Fig 2: CUX1 and ITSN1 are target genes of miR-193a-5p and associated with poor prognosis of colorectal cancer (CRC). (A) The protein expression analysis of CUX1 and ITSN1 in CRC cells after transfection with miR-193a-5p mimics or miR-NC. (B) The mRNA expression analysis of CUX1 and ITSN1 in HCT-8 and SW480 cells after transfection with miR-193a-5p mimics or miR-NC (*p < 0.05, **p < 0.01, ***p < 0.001). (C,D) Dual-luciferase reporter assay. Reporter constructs containing either wild-type CUX1/ITSN1 3′-UTR or CUX1/ITSN1 3′-UTR with mutation at the predicted miR-193a-5p target sequence were transfected into CRC cells, along with miR-193a-5p mimics or miR-NC; the relative luciferase activity was assayed (*p < 0.05, **p < 0.01, ***p < 0.001). (E,F) The overall survival (OS) of CRC patients stratified by CUX1 and ITSN1 expression levels.

Fig 3: β-catenin mediated the effect of CUX1 on EMT process, migration, and invasion.a WB assay showed that β-catenin overexpression could reverse the decreased EMT process induced by siCUX1 in U87 and U251 cells (*P < 0.05). b Transwell and Boyden assays indicated that β-catenin overexpression could reverse the decreased migratory and invasive capacity induced by siCUX1 in U87 and U251 cells (***P < 0.0001).

Fig 4: Absence of neuroserpin does not alter adult neocortical organization. Cortical layers II–IV, V and VI were visualized in the somatosensory cortex using projection neuron markers Cux1, Ctip2, and Tbr1, respectively. Distribution of GABAergic interneurons was assessed with immunostainings against Reelin, SST and PV. No obvious differences could be detected between neuroserpin-deficient mice and wild-type littermates. Scale bars: 100 µm, n = 3.

Fig 5: Neuroserpin expression during cortical development. Representative immunohistochemical staining of the somatosensory cortex with an antibody directed against neuroserpin (Ns) and markers of cortical layer I (Reelin), II–IV (Cux1) and V–VI (Tbr1) at E13 (A), E16 (B), P0 (C) and at 1 week of age (D). At E13, upper cortical layers with Cux1-positive neurons have not yet been generated. Neuroserpin immunoreactivity is homogeneously observed throughout the developing cortical plate, whereas the Reelin-positive marginal zone is devoid of neuroserpin. In the postnatal period, neuroserpin-immunopositive cells are still distributed throughout cortical layer II–VI, around many Cux1-positive nuclei, whereas most Tbr1-positive neurons in the deeper layers do not express neuroserpin (arrows in D). Moreover, colocalization with Reelin-positive interneurons is observed starting from P0 (arrowheads in C). Staining of a cortical section from a Ns-/- mouse demonstrates specificity of the neuroserpin immunohistochemistry. White boxes represent the area shown at higher magnification in the top row. Scale bars: 50 µm in (A–C), 100 µm in (D). n = 3. MZ, marginal zone; CP, cortical plate; SP, subplate.