Sang Kee Kang

Identification of a peptide that interacts with Nestin protein expressed in brain cancer stem cells

Glioma stem cells (GSCs) are presumably major culprits for brain tumor initiation, progression, and recurrence after conventional therapies. Thus, selective targeting and eradication of GSCs may provide a promising and effective therapeutic approach. Here, we isolated a GSC-targeting (GSCT) peptide that demonstrated selective binding affinity for many undifferentiated GSCs using in vitro phage display technology. This GSCT peptide binds to isotypes of Nestin proteins specifically expressed in GSCs, enabling it to target Nestin-positive cells in human glioblastoma tissues. In human glioblastoma tissue specimens, the fluorescence-conjugated GSCT peptide could visualize putative GSC populations, showing its possible use as a diagnostic agent. GSCT peptide is also internalized into undifferentiated GSCs specifically in vitro, and moreover, intravenously injected GSCT peptide effectively penetrated into tissues, specifically accumulated in gliomas that arise from subcutaneous and orthotopic implantation, and predominantly targeted Nestin-positive cells in these tumors. Thus, our GSCT peptide may be useful for the development of more promising therapeutic and diagnostic modalities that target GSCs in brain tumors.

Lactobacillus acidophilus Expressing Recombinant K99 Adhesive Fimbriae Has an Inhibitory Effect on Adhesion of Enterotoxigenic Escherichia coli

The most common enteric colibacillosis in neonatal and newborns is caused by enterotoxigenic Escherichia coli (ETEC). Colonization of ETEC in the small intestine is associated with adhesions using fimbriae, which is known as a specific adhesion factor and provides highly specific means for anchoring and prerequisite for an infectious agent. In the present study we have engineered Lactobacillus acidophilus to produce recombinant K99 fimbriae, which is used for the colonization to the intestine of pigs. The expression of K99 fimbrial protein was confirmed using SDS‐PAGE, immunoblot and agglutination analyses. To evaluate a function of the K99 fimbrial protein, inhibition and competition tests were performed on pre‐screened intestinal brush border from pigs. The tests showed that recombinant L. acidophilus, not control L. acidophilus, had a significant inhibitory effect to and competition against K99+ E. coli in a dose dependent manner. In conclusion, we demonstrated that recombinant K99 fimbriae producing L. acidophilus was able to prevent E. coli binding to intestinal brush border.

Differentially expressed proteins during fat accumulation in bovine skeletal muscle.

The objective of this study was to identify the proteins involved in bovine intramuscular fat (IMF) development. Global proteins were monitored in bovine skeletal muscle at muscle-developing versus IMF-increasing stages and with higher versus lower IMF scores, respectively. We identified two differentially expressed (two-fold or more) proteins at the IMF-increasing stage, up-regulated heat shock protein beta 1 (HSPB1) and down-regulated ATP synthase D chain (ATP5H), and two down-regulated proteins with higher IMF scores, carbonic anhydrase 2 (CA2) and myosin light chain 3 (MYL3). In vitro, after adipogenic differentiation, the mRNA expression of HSPB1 and ATP5H did not be changed, but that of CA2 and MYL3 decreased significantly (P<0.05). After myogenic differentiation, the mRNA expression of HSPB1 increased significantly (P<0.05), but expression of other genes did not vary. We suggested that CA2 and MYL3, which expressed down during adipogenic differentiation, could be indicative markers for negative regulation of IMF development.

Artificial neural network models for prediction of intestinal permeability of oligopeptides

BackgroundOral delivery is a highly desirable property for candidate drugs under development. Computational modeling could provide a quick and inexpensive way to assess the intestinal permeability of a molecule. Although there have been several studies aimed at predicting the intestinal absorption of chemical compounds, there have been no attempts to predict intestinal permeability on the basis of peptide sequence information. To develop models for predicting the intestinal permeability of peptides, we adopted an artificial neural network as a machine-learning algorithm. The positive control data consisted of intestinal barrier-permeable peptides obtained by the peroral phage display technique, and the negative control data were prepared from random sequences.ResultsThe capacity of our models to make appropriate predictions was validated by statistical indicators including sensitivity, specificity, enrichment curve, and the area under the receiver operating characteristic (ROC) curve (the ROC score). The training and test set statistics indicated that our models were of strikingly good quality and could discriminate between permeable and random sequences with a high level of confidence.ConclusionWe developed artificial neural network models to predict the intestinal permeabilities of oligopeptides on the basis of peptide sequence information. Both binary and VHSE (principal components score V ectors of H ydrophobic, S teric and E lectronic properties) descriptors produced statistically significant training models; the models with simple neural network architectures showed slightly greater predictive power than those with complex ones. We anticipate that our models will be applicable to the selection of intestinal barrier-permeable peptides for generating peptide drugs or peptidomimetics.

Expression and characterization of human growth hormone–Fc fusion proteins for transcytosis induction

The major obstacle for oral delivery of administered therapeutic proteins is malabsorption in the intestine. This malabsorption could be overcome by induction of neonatal FcRn [Fc (CH2 and CH3 domains of human IgG1 antibody) receptor]‐mediated transcytosis in the intestine using recombinant fusion of CH2 and CH3 moieties of human IgG to a therapeutic protein. To this end we developed recombinant hGH (human growth hormone) fused to the N‐terminus of Fc moieties [CH2‐CH3 or h (hinge)‐CH2‐CH3] from human IgG1. These recombinant proteins secreted by the methylotrophic yeast Pichia pastoris functionally induced secretion of insulin‐like growth factor 1 by HepG2 cells in the response to hGH moiety in the fusion proteins. In a transport study using polarized T84 cells, 3.7% of added dimeric hGH–h‐Fc was transported in the apical‐to‐basolateral direction within 1 h by FcRn‐mediated transcytosis of 1 cm2 monolayers. However, transport of monomeric hGH–Fc (only 0.43%) was much less effective, yet its transport was 2.3 times higher than that of hGH. Finally, we concluded that, upon recombinant fusion, maintenance of dimeric structure of Fc moieties is crucial for the induction of FcRn‐mediated transcytosis.

Identification of tissue-specific targeting peptide

Using phage display technique, we identified tissue-targeting peptide sets that recognize specific tissues (bone-marrow dendritic cell, kidney, liver, lung, spleen and visceral adipose tissue). In order to rapidly evaluate tissue-specific targeting peptides, we performed machine learning studies for predicting the tissue-specific targeting activity of peptides on the basis of peptide sequence information using four machine learning models and isolated the groups of peptides capable of mediating selective targeting to specific tissues. As a representative liver-specific targeting sequence, the peptide “DKNLQLH” was selected by the sequence similarity analysis. This peptide has a high degree of homology with protein ligands which can interact with corresponding membrane counterparts. We anticipate that our models will be applicable to the prediction of tissue-specific targeting peptides which can recognize the endothelial markers of target tissues.

Heat-shock protein beta 1 regulates androgen-mediated bovine myogenesis

To elucidate the functional significance of heat-shock protein beta 1 (HSPB1) in androgen-mediated myogenesis of bovine cells, we conducted ‘loss and gain of function of HSPB1’ assays by siRNA inhibition and gene overexpression. siRNA inhibition of HSPB1 expression reduced the expression of desmin (a myogenic marker) and repressed the formation of myotubes in cells induced for myogenic differentiation. In contrast, overexpression of HSPB1 enhanced the expression of desmin and accelerated formation of myotubes. The loss and gain of HSPB1 function was closely associated with the expression level of androgen receptor (AR). Our findings suggest that HSPB1 mediates androgen signaling by binding directly to AR and then enhancing androgen-mediated myogenesis in myogenic cells.

Differentially expressed proteins associated with myogenesis and adipogenesis in skeletal muscle and adipose tissue between bulls and steers

The objective of this study was to identify some proteins associated with testosterone-related differences in myogenesis and adipogenesis between bulls and steers. Global proteins were monitored in skeletal muscle and adipose tissue from bulls (nxa0=xa020) and steers (nxa0=xa020), respectively. We identified four differentially expressed (twofold or more) proteins in skeletal muscle from bulls, myosin light chain 1 (MLC1), ankyrin repeat domain-containing protein 1 (ANKRD1) and heat shock protein beta 1 (HSPB1) that were up-regulated and cofilin 2 (CFL2) that was down-regulated, and also identified two down-regulated proteins in adipose tissue, transaldolase 1 (TALDO1) and l-lactate dehydrogenase B chain (LDHB). In vitro, after myogenic differentiation of a bovine cell line, the mRNA expression of HSPB1 not only increased approximately tenfold in response to differentiation but threefold in response to testosterone addition, respectively, but that of ANKRD1 and CFL2 did not significantly change in response to myogenic differentiation or testosterone addition. Likewise, after adipogenic differentiation of a bovine cell line, the mRNA expression of TALDO1 and LDHB did not significantly vary in response to adipogenic differentiation or testosterone addition. Therefore, we suggest that HSPB1 could have an important role during testosterone-related myogenesis.

Identification of a novel peptide ligand targeting visceral adipose tissue via transdermal route by in vivo phage display

To find novel peptide ligands targeting visceral adipose tissue (visceral fat) via transdermal route, in vivo phage display screening was conducted by dermal administration of a phage-peptide library to rats and a peptide sequence, CGLHPAFQC (designated as TDA1), was identified as a targeting ligand to visceral adipose tissue through the consecutive transdermal biopannings. Adipocyte-specific affinity and transdermal activity of the TDA1 were validated in vitro and targeting ability of the dermally administered TDA1 to visceral adipose tissue was also confirmed in vivo. TDA1 was effectively translocated into systemic circulation after dermal administration and selectively targeted visceral adipose tissue without any preference to other organs tested. Fluorescent microscopic analysis revealed that the TDA1 could be specifically localized in the hair follicles of the skin, as well as in the visceral adipose tissue. Thus, we inferred that dermally administered TDA1 would first access systemic circulation via hair follicles as its transdermal route and then could target visceral fat effectively. The overall results suggest that the TDA1 peptide could be potentially applied as a homing moiety for delivery of anti-obesity therapeutics to visceral fat through the convenient transdermal pathway.

Machine learning study for the prediction of transdermal peptide

In order to develop a computational method to rapidly evaluate transdermal peptides, we report approaches for predicting the transdermal activity of peptides on the basis of peptide sequence information using Artificial Neural Network (ANN), Partial Least Squares (PLS) and Support Vector Machine (SVM). We identified 269 transdermal peptides by the phage display technique and use them as the positive controls to develop and test machine learning models. Combinations of three descriptors with neural network architectures, the number of latent variables and the kernel functions are tried in training to make appropriate predictions. The capacity of models is evaluated by means of statistical indicators including sensitivity, specificity, and the area under the receiver operating characteristic curve (ROC score). In the ROC score-based comparison, three methods proved capable of providing a reasonable prediction of transdermal peptide. The best result is obtained by SVM model with a radial basis function and VHSE descriptors. The results indicate that it is possible to discriminate between transdermal peptides and random sequences using our models. We anticipate that our models will be applicable to prediction of transdermal peptide for large peptide database for facilitating efficient transdermal drug delivery through intact skin.