Biology subjects

Xu, H.

Publications and source records attributed to Xu, H..

At least 217 records · Page 12Linked to original sources

bioRxiv2024

Loss of circulating CD8α+ NK cells during human Mycobacterium tuberculosis infection

Natural Killer (NK) cells can recognize and kill Mtb-infected cells in vitro, however their role after natural human exposure has not been well-studied. To identify Mtb-responsive NK cell populations, we analyzed the peripheral blood of healthy household contacts of active Tuberculosis (TB) cases and source community donors in an endemic region of Port-au-Prince, Haiti by flow cytometry. We observed higher CD8 expression on NK cells in putative resistors (IGRA-contacts) with a progressive loss of these circulating cells during household-associated latent infection and disease. In vitro assays and CITE-seq analysis of CD8+ NK cells demonstrated enhanced maturity, cytotoxic gene expression, and response to cytokine stimulation relative to CD8- NK cells. CD8+ NK cells also displayed dynamic surface expression dependent on MHC I in contrast to conventional CD8+ T cells. Together, these results support a specialized role for CD8+ NK cell populations during Mtb infection correlating with disease resistance.

immunology↗

bioRxiv2024

Design of antigens to present a tumor-specific cryptic epitope

In many cancers, the epidermal growth factor receptor (EGFR) gene is amplified, mutated, or both. The monoclonal antibody mAb806 binds selectively to cancer cells that overexpress EGFR or express the truncated mutant EGFRvIII, but not to normal cells. This suggests that a promising avenue for developing cancer vaccines may be to design antigens that elicit mAb806-like antibodies. In this study, we designed antigens that present the mAb806-binding epitope in the same conformation as in overexpressed or truncated EGFR. We first used molecular dynamics simulations to identify conformations of EGFR in which the residues of the mAb806-binding epitope are accessible. We then designed antigens by substituting that epitope in place of a structurally similar loop in a different protein and generating mutants that could potentially stabilize the mAb806-binding conformation in this new context. Two mutants in which the epitope remained stable in subsequent simulations were chosen for evaluation in vitro. Binding kinetics experiments with these designed antigens provided strong evidence that the epitope was successfully stabilized in the mAb806-binding conformation, suggesting that they could potentially form the basis of vaccines that elicit cancer-selective antibodies.

biophysics↗

bioRxiv2024

ACSS2 upregulation enhances neuronal resilience to aging and tau-associated neurodegeneration

Epigenetic mechanisms, including histone acetylation, are pivotal for learning and memory, with a role in neuronal function in Alzheimers disease and Related Dementia (ADRD). Acetyl-CoA synthetase 2 (ACSS2), an enzyme that generates acetyl-CoA, is central to histone acetylation and gene regulation, particularly in neurons, due to their unique metabolic demands and postmitotic state. ACSS2 can be recruited to the nucleus and chromatin, locally supplying acetyl-CoA to directly fuel histone acetyltransferase enzymes and key neuronal gene expression. This regulatory mechanism may be a promising target for therapeutic intervention in neurodegenerative diseases. Previously we showed that systemic ACSS2 deletion in mice, although largely normal in physiology, is greatly impaired in memory. Here we investigated whether increasing ACSS2 levels could protect neurons against disease and age-associated cognitive decline. Given the role of tau in ADRD, we used primary hippocampal neurons that mimic the sporadic development of tau pathology and the P301S transgenic mouse model for tau-induced memory decline. Our results show that ACSS2 upregulation mitigates tau-induced transcriptional alterations, enhances neuronal resilience against tau pathology, improves long-term potentiation, and ameliorates memory deficits. Expanding upon these findings, we reveal that increasing histone acetylation through ACSS2 upregulation improves age-associated memory decline. These findings indicate that increasing ACSS2 is highly effective in countering age- and tau-induced transcriptome changes, preserving elevated levels of synaptic genes, and safeguarding synaptic integrity. We thus highlight ACSS2 as a key player in the epigenetic regulation of cognitive aging and ADRD, providing a foundation for targeted therapeutics to enhance brain resilience and function. SummaryACSS2 upregulation protects neurons from disease and age-related decline by enhancing synaptic and longevity gene expression.

neuroscience↗

bioRxiv2024

Modeling the genomic complexity of human cancer using Cas12a mice

Somatic genome editing in mouse models has increased our understanding of the in vivo effects of genetic alterations in areas ranging from neuroscience to cancer biology and beyond. However, existing models are limited in their ability to create multiple targeted edits. Thus, our understanding of the complex genetic interactions that underlie development, homeostasis, and disease remains incomplete. Cas12a is an RNA-guided endonuclease with unique attributes that enable simple targeting of multiple genes with crRNA arrays containing tandem guides. To accelerate and expand the generation of complex genotypes in somatic cells, we generated transgenic mice with Cre-regulated and constitutive expression of enhanced Acidaminococcus sp. Cas12a (enAsCas12a). In these mice, enAsCas12a-mediated somatic genome editing robustly generated compound genotypes, as exemplified by the initiation of diverse cancer types driven by homozygous inactivation of trios of tumor suppressor genes. We further integrated these modular crRNA arrays with clonal barcoding to quantify the size and number of tumors with each array, as well as the efficiency of each crRNA. These Cas12a alleles will enable the rapid generation of disease models and broadly facilitate the high-throughput investigation of coincident genomic alterations in somatic cells in vivo.

cancer biology↗

bioRxiv2024

Global variation in the ratio of sapwood to leaf area explained by optimality principles

The sapwood area supporting a given leaf area (vH) reflects a coordinated coupling between carbon uptake, water transport and loss at a whole plant level. Worldwide variation in vH reflects diverse plants strategies adapt to prevailing environments, and impact the evolution of global carbon and water cycles. Why such a variation has not been convincingly explained yet, thus hinder its representation in Earth System Models. Here we hypothesize that optimal vH tends to mediate between plant hydraulics and leaf photosynthesis so that leaf water loss matches water supply. By compiling and testing against two extensive datasets, we show that our hypothesis explains nearly 60% of vH variation responding to light, vapour pressure deficit, temperature, and sapwood conductance in a quantitively predictable manner. Sapwood conductance or warming-enhanced hydraulic efficiency reduces the demand on sapwood area for a given total leaf area and, whereas brightening and air dryness enhance photosynthetic capacities consequently increasing the demand. This knowledge can enrich Earth System Models where carbon allocation and water hydraulics play key roles in predicting future climate-carbon feedback.

ecology↗

bioRxiv2024

Efficient cloning of genes for wheat yield component traits from QTLs via sequencing of the RIL population

In wheat (Triticum aestivum L.), yield component traits (YCTs) are the most important yield traits. Only several genes for YCTs have been originally cloned. The efficient cloning of genes for YCTs directly from wheat remains a challenge. Here, we proposed a strategy for cloning genes from quantitative trait loci (QTLs) by sequencing of recombinant inbred lines (RILs) (QTL-Seq-RIL). Using the TN18 x LM6 RIL population as an example, we identified 138 candidate unigenes (CUGs) for YCTs from 77 stable QTLs. The average of CUGs per QTL was 1.8, which enabled us to confirm the CUGs directly. We have confirmed seven CUGs, TaIFABPL, TaDdRp, TaRLK, TaTD, TaTFC3, TaKMT and TaSPL17, via the CRISPR/Cas9 system. Of these, six genes were found firstly to regulate YCTs in crops except for TaSPL17. Five CUGs (include TaSPL17) for which orthologous genes have been cloned previously with the same or similar agronomic functions. It is to say, 11 CUGs were preliminarily validated using a single RIL population. QTL-Seq-RIL provides an efficient method for rapid gene cloning using existing RIL populations.

plant biology↗

bioRxiv2024

Microbial community assembly during aging of food waste-derived hydrochar: key bacterial guilds mediate nutrient dynamics

Hydrochar aging may mitigate the adverse effects of direct application on plant growth. However, little is known about the strategies for accelerating this process and the microbiological mechanisms involved. This study explored the effects of three strategies, including the additions of straw and efficient-degrading microorganisms, on the aging processes of food waste-derived hydrochar, and the key microbial drivers. The bacterial and fungal community compositions in hydrochar were estimated at different aging periods (i.e., day 1, 7, 14 and 35) using 16S and ITS rRNA gene sequencing. The results showed that straw and microbial inoculum addition improved the temperature and shortened the aging time by [~]30%. Three bacterial guilds, mainly including Bacillus-like species, were identified that had significant correlations with the nitrogen (N) and phosphorus (P) dynamics during aging. This study demonstrates the feasibility of manipulating key microbial guilds artificially to achieve efficient harmless-treatment of hydrochar. HighlightsO_LIStraw and microbial inoculum addition reduced ageing duration by 30%. C_LIO_LIMicrobial inoculum addition increases the aging reaction temperature by 13%. C_LIO_LIBacteria, rather than fungi, play the critical role in hydrochar aging. C_LIO_LIBacillus and related species drive nitrogen and phosphorus cycles of aging. C_LI Graphical Abstract O_FIG O_LINKSMALLFIG WIDTH=200 HEIGHT=155 SRC="FIGDIR/small/580608v1_ufig1.gif" ALT="Figure 1"> View larger version (68K): org.highwire.dtl.DTLVardef@b2c2c9org.highwire.dtl.DTLVardef@124ab1org.highwire.dtl.DTLVardef@e5c00borg.highwire.dtl.DTLVardef@1a31deb_HPS_FORMAT_FIGEXP M_FIG C_FIG

microbiology↗

bioRxiv2024

Direct, bisulfite-free 5mC and 5hmC sequencing at single-cell resolution

We report the first direct sequencing methodologies for quantitative detection of 5mC and 5hmC at single-base resolution and single-cell level, termed scTAPS (for 5mC + 5hmC) and scCAPS+ (for 5hmC specifically). With [~]90% mapping efficiency, our methods accurately benchmark 5mC and 5hmC profiles in CD8+ T and mES cells, respectively. Notably, scCAPS+ revealed a global increase in 5hmC within the hippocampus of aging mice, both in neurons and in non-neurons.

genomics↗

bioRxiv2024

A systematic evaluation of computation methods for cell segmentation

Cell segmentation is a fundamental task in analyzing biomedical images. Many computational methods have been developed for cell segmentation and instance segmentation, but their performances are not well understood in various scenarios. We systematically evaluated the performance of 18 segmentation methods to perform cell nuclei and whole cell segmentation using light microscopy and fluorescence staining images. We found that general-purpose methods incorporating the attention mechanism exhibit the best overall performance. We identified various factors influencing segmentation performances, including image channels, choice of training data, and cell morphology, and evaluated the generalizability of methods across image modalities. We also provide guidelines for choosing the optimal segmentation methods in various real application scenarios. We developed Seggal, an online resource for downloading segmentation models already pre-trained with various tissue and cell types, substantially reducing the time and effort for training cell segmentation models.

bioinformatics↗

bioRxiv2024

Genetic and bioinformatic analyses reveal transcriptional networks underlying dual genomic coordination of mitochondrial biogenesis

Mitochondrial biogenesis requires the expression of genes encoded by both the nuclear and mitochondrial genomes. However, aside from a handful transcriptional factors regulating specific subsets of mitochondrial genes, the overall architecture of the transcriptional control of mitochondrial biogenesis remains to be elucidated. The mechanisms coordinating these two genomes are largely unknown. We performed a targeted RNAi screen in developing eyes with reduced mitochondrial DNA content, anticipating a synergistic disruption of tissue development due to impaired mitochondrial biogenesis and mtDNA deficiency. Among 638 transcription factors annotated in Drosophila genome, 77 were identified as potential regulators of mitochondrial biogenesis. Utilizing published ChIP-seq data of positive hits, we constructed a regulatory network revealing the logic of the transcription regulation of mitochondrial biogenesis. Multiple transcription factors in core layers had extensive connections, collectively governing the expression of nearly all mitochondrial genes, whereas factors sitting on the top layer may respond to cellular cues to modulate mitochondrial biogenesis through the underlying network. CG1603, a core component of the network, was found to be indispensable for the expression of most nuclear mitochondrial genes, including those required for mtDNA maintenance and gene expression, thus coordinating nuclear genome and mtDNA activities in mitochondrial biogenies. Additional genetics analyses validated YL-1, a transcription factor upstream of CG1603 in the network, as a regulator controlling CG1603 expression and mitochondrial biogenesis.

genetics↗

bioRxiv2024

Integrative Omics Uncovers Low Tumorous Magnesium Content as A Driver Factor of Colorectal Cancer

Magnesium (Mg) deficiency is associated with increased risk and malignancy of colorectal cancer (CRC), yet the underlying mechanisms remain elusive. Here we used genomic, proteomic, and phosphoproteomic data to elucidate the impact of Mg deficiency on CRC. Genomic analysis identified 160 genes with higher mutation frequencies in Low-Mg tumors, including key driver genes such as KMT2C and ERBB3. Unexpectedly, initiation driver genes of CRC, such as TP53 and APC, displayed higher mutation frequencies in High-Mg tumors. Additionally, proteomics and phosphoproteomics indicated that low tumorous Mg content may activate epithelial-mesenchymal transition (EMT) by modulating inflammation or remodeling the phosphoproteome of cancer cells. Notably, we observed a negative correlation between the phosphorylation of DBN1 at S142 (DBN1S142p) and Mg content. A mutation in S142 to D (DBN1S142D) mimicking DBN1S142p upregulated MMP2 and enhanced cell migration, while treatment with MgCl2 reduced DBN1S142p, thereby reversing this phenotype. Mechanistically, Mg2+ attenuated the DBN1-ACTN4 interaction by decreasing DBN1S142p, which, in turn, enhanced the binding of ACTN4 to F-actin and promoted F-actin polymerization, ultimately reducing MMP2 expression. These findings shed new light on the crucial role of Mg deficiency in CRC progression and suggest that Mg supplementation may offer a promising preventive and therapeutic strategy for CRC.

cancer biology↗

bioRxiv2024

β-Cells as a Cell Factory for On-Demand Recombinant Protein Dosing: Harnessing the Neuroendocrine Cell Secretory Pathway for Controlled Release

This study explores the potential of utilizing {beta}-cells, exemplified with R7T1 {beta}-cell pseudoislets, as a transplantable cell factory for on-demand recombinant protein therapeutic delivery. While mammalian cell lines are widely used for in vitro protein production, the commonly utilized constitutive secretion pathway poses challenges to in vivo cell therapy, especially for delivering proteins requiring precise exposure kinetics. The proposed approach capitalizes on unique aspects of {beta}-cells, including substantial vesicular protein storage capacity and electrochemically-regulated protein release, to facilitate timely and titratable in vivo therapeutic delivery. Examining a variety of strategies to acheive {beta}-cell glucagon or glucagon-like peptide 1 (GLP-1) storage and secretion, we devised a flexible {beta}-cell-based expression platform for efficient cellular peptide production and on-demand release. This platform utilizes the preproinsulin coding sequence as a template, wherein therapeutic peptides of interest (glucagon or GLP-1) are substituted for C-peptide while the A- and B-peptide insulin chains are mutated to prevent bio-active insulin production. This approach overcomes the challenge of efficient bio-active peptide expression by leveraging the endogenous {beta}-cell peptide expression, translation, processing, storage and secretion machinery. Furthermore, {beta}-cells provide a mechanism for scalable electyrochemnically-triggered peptide delivery. This transformative strategy, which may be extended to other proteins and peptide expression cassettes, holds significant promise for targeted and temporally controlled in vivo production and release of recombinant protein therapeutics. The study suggests potential applications in addressing challenges in metabolic disorders, blood disorders, and oncology. Future refinements may focus on optimizing vector design, peptide production, and in vivo adaptation.

bioengineering↗

bioRxiv2024

Microplastics from disposable paper cups are enriched in the placenta and fetus, leading to metabolic and reproductive toxicity during pregnancy

The health implications of microplastics (MPs), especially those originating from hot drinks in disposable paper cups (DPCs), are increasingly alarming. We investigated the accumulation and metabolic and reproductive toxicological effects of MPs from DPCs filled with hot water in various tissues in a pregnant mouse model. Simulating human intake of 0.3, 3.3, and 33.3 cups daily, we found MPs exposure-induced dose-responsive harmful effects on murine fetal development and maternal physiology. MPs were detected in all 13 examined tissues, with the highest accumulation in the cecal contents, followed by significant depositions in the fetus, placenta, kidney, spleen, lung, and heart. A higher proportion of smaller MPs (90.35% < 10 m) was identified in brain tissues. Dose-responsive changes in functional microbiome and gene pathways were observed. Moderate MPs intake of 3.3 cups daily significantly altered cecal microbiome composition and metabolic functions. The transcriptomic functional variations in maternal blood, placenta, and mammary gland underscore the significant impacts of realistic MPs exposure on metabolic and immune health and posing neurodegenerative and miscarriage risks. The benchmark dose framework analysis using tissue-specific gene biomarkers revealed safe exposure limits at 2 to 4 cups/day during pregnancy. Our results indicate selective tissue accumulation and potential metabolic and reproductive toxicities of MPs at exposure levels presumed non-hazardous. Such risks remain unaddressed within current food safety regulations, impacting vulnerable groups such as pregnant women and fetuses. Research HighlightsO_LIMicroplastics released from disposable paper cups filled with hot water showed preferential accumulations in the murine fetus, placenta, kidney, spleen, lung, and heart, with significant adverse impacts on fetal development. C_LIO_LIMicroplastic exposure led to dose-responsive maternal microbiome changes associated with increased fatty acid biosynthesis and elevated expressions of genes related to viral infections, neurodegenerative diseases, oxidative stress, and miscarriage risk. C_LIO_LIA consumption level of 3.3 cups/day was sufficient to elicit systemic metabolic and reproductive toxicity, with a predicted safe exposure limit of 2 to 4 cups/day during pregnancy by benchmark dose framework analysis with molecular biomarkers. C_LI

molecular biology↗

bioRxiv2024

Absolute quantitative and base-resolution sequencing reveals comprehensive landscape of pseudouridine across the human transcriptome

Pseudouridine ({Psi}) is one of the most abundant modifications in cellular RNA. However, its function remains elusive, mainly due to the lack of highly sensitive and accurate detection methods. To address this challenge, we introduced 2-bromoacrylamide-assisted cyclization sequencing (BACS) for quantitative profiling of {Psi} at single-base resolution. Based on novel bromoacrylamide cyclization chemistry, BACS enables a {Psi}-to-C transition. Compared to previous methods, BACS allowed the precise identification of {Psi} positions, especially in densely modified {Psi} regions and consecutive uridine sequences. BACS successfully detected all known {Psi} sites in human rRNA and spliceosomal snRNAs and generated the first quantitative {Psi} map of human snoRNA and tRNA. Furthermore, BACS simultaneously detected adenosine-to-inosine (A-to-I) editing sites and N1-methyladenosine (m1A). Depletion of three key pseudouridine synthases (PUS) enabled us to elucidate the targets and sequence motifs of TRUB1, PUS7, and PUS1 in HeLa cells. We further applied BACS to Epstein-Barr virus (EBV)-encoded small RNAs (EBERs) and identified a highly abundant {Psi}114 site in EBER2. Surprisingly, applying BACS to a panel of RNA viruses demonstrated the absence of {Psi} in their viral transcripts or genomes, shedding light on differences in pseudouridylation between virus families. We anticipate BACS to serve as a powerful tool to uncover the biological importance of {Psi} in future studies.

molecular biology↗

bioRxiv2024

Targeting glioblastoma tumor hyaluronan to enhance therapeutic interventions that regulate metabolic cell properties

Despite extensive advances in cancer research, glioblastoma (GBM) still remains a very locally invasive and thus challenging tumor to treat, with a poor median survival. Tumor cells remodel their microenvironment and utilize extracellular matrix to promote invasion and therapeutic resistance. We aim here to determine how GBM cells exploit hyaluronan (HA) to maintain proliferation using ligand-receptor dependent and ligand-receptor independent signaling. We use tissue engineering approaches to recreate the three-dimensional tumor microenvironment in vitro, then analyze shifts in metabolism, hyaluronan secretion, HA molecular weight distribution, as well as hyaluronan synthetic enzymes (HAS) and hyaluronidases (HYAL) activity in an array of patient derived xenograft GBM cells. We reveal that endogenous HA plays a role in mitochondrial respiration and cell proliferation in a tumor subtype dependent manner. We propose a tumor specific combination treatment of HYAL and HAS inhibitors to disrupt the HA stabilizing role in GBM cells. Taken together, these data shed light on the dual metabolic and ligand - dependent signaling roles of hyaluronan in glioblastoma. SignificanceThe control of aberrant hyaluronan metabolism in the tumor microenvironment can improve the efficacy of current treatments. Bioengineered preclinical models demonstrate potential to predict, stratify and accelerate the development of cancer treatments.

bioengineering↗

bioRxiv2023

Site Effects in Multisite Fetal Brain MRI: Morphological Insights into Early Brain Development

BackgroundMultisite MRI studies have become prevalent given their advantage in revealing reliable biological or clinical findings. Adult and adolescent multisite studies have demonstrated inevitable site-related non-biological effects that introduce bias. However, the site effect on fetal brain MRI remains unknown. PurposeTo identify crucial acquisition factors affecting fetal brain structural measurements and developmental patterns, while assessing the effectiveness of existing harmonization methods in mitigating site effects. Materials and MethodsBetween May 2017 and March 2022, T2-weighted fast spin-echo sequences in-utero MRI were performed in healthy fetuses from prospectively recruited pregnant volunteers on four different scanners at four sites. A generalized additive model (GAM) was used to quantitatively assess site effects, including field strength (FS), manufacture (M), in-plane resolution (R), and slice thickness (ST), on subcortical volume and cortical morphological measurements, including cortical thickness, curvature, and sulcal depth. Growth models were selected to elucidate developmental trajectories of these morphological measurements. Welchs test was performed to evaluate the influence of site effects on developmental trajectories. ComBat-GAM harmonization method was applied to mitigate site-related biases. ResultsThe final analytic sample consisted of 340 MRI scans from 218 fetuses (mean gestational age, 30.1 weeks {+/-} 4.4 [range, 21.7-40 weeks]). GAM results showed that low FS and low spatial resolution led to overestimations in selected brain regions of subcortical volumes and cortical morphological measurements, and cortical measurements were more susceptible to site effects than subcortical volumes. Only the peak cortical thickness in developmental trajectories was significantly influenced by the effects of FS and R. Notably, ComBat-GAM harmonization effectively removed site effects while preserving developmental patterns. ConclusionOur findings pinpointed the key acquisition factors in in-utero fetal brain MRI and underscored the necessity of data harmonization when pooling multisite data for fetal brain morphology investigations.

bioengineering↗

bioRxiv2023

Single cell sequencing as a general variant interpretation assay

The human genome contains [~]70 million possible protein-altering variants, the vast majority of which are of uncertain clinical significance. Closing this gap is essential for accurate diagnosis of disease-causing variants and understanding their mechanisms of action. Towards this goal, we developed a pooled perturbation approach combining saturation mutagenesis with single cell RNA sequencing to map the effects of every single nucleotide variant in a gene. We sequenced [~]440,000 cells expressing variants in CDKN2A (p16INK4a), TP53, and SOD1, observing almost all possible protein-coding variants, with a mean of 61 cells per variant. Using single cell gene expression signatures, we show that each gene may contain multiple types of pathogenic variants that affect distinct downstream pathways. We demonstrate that single cell expression signatures outperform existing bulk experimental assays and computational models for predicting pathogenicity, and summarize both the utility and potential limitations of single cell sequencing as a general variant interpretation assay.

genomics↗

bioRxiv2023

Pressure-Driven Insulin Release Overcomes Limitations of Diffusion for Encapsulated Islet Cell Therapy

Cadaveric islet and stem cell-derived transplantations hold promise as treatments for type 1 diabetes. To tackle the issue of immunocompatibility, numerous cellular macroencapsulation techniques have been developed that utilize diffusion to transport insulin across an immunoisolating barrier. However, despite several devices progressing to human clinical trials, none have successfully managed to attain physiologic glucose control or insulin independence. Based on empirical evidence, macroencapsulation methods with multilayered, high islet surface density are incompatible with homeostatic, on-demand insulin delivery and physiologic glucose regulation, when reliant solely on diffusion. An additional driving force is essential to overcome the distance limit of diffusion. In this study, we present both theoretical proof and experimental validation that applying pressure at levels comparable to physiological diastolic blood pressure significantly enhances insulin flux across immunoisolation membranes--increasing it by nearly three orders of magnitude. This significant enhancement in transport rate allows for precise, sub-minute regulation of both bolus and basal insulin delivery. By incorporating this technique with a pump-based extravascular system, we demonstrate the ability to rapidly reduce glucose levels in diabetic rodent models, effectively replicating the timescale and therapeutic effect of subcutaneous insulin injection or infusion. This advance provides a potential path towards achieving insulin independence with islet macroencapsulation. One Sentence SummaryTowards improved glucose control, applying sub-minute pressure at physiological levels enhances therapeutic insulin transport from macroencapsulated islets.

bioengineering↗