Additional file 1 of Overexpression of Lin28A in neural progenitor cells in vivo does not lead to brain tumor formation but results in reduced spine density

Additional file 1 Figure S1. Phenotype of hGFAP-cre::lsl-Lin28A mice (GL) compared to hGFAP-cre control mice (CTRL). (a) Scheme of mouse breeding and the lsl-Lin28A transgene. Mice carrying the hGFAP-cre promotor are crossed with lsl-Lin28A mice carrying the CAGGS-loxP-PolyA-loxP-LIN28A(3x)-IRES-eGFP sequence. Cre mediated recombination of loxP sites results in removal of the PolyA-Sequence (functional “STOP”-sequence). Consequently, LIN28A is constitutively overexpressed in neural progenitor cells of hGFAP-cre::lsl-Lin28A (GL) mice. (b-q) Fatemapping analyses of hGFAP-positive cells targeted by Cre using hGFAP-cre::R26tdRFPfl/+ mouse brains. Control forebrain (lsl-R26tdRFPfl/+) at P15 with high power images of the hippocampus region (c), stratum pyramidale (d) and isocortex (e). In contrast of the control, hGFAP-cre::R26tdRFPfl/+ mice displayed RFP-positive cells within the forebrain (j-m) at P15. High power images of the hippocampus (k) stratum pyramidale (l) and isocortex (m) are shown. Control cerebellum (lsl-R26tdRFPfl/+) at P15 with high power images of the granule cell layer (g), cerebellar layering (molecular cell layer, purkinje cell layer and granule cell layer) (h) and white matter (i). In contrast of the control, hGFAP-cre::R26tdRFPfl/+ mice displayed RFP-positive cells within the cerebellum (n-q) at P15. High power images of the granule cell layer (o) cerebellar layering (molecular cell layer, purkinje cell layer and granule cell layer) (p) and white matter (q) are shown. GL mice displayed no significant differences in body appearance (r) and body weight (s). GL mice displayed no significant difference in brain macroscopy (t), brain weight (u) and cortical thickness (v). Kaplan Meier analyses showing decreased survival of GL mice compared to CTRL mice (p=0.0028, log-rank test, n=83 for CTRL, n=58 for GL) (w). Scale bar in b is 1000 µm for b and j; scale bar in c is 500 µm for c and k; scale bar in d is 50 µm for d and l; scale bar in e is 200 µm for e and m; scale bar in f is 250 µm for f and n; scale bar in g is 50 µm for g and o; scale bar in h is 100 µm for h and p; scale bar in l is 50 µm for l and q. CTRL n=15, GL n=12 for s, CTRL n=83, GL n=58 for t. FB = forebrain, CB = cerebellum. P values are shown in figures as ns= not significant (p>0.05), * p<0.05, ** p<0.01, *** p<0.001, **** p<0.0001. Figure S2. Organ macroscopy and microscopy of GL mice. (a) Organ macroscopy reveals no obvious macroscopic differences of heart, kidney, spleen, intestine, liver, pancreas and lung morphology (P100). (b-aq) No tumor formation or obvious change in histomorphology in respective organs of GL mice (w-ac) compared to CTRL (P30) (b-h). LIN28A staining of CTRL (i-v) and GL (ad-aq) showed no positive cells both in heart and kidney in CTRL (i-l) and GL (ad-ag). Positive cells in spleen and intestine of CTRL (m-p) and GL (ah-ak) mice. In liver, pancreas and lung, single LIN28A-positive cells were detected in GL mice only (al-aq) when compared to CTRL mice (q-v), indicating minor Cre-activity. Scale bar in b is 2,5 mm for i, w and ad; scale bar in c is 250 µm for d-h, k, m, o, q, s, u (CTRL), and x-ac, af, ah, aj, al, an, ap (GL); scale bar in insert j is 100 µm for j, l, n, p, r, t, v (CTRL) and ae, ag, ai, ak, am, ao, aq (GL); arrows mark LIN28A positive cells. (ar) Differential blood counts of 16-week-old CTRL and GL mice. GL mice showed significant higher hemoglobin (p=0.0286, unpaired t-test, n=4 for CTRL, n=4 for GL) and hematocrit (p=0.0197, unpaired t-test, n=4 for CTRL, n=4 for GL) values, significant lower absolute (p=0.0286, unpaired t-test, n=4 for CTRL, n=4 for GL) and relative (p=0.0286, unpaired t-test, n=4 for CTRL, n=4 for GL) neutrophil granulocyte values and significant higher monocyte (p=0.0129, unpaired t-test, n=4 for CTRL, n=4 for GL) and lymphocyte (p=0.0459, unpaired t-test, n=4 for CTRL, n=4 for GL) values. Pap-staining on P15 displays no significant changes in cytomorphology of GL blood cells when compared to CTRL (as, at). CTRL n=4, GL n=4 for ar. Scale bar in inset of as is 5 µm for blood cell insets in as and at Figure S3. Analysis of mitotic phases and mitotic spindle orientation. (a) Representative pictures of mitotic phases. (b) Distribution of pro-/prometa-/metaphase, anaphase and telophase at the ventricular lining on E14.5 compared between CTRL and GL (p=0.6489, chi square test, each n=4). (c) Representative pictures of mitotic spindle orientation. (d) Scheme of mitotic spindle orientation by “division angle relative to the VZ”. (e) Distribution of mitotic spindle orientation in anaphase and telophase by vertical (60-90°), oblique (30-60°) and horizontal (0-30°) orientation at the ventricular lining on E14.5 compared between CTRL and GL (p=0.3325, chi square test, each n=4) (e). p/pm/m = prophase/prometaphase/metaphase, ana = anaphase, telo = telophase. Scale bar in a is 10 µm for a and c. Figure S4. Cortical layering in GL mice. (a-aj) Immunostainings of CTRL and GL cortices at E14.5, P0 and P15. During development (E14.5) and at P0, SOX2 positive are found in the ventricular zone both in CTRL (a, m) and GL (d, p) mice. At P15 glia cells are also positive (y, ab). Both in CTRL (b, n) and GL (e, q) mice TBR2-positive cells were detected in the subventricular zone at E14.5 and P0, but not at later stages (z, ac). TBR1 marked early born neurons in in CTRL (c) and GL (f) mice at E14.5 and layer VI cortical neurons both in CTRL (o, aa) and GL (r, ad) mice at postnatal stages. CTIP2-positive layer V cortical neurons were detected both in CTRL (s, ae) and GL (v, ah) mice at P0 and P15. REEL (Reelin) positive Cajal-Retzius cells were found in the marginal zone layer both in CTRL (h, t, af) and GL (k, w, ai) mice. NeuN was not detected on E14.5 (i, l) but is seen in mature cortical neurons both in CTRL (u, ag) and GL (x, aj) mice at P0 and P15. Scale bar in a is 50 µm for a-l; scale bar in m is 100 µm for m-x; scale bar in y is 250 µm for y-aj; scale bar for panel a is 25 µm for all panels from a-aj. Figure S5. Immunohistochemical analyses of hippocampal architecture in GL mice. (a-l) WFS and HUB stainings demonstrated similar formation of CA1 (WFS) and CA3 (HUB) areas in CTRL mice (a-f) and GL mice (g-l) at P15 despite granule cell dispersion in these regions. (m-aj) Immunohistochemical stainings of LIN28A, NeuN, REEL (Reelin), CTIP2, TBR2 and SOX2 of the stratum pyramidale in the CA1 region of CTRL (m-r) and GL (y-ad) and DG of CTRL (s-x) and GL (ae-aj). Strong LIN28A expression in the stratum pyramidale of CA1 region in GL (y, ae) mice in contrast to CTRL (m, s) mice. Similar to CTRL mice (n, t), GL mice (z, af) showed NeuN-positive neurons in the pyramidal cell layer and gyrus dentatus. Similarly, REEL-positive cells were detected in the molecular layer and Stratum oriens in CTRL (o, u) and GL mice (aa, ag). Strong CTIP2 expression in the DG of GL (ah) as well as in CTRL (v). Similar distribution of TBR2 positive cells in GL (ac, ai) mice compared to CTRL (q, w). Similar distribution of SOX2-positive cells in the subgranular zone of CTRL (r, x) and GL (ad, aj) mice. Similar distribution of Ki67-positive cells in the subgranular zone of CTRL (ak) and GL (al) mice on P15 and P30. Scale bar in a is 500 µm for a,d, g, j; scale bar in b is 200 µm for b, c, e, f, h, i, k, l; scale bar in m is 100 µm for m-r and y-ad; scale bar in s is 100 µm for s-x and ae-aj; scale bar in ak is 100 µm for ak and al. CA = cornu ammonis, DG = dentate gyrus; lines emphasize intactness of stratum pyramidale, arrows mark dispersion; PCL = pyramidal cell layer, GCL = granular cell layer, SGCL = subgranular cell layer Figure S6. GL mice show hyperkinetic activity. Modified open field test of CTRL (a) and GL (b) mice on P35 analyzed with the animal tracking software ToxTrac displays significant higher values for the parameters distance travelled (c, p=0.0054, unpaired t-test, n=12 for CTRL, n=10 for GL), average speed (d, p=0.0042, unpaired t-test, n=12 for CTRL, n=10 for GL) and average acceleration (e, p=0.0099, unpaired t-test, n=12 for CTRL, n=10 for GL) in GL mice when compared to CTRL. Anxiety parameters such as exploration rate (f), time spent in center (g) and vertical activity (h) did not reveal significant differences. CTRL n=12, GL n=10 for c-h. P values are shown in figures as ns= not significant (p>0.05), * p<0.05, ** p<0.01, *** p<0.001, **** p<0.0001 Figure S7. Western Blot Analyses of selected Shh and Wnt targets. (a, b) Western blot and quantification of APC, GLI2, AXIN2, MYCN, MYC and LIN28A in cortices of GL and CTRL mice at P15. HSC70 and a-TUBULIN served as housekeeping proteins. Every sample number represents an individual mouse. Significantly increased LIN28A protein expression in GL mice compared to CTRL (p=0.0001, unpaired t-test, n=4 for CTRL, n=4 for GL). No significant changes in APC, GLI2, AXIN2, MYCN and MYC in GL mice compared to CTRL (p=0.2863, p=0.2859, p=0.1794, p=0.8252 and p=0.9872, respectively). P values are shown in figures as ns= not significant (p>0.05), * p<0.05, ** p<0.01, *** p<0.001, **** p<0.0001