Identifying key molecules that launch regeneration has been a long-sought goal. Multiple regenerative animals show an initial wound-associated proliferative response that transits into sustained proliferation if a considerable portion of the body part has been removed1,2,3. In the axolotl, appendage amputation initiates a round of wound-associated cell cycle induction followed by continued proliferation that is dependent on nerve-derived signals4,5. A wound-associated molecule that triggers the initial proliferative response to launch regeneration has remained obscure. Here, using an expression cloning strategy followed by in vivo gain- and loss-of-function assays, we identified axolotl MARCKS-like protein (MLP) as an extracellularly released factor that induces the initial cell cycle response during axolotl appendage regeneration. The identification of a regeneration-initiating molecule opens the possibility of understanding how to elicit regeneration in other animals.
This is a preview of subscription content, access via your institution
Open Access articles citing this article.
Cell Regeneration Open Access 25 July 2023
Marcksl1 modulates endothelial cell mechanoresponse to haemodynamic forces to control blood vessel shape and size
Nature Communications Open Access 30 October 2020
Proteome Science Open Access 20 November 2019
Subscribe to this journal
Receive 51 print issues and online access
$199.00 per year
only $3.90 per issue
Rent or buy this article
Prices vary by article type
Prices may be subject to local taxes which are calculated during checkout
Wenemoser, D. & Reddien, P. W. Planarian regeneration involves distinct stem cell responses to wounds and tissue absence. Dev. Biol. 344, 979–991 (2010)
Knapp, D. et al. Comparative transcriptional profiling of the axolotl limb identifies a tripartite regeneration-specific gene program. PLoS ONE 8, e61352 (2013)
Tassava, R. A. & Mescher, A. L. The roles of injury, nerves, and the wound epidermis during the initiation of amphibian limb regeneration. Differentiation 4, 23–24 (1975)
Endo, T., Bryant, S. V. & Gardiner, D. M. A stepwise model system for limb regeneration. Dev. Biol. 270, 135–145 (2004)
Holtzer, S. The inductive activity of the spinal cord in urodele tail regeneration. J. Morphol. 99, 1–39 (1956)
Julius, D., MacDermott, A. B., Axel, R. & Jessell, T. M. Molecular characterization of a functional cDNA encoding the serotonin 1c receptor. Science 241, 558–564 (1988)
Yang, Y. C. et al. Human IL-3 (multi-CSF): identification by expression cloning of a novel hematopoietic growth factor related to murine IL-3. Cell 47, 3–10 (1986)
Tanaka, E. M., Gann, A. A., Gates, P. B. & Brockes, J. P. Newt myotubes reenter the cell cycle by phosphorylation of the retinoblastoma protein. J. Cell Biol. 136, 155–165 (1997)
Khattak, S. et al. Foamy virus for efficient gene transfer in regeneration studies. BMC Dev. Biol. 13, 17 (2013)
Khattak, S. et al. Germline transgenic methods for tracking cells and testing gene function during regeneration in the axolotl. Stem Cell Reports 1, 90–103 (2013)
Schnapp, E. & Tanaka, E. M. Quantitative evaluation of morpholino-mediated protein knockdown of GFP, MSX1, and PAX7 during tail regeneration in Ambystoma mexicanum. Dev. Dyn. 232, 162–170 (2005)
Gruber, C. E. Production of cDNA libraries by electroporation. Methods Mol. Biol. 47, 67–79 (1995)
Sundaram, M., Cook, H. W. & Byers, D. M. The MARCKS family of phospholipid binding proteins: regulation of phospholipase D and other cellular components. Biochem. Cell Biol. 82, 191–200 (2004)
Aderem, A. The MARCKS brothers: a family of protein kinase C substrates. Cell 71, 713–716 (1992)
Sandoval-Guzmán, T. et al. Fundamental differences in dedifferentiation and stem cell recruitment during skeletal muscle regeneration in two salamander species. Cell Stem Cell 14, 174–187 (2014)
Berg, D. A. et al. Efficient regeneration by activation of neurogenesis in homeostatically quiescent regions of the adult vertebrate brain. Development 137, 4127–4134 (2010)
Maden, M., Manwell, L. A. & Ormerod, B. K. Proliferation zones in the axolotl brain and regeneration of the telencephalon. Neural Dev. 8, 1 (2013)
Seykora, J. T., Myat, M. M., Allen, L. A., Ravetch, J. V. & Aderem, A. Molecular determinants of the myristoyl-electrostatic switch of MARCKS. J. Biol. Chem. 271, 18797–18802 (1996)
Rodrigo Albors, A. et al. Planar cell polarity-mediated induction of neural stem cell expansion during axolotl spinal cord regeneration. eLife 4, e10230 (2015)
Kragl, M. et al. Cells keep a memory of their tissue origin during axolotl limb regeneration. Nature 460, 60–65 (2009)
Sive, H. L., Grainger, R. M. & Harland, R. M. Early Development of Xenopus laevis: A Laboratory Manual (Cold Spring Harbor Laboratory, 2000)
Habermann, B. et al. An Ambystoma mexicanum EST sequencing project: analysis of 17,352 expressed sequence tags from embryonic and regenerating blastema cDNA libraries. Genome Biol. 5, R67 (2004)
Ferretti, P. & Brockes, J. P. Culture of newt cells from different tissues and their expression of a regeneration-associated antigen. J. Exp. Zool. 247, 77–91 (1988)
Lo, D. C., Allen, F. & Brockes, J. P. Reversal of muscle differentiation during urodele limb regeneration. Proc. Natl Acad. Sci. USA 90, 7230–7234 (1993)
Roensch, K., Tazaki, A., Chara, O. & Tanaka, E. M. Progressive specification rather than intercalation of segments during limb regeneration. Science 342, 1375–1379 (2013)
Zarzosa, A. et al. Axolotls with an under- or oversupply of neural crest can regulate the sizes of their dorsal root ganglia to normal levels. Dev. Biol. 394, 65–82 (2014)
Rodrigo-Albors, A. & Tanaka, E. M. in Salamanders in Regeneration Research: Methods and Protocols (eds Kumar, A. & Simon, A. ) 115–126 (Springer, 2015)
We thank A. Tazaki, Y. Taniguchi, I. Wagner, A. Rodrigo-Albors, D. Knapp, P. Murawala, B. Borgonovo and D. Drechsel for technical advice and important discussions; M. Schuez, A. Telzerow and Y. Taniguchi for assistance; B. Gruhl, A. Wagner, S. Mögel for animal care; and J. Currie, T. Sandoval-Guzman and E. Nacu for comments on the manuscript. E.M.T. was supported by a German Federal Ministry of Education and Research (BMBF) Biofutures grant, German Research Foundation (DFG) grant TA274/5-1, European Research Council Advanced Grant, Institutional funding from the Max Planck Institute of Molecular Cell Biology and Genetics (MPI-CBG), and the DFG Research Center for Regenerative Therapies Dresden (CRTD) and A.S. by the Swedish Research Council and Cancerfonden.
The authors declare no competing financial interests.
Extended data figures and tables
a, 110,592 clones from a 6-day tail blastema library were arrayed on 288 × 384-well plates. b, One 384-well plate was pooled into one conical tube and called a ‘pool’. In total, 288 pools were prepared from the library. c, Twenty-four pools were combined in one conical tube and called a ‘superpool’ (SP) containing 9,216 clones. In total 12 superpools were prepared. d, Bacteria of each superpool was cultured and plasmid was prepared. e, The superpool plasmids were transfected into HEK293 cells. f, Individual supernatants were tested on A1 myotubes for cell cycle re-entry activity (myotube assay) (see Fig. 1b). Positive superpools were successively subfractionated and the assay process was repeated back from the positive superpool (first screen) to come to a single clone (fourth screen) (a–c, right) (see Extended Data Fig. 2a–c).
a, The results of the second-round screen of superpool 9 (see Fig. 1b and Extended Data Fig. 1) and its sub-pooling diagram (right). Sub-pool D and sub-pool 2 showed higher BrdU incorporation activity than the others, identifying pool number 212 as positive (n = 12: 4 biological, 3 technical replicates each; mean ± s.d.). b, The result of the third-round screen of pool number 212 from superpool 9 and its sub-pooling diagram (right). Sub-pool A1 showed activity (n = 6: 2 biological, 3 technical replicates each; mean ± s.d.). c, Fourth-round screen of SP9 identified a single active clone (c1), AxMlp (n = 12: 4 biological, 3 technical replicates each; mean ± s.d.). The pooling diagram is shown on the right side. d, AxMLP supernatant induces an S-phase response in a dose-dependent manner in the newt myotube assay. Different amounts of AxMLP-containing supernatant (30 μl, 20 μl, 10 μl and 5.0 μl, respectively) were provided to the myotube cell culture medium. The myotube BrdU incorporation correlated with the amount of supernatant provided, whereas pCMV-SPORT6 supernatant did not provoke cell cycle entry at any dose (n = 6: 2 biological, 3 technical replicates each; mean ± s.d.). e, f, Newt myotubes treated with purified AxMLP (e) or flow-through (f) were immunostained for BrdU and MHC. More BrdU-incorporated nuclei (red) in myotubes (green) were observed in culture supplied with purified AxMLP compared with flow-through-treated cultures. Scale bar, 1 mm.
Extended Data Figure 3 AxMLP is classified as a member of the MARCKS family and characterization of its extracellular release in HEK293 cells.
a, Amino-acid sequence alignment of AxMLP with sequences from other vertebrates, human, mouse, rat, chick, newt, Xenopus and zebrafish. AxMLP contains three conserved domains: (1) myristoylated N terminus domain; (2) MARCKS homology domain; and (3) effector domain. b, A phylogenetic tree of vertebrate MARCKS family proteins. The tree was constructed by the neighbour-joining method with the ClustalW program. The percentage beside the nodes shows that a node was supported in 1,000 bootstrap pseudo replications. The scale bar indicates evolutionary distance. c, Schematic illustration of His-tagged AxMlp (left) and eGFP-fused AxMlp (right). 3C protease PreScission site was inserted between AxMlp and the tag for both constructs. d, e, AxMLP does not induce significant cell death. The percentage of GFP-expressing HEK293 cells (d) and absolute number of the cells (e) (n = 16: 4 biological, 4 technical replicates each; mean ± s.d.; centre values as median; whiskers as maximum and minimum, respectively) at the indicated time points of culture. There was no significant difference with Student’s t-test between AxMlp-transfected cells and the control in any time points. f, Characterization of anti-AxMLP antibodies by western blot. Cell lysates from HEK293 cells transfected with the indicated plasmids were tested for the full-length AxMLP polyclonal antibody (left) and C-terminal AxMLP polyclonal antibody (right). g, Silver staining of the fractions from AxMLP–His purification. Bovine serum albumin (BSA) was added to purified fraction as a carrier protein. h, AxMLP–His purification analysed by anti-His-tag western blotting. NS, not significant.
a–d, Sections from AxMLP-injected tails immunostained for BrdU/PAX7 (a, b) and BrdU/MEF2C (c, d) (refers to data in Fig. 2). Scale bars, 100 μm. e, Schematic illustration of the protein injection into axolotl limb. f, Quantification of BrdU+ cells in the limbs injected with PBS, flow-through or purified AxMLP (n = 4: biological replicates; centre values as median; points represent each sample). g–n, Transverse sections from purified AxMLP-injected (g–j) or flow-through-injected limbs (k–n). Scale bars: lower-magnification images, 200 μm; higher-magnification images, 50 μm. Sections were immunostained for BrdU (g, k), BrdU/myelin basic protein (MBP) (h, l), BrdU/PAX7 (i, m) and BrdU/GFP (j, n). GFP+ cells represent connective tissues in lateral plate mesoderm (LPM)-GFP transplanted axolotls. All molecular markers used except MBP had nuclear expression, and therefore allowed one-to-one colocalization of nuclear BrdU with nuclear staining of the marker. Therefore, we refer to the MBP data as ‘MBP-associated’. White boxes highlight the magnified images. Yellow circles indicate two bones in the lower limb. NS, not significant; *P < 0.05, **P < 0.005, ***P < 0.0005, ****P < 0.00005 with Student’s t-test. White arrowheads indicate marker+/BrdU+ cells.
Extended Data Figure 5 Upregulation of AxMlp transcript during early regeneration and alteration of AxMLP protein localization in wound epidermis cells.
a, e, Measurement of AxMlp expression by qPCR at the indicated time points during tail (a) and limb (e) regeneration (n = 3: biological replicates; mean ± s.d.). To obtain the values of fold-change for each time point, the relative concentrations of the PCR products were calculated by the standard curve method. The concentration of AxMlp was normalized to that of large ribosomal protein 4 (Rpl4). b–h, Immunostaining with anti-AxMLP antibody (white) on tails (b–d) and limbs (f–h) of intact (b, f: transverse sections), 1 dpa (c: sagittal; g: horizontal) and 6 dpa (d: sagittal; h: horizontal) samples. By 6 dpa, the epithelial organization and AxMLP expression appeared to be returning to a less tightly adherent, less membrane-associated appearance (d, h). Scale bars: left, 200 μm; right 50 μm. Red arrowheads indicate spinal cord; green arrowheads indicate wound epidermis; yellow arrowheads indicate normal epidermis; yellow circles indicate two bones in the lower limb.
Extended Data Figure 6 AxMLP morpholinos specifically and efficiently reduce AxMLP translation in cultured cells.
a, Schematic illustration of wild-type (WT) AxMlp (top) and N-terminal deletion AxMlp (bottom) constructs used to characterize AxMlp morpholino 1. The N-terminally deleted AxMlp lacks the morpholino-binding site. Both constructs have a His-tag on their C terminus (a). ED, effector domain; M, myristoylated N terminus domain; MH, MARCKS homology domain. b–i, Electroporated A1 myoblasts were stained with the indicated markers. b–e, Wild-type AxMlp plasmid was co-electroporated with the control morpholino (c) or the AxMlp-specific morpholino 1 (d), whereas wild-type AxMlp plasmid only (b) or wild-type AxMlp only without any primary antibody staining were used as negative controls (e). f–h, ΔN-AxMlp plasmid was co-electroporated with the control morpholino (g) or the AxMlp-specific morpholino 1 (h), whereas ΔN-AxMlp plasmid only (f) or pCMV-SPORT6-3C-His (empty vector) plasmid only served as negative controls (i). j, Western blotting for the cell lysates from the experiment above. AxMLP morpholino 1 specifically reduced AxMLP protein expression. k, Schematic illustration of the constructs used to characterize AxMlp morpholino 2. The original AxMlp expression clone from the cDNA library (BL212a101: top) was used as it included the 5′ untranslated region (UTR) target site for AxMlp morpholino 2. The subcloned AxMlp-His construct lacks the binding site for AxMlp morpholino 2 and was used as the control construct. l–r, Electroporated A1 myoblasts were stained with the indicated markers. l–n, BL212a101 plasmid was co-electroporated with the five-mismatch control morpholino (m) or the AxMLP-specific morpholino 2 (n), or BL212a101 plasmid only (l) or pCMV-SPORT6-3C-His (empty vector) plasmid only served as negative controls (o). AxMLP was detected using an anti-AxMLP antibody (red), and morpholinos were detected via FITC conjugation (green). p–r, AxMlp-3C-His plasmid was co-electroporated with the five-mismatch control morpholino (q) or the AxMlp-specific morpholino 2 (r) or AxMlp-3C-His plasmid only (p). AxMLP was detected using an anti-His-tag antibody (red), and morpholinos were detected via FITC conjugation (green). s, Western blotting for the cell lysates from the experiment above. AxMlp morpholino 2 specifically reduced AxMLP protein expression. Scale bars, 100 μm.
a–j, The morpholinos shown in were used in Fig. 4 and Extended Data Figs 6a–j,8a–f. k–t, The morpholinos shown were used in Extended Data Figs 6k–s, 8g–j. Transverse sections from AxMLP-specific morpholino 1 (a–e) or control morpholino (f–j) electroporated tail. b, The spinal cord (SC) boxed in a. c, The higher-magnification images of the spinal cord boxed in b. AxMLP expression was detected in morpholino-negative cells (red asterisks), whereas it was reduced in morpholino-positive cells (yellow asterisks). d, The epidermis boxed in a. e, AxMLP expression was unaffected in morpholino-negative cells (red asterisks), whereas it was reduced in morpholino-positive cells (yellow arrowheads). In the control morpholino-electroporated tail (f) there was no morpholino-specific knockdown phenotype in either spinal cord (g, h) or epidermis (i, j). The same experiments were performed with AxMLP-specific morpholino 2 (k–o) and the corresponding five-mismatch control morpholino (p–t). k–t, The data sets were the same as a–j. Scale bars, 200 μm (a, f, k, p); 50 μm (b–e, g–j, l–o, q–t).
a–d, Representative transverse sections of the morpholino-electroporated/protein-injected blastemas that were used for quantification of BrdU incorporation in Fig. 4d. Rhodamine was co-injected with the protein samples. e, Quantification of BrdU+ cells in blastema sections of morpholino-electroporated/protein-injected tails at 3 dpa (n = 4: biological replicates; centre values as median; points represent each sample). f, The length of the blastema during tail regeneration. The data at 6 dpa were plotted in Fig. 4c. By 14 days the difference in total regenerate length among the samples was not statistically significant. g–j, The same experimental scheme (shown in Fig. 4a) as was used for AxMLP morpholino 1 was implemented for a second specific morpholino (AxMLP-specific morpholino 2). g, Bright-field images of the morpholino-2-electroporated/protein-injected tails at 6 dpa. Red bars indicate amputation planes. Dashed lines delineate the shape of the mesenchymal blastema. h, Blastema length at 6 dpa (n = 4: biological replicates; centre values as median; points represent each sample). i, The length of the blastema during tail regeneration. The data at 6dpa were plotted in h. j, Transverse sections immunostained for BrdU from morpholino-electroporated/protein-injected tails at 3 dpa. AxMLP-specific morpholino 2 combined with flow-through (FT) injection shows reduction of BrdU incorporation, whereas AxMLP protein injection rescues the phenotype. The corresponding five-mismatch control morpholino does not affect BrdU incorporation. Yellow circles indicate spinal cord (top) and notochord/cartilage (bottom). NS, not significant; **P < 0.005, ***P < 0.0005, ****P < 0.00005 with Student’s t-test. Scale bars, 200 μm (a–d, j); 500 μm (g).
Extended Data Figure 9 Anti-AxMLP antibody significantly blocks BrdU incorporation during tail regeneration.
a, Schematic illustration of antibody injection into axolotl tail. b, Quantification of BrdU+ cells in blastema sections of antibody-injected tails at 3 dpa (n = 4: biological replicates; centre values as median; points represent each sample). NS, not significant; **P < 0.005, ***P < 0.0005, ****P < 0.00005 with Student’s t-test.
a, Schematic illustration of the protein injection into axolotl tail and blastema. b, Bright-field images of the protein-injected tails at 4 dpa. c, Blastema length at 4 dpa (n = 6: PBS, FT; n = 8: AxMLP, biological replicates; centre values as median; points represent each sample). The blastema from purified AxMLP injected tails significantly increased the regenerate length. Scale bar, 500 μm. Red bars indicate amputation planes; dashed lines delineate the shape of the mesenchymal blastema. NS, not significant; ***P < 0.0005 with Student’s t-test.
This table contains raw values of qPCR in tail regeneration time course (a, ExFig. 5a) and limb (b, ExFig. 5e). (XLSX 48 kb)
A list of oligos used in this manuscript, see “Plasmid construction” section in the Online Methods. (XLSX 51 kb)
Time-lapse videos of HEK293 cells transfected with AxMLP-3C-His plasmid between 5 and 72 hours after transfection. The culture medium was changed at 24 hours post transfection from 10% serum medium to the serum free, FreeStyle 293 medium. (MOV 4861 kb)
Time-lapse videos of pCMV-SPORT6-3C-His as a negative control to compare to AxMLP transfected HEK293 cells (Video 1) between 5 and 72 hours after transfection. The culture medium was changed at 24 hours post transfection from 10% serum medium to the serum free, FreeStyle 293 medium. (MOV 4859 kb)
About this article
Cite this article
Sugiura, T., Wang, H., Barsacchi, R. et al. MARCKS-like protein is an initiating molecule in axolotl appendage regeneration. Nature 531, 237–240 (2016). https://doi.org/10.1038/nature16974
This article is cited by
Cell Regeneration (2023)
Marcksl1 modulates endothelial cell mechanoresponse to haemodynamic forces to control blood vessel shape and size
Nature Communications (2020)
Proteome Science (2019)
Using transcriptomics to enable a plethodontid salamander (Bolitoglossa ramosi) for limb regeneration research
BMC Genomics (2018)