Cellular and genetic mechanisms that shape the development and evolution of tail vertebral proportion in mice and jerboas

Abstract

Limbs and vertebrae elongate by endochondral ossification, but local growth control is highly modular such that not all bones are the same length. Compared to limbs, which have a different evolutionary and developmental origin, far less is known about how individual vertebrae establish proportion. Using the jerboa and mouse tail skeletons, we find that cell number is a common driver of limb and vertebral proportion in both species. However, chondrocyte hypertrophy, which is a major driver of proportion in all mammal limbs, is limited to the extreme disproportionate growth of jerboa mid-tail vertebrae. The genes associated with differential growth in the vertebral skeleton overlap significantly, but not substantially, with genes associated with limb proportion. Among shared candidates, loss of Natriuretic Peptide Receptor 3 in mice causes disproportionate elongation of the proximal and mid-tail vertebrae, in addition to the proximal limb. Our findings therefore, reveal cellular processes that tune the growth of individual vertebrae while also identifying natriuretic peptide signaling among genetic control mechanisms that shape the entire skeleton.

Fig. 1. Development of mouse and jerboa tail proportion.

A, B Diagram of adult mouse (A) and jerboa (B) skeletons adapted from Moore et al. 2015. C Alcian-stained neonates show all proximal vertebral elements are present at birth (P0) in both species. Adult proximal tail skeletons have similar vertebral morphologies despite differences in size and proportion. D Mouse (teal) and jerboa (pink) tails are approximately half the naso-anal length at birth. Tail proportion diverges by P21; the mouse tail remains about equal to body length, while the jerboa tail elongates to 1.5-times the body length. E–H µCT scans were used to measure vertebral centra lengths weekly from birth to six weeks, normalized to the naso-anal length of each mouse (E) and jerboa (F). E, F Points represent the mean vertebral lengths at each position and timepoint with error bars showing the standard deviation of these measurements. Each timepoint from P0 to postnatal day 42 (P42) is represented by a different shape and color for mouse (blues) and jerboa (pinks). The weekly relative change in length of each vertebra, normalized to naso-anal length, is represented in a heat map. The greatest rate of change is yellow and least in dark blue with the scale equivalent for mouse (G) and jerboa (H). D–H Six male and female animals were measured for each time point, except jerboa P0 and P42 where n = 5. Source data are provided as a Source Data file.

Fig. 2. Cellular parameters of growth during the greatest difference in vertebral elongation rate.

A Schematic of tail vertebrae with cranial and caudal growth cartilage color schemes used throughout. B, C Hematoxylin and eosin staining of mouse and jerboa cranial and caudal growth cartilages of tail vertebra 1 (TV1) (B) and TV6 (C) during rapid elongation. (RZ resting zone, PZ proliferative zone, HZ hypertrophic zone). Scale bar is 100 µm. D–I Growth cartilage parameters in each cranial and caudal cartilage of mouse and jerboa TV1 and TV6. D Daily elongation rate, n = 8 for mouse TV1 caudal and mouse TV6 cranial and caudal, n = 9 for mouse TV1 cranial and all jerboa TV1 and TV6 cartilages. E Height of the growth cartilage. F Height of the proliferative zone. G Proliferative index calculated as the fraction of EdU+ cells of all cells in an ROI. For mouse cartilages n = 9, and n = 8 for all jerboa cartilages. H Height of the hypertrophic zone. I Maximum height of hypertrophic chondrocytes in the direction of bone elongation. D–I Welch’s t-test, two-tailed, * ≤ 0.05, ** ≤ 0.01, *** ≤ 0.001, **** ≤ 0.0001. Exact p-values are available in Source Data file. E, F, H, I n = 8 for mouse TV1 cranial and caudal cartilages, n = 9 for mouse TV6 and all jerboa TV1 and TV6 cartilages, all mixed male and female animals. D–F, H, I Error bar centers are the data means and error bands show standard deviation. G Box and whisker plot boxes extended from the 25th to 75th percentiles with whiskers showing the minimum and maximum values, the center line is the median, and all points are plotted. Source data are provided as a Source Data file.

Fig. 3. Design and analysis of intersectional interspecies transcriptomics.

A–C Schematics of the interspecies and intraspecies comparison approaches of differential expression analyses. D 1864 genes are significantly differentially expressed (adjusted p-value using Benjamini–Hochberg (BH) method; padj < 0.05) between mouse and jerboa TV6 but not between species in TV1. Genes that are expressed higher in jerboa TV6 are orange; lower are purple. E 6786 genes are differentially expressed (padj < 0.05) between jerboa and mouse both in TV6 (y-axis) and in TV1 (x-axis); most are equivalently differentially expressed (gray points, slope = 0.883), but 421 genes are outside of the 95% prediction interval and designated “disproportionately differentially expressed”. F Of all 2285 genes highlighted in (D and E), 1454 are also differentially expressed in jerboa TV6 versus TV1, consistently in the same direction in the longest versus shortest element. Genes in orange are expressed higher in jerboa TV6 compared to mouse TV6 and in jerboa TV6 compared to jerboa TV1, while those lower in both comparisons are in purple. G Selection of GO terms relevant to cartilage that are enriched among the 1454 candidate genes. clusterProfiler uses hypergeometric distribution to calculate gene enrichment, adjusted p-values are calculated using the BH method. Source data are provided in Supplementary Data file.

Fig. 4. Natriuretic peptide signaling affects mouse tail vertebrae proportion.

A Ratio of tail length to naso-anal distance in postnatal (P7) and adult (P42) Npr3^−/− mice (triangles, yellow/gold) and wild-type Npr3^+/+ siblings (squares, blue/turquoise), where solid lines indicate means, compared to average relative tail lengths in CD1 mice (gray circles, error bars show the standard deviation). Two-way ANOVA performed for P42, p = 0.0491. B µCT scans were used to measure the lengths of vertebral centra in the same animals at P7 (yellow; blue) and P42 (gold; turquoise), normalized to the naso-anal length of each Npr3^−/− (triangles) and wild-type Npr3^+/+ siblings (squares). Points represent the mean vertebral lengths at each position and timepoint with error bars showing the standard deviation of these measurements. For Npr3^+/+ and Npr3^−/− P42 n = 3 of mixed male and female animals, for CD1 P42 n = 3 males and 3 females. Due to one mutant having a tail kink, n = 2 for Npr3^−/− at P7 for total tail proportion. C Npr3^−/− mouse with wild-type sibling aged P42. White arrowhead points to “cone-shaped implantation of the tail”. D Immunofluorescent expression of NPR3 in Npr3^−/− and Npr3^+/+ TV1 and (E) TV6 cranial growth cartilages at P7. F Immunofluorescent staining of NPR3 in jerboa TV1 and G TV6 cranial growth cartilages at P16. For each of Npr3^+/+ and Npr3^−/−, n = 3 of mixed male and female animals. The panel in (G) has been subdivided into growth zone insets to show expression in resting, proliferative, and hypertrophic zones. Solid white lines show the edge of the intervertebral disc and dotted white lines demark the chondro-osseous junction. DAPI stain is shown in cornflower blue and NPR3 in yellow. Scale bar is 100 µm. Source data are provided as a Source Data file.

Fig. 5. Loss of NPR3 drives rapid cartilage elongation in part by increasing chondrocyte hypertrophy.

A, B Histology of mouse Npr3^+/+ and Npr3^−/− cranial and caudal growth cartilages of TV1 (A) and TV6 (B) during rapid elongation at P7. Scale bar is 100 µm. C–F Growth cartilage parameters in each cranial and caudal cartilage of Npr3^+/+ and Npr3^−/− TV1 and TV6. C Height of the growth cartilage. D Height of the proliferative zone. E Height of the hypertrophic zone. F Maximum height of hypertrophic chondrocytes in the direction of bone elongation. Welch’s t-test, two-tailed, n = 3 for all Npr3^+/+ cartilages and Npr3^−/− TV1, n = 4 for Npr3^−/− TV6 cartilages, mixed male and female animals, * ≤ 0.05, ** ≤ 0.01, *** ≤ 0.001. Exact p-values are available in Source Data file. C–F Error bar centers are the data means and error bands show standard deviation. Source data are provided as a Source Data file.