Logo of nihpaAbout Author manuscriptsSubmit a manuscriptNIH Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
Dev Dyn. Author manuscript; available in PMC Oct 1, 2012.
Published in final edited form as:
PMCID: PMC3180884

Identity and fate of Tbx4-expressing cells reveal developmental cell fate decisions in the allantois, limb, and external genitalia


T-box gene Tbx4 is critical for the formation of the umbilicus and the initiation of the hindlimb. Previous studies show broad expression in the allantois, hindlimb, lung and proctodeum. We have examined the expression of Tbx4 in detail and used a Tbx4-Cre line to trace the fates of Tbx4-expressing cells. Tbx4 expression and lineage reveal that various distinct appendages, such as the allantois, hindlimb, and external genitalia, all arise from a single mesenchymal expression domain. Additionally, although Tbx4 is associated primarily with the hindlimb, we find two forelimb expression domains. Most notably, we find that, despite the requirement for Tbx4 in allantoic vasculogenesis, the presumptive endothelial cells of the allantois do not express Tbx4 and lineage tracing reveals that the umbilical vasculature never expresses Tbx4. These results imply that endothelial lineages are segregated prior to the onset of vasculogenesis, and demonstrate a role for the peri-vascular tissue in vasculogenesis.

Keywords: Tbx4, allantois, limb, hindlimb, lung, genital tubercle, mouse, development


T-box genes code for a family of DNA-binding transcription factors known to play critical roles in the formation of numerous embryonic structures and maintenance of certain adult tissues (Showell et al., 2004; Naiche et al., 2005; King et al., 2006; Wardle and Papaioannou, 2008). Mutations of T-box genes in vertebrate organisms result in dramatic developmental defects including peri-implantation lethality, embryonic axis truncation, and debilitating heart and skeletal defects. T-box gene mutations are also known to be causative for a number of human developmental syndromes, including Holt-Oram, ulnar-mammary, and DiGeorge syndromes (Naiche and Papaioannou, 2007b).

T-box gene Tbx4 is primarily known for its role in hindlimb development. Tbx4 was originally described as one of the few genes expressed specifically in hindlimbs and not forelimbs (Chapman et al., 1996; Gibson-Brown et al., 1996) and has been shown to be required for early stages of hindlimb development and for the development of hindlimb skeletal elements, such as the femur (Naiche and Papaioannou, 2003; Naiche and Papaioannou, 2007c). The hindlimb specificity of Tbx4 has been observed throughout the gnathostomes, including mouse, chick, opossum, zebrafish, and shark (Gibson-Brown et al., 1996; Gibson-Brown et al., 1998a; Tamura et al., 1999; Ruvinsky et al., 2000; Tanaka et al., 2002; Horton et al., 2008; Keyte and Smith, 2010). Direct evidence for hindlimb-specific expression has not been shown in humans, but human mutations in TBX4 cause small patella syndrome, which results in abnormalities of the pelvis, knee, and foot (Bongers et al., 2004). Hindlimb specific Tbx4 expression may have evolved concomitantly with the evolution of pelvic fins in fish, which is the origin of the posterior limb pair (Horton et al., 2008).

Whether the hindlimb specificity of Tbx4 confers hindlimb, or leg-like, identity and morphology to the hindlimb is still controversial. Studies in chick have indicated that ectopic Tbx4 can induce limbs with leg-like characteristics (Rodriguez-Esteban et al., 1999; Takeuchi et al., 1999), while loss-of-function studies in the mouse showed no change in hindlimb identity or morphology (Naiche and Papaioannou, 2007c). Gene-swap experiments with the closely related forelimb-specific gene, Tbx5, in the mouse forelimb have shown that Tbx4 can replace Tbx5 in driving limb outgrowth and Pitx1 is the actual determinant of hindlimb morphology (Minguillon et al., 2005). Conversely, similar experiments in the hindlimb have shown that only Tbx4 can support normal morphology of the hindlimb, and that Tbx4 and Tbx5 have different transcriptional activities (Ouimette et al., 2010).

In addition to the hindlimb, Tbx4 is expressed in the allantois (Chapman et al., 1996), an outgrowth from the posterior of the embryo which is the precursor to the umbilical cord and the vascular portions of the fetal placenta (Inman and Downs, 2007). The allantois is one of only three regions known to undergo vasculogenesis, the de novo differentiation of endothelial cells and formation of blood vessels out of undifferentiated mesenchyme, as opposed to the more common angiogenesis, which entails sprouting and extension of existing blood vessels (Baron and Fraser, 2005). Tbx4 null embryos show severe defects in growth of the allantois and in vasculogenesis, resulting in the death of such embryos due to their failure to establish an umbilical connection (Naiche and Papaioannou, 2003).

Tbx4 expression has also been noted in the lung (Chapman et al., 1996). Loss of Tbx4 alone does not appear to alter lung development (Naiche and Papaioannou, 2003; Naiche and Papaioannou, 2007c), but Tbx4 may play a role in lung bud branching in concert with Tbx5, which is co-expressed in this tissue (Cebra-Thomas et al., 2003; Sakiyama et al., 2003).

By examining the expression of Tbx4 in greater detail and developmental range than previous studies, we have found several previously unreported aspects of this gene’s expression. We find that tracing the lineage of Tbx4 expressing cells reveals previously unrecognized developmental cell fate decisions, in particular the pre-patterning of vascular and perivascular fates in vasculogenic mesenchyme, continuity between multiple outgrowths of the posterior mesenchyme, and the origin of the pelvic girdle. We also describe the regions of recombination of the Tbx4-Cre allele, which will be of great use as a tool to conditionally excise genes throughout the limb, lung, and genital mesenchyme from their earliest inceptions, and to selectively ablate perivascular tissue in the allantois.


Tbx4 expression in hindlimb

As previously reported, Tbx4 expression in the lateral plate hindlimb mesenchyme initiates at embryonic day (E) 9.5, prior to hindlimb outgrowth (Fig. 1A). At E10.5-11.5, expression was observed throughout the extending hindlimb, with the proximal boundary of expression at the body wall (Fig. 1B-C). At later stages, Tbx4 expression shifted distally but remained widespread in the hindlimb through E14.5 (Fig. 1D-F).

Fig. 1
Tbx4 expression and lineage in the hindlimb. A-F: Whole mount in situ hybridization of Tbx4 in the hindlimb at indicated stages. hlf = hindlimb field, a = allantois, hl = hindlimb. G-H: Comparison of Tbx4-Cre lineage observed using the R26R-lacZ reporter ...

Lineage tracing of Tbx4-expressing cells in hindlimb

To explore the role of Tbx4-expressing cells in the limb, we examined the fate of cells that have expressed Tbx4 using a Tbx4-Cre mouse line, where an internal ribosomal entry site (IRES) followed by the Cre recombinase gene has been inserted into the 3’ UTR of the Tbx4 gene (Luria et al., 2008). Tbx4-Cre homozygous mice are viable and fertile, confirming that Tbx4 function is not disrupted, and Tbx4-Cre embryos do not show abnormal apoptosis in regions of Cre expression, as has been reported for some Cre lines (Naiche and Papaioannou, 2007a). Tbx4-Cre mice were mated to the R26R-lacZ and R26R-YFP Cre reporter lines (Soriano, 1999; Srinivas et al., 2001), which undergo excision of a transcriptional stop in Cre-expressing cells and irreversibly activate expression of the reporter, allowing us to trace the eventual distribution of all cells that ever express Tbx4, hereafter referred to as the Tbx4-lineage or Tbx4-lineage cells.

Both reporter lines produced similar results (Fig. 1G-H), but we noted that detection of the YFP florescence lagged behind the appearance of endogenous Tbx4 expression by approximately half a day (Fig. 1I-J). No such delay was observed using the lacZ reporter, possibly due to the enzymatic amplification inherent in the assay, and we therefore conducted most of the following studies with R26R-lacZ.

Whole mount ß-galactosidase staining of Tbx4-Cre; R26R-lacZ embryos showed apparently complete constitution of the hindlimb by Tbx4-lineage cells (Fig. 1G, K, L). Additionally, it was clear at postnatal day 1 (P1) that tissues associated with hindlimb mesenchyme but medial to the limb itself were also part of the Tbx4-lineage; the dermis showed ß-galactosidase staining over most of the haunch, and the gluteal muscles that extend nearly to the spine were also strongly stained (Fig. 1L).

The limb is derived from two tissues sources: the lateral plate mesoderm of the hindlimb field gives rise to the cartilage, bone, perichondrium, tendons, ligaments, and connective tissue, whereas muscle and blood vessels arise from somitic cells that migrate into the limb (Chevallier et al., 1977). Sections through Tbx4-Cre; R26R-lacZ embryos showed that the Cre reporter is completely recombined in the hindlimb field mesenchyme as early as E9.5 (Fig. 1M). By E13.5, when the skeletal elements of the hindlimb can be discerned and the muscle primordia have entered the limb, all tissues of the hindlimb appear to be Tbx4-lineage cells (Fig. 1N), suggesting that much of the somite-derived tissue of the hindlimb also upregulates Tbx4 at some point in its migration into the limb.

However, at P1 it is clear that the Tbx4 lineage does not form the entirety of the hindlimb; there is a clear proximo-distal gradient of Tbx4-lineage tissue (Fig. 1O). In the hindlimb autopod, all tissues other than the epidermis appeared to be composed of Tbx4-lineage cells (Fig. 1P-Q). In the stylopod and zeugopod, the lateral plate-derived tissues, particularly bone, tendon, and perichondrium, appear to be completely composed of Tbx4-lineage cells (Fig. 1R-S), while the major muscles of the leg show more modest Tbx4-lineage contribution (Fig. 1 O, R, T). The low contribution of the Tbx4 lineage to proximal muscle reflects the case of Tbx5, which is expressed only at low levels in migrating myocytes (Hasson et al., 2010).

Interestingly, even though Tbx4 expression is not observed within the body wall after E9.5, all of the elements of the pelvic girdle and its surrounding perichondrium, which lie deep within the body, are Tbx4-lineage tissues (Fig.1 U-W). This data supports previous studies in chicks suggesting that the pelvis derives from lateral plate mesoderm of the hindlimb field (Malashichev et al., 2008). However, our data cannot rule out somitic contribution to the pelvic girdle, since we observe that some somite-derived tissue becomes part of the Tbx4 lineage. Indeed, the hindlimb muscles that attach to the body core (i.e., the gluteal and groin muscles) contain significant Tbx4 lineage contribution (Fig. 1L). This would suggest that the progenitors of the proximal hindlimb muscles migrate laterally into the hindlimb, upregulate Tbx4, and then reverse their migration to form muscles within the body wall, in a process similar to that which has been demonstrated for the muscles of the urogenital region (Valasek et al., 2005).

The early and complete recombination afforded by Tbx4-Cre in the hindlimb demonstrates that this is an excellent tool for limb-specific ablation of genes, including those involved in the very earliest stages of limb development and pelvic girdle formation.

Tbx4 expression and lineage in forelimb and sternum

Despite the predominantly hindlimb-specific expression of Tbx4, the earliest descriptions of Tbx4 expression noted faint forelimb expression at E9.5 (Gibson-Brown et al., 1996). Similarly, we observe a small domain in the proximal core of the forelimb from E10.5-11.5 (Fig. 2A-B) that expresses Tbx4 at lower levels than in the hindlimb (compare forelimbs in Fig. 2A-D with corresponding hindlimbs in Fig. 1B-E, which originated from the same embryos). This proximal forelimb expression disappears by E12.5. A novel domain of Tbx4 expression in the forelimb autopod appears faintly at E12.5 and becomes more intense by E13.5, disappearing gradually thereafter (Fig. 2C-D).

Fig. 2
Tbx4 in the forelimb and sternum. A-B: Tbx4 expression in core mesenchyme (c) of the forelimb. C-D: Tbx4 expression in the forelimb autopod (au). E-H: Whole mount ß-galactosidase staining marking Tbx4 lineage in the forelimb. E-F: Side view of ...

When we examined the Tbx4 lineage in the forelimb at E11.5-13.5, we observed the appearance of scattered lineage-marked cells primarily near the prospective elbow, corresponding to the proximal core expression domain noted above (Fig. 2E). At later stages, in addition to cells near the elbow, dense accumulations of Tbx4-lineage cells were observed in the ventral hand, corresponding to the autopod expression domain observed at E13.5 (Fig. 2F-H).

Sections of E13.5 embryos showed that, while there were a few Tbx4-lineage cells in the stylopod near the shoulder, the bulk of the early forelimb Tbx4 lineage was concentrated in the bone and nearby tissues of the elbow joint (Fig. 2I-J). By P1, sections of the corresponding region show that these cells contribute mostly to the tendons around the elbow and to the periphery of the bone (Fig. 2K). In the autopod, Tbx4-lineage cells are likewise concentrated in the tendons and bone periphery, but scattered cells also appear in the muscle and epidermis (Fig. 2L-N).

At E13.5, Tbx4-lineage cells were also scattered throughout the ventral body wall and by P1 this lineage domain resolved into the cartilage of the anterior portion of the sternum and the xyphoid process (Fig. 2O-P). This region probably corresponds to the previously reported Tbx4 expression in the thoracic body wall (Chapman et al., 1996).

The developmental significance of forelimb expression of Tbx4 is unclear. Embryos with partial or complete loss of Tbx4 function show no obvious defects in overall forelimb size or skeletal formation, but detailed analysis of forelimb tendons, where most Tbx4 expression occurs, was not performed (Naiche and Papaioannou, 2007c; Menke et al., 2008). It has been suggested that forelimb expression of Tbx4 may impose features of hindlimb morphology in the forelimb (Wang et al., 2010). In bats, Tbx4 is expressed in the first digit of the forelimb, which is the only digit that retains a more hindlimb-like morphology and does not elongate to form part of the wing (Wang et al., 2010). However, in mouse there is no obvious correlation between the regions of forelimb Tbx4 expression and hindlimb-like morphology; on the contrary, the carpal bones and tendons, where Tbx4 expression is concentrated, are some of the most divergent structures between limb pairs.

Characterization of the Tbx4 locus has identified an enhancer element which drives Tbx4 expression in the forelimb, but this element is poorly conserved outside of the mammalian class and is dispensable for viability and forelimb morphology in mouse (Menke et al., 2008). Collectively, these data suggest that the forelimb expression of Tbx4 is not evolutionarily conserved and may have little developmental relevance, but this domain should be noted by anyone using the Tbx4-Cre allele for hindlimb-specific gene recombination.

The Tbx4 lineage in tissues derived from posterior ventral mesenchyme

Expression of Tbx4 can also be seen in the developing external genitalia throughout development (Fig. 3A-F). Tbx4 expression in this region initiates at E9.5 in the mesenchyme surrounding the cloacal slit, in the region that will form the genital folds and the genital tubercle (Fig. 3A-B). Interestingly, at this early time point, Tbx4 expression exists as a single contiguous expression domain in the ventral mesenchyme that extends laterally to the hindlimbs, posteriorly to the genital folds, and anteriorly to the allantois. During later development, Tbx4 expression in each of these three regions becomes more distal and becomes separate expression domains (Fig 3C-F).

Fig. 3
Tbx4 expression and lineage in developing urogenital organs. A-F: Tbx4 expression in the posterior ventral mesenchyme at indicated stages. View is ventral in A-E, anterior in F. Tails have been removed for clarity. A-B: Expression is contiguous between ...

Lineage tracing of Tbx4 reflects this early continuity: at E13.5 sections through the posterior of the embryo show a broad swath of Tbx4-lineage tissue that encompasses the mesenchyme of the hindlimb, genital tubercle, and urogenital sinus (Fig. 3G-I). In neonates, much of the urogenital system is derived partially or wholly from the Tbx4 lineage (Fig. 3J). The mesenchyme (but not the epithelium) of the bladder, which arises from the urogenital sinus, is entirely composed of Tbx4-lineage cells (Fig. 3K). The urachus, which helps hold the bladder in place and is the post-natal remnant of the allantois, is also wholly composed of Tbx4-lineage cells (Fig. 3L), while the uterus, cervix, and hindgut mesenchyme show partial contribution of Tbx4-lineage tissues (Fig. 1N, W, Fig. 3G, J, M). All mesenchyme of the male and female external genitalia, except for the proximal ventral border, is comprised of Tbx4-lineage cells, including the prepuce, glans, and erectile tissues (Fig. 3N-O). Also in the posterior embryo, we note Tbx4 lineage contribution at low levels in both the ovary and the testes (Fig. 3P-Q).

Tbx4 expression and lineage in heart and lung

Expression of Tbx4 can be seen in the lung by early E10.5, when the first bifurcation of the lung bud is observed (Fig. 4A). Uniform expression is seen throughout the mesenchyme of the lung at all stages examined (Fig. 4A-C). Lineage tracing confirms the expression data; the entire lung mesenchyme is positive for the Tbx4 lineage reporter (Fig. 4D). At E13.5, the rostral-most extent of the Tbx4 lineage is seen in scattered cells around the trachea at the level of the laryngeal aditus (Fig. 4E). Caudal to the pharyngeal opening, the mesenchyme of the trachea and branching lung is entirely composed of Tbx4-lineage cells, while the epithelium shows no reporter expression (Fig. 4F-G). Similar results were observed at P1 (Fig. 4H). Interestingly, the pulmonary vein and artery are also composed of Tbx4-lineage cells, suggesting that these vessels arise from the surrounding lung mesenchyme (Fig. 4H).

Fig. 4
Tbx4 expression and lineage in lung and heart. A-C: Tbx4 expression in the lung at indicated stages. lb = lung bud, h = heart, li = liver, t = trachea. D: Whole mount ß-galactosidase staining to mark the Tbx4 lineage. E-H: Sectioned embryos stained ...

Conflicting data exist regarding the expression of Tbx4 in the heart. One study reports expression in the atria (Chapman et al., 1996), another reports no expression in the atria but weak expression in the outflow tract and right ventricle (Krause et al., 2004), while yet another found no expression of Tbx4 in the heart (Gibson-Brown et al., 1998b). In E12.5-P1 mice, we found no Tbx4 expression or Tbx4-lineage cells in any region of the heart (Fig. 4B-D, G, H), suggesting that Tbx4 is not expressed at physiological levels at any stage of development in this organ.

Tbx4 expression in allantois

Previous work has shown Tbx4 expression in the allantois, the primordium of the umbilicus (Inman and Downs, 2007), at E7.5, but little is known about the localization or temporal limits of its expression (Chapman et al., 1996). Whole mount in situ hybridization of wildtype embryos shows that faint Tbx4 expression initiates in the extraembryonic mesoderm as early as the late streak stage (Downs and Davies, 1993), before the morphological allantoic bud can be seen in unstained embryos (Fig. 5A, A’), and strong expression can be observed throughout the allantois by the late bud stage (Fig. 5B). When examined by whole mount in situ hybridization, Tbx4 expression encompasses the entire allantois through all stages of its growth and attachment and fusion to the chorion (bud stage-E8.5, Fig. 5C-F). After remodeling of the allantois into the umbilical cord, Tbx4 expression is observed in the umbilical vessels and the fetal placenta until E11.5, after which Tbx4 expression is faint or absent in the umbilicus and placenta (Fig. 5G-I and data not shown).

Fig. 5
Expression of Tbx4 in the allantois. A-I: Whole mount in situ hybridization with Tbx4 at stages indicated. Posterior is to the right in all panels except C’, which is a posterior view of the embryo in C. Red arrowheads = allantois, LS = late streak ...

Despite the apparently homogenous expression seen in whole mount preparations, sectioning reveals that Tbx4 is expressed only in a subset of the cells of the allantois (Fig. 5J-M). At all stages examined, Tbx4 is expressed in all cells of the mesothelial (outer) layer of the allantois (Fig. 5J-L and data not shown). In pre-fusion allantoises, Tbx4 is expressed in most of the allantois, but a few cells that do not express Tbx4 (Tbx4-negative cells) are seen scattered throughout the allantois with no obvious pattern (data not shown). In 8-10 somite allantoises, clear organization can be seen in the Tbx4-positive and negative cells (Fig. 5J-L and Supp. Fig. 1). At the base of the allantois, where cells have recently emerged from the primitive streak (Downs and Harmann, 1997; Kinder et al., 1999), most cells are Tbx4 positive and only a few scattered Tbx4-negative cells are observed (Fig. 5J). In the middle of the allantois, the field of Tbx4-positive cells is interspersed with Tbx4-negative cells, which appear in small clusters frequently located immediately subjacent to the mesothelium (Fig. 5K). At the tip of the allantois, which contains cells that have been in the allantois for the longest period, the Tbx4 negative cells are clearly localized to large clusters subjacent to the mesothelium and to the central umbilical vessel (Fig. 5L). After remodeling of the allantois, Tbx4 expression surrounds both umbilical vessels, but the endothelium is Tbx4-negative (Fig. 5M).

The observation that the endothelium of E9.5 umbilical vessels does not express Tbx4 led us to investigate whether the Tbx4-negative cells in the allantois might also be developing endothelial cells. In 2-4 somite embryos, Pecam protein, which marks developing endothelial cells, is found in small clumps of cells scattered throughout the mesoderm of the allantois (Downs et al., 1998), but excluded from the mesothelium (data not shown). By 8-10 somites, the base of the allantois contains scattered Pecam-positive cells, while the distal allantois shows Pecam staining in cell clusters subjacent to the mesothelium and in the central vessel (Fig. 5N-P). Pecam also marks circulating erythrocytes, which can be observed in the lumen of some sections of the central vessel (Fig. 5P, Supp. Fig. 1C). This pattern appears to be complementary to the pattern of Tbx4 expression (Fig. 5Q), suggesting that Tbx4 and Pecam expression is mutually exclusive. Unfortunately, double labeling was not possible due to the lack of a suitable Tbx4 antibody. Additionally, it could not be determined from expression data alone whether cells in the allantois mesenchyme merely downregulate Tbx4 as they differentiate into endothelium, or if potential endothelial cells never express Tbx4, so we examined the lineage fates of Tbx4 expressing tissues.

Fate of the Tbx4 lineage in the allantois

As expected, the Tbx4 lineage comprised most of the allantois and umbilicus at E8.5-13.5, as well as tracing the vasculature and other allantois-derived tissues of the fetal placenta (Fig. 6A-D). In sections of early stage allantoises, the majority of the allantois was derived from the Tbx4 lineage, although scattered cells could be seen which were not part of the Tbx4-lineage and whose identity could not be clearly determined (Fig. 6E). No Tbx4 lineage was observed in other extraembryonic tissues, such as the yolk sac, amnion, or unfused portions of the chorion (data not shown).

Fig. 6
Tbx4 lineage in the allantois and umbilicus. A-D: Whole mount florescence (A) or lacZ staining (B-D) shows where Tbx4-cre has excised R26R-YFP (A) or R26R-lacZ (B-G). In C-D, the placenta was bisected near the umbilicus to expose a cross section. The ...

After remodeling of the allantois into the umbilical cord it became clear that, while most of the umbilicus is part of the Tbx4 lineage, the endothelial lining of the umbilical arteries and veins show no contribution of Tbx4-lineage tissues (Fig. 6F). Within the fetal placenta, this situation is more variable, with both vessels composed of Tbx4-lineage cells and vessels with no Tbx4-lineage (Fig. 6G). We examined whether the tissue that was negative for the Tbx4 lineage trace was endothelium by double labeling with antibodies against ß-galactosidase and Pecam (Fig. 6H-M), and thus confirmed that the endothelium of the umbilicus and parts of the placenta are derived from cells that are not part of the Tbx4 lineage.

The vessels of the allantois/umbilicus are created via a process called vasculogenesis, which differs from angiogenesis, in which new blood vessels form by sprouting from preexisting vessels (Baron and Fraser, 2005). In vasculogenesis, a previously unpatterned mesenchyme forms blood vessels via the de novo differentiation of individual endothelial cells from the vasculogenic mesenchyme, which then coalesce into tubules and eventually mature vessels. Vasculogenesis primarily occurs in the allantois, yolk sac, and the aorta-gonad-mesonephros (AGM) region of the embryo, and the vasculogenic endothelium is the tissue that later becomes capable of producing all of the hematopoietic stem cells (HSCs) of the body (Baron and Fraser, 2005).

Our results indicate that the scattered Tbx4-negative cells observed in the expression data are, in fact, the prospective endothelial cells, and that unlike the bulk of the allantois mesenchyme, these cells do not express Tbx4 at any point in their development. This implies that the vasculogenic mesenchyme of the allantois is not naive; in fact, it contains a compartment of cells which do not at any point express the same genes as the surrounding mesenchyme and which are solely capable of responding to vasculogenic signals.

It is not clear whether this prepatterning is a feature unique to the allantois. The allantois is unlike other vasculogenic regions in that it can initiate vasculogenesis independent of Hedgehog signaling (Astorga and Carlsson, 2007), and the allantois-derived endothelium of the umbilicus and placenta may produce HSCs earlier and more abundantly than other tissues (Gekas et al., 2005; Ottersbach and Dzierzak, 2005; Zeigler et al., 2006; Yokomizo and Dzierzak, 2010). However, it is equally possible that other vasculogenic tissues possess the same compartments, the discovery of which awaits a similar fortuitous lineage analysis.

Interestingly, our observations also imply that the severe defect in vasculogenesis in the Tbx4 null mutant embryos is due to non-cell autonomous defects, since the endothelium does not express Tbx4 at any point. This illustrates an important role for the surrounding mesenchyme in vasculogenesis, and identifies the allantois as a potential model system for investigating these interactions.

While not previously emphasized, expression of Tbx4 in a single ventral domain that encompasses the allantois, hindlimb, and proctodeum appears to be an evolutionarily conserved feature and can be observed in possum, chick, and in the ancestral chordate tbx4/5 gene of amphioxus (Gibson-Brown et al., 1998a; Horton et al., 2008; Keyte and Smith, 2010). These data might suggest that the Tbx4-positive mesenchyme has a particular facility for generating appendages, and/or that appendages in this region might share similar characteristics. It has long been known that the limb and genital tubercle show a number of regulatory similarities, including requirements for hedgehog and fibroblast growth factor (FGF) signaling for their growth and patterning, supporting this speculation (Kondo et al., 1997; Minelli, 2002; Yamada et al., 2006). Conversely, the allantois does not require hedgehog signaling for its growth or vascular patterning (Astorga and Carlsson, 2007), and while our recent work has shown that loss of FGF signaling impairs allantois growth (see supplemental material in Naiche et al., 2011), it is currently unclear whether the growth defect is secondary to defects in the primitive streak. Current evidence suggests that the allantois is actually not an appendage, but rather may be an extension of the primitive streak (Downs et al., 2009). Further study will elucidate whether all of the outgrowths of the Tbx4 mesenchyme share common regulation.


In situ hybridization

Timed pregnancies from ICR females mated to ICR males were harvested at indicated developmental stages. For whole-mount lung staining, the lungs, still attached to the heart, were dissected out of fresh embryos and fixed separately. All samples were fixed in 4% paraformaldehyde (PFA) in phosphate-buffered saline (PBS) at 4° for 24-48 hours, then dehydrated stepwise into methanol. Whole mount in situ hybridizations were performed according to established protocols (Nagy et al., 2003).

Lineage tracing

Timed matings were set up between heterozygous Tbx4-Cre (Luria et al., 2008) males and R26R-lacZ (Soriano, 1999) or R26R-YFP (Srinivas et al., 2001) homozygous females. For whole mount ß-galactosidase staining, embryos were fixed briefly in 4% PFA (30-120 minutes) at 4° and transferred to X-gal stain at 37°overnight. For sections, E9.5 embryos were stained in whole mount and then paraffin embedded for sectioning; older embryos were frozen without fixation in OTC media (Tissue-Tek 4583), cryosectioned at 20μm, fixed briefly in PFA, and transferred to X-gal stain overnight. Counterstain was either neutral red or nuclear fast red. Controls for background staining were conducted on Cre-negative littermates in all experiments, but no background was observed except in the post-natal gut (Fig. 3J-M and data not shown).

Antibody staining

For antibody staining, embryos were dehydrated in 30% sucrose in PBS overnight at 4°, then embedded in OTC media and cryosectioned at 8-10μm. Immunohistochemistry was performed using standard protocols and anti-Pecam (BD Pharmingen # 550274) and anti-ß-galactosidase (MP Biomedicals # 55976).

Supplementary Material

Supp Fig S1

Supporting Figure 1:

Pattern of Tbx4 expression in the allantois. A-I: Sections from the base (A, D, G), middle (B, E, H), tip (C, F, I) of three additional 8-10 somite stage allantoises hybridized with Tbx4, showing the reproducible staining pattern. Each row contains sections from a single embryo. cv = central vessel, am = amnion, thin line divides allantois from amnion.


We thank Thomas Jessell for material and mentoring assistance, Xin Sun and Bin Zhou for pathology assistance, Jeremy Gibson-Brown for commenting on the manuscript, and the NCI histopathology lab for their help with sectioning.

Funding: NCI intramural funding (M. L.), NIH grant number RO1-HD033082 (V. E. P.)


  • Astorga J, Carlsson P. Hedgehog induction of murine vasculogenesis is mediated by Foxf1 and Bmp4. Development. 2007;134:3753–3761. [PubMed]
  • Baron MH, Fraser ST. The specification of early hematopoiesis in the mammal. Current Opinion in Hematology. 2005;12:217–221. [PubMed]
  • Bongers EMHF, Duijf PHG, Beersum SEMv, Schoots J, Kampen Av, Burckhardt A, Hamel BCJ, Lošan F, Hoefsloot LH, Yntema HG, Knoers NVAM, Bokhoven Hv. Mutations in the human TBX4 gene cause small patella syndrome. Am J Hum Genet. 2004;74:1239–1248. [PMC free article] [PubMed]
  • Cebra-Thomas JA, Bromer J, Gardner R, Lam GK, Sheipe H, Gilbert SF. T-box gene products are required for mesenchymal induction of epithelial branching in the embryonic mouse lung. Dev Dyn. 2003;226:82–90. [PubMed]
  • Chapman DL, Garvey N, Hancock S, Alexiou M, Agulnik SI, Gibson-Brown JJ, Cebra-Thomas J, Bollag RJ, Silver LM, Papaioannou VE. Expression of the T-box family genes, Tbx1-Tbx5, during early mouse development. Developmental Dynamics. 1996;206:379–390. [PubMed]
  • Chevallier A, Kieny M, Mauger A. Limb-somite relationship: origin of the limb musculature. J Embryol Exp Morphol. 1977;41:245–258. [PubMed]
  • Downs KM, Davies T. Staging of gastrulating mouse embryos by morphological landmarks in the dissecting microscope. Development. 1993;118:1255–1266. [PubMed]
  • Downs KM, Gifford S, Blahnik M, Gardner RL. Vascularization in the murine allantois occurs by vasculogenesis without accompanying erythropoiesis. Development. 1998;125:4507–4520. [PubMed]
  • Downs KM, Harmann C. Developmental potency of the murine allantois. Development. 1997;124:2769–2780. [PubMed]
  • Downs KM, Inman KE, Jin DX, Enders AC. The Allantoic Core Domain: new insights into development of the murine allantois and its relation to the primitive streak. Dev Dyn. 2009;238:532–553. [PMC free article] [PubMed]
  • Gekas C, Dieterlen-Lièvre F, Orkin SH, Mikkola HKA. The placenta is a niche for hematopoietic stem cells. Developmental Cell. 2005;8:365–375. [PubMed]
  • Gibson-Brown JJ, Agulnik SI, Chapman DL, Alexiou M, Garvey N, Silver LM, Papaioannou VE. Evidence of a role for T-box genes in the evolution of limb morphogenesis and the specification of forelimb/hindlimb identity. Mechanisms of Development. 1996;56:93–101. [PubMed]
  • Gibson-Brown JJ, Agulnik SI, Silver LM, Niswander L, Papaioannou VE. Involvement of T-box genes Tbx2-Tbx5 in vertebrate limb specification and development. Development. 1998a;125:2499–2509. [PubMed]
  • Gibson-Brown JJ, I A S, Silver LM, Papaioannou VE. Expression of T-box genes Tbx2-Tbx5 during chick organogenesis. Mechanisms of Development. 1998b;74:165–169. [PubMed]
  • Hasson P, DeLaurier A, Bennett M, Grigorieva E, Naiche LA, Papaioannou VE, Mohun TJ, Logan MP. Tbx4 and Tbx5 acting in connective tissue are required for limb muscle and tendon patterning. Dev Cell. 2010;18:148–156. [PMC free article] [PubMed]
  • Horton AC, Mahadevan NR, Minguillon C, Osoegawa K, Rokhsar DS, Ruvinsky I, de Jong PJ, Logan MP, Gibson-Brown JJ. Conservation of linkage and evolution of developmental function within the Tbx2/3/4/5 subfamily of T-box genes: implications for the origin of vertebrate limbs. Dev Genes Evol. 2008;218:613–628. [PubMed]
  • Inman KE, Downs KM. The murine allantois: emerging paradigms in development of the mammalian umbilical cord and its relation to the fetus. Genesis. 2007;45:237–258. [PubMed]
  • Keyte AL, Smith KK. Developmental origins of precocial forelimbs in marsupial neonates. Development. 2010;137:4283–4294. [PubMed]
  • Kinder SJ, Tsang TE, Quinlan GA, Hadjantonakis AK, Nagy A, Tam PP. The orderly allocation of mesodermal cells to the extraembryonic structures and the anteroposterior axis during gastrulation of the mouse embryo. Development. 1999;126:4691–4701. [PubMed]
  • King M, Arnold JS, Shanske A, Morrow BE. T-genes and limb bud development. Am J Med Genet A. 2006;140:1407–1413. [PubMed]
  • Kondo T, Zakany J, Innis JW, Duboule D. Of fingers, toes and penises. Nature. 1997;390:29. [PubMed]
  • Krause A, Zacharias W, Camarata T, Linkhart B, Law E, Lischke A, Miljan E, Simon HG. Tbx5 and Tbx4 transcription factors interact with a new chicken PDZ-LIM protein in limb and heart development. Dev Biol. 2004;273:106–120. [PubMed]
  • Luria V, Krawchuk D, Jessell TM, Laufer E, Kania A. Specification of motor axon trajectory by ephrin-B:EphB signaling: symmetrical control of axonal patterning in the developing limb. Neuron. 2008;60:1039–1053. [PubMed]
  • Malashichev Y, Christ B, Prols F. Avian pelvis originates from lateral plate mesoderm and its development requires signals from both ectoderm and paraxial mesoderm. Cell Tissue Res. 2008;331:595–604. [PubMed]
  • Menke DB, Guenther C, Kingsley DM. Dual hindlimb control elements in the Tbx4 gene and region-specific control of bone size in vertebrate limbs. Development. 2008;135:2543–2553. [PubMed]
  • Minelli A. Homology, limbs, and genitalia. Evol Dev. 2002;4:127–132. [PubMed]
  • Minguillon C, Del Buono J, Logan MP. Tbx5 and Tbx4 are not sufficient to determine limb-specific morphologies but have common roles in initiating limb outgrowth. Dev Cell. 2005;8:75–84. [PubMed]
  • Nagy A, Gertsenstein M, Vintersten K, Behringer R. Manipulating the Mouse Embryo. Cold Spring Harbor, New York: Cold Spring Harbor Laboratory Press; 2003.
  • Naiche LA, Harrelson Z, Kelly RG, Papaioannou VE. T-box genes in vertebrate development. Annu Rev Genet. 2005;39:219–239. [PubMed]
  • Naiche LA, Holder N, Lewandoski M. FGF4 and FGF8 comprise the wavefront activity that controls somitogenesis. Proc Natl Acad Sci U S A. 2011;108:4018–4023. [PMC free article] [PubMed]
  • Naiche LA, Papaioannou VE. Loss of Tbx4 blocks hindlimb development and affects vascularization and fusion of the allantois. Development. 2003;130:2681–2693. [PubMed]
  • Naiche LA, Papaioannou VE. Cre activity causes widespread apoptosis and lethal anemia during embryonic development. Genesis. 2007a;45:768–775. [PubMed]
  • Naiche LA, Papaioannou VE. Multiple roles of T-box genes. In: Moody S, editor. Principles of Developmental Genetics. Elsevier Ltd; 2007b. pp. 341–359.
  • Naiche LA, Papaioannou VE. Tbx4 is not required for hindlimb identity or post-bud hindlimb outgrowth. Development. 2007c;134:93–103. [PubMed]
  • Ottersbach K, Dzierzak E. The murine placenta contains hematopoietic stem cell within the vascular labyrinth region. Developmental Cell. 2005;8:377–387. [PubMed]
  • Ouimette JF, Jolin ML, L’Honore A, Gifuni A, Drouin J. Divergent transcriptional activities determine limb identity. Nat Commun. 2010;1:1–9. [PMC free article] [PubMed]
  • Rodriguez-Esteban C, Tsukui T, Yonei S, Magallon J, Tamura K, Izpisua Belmonte JC. The T-box genes Tbx4 and Tbx5 regulate limb outgrowth and identity. Nature. 1999;398:814–818. [PubMed]
  • Ruvinsky I, Oates AC, Silver LM, Ho RK. The evolution of paired appendages in vertebrates: T-box genes in the zebrafish. Development Genes & Evolution. 2000;210:82–91. [PubMed]
  • Sakiyama J, Yamagishi A, Kuroiwa A. Tbx4-Fgf10 system controls lung bud formation during chicken embryonic development. Development. 2003;130:1225–1234. [PubMed]
  • Showell C, Binder O, Conlon FL. T-box genes in early embryogenesis. Dev Dyn. 2004;229:201–218. [PMC free article] [PubMed]
  • Soriano P. Generalized lacZ expression with the ROSA26 Cre reporter strain. Nat Genet. 1999;21:70–71. [PubMed]
  • Srinivas S, Watanabe T, Lin CS, William CM, Tanabe Y, Jessell TM, Costantini F. Cre reporter strains produced by targeted insertion of EYFP and ECFP into the ROSA26 locus. BMC Dev Biol. 2001;1:4. [PMC free article] [PubMed]
  • Takeuchi JK, Koshiba-Takeuchi K, Matsumoto K, Vogel-Hopker A, Naitoh-Matsuo M, Ogura K, Takahashi N, Yasuda K, Ogura T. Tbx5 and Tbx4 genes determine the wing/leg identity of limb buds. Nature. 1999;398:810–814. [PubMed]
  • Tamura K, Yonei-Tamura S, Izpisua Belmonte JC. Differential expression of Tbx4 and Tbx5 in Zebrafish fin buds. Mech Dev. 1999;87:181–184. [PubMed]
  • Tanaka M, Munsterberg A, Anderson WG, Prescott AR, Hazon N, Tickle C. Fin development in a cartilaginous fish and the origin of vertebrate limbs. Nature. 2002;416:527–531. [PubMed]
  • Valasek P, Evans DJ, Maina F, Grim M, Patel K. A dual fate of the hindlimb muscle mass: cloacal/perineal musculature develops from leg muscle cells. Development. 2005;132:447–458. [PubMed]
  • Wang Z, Dong D, Ru B, Young RL, Han N, Guo T, Zhang S. Digital gene expression tag profiling of bat digits provides robust candidates contributing to wing formation. BMC Genomics. 2010;11:619. [PMC free article] [PubMed]
  • Wardle FC, Papaioannou VE. Teasing out T-box targets in early mesoderm. Curr Opin Genet Dev. 2008;18:418–425. [PMC free article] [PubMed]
  • Yamada G, Suzuki K, Haraguchi R, Miyagawa S, Satoh Y, Kamimura M, Nakagata N, Kataoka H, Kuroiwa A, Chen Y. Molecular genetic cascades for external genitalia formation: an emerging organogenesis program. Dev Dyn. 2006;235:1738–1752. [PubMed]
  • Yokomizo T, Dzierzak E. Three-dimensional cartography of hematopoietic clusters in the vasculature of whole mouse embryos. Development. 2010;137:3651–3661. [PMC free article] [PubMed]
  • Zeigler BM, Sugiyama D, Chen M, Guo Y, Downs KM, Speck NA. The allantois and chorion, when isolated before circulation or chorio-allantoic fusion, have hematopoietic potential. Development. 2006;133:4183–4192. [PubMed]
PubReader format: click here to try


Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...


Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...