Figure 1.
Visualization of discrete bcd mRNA particles with fluorescently labeled oligonucleotides.
(A–G) Confocal slices through the midsagittal plane at the anterior (A) and posterior (G) of a wild-type embryo during nuclear cycle (n.c.) 4. Copper-shaded boxes in (A) and (G) indicate regions shown in magnified views (B–F); top row, near embryo surface; lower row, center views; arrowheads indicate selected dimly fluorescent particles. (G inset) Low magnification image of whole embryo (mRNA in green, DNA in blue), boxed regions are shown in (A) and (G). (H) Six sequential confocal z-slices beginning 0.8 µm below (leftmost panel) and ending 1.2 µm above (rightmost panel) the image shown in (C); lower panel in (C) is identical to third panel from the left in (H). Arrows: bright particles appearing on five slices. Arrowheads: weak particles appearing on three slices. (I–K) Confocal slices of the same embryo imaged ∼5 µm beneath cortical surface. (J) Magnified views of selected regions indicated by six shaded boxes in (I) and (K). Here and in Figure 2, posterior regions containing particles were deliberately chosen for zoomed views; however, the majority of the posterior does not contain detectable particles. Scale bars: 25 µm (A), 2 µm (B, H).
Figure 2.
bcd mRNA particles surrounding surface nuclei during syncytial blastoderm stages.
(A–F) n.c. 11 embryo near the midsagittal plane at the anterior (A) and posterior (F). Shaded boxes indicate magnified views (B–E) corresponding to cortex (upper) or core (lower); arrowheads indicate selected faint particles. (F inset) Low magnification image of whole embryo (mRNA in green, DNA in blue), boxes indicate regions shown in (A) and (F). (G–L) Same embryo imaged at the nuclear layer. (H–K) Selected z-slices of boxed regions indicated in (G) and (L) showing apical (upper panels) and basal (lower panels) planes surrounding the nuclear layer (middle panels). Arrowheads indicate selected dimly fluorescent particles. Scale bars: 25 µm (A), 2 µm (B).
Figure 3.
Quantification of diffraction-limited bcd mRNA particles.
Individual particles identified in anterior stacks of 32 embryos hybridized with fluorescently labeled bcd oligonucleotides ranging from n.c. 4 to n.c. 14 (see color code in legend). (A) bcd particle eccentricity e calculated as the ratio of the difference of the major and minor particle axes (rx and ry) to their sum; e = 0 and e = 0.3 correspond to rx/ry = 1 (perfectly circular) and rx/ry = 2, respectively. Shown is the probability distribution of e across all identified particles. (B) Probability distribution of bcd particle diameters d with a mean of 3.05±0.10 pxl and a standard deviation of 0.42±0.05 pxl (1 pixel = 75.7 nm). (C) Distribution of bcd particle intensities (note log units on x-axis). (D) Average bcd particle intensity as a function of distance along the AP axis in fractional egg length x/L. (E) Distribution of fluorescence intensities in bcd mRNA particles along the AP axis; error bars are standard deviations across embryos (note difference from total bcd particle distribution in Figure S4). (F) Cumulative distribution of bcd mRNA fluorescence intensity (cumulative plot of (E)).
Figure 4.
bcd mRNA particle dynamics, disassembly, and stability.
All panels are generated on same data set as in Figure 3, with identical color labeling for different nuclear cycles. (A) bcd particle density as a function of fractional distance from the cortex d/L (d is the red line in inset of partial image of DAPI stained embryo) in a single midsagittal plane. (B) Probability distribution of bcd particle intensities along the z-axis within a confocal stack (error bars are standard deviations across embryos). (C) Absolute number of bcd particles as a function of n.c. Each dot corresponds to total number count in full anterior stack of an individual embryo (stack encompasses left or right embryo-half; due to left-right symmetry, the full embryo contains twice the number of particles indicated). Black squares and error bars correspond to averages and standard deviations across embryos at a given n.c., respectively (black line is a guide to the eye). Note the sharp drop during n.c. 14 (time point corresponds to 15 min after the end of mitosis 13; the actual average particle count below 50). (D) Average total bcd particle intensity as a function of n.c. (black squares; error bars are standard deviations across embryos in a given n.c.). In color, for individual embryos, mean bcd particle intensity as a function of n.c. (green scale on right).
Figure 5.
Uniform bcd mRNA particle degradation at the onset of cellularization.
Each row shows an embryo during early (0–5 min, top row), mid- (5–10 min, middle row), or late (10–20 min, bottom row) n.c. 14 (times are approximate after mitosis 13). (A–C) Anterior midsagittal views. Boxed regions are magnified in (D–F). (G–O) Images of nuclear layer at embryo cortex; (G–I), anterior; (M–O), posterior; (J–L) magnified views of boxed regions apical (top panels) and basal (bottom panels) to nuclei (middle panels). Arrowheads indicate selected faint particles. Box in (F) highlights a cluster of particles. Scale bars: 25 µm (A), 5 µm (B), and 2 µm (J).
Figure 6.
Presyncytial blastoderm stage (A–D) and syncytial stage (E–I) embryos expressing Bcd-GFP. Images are maximum-projections of confocal stacks to display nuclei residing in several focal planes (A–D) or midsagittal confocal slices (E–I, 7 µm thick sections). Nuclear cycle (nc) is indicated. DAPI staining shown (A′–C′) to indicate positions of nuclei. Scale bar: 50 µm. Arrowheads in (A) show nuclear Bcd-GFP in n.c. 6, i.e. the earliest nuclear Bcd-GFP detected in any examined embryo.
Figure 7.
(A–B) Mean attenuation-corrected nuclear Bcd-GFP gradients as a function of relative position along the AP axis extracted from 35 embryos expressing Bcd-GFP fixed in n.c. 6 to n.c. 14; each dot corresponds to an individual nucleus (see Materials and Methods for image analysis procedures); color indicates nuclear cycle (see legend). Insets show log-linear plots of same data; slopes of fitted lines indicate a length constant of ∼0.15 EL from n.c. 8 onward. From n.c. 10 onward, GFP intensities for cortically localized dorsal nuclei only are shown for clarity. (C–D) Same data as in (A–B) but plotted as a function of developmental time (nuclear cycles) with copper shading indicating AP position as fraction of EL. (C) Nuclear GFP intensities as a function of time. Lines connect binned means of 50 equally spaced intervals along the AP axis. (D) Binned means normalized to the maximum attained in each bin during development, plotted as a function of time.
Figure 8.
Bcd-GFP gradients in fixed and living embryos.
Living Bcd-GFP (A–B) and Histone-RFP embryos (C–D) were imaged at blastoderm stages using confocal microscopy. Embryos were fixed within 3 min of live imaging, then subsequently re-imaged under the same microscopy settings (midsagittal slices of the same embryo at n.c. 13 are shown live in (A, C) and fixed in (B, D)). (E–F) Mean nuclear fluorescence intensities in living (red) and fixed (blue) embryos expressing Bcd-GFP (E) and Histone-RFP (F) as a function of position along AP axis. (G) Live and fixed Bcd-GFP gradients rescaled by 3-fold. (H) Scatter plot of fixed (x-axis) and 3-fold rescaled live (y-axis) mean intensities in equally spaced bins along the AP axis in embryos expressing Bcd fusions to GFP, Venus, or RFP. Dashed line indicates 1-to-1 correspondence. Error bars for Bcd-GFP are standard deviations within each bin. To plot curves on the same axes, values were normalized to maximum.
Figure 9.
Extended SDD models with realistic 3d geometry and source distribution.
(A) A well-fitting model with parameters as in Figure S10: solid lines are measured Bcd protein gradients from n.c. 7 (dark blue) to n.c. 14 (red); dashed line come from a single model fit. (B) Log-log plot of the measurement versus the model prediction at different developmental times (colors as in (A)); dashed black line corresponds to perfect overlap. (C) The best-fitting model when an mRNA point source at the anterior pole is assumed. Despite having the same freedom to pick the model parameters as in (B), the best-found model for the point source deviates from the data substantially.