Natural Language Processing for Smart Healthcare.

Smart healthcare has achieved significant progress in recent years. Emerging artificial intelligence (AI) technologies enable various smart applications across various healthcare scenarios. As an essential technology powered by AI, natural language processing (NLP) plays a key role in smart healthcare due to its capability of analysing and understanding human language. In this work, we review existing studies that concern NLP for smart healthcare from the perspectives of technique and application. We first elaborate on different NLP approaches and the NLP pipeline for smart healthcare from the technical point of view. Then, in the context of smart healthcare applications employing NLP techniques, we introduce representative smart healthcare scenarios, including clinical practice, hospital management, personal care, public health, and drug development. We further discuss two specific medical issues, i.e., the coronavirus disease 2019 (COVID-19) pandemic and mental health, in which NLP-driven smart healthcare plays an important role. Finally, we discuss the limitations of current works and identify the directions for future works.

Robust drug bioavailability and safety for rheumatoid arthritis therapy using D-amino acids-based supramolecular hydrogels.

Long-term use of disease-modifying anti-rheumatic drugs (DMARDs) such as methotrexate (MTX) shows clinical benefits for rheumatoid arthritis (RA) treatment. However, there are growing concerns over the adverse effects of systemic drug administration. Therefore, a strategy that can enhance drug bioavailability while minimizing side effects is urgently needed, but remains a challenge in RA therapy. To this end, here we conjugated MTX with a supramolecular self-assembling hydrogel composed of d-amino acids with a sequence of GDFDFDY. It was shown that MTX-GDFDFDY hydrogels exhibited a favorable drug selectivity behavior that they increased MTX toxicity toward RA synoviocytes, but reduce toxicity toward normal cells. Moreover, MTX-GDFDFDY hydrogels not only effectively inhibited the proliferation and migration of RA synoviocytes, but also inhibited the polarization of proinflammatory M1 type macrophages to reduce inflammation. After intra-articularly injected the hydrogels into the joints of adjuvant induced arthritis (AIA) mice, we found that MTX-GDFDFDY hydrogels significantly alleviated RA syndromes of joint swelling and fever compared to L-configuration MTX-GFFY hydrogels and free MTX. Furthermore, MTX-GDFDFDY hydrogels successfully protected cartilage though inhibiting synovial invasion and inflammation without causing systematic side effects. Therefore, d-amino acids supramolecular hydrogels can serve as an efficient and safe drug delivery system, showing a promising potential to improve RA therapy.

Interpretable Temporal Attention Network for COVID-19 forecasting.

The worldwide outbreak of coronavirus disease 2019 (COVID-19) has triggered an unprecedented global health and economic crisis. Early and accurate forecasts of COVID-19 and evaluation of government interventions are crucial for governments to take appropriate interventions to contain the spread of COVID-19. In this work, we propose the Interpretable Temporal Attention Network (ITANet) for COVID-19 forecasting and inferring the importance of government interventions. The proposed model is with an encoder-decoder architecture and employs long short-term memory (LSTM) for temporal feature extraction and multi-head attention for long-term dependency caption. The model simultaneously takes historical information, a priori known future information, and pseudo future information into consideration, where the pseudo future information is learned with the covariate forecasting network (CFN) and multi-task learning (MTL). In addition, we also propose the degraded teacher forcing (DTF) method to train the model efficiently. Compared with other models, the ITANet is more effective in the forecasting of COVID-19 new confirmed cases. The importance of government interventions against COVID-19 is further inferred by the Temporal Covariate Interpreter (TCI) of the model.

Organelle-inspired supramolecular nanomedicine to precisely abolish liver tumor growth and metastasis.

Organelles are responsible for the efficient storage and transport of substances in living systems. A myriad of extracellular vesicles (EVs) acts as a bridge to exchange signaling molecules in cell-cell communication, and the highly dynamic tubulins and actins contribute to efficient intracellular substance transport. The inexhaustible cues of natural cargo delivery by organelles inspire researchers to explore the construction of biomimetic architectures for "smart" delivery carriers. Herein, we report a 10-hydroxycamptothecin (HCPT)-peptide conjugate HpYss that simulates the artificial EV-to-filament transformation process for precise liver cancer therapy. Under the sequential stimulus of extracellular alkaline phosphatase (ALP) and intracellular glutathione (GSH), HpYss proceeds via tandem self-assembly with a morphological transformation from nanoparticles to nanofibers. The experimental phase diagram elucidates the influence of ALP and GSH contents on the self-assembled nanostructures. In addition, the dynamic transformation of organelle-mimetic architectures that are formed by HpYss in HepG2 cells enables the efficient delivery of the anticancer drug HCPT to the nucleus, and the size-shape change from extracellular nanoparticles (50-100 nm) to intracellular nanofibers (4-9 nm) is verified to be of key importance for nuclear delivery. Nuclear targeting of HpYss amplifies apoptosis, thus significantly enhancing the inhibitory effect of HCPT (>10-fold) to HepG2 cells. Benefitting from the spatiotemporally controlled nanostructures, HpYss exhibited deep penetration, enhanced accumulation, and long-term retention in multicellular spheroid and xenograft models, potently abolishing liver tumor growth and preventing lung metastasis. We envision that our organelle-mimicking delivery strategy provides a novel paradigm for designing nanomedicine to cancer therapy.

Correction: Furin-instructed molecular self-assembly actuates endoplasmic reticulum stress-mediated apoptosis for cancer therapy.

Correction for 'Furin-instructed molecular self-assembly actuates endoplasmic reticulum stress-mediated apoptosis for cancer therapy' by Chenxing Fu et al., Nanoscale, 2020, 12, 12126-12132, DOI: .

Supramolecular Self-Assembled Nanofibers Efficiently Activate the Precursor of Hepatocyte Growth Factor for Angiogenesis in Myocardial Infarction Therapy.

The reconstruction of blood perfusion is a crucial therapeutic method to save and protect cardiac function after acute myocardial infarction (AMI). The activation of the hepatocyte growth factor precursor (pro-HGF) has a significant effect on promoting angiogenesis and antiapoptosis. The oxygen/glucose deprivation (OGD) caused by AMI could induce vascular adventitia fibroblasts to differentiate into myofibroblasts and secrete the pro-HGF. Meanwhile, the specific Met receptor of the hepatocyte growth factor (HGF) is upregulated in endothelial cells during AMI. However, the poor prognosis of AMI suggests that the pro-HGF is not effectively activated. Improving the activation efficiency of the pro-HGF may play a positive role in the treatment of AMI. Herein, we designed supramolecular nanofibers self-assembled by compound 1 (Comp.1, Nap-FFEG-IVGGYPWWMDV), which can strongly activate the pro-HGF and initiate HGF-Met signaling. Studies have proven that Comp.1 possesses a better ability to activate the pro-HGF to perform antiapoptosis and pro-angiogenesis. In vivo results have confirmed that the retention time of Comp.1 and its accumulation in the infarct area of the heart are promoted. Moreover, Comp.1 plays an effective role in promoting angiogenesis in the marginal area of AMI, reducing myocardial fibrosis, and protecting cardiac function. Herein, we will optimize the structure of bioactive peptides through supramolecular self-assembly and amplify their therapeutic effect by improving their efficiency, providing a new strategy for the therapy of AMI.

Dual-ligand supramolecular nanofibers inspired by the renin-angiotensin system for the targeting and synergistic therapy of myocardial infarction.

Rationale: The compensatory activation of the renin-angiotensin system (RAS) after myocardial infarction (MI) plays a crucial role in the pathogenesis of heart failure. Most existing studies on this subject focus on mono- or dual-therapy of blocking the RAS, which exhibit limited efficacy and often causes serious adverse reactions. Few studies have been conducted on targeted therapy based on the activated RAS post-MI. Thus, the development of multiple-functional nanomedicine with concurrent targeting ability and synergistic therapeutic effect against RAS may show great promise in improving cardiac function post-MI. Methods: We utilized a cooperative self-assembly strategy constructing supramolecular nanofibers- telmisartan-doped co-assembly nanofibers ( TDCNfs ) to counter-regulate RAS through targeted delivery and combined therapy. TDCNfs were prepared through serial steps of solvent exchange, heating incubation, gelation, centrifugation, and lyophilization, in which the telmisartan was doped in the self-assembly process of Ang1-7 to obtain the co-assembly nanofibers wherein they act as both therapeutic agents and target-guide agents. Results: TDCNfs exhibited the desired binding affinity to the two different receptors, AT1R and MasR. Through the dual ligand-receptor interactions to mediate the coincident downstream pathways, TDCNfs not only displayed favorably targeted properties to hypoxic cardiomyocytes, but also exerted synergistic therapeutic effects in apoptosis reduction, inflammatory response alleviation, and fibrosis inhibition in vitro and in vivo, significantly protecting cardiac function and mitigating post-MI adverse outcomes. Conclusion: A dual-ligand nanoplatform was successfully developed to achieve targeted and synergistic therapy against cardiac deterioration post-MI. We envision that the integration of multiple therapeutic agents through supramolecular self-assembly would offer new insight for the systematic and targeted treatment of cardiovascular diseases.

Enzyme-Instructed Self-Assembly Enabled Monomer-Excimer Transition to Construct Higher Ordered Luminescent Supramolecular Assembly for Activity-based Bioimaging.

It is challenging to construct high-performing excimer-based luminescent analytic tools at low molecular concentrations. We report that enzyme-instructed self-assembly (EISA) enables the monomer-excimer transition of a coumarin dye (Cou) at low molecular concentrations, and the resulting higher ordered luminescent supramolecular assemblies (i.e., nanofibers) efficiently record the spatiotemporal details of alkaline phosphatase (ALP) activity in vitro and in vivo. Cou was conjugated to short self-assembly peptides with a hydrophilic ALP-responsive group. By ALP triggering, EISA actuated a nanoparticles-nanofibers transition at low peptide concentrations followed by monomer-excimer transition of Cou. Analysis of structure-property relationships revealed that the self-assembly motif was a prerequisite for peptides to induce the monomer-excimer transition of Cou. Luminescent supramolecular nanofibers of pYD (LSN-pYD) illuminated the intercellular bridge of cancer cells and distinguished cancer cells (tissues) from normal cells (tissues) efficiently and rapidly, promising potential use for the early diagnosis of cancer. This work extends the functions of EISA and provides a new application of supramolecular chemistry.

Furin-instructed molecular self-assembly actuates endoplasmic reticulum stress-mediated apoptosis for cancer therapy.

Protein quality control and proteostasis are essential to maintain cell survival as once disordered, they will trigger endoplasmic reticulum (ER) stress and even initiate apoptosis. Severe ER stress-mediated apoptosis is the cause of neurodegenerative diseases and expected to be a new target for cancer therapy. In this study, we designed a small molecule of 1-Nap to execute furin-instructed molecular self-assembly for selectively inhibiting the growth of MDA-MB-468 cells in vitro and in vivo. According to the results of transmission electron microscopy (TEM) and HPLC tracing analysis, 1-Nap is capable of self-assembling upon furin-instructed cleavage that transforms 1-Nap nanoparticles to 1-Nap nanofibers. Fluorescence imaging and Western-blot analysis results indicate that the furin-instructed self-assembly of 1-Nap rather than its ER-targeting interaction is indispensable for the ER stress and activation of apoptosis. The furin-instructed self-assembly of 1-Nap is associated with both the ER (1-Nap's targeting location) and the trans-Golgi network (furin's location); this inspired us to reasonably believe that the blocking of ER-to-Golgi traffic in the secretory pathway by molecular self-assembly may be the intrinsic motivation for controlling cell fate. This work provides a new way for the targeted disturbance of the proteostasis of cells through molecular self-assembly for developing cancer therapeutics.

Dual-responsive self-assembly in lysosomes enables cell cycle arrest for locking glioma cell growth.

Herein we first report a dual-responsive peptide substrate (Comp. 1) for preparing self-assembled nanomaterials triggered by pH and legumain. The dual-responsive self-assembly of Comp. 1 in glioma cells enables its long retention time in lysosomes, S phase arrest, and cell growth locking. We verified that the blocked degradation of HIF-1α in lysosomes played a key role in cell cycle arrest and decreased DNA replication. This work illustrates the disturbance of lysosomal function by self-assembled nanomaterials as a promising strategy for inhibiting glioma cell growth.