Superparamagnetic MnFe2O4 nanocrystals of different sizes were synthesized in high-boiling ether solvent and transferred into water using three different approaches. First, we applied a ligand exchange in order to form a water soluble polymer shell. Second, the particles were embedded into an amphiphilic polymer shell. Third, the nanoparticles were embedded into large micelles formed by lipids. Although all approaches lead to effective negative contrast enhancement, we observed significant differences concerning the magnitude of this effect. The transverse relaxivity, in particular r2*, is greatly higher for the micellar system compared to the polymer-coated particles using same-sized nanoparticles. We also observed an increase in transverse relaxivities with increasing particle size for the polymer-coated nanocrystals. The results are qualitatively compared with theoretical models describing the dependence of relaxivity on the size of magnetic spheres.

Semiconductor quantum dots and superparamagnetic iron oxide nanocrystals have physical properties that are well suited for biomedical imaging. Previously, we have shown that iron oxide nanocrystals embedded within the lipid core of micelles show optimized characteristics for quantitative imaging. Here, we embed quantum dots and superparamagnetic iron oxide nanocrystals in the core of lipoproteins--micelles that transport lipids and other hydrophobic substances in the blood--and show that it is possible to image and quantify the kinetics of lipoprotein metabolism in vivo using fluorescence and dynamic magnetic resonance imaging. The lipoproteins were taken up by liver cells in wild-type mice and displayed defective clearance in knock-out mice lacking a lipoprotein receptor or its ligand, indicating that the nanocrystals did not influence the specificity of the metabolic process. Using this strategy it is possible to study the clearance of lipoproteins in metabolic disorders and to improve the contrast in clinical imaging.

Here we present a biophysical, structural, and computational analysis of the directed evolution of the human DNA repair protein O 6 -alkylguanine-DNA alkyltransferase (hAGT) into SNAP-tag, a self-labeling protein tag. Evolution of hAGT led not only to increased protein activity but also to higher stability, especially of the alkylated protein, suggesting that the reactivity of the suicide enzyme can be influenced by stabilizing the product of the irreversible reaction. Whereas wild-type hAGT is rapidly degraded in cells after alkyl transfer, the high stability of benzylated SNAP-tag prevents proteolytic degradation. Our data indicate that the intrinstic stability of a key α helix is an important factor in triggering the unfolding and degradation of wild-type hAGT upon alkyl transfer, providing new insights into the structure−function relationship of the DNA repair protein.T he specific labeling of proteins with synthetic probes is a powerful approach for studying protein function. One way to achieve such a specific labeling is based on so-called selflabeling protein tags.1 In this approach, the protein of interest is expressed as a fusion protein with a peptide or protein (i.e., tag) whose role is to specifically bind to a synthetic probe in vitro or in vivo. A well-established example of a self-labeling protein tag is SNAP-tag.2 SNAP-tag specifically reacts with substituted O 6 -benzylguanine derivatives and thereby permits the labeling of SNAP-tag fusion proteins with a wide variety of different synthetic probes. Recent applications include its use for the analysis of protein complexes, 3 super-resolution microscopy, 4 the identification of protein−protein interactions, 5 drug target identification, 6 and the determination of protein half-life in animals. 7 The appeal of self-labeling tags such as SNAP-tag is the ease with which fusion proteins can be labeled with synthetic probes even in living cells. A conceptual limitation of the approach is the fact that the tag can affect the properties of its fusion partner. It is therefore important that the properties of the tag be as thoroughly characterized as possible.SNAP-tag was generated in a stepwise manner from human O 6 -alkylguanine-DNA alkyltransferase (hAGT) by introduction of a total of 19 point mutations (Figure 1) and deletion of 25 C-terminal residues. Saturation mutagenesis of four active-site residues followed by phage display and selection for activity against BG derivatives resulted in GE AGT, a mutant with 20-fold increased activity toward such substrates ( Figure 1B). 8Subsequent saturation mutagenesis of four additional residues involved in substrate binding followed by phage selection resulted in AGT-54, a mutant with 1.5-fold higher activity than GE AGT. To further optimize the protein for applications in protein labeling, mutations were introduced to suppress DNA binding and reactivity toward nucleosides, to remove nonessential cysteines, and to truncate the last 25 residues. 9The resulting mutant M AGT displayed relatively low activ...