In recent years, multiple theories have been proposed to explain how the proximodistal (PD) axis of the tetrapod limb (i.e. from shoulder to digit tips) is patterned. They can be grouped in two broad categories based on the order of specification: progressive and early patterned. The common ground in both categories is that the ∼250 µm region under the distal epithelium of the limb bud is considered undifferentiated, such that the very early bud is fully undifferentiated and, as the limb grows out, the cells that remain in the proximal side of that limit start to differentiate (Tabin and Wolpert, 2007). In progressive models, this fate is progressively specified, first stylopod (arm), then zeugopod (forearm), then autopod (hand). This process is often divided into two phases, a first one driven by diffusible signals, either proximal and distal or only distal (Berenguer and Duester, 2021; Delgado et al., 2020; Mercader et al., 2000; Roselló-Díez et al., 2011), followed by a second phase driven by an intrinsic timer (Roselló-Díez et al., 2014; Saiz-Lopez et al., 2015), mediated in part by the progressive degradation of MEIS transcription factors (Delgado et al., 2020). In early specification models, by contrast, the proximal and distal regions are formed early on, with the intermediate segment arising either at the same time as the other two (Dudley et al., 2002), or slightly later, as a consequence of the interaction between proximal and distal domains (Mariani et al., 2008). The latter model bears significant similarity to the classic ‘distal first’ (aka intercalary) model of limb regeneration, which was conceived by Gardiner and Bryant in the 1990s following the observation that the undifferentiated early blastema (the adult equivalent of the limb bud) expresses Hox genes associated with both proximal (HoxA9) and distal (HoxA13) locations (Gardiner et al., 1993). However, grafting studies on regenerating axolotl from the Tanaka lab found that early blastema cells are not committed to an autopod fate (Roensch et al., 2013), suggesting that patterning of regenerating limb segments occurs through progressive specification, rather than intercalation.

The waters were again muddied when it was shown that, under the right conditions, early limb cells show plasticity and adaptability to the environment into which they are grafted (Dudley et al., 2002; McCusker and Gardiner, 2013; Roselló-Díez et al., 2011; Roselló-Díez and Torres, 2011), making it impossible to distinguish between progressive versus a two-signal/distal-first mechanism. Luckily, the advent of single-cell sequencing strategies, as performed on developing mouse limb buds in a recent preprint by Markman et al. (2022), now provides tools to tackle this question from a new angle. Their findings put forward a completely new differentiation model that integrates progressive, early specification and intercalary modes of PD limb patterning.

To better understand how the PD forelimb axis is specified in the mammalian embryo, Markman et al. (2022) started by using single-cell RNA-seq to characterise the molecular signature of each limb cell across five stages of mouse limb development spanning embryonic day (E) 10.5 to 14.5. The initial analysis identified robust and homogenous groups of cells referred to as ‘metacells’ (Baran et al., 2019). Clustering of these metacells led to the identification of six chondrocyte populations, three connective-tissue fibroblast subgroups and three progenitor cell states (P1-P3).

Further analysis of the progenitor states revealed that the P2 population exhibits a ‘proximal’ gene expression signature, P3 has an ‘autopod’ signature and P1 shows high expression of limb patterning genes, but lacks a spatial signature, meaning that, surprisingly, a zeugopod (forearm) progenitor was not identified. This observation is consistent with the intercalary model of specification. Interestingly, both P1 and P2 cells are most abundant in early limb buds, and decrease to a minimum by E14.5, which suggests progressive differentiation of these cells over time.

To evaluate the spatial dynamics of these progenitor populations, Markman et al. (2022) performed light sheet microscopy on different staged limb buds stained with unique markers for each progenitor type using mRNA in situ hybridisation. They found that in the earliest limb bud stage analysed (E10.5), all three progenitor cell types are present in spatially restricted domains. These results suggest that, at least at this stage, there is not a single distally located progenitor, which goes against the progressive specification/differentiation models.

Using clever pulse-chase lineage tracing strategy, the authors established that Msx1 marks naïve progenitors (P1) of the limb, and using this marker they further showed that: (1) naïve progenitors transition to other fates (P2, P3 or differentiated cells) in a continuous manner; (2) proximal progenitors keep being generated at advanced stages of limb development; and (3) differentiation of naïve progenitors into chondroprogenitors (that generate the limb skeleton) occurs simultaneously in all limb segments along the PD axis.

As the current dogma is that differentiation of limb cells occurs consecutively starting in the most-proximal limb bud locations, this last observation prompted the authors to revaluate the order in which these segments form by performing pulse-chase genetic linage tracing using the Msx1 reporter (to mark the progeny of naïve cells) combined with staining for Sox9 (a marker for skeleton), allowing them to track the timing of when each skeletal element lost the expression of Msx1. Remarkably, the differentiation process was found not to be restricted spatially, but rather occurred simultaneously along all segments of the developing skeleton, in a complex but stereotypical three-dimensional manner.

Overall, this study suggests that the separation between proximal and distal limb fates occurs earlier than anticipated, redefining the spatiotemporal location and properties of naïve limb mesenchyme progenitors. Most remarkably, the work shows that the differentiation process does not occur in a simple proximal-to-distal manner. One important pending question is whether, at an earlier stage (E9.5), the limb bud is composed of only naïve P1 cells, which begin differentiating into P2 and P3 populations by E10.5. The Msx1 pulse-chase experiment starting at E9.5 in the current study, as well as a similar study by the Torres lab pulse-chasing Meis1 at E8.5 (Delgado et al., 2020), support this idea because both lineages show contribution to the entire PD axis when the reporters are activated very early, and become more restricted to the proximal structures when the labels are activated progressively later. Furthermore, it also remains to be determined what makes cells in different positions in the limb differentiate at distinct times, even when located within the same skeletal element. Last, it will be interesting to determine whether this mode of PD specification and differentiation is conserved in regenerating tetrapod limbs, which are initially composed of dedifferentiated connective tissue cells (as opposed to naïve stem-like cells) connected to a differentiated limb stump. Luckily, the future is now at hand.

C.M. thanks members of the McCusker lab for the discussions about limb patterning during regeneration. A.R.-D. thanks M. Torres and his group for discussions on limb patterning.

The Rosello-Diez lab is funded by a Human Frontier Science Program Career Development Award (CDA00021/2019) and a National Health and Medical Research Council Ideas grant (2002084). C.M. is supported by the National Institutes of Health/National Institute of Child Health and Human Development (2R15HD0921-02).

Baran
,
Y.
,
Bercovich
,
A.
,
Sebe-Pedros
,
A.
,
Lubling
,
Y.
,
Giladi
,
A.
,
Chomsky
,
E.
,
Meir
,
Z.
,
Hoichman
,
M.
,
Lifshitz
,
A.
and
Tanay
,
A.
(
2019
).
MetaCell: analysis of single-cell RNA-seq data using K-nn graph partitions
.
Genome Biol.
20
,
206
.
Berenguer
,
M.
and
Duester
,
G.
(
2021
).
Role of retinoic acid signaling, FGF signaling and meis genes in control of limb development
.
Biomolecules
11
,
80
.
Delgado
,
I.
,
López-Delgado
,
A. C.
,
Roselló-Díez
,
A.
,
Giovinazzo
,
G.
,
Cadenas
,
V.
,
Fernández-de-Manuel
,
L.
,
Sánchez-Cabo
,
F.
,
Anderson
,
M. J.
,
Lewandoski
,
M.
and
Torres
,
M.
(
2020
).
Proximo-distal positional information encoded by an Fgf-regulated gradient of homeodomain transcription factors in the vertebrate limb
.
Sci. Adv.
6
,
eaaz0742
.
Dudley
,
A. T.
,
Ros
,
M. A.
and
Tabin
,
C. J.
(
2002
).
A re-examination of proximodistal patterning during vertebrate limb development
.
Nature
418
,
539
-
544
.
Gardiner
,
D. M.
,
Blumberg
,
B.
and
Bryant
,
S. V.
(
1993
).
Expression of homeobox genes in limb regeneration
.
Prog. Clin. Biol. Res.
383A
,
31
-
40
.
Mariani
,
F. V.
,
Ahn
,
C. P.
and
Martin
,
G. R.
(
2008
).
Genetic evidence that FGFs have an instructive role in limb proximal-distal patterning
.
Nature
453
,
401
-
405
.
McCusker
,
C. D.
and
Gardiner
,
D. M.
(
2013
).
Positional information is reprogrammed in blastema cells of the regenerating limb of the axolotl (Ambystoma mexicanum)
.
PLoS ONE
8
,
e77064
.
Markman
,
S.
,
Zada
,
M.
,
David
,
E.
,
Giladi
,
A.
,
Amit
,
I.
and
Zelzer
,
E.
(
2022
).
Single-cell atlas of mouse limb development reveals a complex spatiotemporal dynamics of skeleton formation
.
bioRxiv
.
Mercader
,
N.
,
Leonardo
,
E.
,
Piedra
,
M. E.
,
Martinez-A
,
C.
,
Ros
,
M. A.
and
Torres
,
M.
(
2000
).
Opposing RA and FGF signals control proximodistal vertebrate limb development through regulation of Meis genes
.
Development
127
,
3961
-
3970
.
Roensch
,
K.
,
Tazaki
,
A.
,
Chara
,
O.
and
Tanaka
,
E. M.
(
2013
).
Progressive specification rather than intercalation of segments during limb regeneration
.
Science
342
,
1375
-
1379
.
Roselló-Díez
,
A.
and
Torres
,
M.
(
2011
).
Regulative patterning in limb bud transplants is induced by distalizing activity of apical ectodermal ridge signals on host limb cells
.
Dev. Dyn.
240
,
1203
-
1211
.
Roselló-Díez
,
A.
,
Ros
,
M. A.
and
Torres
,
M.
(
2011
).
Diffusible signals, not autonomous mechanisms, determine the main proximodistal limb subdivision
.
Science
332
,
1086
-
1088
.
Roselló-Díez
,
A.
,
Arques
,
C. G.
,
Delgado
,
I.
,
Giovinazzo
,
G.
and
Torres
,
M.
(
2014
).
Diffusible signals and epigenetic timing cooperate in late proximo-distal limb patterning
.
Development
141
,
1534
-
1543
.
Saiz-Lopez
,
P.
,
Chinnaiya
,
K.
,
Campa
,
V. M.
,
Delgado
,
I.
,
Ros
,
M. A.
and
Towers
,
M.
(
2015
).
An intrinsic timer specifies distal structures of the vertebrate limb
.
Nat. Commun.
6
,
8108
.
Tabin
,
C.
and
Wolpert
,
L.
(
2007
).
Rethinking the proximodistal axis of the vertebrate limb in the molecular era
.
Genes Dev.
21
,
1433
-
1442
.

Competing interests

The authors declare no competing or financial interests.