Biology subjects

Nauman, M.

Publications and source records attributed to Nauman, M..

5 recordsLinked to original sources

bioRxiv2026

Absence of EOGT Precludes Defective Development in Fringe-null Mouse Intestine

Identifying biological roles for glycosyltransferases is a continuing challenge and important for defining morbidities associated with congenital disorders of glycosylation. Here we investigate the consequences to intestinal development of conditionally deleting Lfng alone or Lfng, Mfng and Rfng together in a mixed or Eogt-null genetic background. Each Fringe transfers N-acetylglucosamine (GlcNAc) to fucose (Fuc) attached to Ser or Thr by POFUT1 in a consensus sequence found in certain epithelial growth factor-like (EGF) repeats. EOGT transfers GlcNAc directly to Ser/Thr in a separate consensus sequence of the EGF repeat. Notch receptors and Notch ligands contain the largest number of EGF repeats with consensus sites for these O-glycans. Conditional deletion of Pofut1 in mouse intestine causes similar developmental defects to deletion of Notch1 and Notch2 or Dll1 and Dll4. LFNG also contributes to optimal Notch signaling in mouse intestine. In this work, we generated Lfng[F/F]:Villin-Cre and Lfng[F/F]Mfng[-/-]Rfng[-/-]:Villin-Cre mice in which extension of O-Fuc on EGF repeats was inhibited or prevented in intestinal epithelium. Conditional deletion of either Lfng alone or all three Fringe activities together led to defective intestinal development with a marked increase in goblet and Paneth cells, increased crypt width and reduced villus length. Unexpectedly, in mice globally lacking EOGT, conditional inactivation of the three Fringe genes did not lead to defective intestinal development. Thus, the absence of EOGT prevented disruption of development in Fringe-null intestine, identifying a novel role for EOGT in regulating intestinal development.

cell biology↗

bioRxiv2026

Metastatic dissemination of breast cancer stem cells requires MenaINV for lung extravasation but not survival

Cancer stemness is a pivotal driver of tumor initiation, treatment resistance, and tumor cell survival. Cancer stem cells (CSCs), though constituting only a small fraction of primary tumor cells, are progressively enriched during metastatic progression: from circulating tumor cells traveling in the bloodstream, to disseminated tumor cells lodged in the lung vasculature, to extravasated tumor cells that have entered tissue parenchyma. However, whether CSCs have an intrinsic advantage for extravasation over cancer non-stem cells (CnSCs), or simply their increased representation in circulation renders them more likely to extravasate, remains unresolved. MenaINV, an invasive isoform of the actin regulatory protein Mena, promotes tumor cell transendothelial migration in primary and secondary sites, yet the direct mechanistic link between stemness and MenaINV in lung metastasis remains unresolved. Here, using a validated fluorescent stemness reporter (SORE6) to identify CSCs, we found that CSCs display elevated MenaINV expression relative to CnSCs. High-resolution intravital imaging showed that CSCs extravasate efficiently into lung parenchyma and survive at higher levels, robustly forming metastatic lesions, while CnSCs show limited extravasation, low survival, and poor colonization. Mechanistically, MenaINV disruption in CSCs specifically impaired extravasation without affecting survival, demonstrating that MenaINV is the key extravasation effector downstream of stemness, whereas stemness-associated factors independently confer survival advantages. Moreover, reintroduction of MenaINV in CnSCs restores their extravasation ability upon which extravasated CnSCs reactivate stem program and form metastases. Overall, we discovered a hierarchical framework where stemness regulates both survival and extravasation capacity, with MenaINV as the key CSC extravasation effector. Significance: This study reveals how breast cancer stem cells achieve metastatic dominance through separable pathways: MenaINV-dependent extravasation and MenaINV-independent survival, providing rationale for targeting stem program to improve patient outcome.

cancer biology↗

bioRxiv2023

EOGT Enables Residual Notch Signaling in Mouse Intestinal Cells Lacking POFUT1

Notch signaling determines cell fates in mouse intestine. Notch receptors contain multiple epidermal growth factor-like (EGF) repeats modified by O-glycans that regulate Notch signaling. Conditional deletion of protein O-fucosyltransferase 1 (Pofut1) substantially reduces Notch signaling and markedly perturbs lineage development in mouse intestine. However, mice with inactivated Pofut1 are viable, whereas complete elimination of Notch signaling in intestine is lethal. Here we investigate whether residual Notch signaling enabled by EOGT permits mice lacking Pofut1 in intestine to survive. Mice globally lacking Eogt alone were grossly unaffected in intestinal development. In contrast, mice lacking both Eogt and Pofut1 died at [~]28 days after birth with greater loss of body weight, a greater increase in the numbers of goblet and Paneth cells, and greater downregulation of Notch target genes, compared to Pofut1 deletion alone. These data establish that both O-fucose and O-GlcNAc glycans are fundamental to Notch signaling in the intestine and provide new insights into roles for O-glycans in regulating Notch ligand binding. Finally, EOGT and O-GlcNAc glycans provide residual Notch signaling and support viability in mice lacking Pofut1 in the intestine.

developmental biology↗

bioRxiv2020

Contrasting population genetics of cattle- and buffalo- derived Theileria annulata causing tropical theileriosis

The present study was designed to improve understanding of Theileria annulata in sympatric water buffalo and cattle in the Punjab province of Pakistan. The prevalence of tropical theileriosis is high, buparvaquone resistance is widespread, and vaccine protection is poor in the field. Better understanding is, therefore, needed of the factors that influence the genetics of T. annulata populations both within its hosts and in its overall populations. Here we utilise a panel of six satellites and a mitochondrial cytochrome b marker to explore the multiplicity of T. annulata infection and patterns of emergence and spread of different parasite genotypes. Parasite materials were collected from infected animals in defined regions, where water buffalo and cattle are kept together. Our results show that T. annulata is genetically more diverse in cattle- than in water buffalo-derived populations (the mean numbers of unique satellite alleles were 13.3 and 1.8 and numbers of unique cytochrome b locus alleles were 65 and 27 in cattle- and water buffalo-derived populations, respectively). The data show a high level of genetic diversity among the individual host-derived populations (the overall heterozygosity (He) indices were 0.912 and 0.931 in cattle, and 0.874 and 0.861 in buffalo, based on satellite and cytochrome b loci, respectively). When considered in the context of high parasite transmission rates and frequent animal movements between different regions, the predominance of multiple T. annulata genotypes, with multiple introductions of infection in the hosts from which the parasite populations were derived, may have practical implications for the spread of parasite genetic adaptations; such as those conferring vaccine cross-protection against different strains affecting cattle and buffalo, or resistance to antiprotozoal drugs.

genetics↗

bioRxiv2017

Beyond Homology Transfer: Deep Learning for Automated Annotation of Proteins

Accurate annotation of protein functions is important for a profound understanding of molecular biology. A large number of proteins remain uncharacterized because of the sparsity of available supporting information. For a large set of uncharacterized proteins, the only type of information available is their amino acid sequence. In this paper, we propose DeepSeq - a deep learning architecture - that utilizes only the protein sequence information to predict its associated functions. The prediction process does not require handcrafted features; rather, the architecture automatically extracts representations from the input sequence data. Results of our experiments with DeepSeq indicate significant improvements in terms of prediction accuracy when compared with other sequence-based methods. Our deep learning model achieves an overall validation accuracy of 86.72%, with an F1 score of 71.13%. Moreover, using the automatically learned features and without any changes to DeepSeq, we successfully solved a different problem i.e. protein function localization, with no human intervention. Finally, we discuss how this same architecture can be used to solve even more complicated problems such as prediction of 2D and 3D structure as well as protein-protein interactions.

bioinformatics↗