterized (6). The wild-type protein is very stable (7, 8) and can spontaneously insert into lipid bilayers to adopt its functional fold (9, 10). Paradoxically, DAGK resembles many disease-linked human membrane proteins because it is highly susceptible to mutation-induced misfolding (1012). DAGK functions as a 40-kD homotrimer, with a total of nine transmembrane (TM) helices and three active sites (13).

The structure of DAGK was determined using solution nuclear magnetic resonance (NMR) methods (14) under conditions in which the enzyme is a functional homotrimer solubilized in ~100-kD dodecylphosphocholine micelles; this work extends other solution NMR studies of >20-kD multispan membrane proteins (1522). The backbone structure of the helical TM domain of DAGK (residues 26 to 121) was precisely determined by the data (Fig. 1, fig. S1, and table S4); however, motions associated with the N terminus (residues 1 to 25) have hindered determination of its conformation beyond confirming the presence of two stable amphipathic helices.

Fig. 1

Structure of diacylglycerol kinase. All panels represent the same conformer from the ensemble of structures (fig. S1). Omitted from this figure is the N terminus (residues 1 to 25), which was not precisely determined, although it is known to comprise two short α helices that extend over residues 6 to 14 and 17 to 23. (A) Ribbon diagram of DAGK with inscribed portico viewed from the membrane plane. DAGK’s five most highly conserved residues are shown (for a single portico only). (B) Ribbon diagram viewed from the membrane plane, showing a close-up of the region of the active site containing DAGK’s most highly conserved residues (35). (C) Ribbon diagram viewed from the cytosol, showing side chains for DAGK’s five most highly conserved residues (for a single portico only). The connecting loops between the first and second TM segments have been omitted from view so that the organization of the TM helices can more easily be discerned.

DAGK’s structure bears no resemblance to that of the water-soluble DAGK (23) (Fig. 1). The three-fold symmetry axis lies at the center of a parallel left-handed bundle formed by the second transmembrane (TM2) helices of the three subunits. TM2 has previously been proposed to play a central role in DAGK’s folding and stability (24, 25) and contains several highly conserved residues (Fig. 1), particularly near the membrane-cytoplasm interface. Characterization of a series of cysteine-replacement mutants for sites in TM2 showed that mutations in this segment often resulted in a marked reduction in catalytic function (Fig. 2 and table S1), underscoring the importance of TM2 to DAGK folding and catalysis.

Fig. 2

Identification of sites critical for catalysis and/or for folding, based on the functional analysis of mutants generated by systematically replacing each residue in a cysteineless mutant form of DAGK (7) with cysteine (35). (A) Sites labeled blue designate cysteine mutants that exhibited at least 20% of wild-type activity. Yellow labels indicate sites for which mutants were observed to fold but exhibited less than 20% catalytic activity. Mutation to cysteine at the pink sites resulted in forms of DAGK that were both inactive and misfolded, both in native E. coli membranes and after purification and application of normally effective refolding protocols. (B) Surface-filled representation of DAGK. Highlighted residues are for a single portico site only (residues are highlighted only for TM1, TM3, and TM2′). Omitted from the surface representations are the imprecisely determined N termini (residues 1 to 25). Specific activities for each mutant are listed in table S1. Details of the folding behavior of mutants that exhibit <15% of wild-type activity are given in table S2. Sites that are indicated as being conserved with >95% identity were identified on the basis of multiple sequence alignment (figs. S2 and S3).

DAGK is seen to be a domain-swapped homotrimer (Fig. 1C). TM3 makes virtually no contact with TM2 from the same subunit. Instead, it packs against the hairpin formed by the first and second TM segments from an adjacent subunit (TM1′ and TM2′). Domain swapping may contribute to the high stability of wild-type DAGK (7, 8). Conversely, domain swapping may illuminate why DAGK mutants are often difficult to refold after denaturation (11).

A distinctive feature of the DAGK structure is a membrane-submerged cavity that resembles a portico (Figs. 1A and 2B and fig. S1). TM1 and TM3 serve as the pillars that bound the entry, terminating at an inner wall formed by TM2′ from a neighboring subunit, and crowned by an overhanging cornice comprising the connecting loop between TM2 and TM3. The portico, and in particular the cornice, contains a majority of the residues seen to be functionally essential in our mutagenesis studies (Fig. 2). Each portico contains critical residues that are contributed by two different subunits, consistent with a previous mutagenesis study that showed that each active site is shared between subunits (13). Titrations of DAGK with its substrates (MgATP and DAG), product (phosphatidic acid), and a nonhydrolyzable ATP analog (β,γ-methyleneadenosine 5′-triphosphate, AMP-PCP) were monitored by NMR. Saturable changes in NMR resonances upon ligand binding (Fig. 3 and fig. S5) confirmed that the portico includes the lipid substrate binding site of DAGK. The overhanging cornice likely blocks lateral diffusion into the active site of lipids with head groups larger than DAG and phosphatidic acid. On the other hand, the spaciousness of the lower part of the portico explains the lack of specificity of DAGK with respect to the acyl chain composition of diglyceride substrates (26). Interactions of membrane enzymes with membranous substrates more typically involve gated internal cavities within the protein or well-structured grooves that confer high specificity for specific substrates (2729), although the active site of the homodimeric β barrel outer-membrane phospholipase A also exhibits a portico-like architecture, consistent with the lack of specificity of this enzyme in terms of substrate acyl chains (27, 30).

Fig. 3

NMR mapping of the substrate binding sites (35). (A) Example of 800-MHz 1H,15N-TROSY (transverse relaxation-optimized spectroscopy) NMR spectra used to monitor titration of wild-type DAGK by a substrate, product, or substrate analog. Shown are the overlaid spectra for titration of DAGK by MgAMP-PCP at concentrations of 0 mM (black), 2 mM (red), 4 mM (yellow), 8 mM (green), and 16 mM (blue). Resonances are labeled on the basis of their residue assignments (36). Unlabeled peaks reflect the 10% of resonances for which assignments were not completed. NMR data for titrations of DAGK with DAG, phosphatidic acid, and MgATP are shown in fig. S5. Insets: Selected peaks from the MgAMP-PCP titration (upper panel) and from a separate DAG titration (lower panel; see full spectra in fig. S5). The MgAMP-PCP data illustrate fast-exchange NMR conditions in which DAGK peaks gradually change position during the titration, reflecting population-weighted averages between free and complexed forms. In contrast, the data from the DAG titration (lower panel) illustrate slow exchange behavior, where the free enzyme peaks disappear and peaks representing the complex appear at new positions, with contour intensities reflecting the relative populations of the free and complexed states. (B) Summary of active-site mapping, highlighting residues associated with TROSY resonances that exhibit large and saturable changes in positions in response to substrate, product, or analog binding. (C) Structure of DAGK with residues highlighted for which TROSY resonances exhibited significant and saturable changes in chemical shift upon binding. In blue are residues that are perturbed by addition of MgATP or MgAMP-PCP (Δδ > 0.02 ppm); in orange are those perturbed by addition of DAG (Δδ > 0.025 ppm). Residues for which resonance positions are perturbed by both nucleotide and lipid titrations are in green.

NMR resonances for three residues located just under the cornice on TM1 and TM2′ exhibited shifts in response to the binding of nucleotides and lipids (Fig. 3). These sites are likely proximal to the site of ATP-to-DAG phosphoryl transfer, consistent with the observed proximity to the cornice of many of the highly conserved and functionally critical residues. We suggest that a second role for the cornice in the active site may be to seal off the substrate-filled active site from water (to avoid ATP hydrolysis), a catalytic imperative that is usually satisfied in water-soluble phosphotransferases, such as adenylate kinase, by global-domain motions that close the active site upon substrate binding (31). DAGK appears not to undergo a global conformational change, as revealed by observation of only modest and highly localized changes in NMR resonance positions for DAGK upon forming either binary complexes with its substrates or a ternary complex with DAG and MgAMP-PCP (Fig. 3 and fig. S5).

As alluded to above, we systematically examined the functional consequences of mutating each residue in DAGK to cysteine to gain insight into structure-function relationships beyond what can be deduced either from multiple sequence alignment or from a previous phenotype-based random mutagenesis study (32). Mutants that resulted in >85% loss of catalytic activity (table S1) were further characterized to reveal two categories of defects (Fig. 2 and table S2). One set of mutations do not disrupt protein folding but result in loss of catalytic function. These residues are either directly in the active site or play specific local structural roles in support of the active site. Other mutations to cysteine result in misfolding so severe that in vivo expression levels were seen to be reduced (Fig. 2 and table S2), often to an undetectable level. In cases where purification was possible, such mutants were typically observed to be aggregation-prone (table S2). Although cysteine-scanning mutagenesis is a commonly used in studies of membrane proteins (33), we are unaware of previous documentation of a set of mutants analogous to the misfolding-prone variants described in this work.

A remarkable feature of the DAGK structure is the close proximity between the active site and a majority of residues for which mutations to cysteine result in severe misfolding (Fig. 2 and table S2). This is in contrast with enzymes, such as those of the TIM barrel superfamily (34), where the active site is set within a robust structural framework that is tolerant of major remodeling of catalytic residues. Given that numerous diseases, such as cystic fibrosis, are associated with inherited or sporadic mutations that result in single–amino acid changes in membrane protein sequence (12), our results serve as a warning that even for a structurally well-characterized membrane protein, knowledge regarding the location of a disease-promoting mutation site cannot a priori be used to reliably predict the exact nature of the molecular defect that is linked to disease etiology (i.e., local active-site disruption versus catastrophic misfolding).

The observation that most of the residues important for avoiding misfolding of DAGK are located at the DAGK active site offers insight regarding why the portico has not been adapted by evolution to orthologous functions. Although DAGK has been shown to be a highly efficient catalyst in its microbial physiological niche (3), intimate linkage between activity and folding may help to explain why this enzyme is an evolutionary orphan.

Supporting Online Material

Materials and Methods

Figs. S1 to S8

Tables S1 to S4


  • * These authors contributed equally to this work.

References and Notes

  1. See supporting material on Science Online. Briefly, the backbone structure of micellar DAGK at pH 6.5 was determined using 1H-1H nuclear Overhauser effect–derived distances, backbone torsion angle restraints from chemical shifts, residual dipolar couplings, and paramagnetic relaxation effect–derived distances. These NMR restraints were also supplemented by a limited number of intersubunit inter-residue distance restraints derived from disulfide mapping measurements that used single-cysteine mutant forms of DAGK. Care was taken to avoid possible motional complications to the disulfide mapping–derived distances. The structure was calculated using standard restrained molecular dynamics and simulated annealing protocols.
  2. Single-letter abbreviations for amino acid residues: A, Ala; C, Cys; D, Asp; E, Glu; F, Phe; G, Gly; H, His; I, Ile; K, Lys; L, Leu; M, Met; N, Asn; P, Pro; Q, Gln; R, Arg; S, Ser; T, Thr; V, Val; W, Trp; Y, Tyr.
  3. We thank J. Bowie for providing the DAGK expression system, many of the mutants used in this work, and much discussion; M. Voehler, B. Gorzelle, P. Power, C. Kang, and J. Jacob for technical assistance; and K. Oxenoid, L. Kay, S. Prosser, M.-D. Tsai, T. Iverson, J. Meiler, A. Forster, J. Battiste, W. Chazin, and members of the Sanders lab for discussion. Supported by NIH training grant T32 NS007491 (W.V.H.) and National Institute of General Medical Sciences grant RO1 GM47485. C.R.S. is a consultant for Anatrace Inc. (now owned by Affymetrix), a company that markets high-purity biological detergents, including some of the detergents used in sample preparation in this work. The coordinates for DAGK have been deposited as PDB entry 2kdc in the Protein Data Bank. This work is dedicated to the fond memory of Anne Karpay.
View Abstract

Stay Connected to Science

Navigate This Article