Chaomin Ke

Cancer is one of the leading causes of morbidity and mortality in the world, but more cancer therapies are needed to complement existing regimens due to problems of existing cancer therapies. Herein, we term ferroptosis therapy (FT) as a form of cancer therapy and hypothesize that the FT efficacy can be significantly improved via accelerating the Fenton reaction by simultaneously increasing the local concentrations of all reactants (Fe2+, Fe3+, and H2O2) in cancer cells. Thus, Fenton-reaction-acceleratable magnetic nanoparticles, i.e., cisplatin (CDDP)-loaded Fe3O4/Gd2O3 hybrid nanoparticles with conjugation of lactoferrin (LF) and RGD dimer (RGD2) (FeGd-HN@Pt@LF/RGD2), were exploited in this study for FT of orthotopic brain tumors. FeGd-HN@Pt@LF/RGD2 nanoparticles were able to cross the blood–brain barrier because of its small size (6.6 nm) and LF-receptor-mediated transcytosis. FeGd-HN@Pt@LF/RGD2 can be internalized into cancer cells by integrin αvβ3-mediated endocytosis and then release Fe2+, Fe3+, and CDDP upon endosomal uptake and degradation. Fe2+ and Fe3+ can directly participate in the Fenton reaction, whereas the CDDP can indirectly produce H2O2 to further accelerate the Fenton reaction. The acceleration of Fenton reaction generates reactive oxygen species to induce cancer cell death. FeGd-HN@Pt@LF/RGD2 successfully delivered reactants involved in the Fenton reaction to the tumor site and led to significant inhibition of tumor growth. Finally, the intrinsic magnetic resonance imaging (MRI) capability of the nanoparticles was used to assess and monitor tumor response to FT (self-MRI monitoring).

The second near-infrared (NIR-II) window is a fundamental modality for deep-tissue in vivo imaging. However, it is challenging to synthesize NIR-II probes with high quantum yields (QYs), good biocompatibility, satisfactory pharmacokinetics, and tunable biological properties. Conventional long-wavelength probes, such as inorganic probes (which often contain heavy metal atoms in their scaffolds) and organic dyes (which contain large π-conjugated groups), exhibit poor biosafety, low QYs, and/or uncontrollable pharmacokinetic properties. Herein, we present a bioengineering strategy that can replace the conventional chemical synthesis methods for generating NIR-II contrast agents. We use a genetic engineering technique to obtain a series of albumin fragments and recombinant proteins containing one or multiple domains that form covalent bonds with chloro-containing cyanine dyes. These albumin variants protect the inserted dyes and remarkably enhance their brightness. The albumin variants can also be genetically edited to develop size-tunable complexes with precisely tailored pharmacokinetics. The proteins can also be conjugated to biofunctional molecules without impacting the complexed dyes. This combination of albumin mutants and clinically-used cyanine dyes can help widen the clinical application prospects of NIR-II fluorophores.

Graphs have been widely used to represent objects and object connections in applications such as the Web, social networks, and citation networks. Mining influence relationships from graphs has gained interests in recent years because providing influence information about the object connections in graphs can facilitate graph exploration, graph search, and connection recommendations. In this paper, we study the problem of detecting influence aspects, on which objects are connected, and influence degree (or influence strength), with which one graph node influences another graph node on a given aspect. Existing techniques focus on inferring either the influence degrees or influence types from graphs. We propose two generative Aspect Influence Models, OAIM and LAIM, to detect both influence aspects and influence degrees. These models utilize the topological structure of the graphs, the text content associated with objects, and the context in which the objects are connected. We compare these two models with one baseline approach which considers only the text content associated with objects. The empirical studies on citation graphs and networks of users from Twitter show that our models can discover more effective results than the baseline approach.