Most bones in the vertebrate skeleton are made in the same way – endochondrial ossification – yet they display a variety of shapes and sizes. The question of how these unique bone morphologies, including the superstructures that protrude from their surfaces, arise during development is still unclear, and the subject of a new paper in Development. We caught up with first author Shai Eyal and his supervisor Elazar Zelzer, Professor in the Department of Molecular Genetics at the Weizmann Institute of Science in Rehovot, Israel, to find out more about the story.

Elazar Zelzer (L) and Shai Eyal (R).

Elazar, can you give us your scientific biography and the questions your lab is trying to answer?

EZ We walk, run and jump using the complex and ingenious musculoskeletal system. It is therefore puzzling that, although each of its components has been extensively studied, research on the musculoskeleton as an integrated system, and in particular on how it is assembled, has been scarce. My vision as a scientist is to demonstrate the centrality of molecular and biomechanical crossregulation between musculoskeletal tissues in the development and function of this system. In 2004, I established my independent research group at the Weizmann Institute. Over the years, I was joined by a group of extremely talented students. Together, we have made interesting discoveries about various aspects of musculoskeletal biology, including the regulatory interactions between the developing skeleton and its vasculature, the regulation of bone morphogenesis and joint formation by muscle-induced mechanical signals, and the development of secondary structures such as bone eminences and sesamoid bones.

One of the cornerstones of my philosophy as a mentor of young creative people is to allow team members space for individual expression. I believe that this has been key to our ability to develop new avenues of research. Indeed, we have recently opened a new and exciting direction by introducing the significance of proprioceptive signalling in skeletal regulation, specifically in spine alignment, bone regeneration and joint development.

And Shai, how did you come to work in the Zelzer lab, and what drives your research today?

SE I majored in Life Sciences at the Hebrew University of Jerusalem. During that time, I was exposed to various fields of biological research; however, the field that really caught my attention was developmental biology. When I think of how a single cellular unit, a fertilized egg, can develop into endless forms and shapes, it is like biological magic to me. Therefore, when I moved to the Weizmann Institute of Science, I looked for labs in which I could continue studying concepts in developmental biology. Looking into the research that was carried out by the Zelzer lab, I was very intrigued. Skeletogenesis, in particular how the skeletal and muscular systems are integrated, is a very interesting and largely unknown process. With Eli's superb mentorship and the supportive lab atmosphere, I was able to highlight several key processes taking place during sesamoid bone formation and long bone morphogenesis. Today, my interest has shifted away from skeletogenesis, but I am still as hooked on developmental biology as I was when I began my research more than a decade ago.

Prior to your work, what was known about how each bone gets its distinctive shape?

EZ Although the generic mechanisms of long bone development, in particular elongation, have been extensively studied, far less was known about how each bone gets its distinctive shape. Nevertheless, several intriguing discoveries have been made in recent years. Muscle force was shown to regulate appositional growth, which forms the circumferential shape of the bone (Sharir et al., 2011), and bone superstructures were shown to develop modularly from a distinct population of Sox9 and scleraxis (Scx) double-positive progenitor cells under regulation of TGFβ and BMP signalling (Blitz et al., 2009, 2013). Lastly, the relative positions of superstructures along the bone shaft was found to be determined both by bone modelling and by a unique ratio between growth rates at the two bone ends (Stern et al., 2015).

SE The process of endochondral ossification was described a long time ago. In short, it is a process in which pools of Sox9-expressing chondroprogenitors condense into discrete cartilage templates, which later will ossify into mineralized bones. Until recently, the dogma in the field was that long bones acquire their specific morphologies at the stage of cartilage condensation, with some changes occurring after ossification by bone modelling. However, studies that were done by a former graduate student, Einat Blitz, challenged this dogma and presented evidence of a secondary population of chondroprogenitors that co-express both Sox9 and Scx. These cells are specified after the primary condensation stage and their addition forms superstructures along the bone shaft and thereby contributes to the final bone morphology. This finding served as our starting point and raised the question of how the addition of the Sox9+/Scx+ cells to developing bones at specific sites is controlled.

Sagittal sections through the proximal humeri of limbs from control Prx1-Cre;Pbx1floxed mutant embryos, stained against SOX9 (red), COL2A1 (green) and DAPI (blue).

Sagittal sections through the proximal humeri of limbs from control Prx1-Cre;Pbx1floxed mutant embryos, stained against SOX9 (red), COL2A1 (green) and DAPI (blue).

Can you give us the key results of the paper in a paragraph?

SE & EZ First, we show that the Sox9+/Scx+ progenitor cells contribute extensively to the formation of bone superstructures in mouse. Importantly, we show that they contribute not only to bone eminences, but also to condyles and sesamoid bones, which are auxiliary bones that are important for proper locomotion. Then, we show that the genetic programme that controls the patterning of these progenitors contains both global and regional regulatory modules, as Gli3 regulated this process globally, whereas Pbx1, Pbx2, Hoxa11 and Hoxd11 acted as either proximal or distal regulators. Finally, by demonstrating a dose-dependent pattern regulation in Gli3 and Pbx1 compound mutations, we show that the global and regional regulatory modules work in tandem. Together, these results demonstrate genetic regulation of superstructure patterning, thereby supporting the notion that long bone development is a modular process.

How do your findings contribute to the modular model of bone morphogenesis?

EZ Previously, we showed that bone superstructures originate from a dedicated pool of progenitors. Now, we expand the modular model of bone development by identifying part of the genetic programme that controls the patterning of these progenitors at specific locations along the bone. Furthermore, we show that modularity exists within the process of bone superstructure patterning, as some components of this genetic programme regulate all superstructures, whereas others regulate only proximal or distal superstructures. From an evolutionary perspective, this strategy allows changing the position of superstructures and, thereby, the locomotor abilities of the organism without rewriting the entire skeletogenic programme.

SE What allows modularity in long bone morphogenesis is the appearance of the secondary Sox9/Scx progenitors and the fact that they can be regulated independently from the sox9-expressing cells of the bone shaft. Our findings provide further evidence of the abundance and contribution of these progenitors and highlight several genetic components and programmes that regulate their patterning both globally and locally. These are missing from the present dogma of a rigid patterning programme executed by a single condensation step of a single population of progenitors, therefore changing our view of bone morphogenesis.

How do you think that Sox9+/Scx+ chondroprogenitors get to the right sites to form the superstructures?

SE & EZSox9+/Scx+ chondroprogenitors could get to their locations in one of two ways. The first option is their migration to specific condensation sites, where they will form superstructures. Alternatively, it is possible that the Sox9+/Scx+ chondroprogenitors are specified from cells that are already present at the designated superstructure formation sites. In the latter scenario, the mechanism would involve activation of a chondrogenic programme in selected cell subpopulations at specific spatiotemporal positions. Either way, it appears that the mechanism that decides where superstructure development should take place involves Gli3 and the Pbx and Hox genes.

When doing the research, did you have any particular result or eureka moment that has stuck with you?

SE It is hard to choose a specific eureka moment. I think every experiment we performed presented some advancement toward our findings. Perhaps such a moment would be when we first put together the results from our Sox9/Col2A1-CreER;tdTomato 3D reconstructions and the SOX9/ScxGFP-labelled sections. It was then that we saw clearly for the first time how abundant these progenitors were. That was a very reassuring moment that gave me the confidence to move forward with studying the patterning of superstructures.

And what about the flipside: any moments of frustration or despair?

SE I think that working with a model organism, and especially with mutant or transgenic lines, can sometime be a source of frustration. For example, we had to postpone the revision of two papers twice by more than 6 months because we had to revive our double-mutant colonies sufficiently to carry out additional experiments that were requested by the reviewers. Another experiment required us to cross the compound HOX11 mutants to a ScxGFP reporter to follow the Sox9+/Scx+ progenitors. The experiment was straightforward but the genetic ratio was against us and, after two years of trying, we had to submit the manuscript without these results.

So what next for you after this paper – I understand you are now in the US?

SE That is correct. I am now doing my postdoctorate in Professor David Traver's lab at the University of California, San Diego (UCSD). The main areas researched in the lab are the formation and development of the haematopoietic stem cells and their niche in both juvenile and adult zebrafish. The advantage of working with zebrafish is that their embryonic development is rapid and the developing embryos are transparent. That gives us the opportunity to follow cellular behaviours over long periods of time with high-resolution imaging. Specifically, for my projects I am working on setting up a pipeline that will use cutting-edge labelling and imaging technologies to allow me to label and follow single cells and their cellular fates and lineages. That system will allow me to study heterogeneous populations within the blood system by following individual cells within such populations.

Where will this work take the Zelzer lab?

EZ We are interested in further understanding the process of superstructure development, both upstream and downstream to the

current study. We are investigating the unique identity of the progenitor cells that give rise to superstructures, focusing on their single cell transcriptome and chromatin landscape. At the same time, we are pursuing the molecular and biomechanical signals that regulate superstructure development. These studies will shed light not only on bone morphogenesis, but also on the connection of tendons to bones during musculoskeletal assembly. To achieve these goals, we would welcome new partners. We are always looking for curious students and postdocs on the quest for self-fulfilment, who are interested in musculoskeletal biology, organ shaping or the assembly of complex systems.

In my view, the field of developmental biology is currently at a crossroads. Pushing the field back to centre stage may require the integration of other fields such as physiology, engineering and mechanobiology. The musculoskeleton is the ideal model system for such integration. An example for that is the new direction we have recently taken in studying the regulatory effect of the proprioceptive system on skeletal development and function.

The field of developmental biology is currently at a crossroads

Finally, let's move outside the lab – what do you like to do in your spare time in Rehovot and San Diego?

EZ I like mountain biking and practicing jujutsu.

SE I have moved to San Diego with my wife and two amazing children. Coming to San Diego opened up many new experiences for all of us. The city is beautiful and there are endless family attractions, especially in the larger area of Southern California. We also like to travel and camp on long weekends. For myself, I started rock climbing during my PhD in Israel and continue to climb in San Diego. In addition, as I am in California, it didn't take long before I took up surfing, too. It is amazing to have the opportunity to start the day by surfing among sea lions and dolphins.

Weizmann Institute of Science, Department of Molecular Genetics, Rehovot 76100,Israel.

E-mail: eli.zelzer@weizmann.ac.il

Blitz
,
E.
,
Viukov
,
S.
,
Sharir
,
A.
,
Shwartz
,
Y.
,
Galloway
,
J. L.
,
Pryce
,
B. A.
,
Johnson
,
R. L.
,
Tabin
,
C. J.
,
Schweitzer
,
R.
and
Zelzer
,
E.
(
2009
).
Bone ridge patterning during musculoskeletal assembly is mediated through SCX regulation of Bmp4 at the tendon-skeleton junction
.
Dev. Cell
17
,
861
873
.
Blitz
,
E.
,
Sharir
,
A.
,
Akiyama
,
H.
and
Zelzer
,
E.
(
2013
).
Tendon-bone attachment unit is formed modularly by a distinct pool of Scx- and Sox9-positive progenitors
.
Development
140
,
2680
2690
.
Eyal
,
S.
,
Kult
,
S.
,
Rubin
,
S.
,
Krief
,
S.
,
Felsenthal
,
N.
,
Pineault
,
K. M.
,
Leshkowitz
,
D.
,
Salame
,
T.-M.
,
Addadi
,
Y.
,
Wellik
,
D. M.
, et al. 
(
2019
).
Bone morphology is regulated modularly by global and regional genetic programs
.
Development
146
,
dev167882
.
Sharir
,
A.
,
Stern
,
T.
,
Rot
,
C.
,
Shahar
,
R.
and
Zelzer
,
E.
(
2011
).
Muscle force regulates bone shaping for optimal load-bearing capacity during embryogenesis
.
Development
138
,
3247
3259
.
Stern
,
T.
,
Aviram
,
R.
,
Rot
,
C.
,
Galili
,
T.
,
Sharir
,
A.
,
Kalish Achrai
,
N.
,
Keller
,
Y.
,
Shahar
,
R.
and
Zelzer
,
E.
(
2015
).
Isometric scaling in developing long bones is achieved by an optimal epiphyseal growth balance
.
PLoS Biol.
13
,
e1002212
.