All posts by lijingyi

Brief Review of Pigment Generation, Part IV (BIOL 5104 blog 10)

The settlement of melanoblast and the generation of pigment

After dorsolaterally migration, the melanoblast finds its destinations; and settled down surrounded by epithelial cells. What attracts melanoblast to settle in these particular regions in molecular level?

The role of the apical ectodermal ridge and of fibroblast growth factors FGF-2 and FGF-4 and of the insulin-like growth factor I (IGF-I) in the control of the migration of epidermal melanoblast into wing buds was investigated using quail–chicken chimeras. As the results, both the apical ectodermal ridge and the growth factors invariably caused the migration of epidermal melanoblasts towards them. Furthermore, FGF-2 and IGF-I and to a lesser extent FGF-4 play a decisive role in directing the migration of epidermal melanoblasts within chicken wing buds and are likely to be involved in the molecular cascade by means of which the apical ectodermal ridge controls the migration of epidermal melanoblasts. Although IGF-1 acts as upstream regulator of FGFs; FGFs have been shown to promote survival and proliferation of melanoblasts, further study has shown that the proliferation of melanoblasts can also be promoted by IGF-I, independent with FGFs (Schofer, et al., 2001).

Reaching to the destination is not sufficient for melanoblast settlement. As a kind of steam cell, melanoblast needs a population of cells (known as a “niche”) surrounding it to keep its capacity to self-renew and generate differentiated progeny (Nishimura, et al., 2002). At specific cutaneous sites, epithelial cells use Foxn1 (a transcription factor) to recruit melanocytes and induce their own pigmentation. Foxn1 thus defines a distinct cell population that ultimately controls the targeting of pigment in the skin (Weiner, et al., 2007).

After settlement, the melanoblast will start to exert its role, differentiated into melanocyte and generate pigmentation.

Mammals develop most of their coloration through a system comprised of two types of cells (reviewed by Slominski et al., 2004), referred to here as pigment donors and recipients. The pigment donors are melanocytes, which synthesize melanin in distinct organelles called melanosomes. The pigment recipients are epithelial cells, which acquire and hold most cutaneous melanin. As the system forms, each melanocyte extends dendrites and contacts multiple epithelial cells, creating a ‘‘pigmentary unit.’’ Melanosomes are then transported along the dendrites and into the epithelial cells, which may internalize the melanosomes via phagocytosis (Weiner, et al., 2007).

But the relationship between the donor melanocytes and the recipient epithelial cells are not changeless. Some differentiated melanocytes can return to the “niche” and become stem-cell again (Nishimura, et al., 2002).

Studies of pigment generation in melanocytes always focus on the function of the three transcription factors MITF, PAX3, and SOX10. The pathways of these genes interact to regulate crucial aspects of melanocyte development and function. Microphthalmia-associated Transcription Factor (MITF) has been termed the “melanocyte master regulator”, because it plays such a central role in melanocyte development and function. It is required for melanocyte differentiation and survival, activates transcription of melanogenic enzymes and melanogenic proteins, and is associated with melanoma progression. MITF also governs numerous other cellular functions in the melanocyte, including environmental response, cell survival, cell motility, and cell cycle progression. MITF itself undergoes complex post-transcriptional regulation, including phosphorylation, sumoylation, ubiquitination, and caspase cleavage. Paired box gene 3 (PAX3) has a broader expression pattern than MITF, regulating neural tube closure, early development of myoblast and neural crest lineages, and the formation of nervous, muscular, cardiovascular and melanocyte systems. PAX3 regulation of early neural crest development appears to maintain neural crest stem cell properties via inhibition of apoptosis, and PAX3 has been proposed to inhibit apoptosis in melanoma. In melanocytes, PAX3 activates transcription of MITF and plays a crucial role in maintaining melanocyte stem cells. SRY-box containing gene 10 (SOX10) regulates specification of neural crest-derived melanocytes, neurons, and glia. SOX10 strongly activates MITF and regulates expression of melanogenic enzymes. Upstream regulation of SOX10, as well as additional downstream targets, interacting factors, and posttranslational modifications are only beginning to be ascertained. In summary, MITF, PAX3, and SOX10 serve to illustrate that the extensive data currently known on cellular processes governing melanocyte biology will provide an ideal foundation for future systems biology analyses in these cells (reviewed by Baxter et al., 2009).


Schofer, C., Frei, K., Weipoltshammer, K., and Wachtler, F. The apical ectodermal ridge, fibroblast growth factors (FGF-2 and FGF- 4) and insulin-like growth factor I (IGF-I) control the migration of epidermal melanoblasts in chicken wing buds. Anat. Embryol. (Berl.)  2001. 203, 137–146.

Emi K. Nishimura, Siobhan A. Jordan, Hideo Oshima, Hisahiro Yoshida, Masatake Osawa, Mariko Moriyama, Ian J. Jackson§, Yann Barrandonk, Yoshiki Miyachi, Shin-Ichi Nishikawa. Dominant role of the niche in melanocyte stem-cell fate determination. Nature, 2002, Vol 416, 25 April. 854-860.

Lorin Weiner, Rong Han, Bianca M. Scicchitano, Jian Li, Kiyotaka Hasegawa, Maddalena Grossi, David Lee, Janice L. Brissette. Dedicated Epithelial Recipient Cells Determine Pigmentation Patterns. Cell. 2007. 07. 024.

Slominski, A., Tobin, D.J., Shibahara, S., and Wortsman, J. Melanin pigmentation in mammalian skin and its hormonal regulation. Physiol. Rev. 2004. 84, 1155–1228.

Laura L Baxter, Stacie K Loftus, William J Pavan. Networks and pathways in pigmentation, health, and disease. Wiley Interdiscip Rev Syst Biol Med. 2009 November 1; 1(3): 359–371.

Hox genes are involved in evolution of two-winged Drosophila. (BIOL 5104 blog 9)

As we learnt from the text book, the four-winged Drosophila is the phenotype (a homeotic transformation of the halteres of the third thoracic segment changed into the wings normally associated with the second thoracic segement) that results from a Ubx gene mutation. But, of course, there are many insects, such as dragonflies, butterflies, and bees, which normally have four wings. Is two-winged flies evolved from four-winged ancestors, or opposite?

To investigate this question, I look up the text book of evolution. In fact, all the phylogenetic relationship of various insect groups and evidence from the fossil record indicate that the ancestor of all these winged insects was one with four wings. During the lineage leading to files, the wings of the third thoracic segment were modified into halteres. This is not surprising, at least for me. In my point of view, flies are very successful animals and should be more evolutionary than most insects. And also I think evolution of wings from halteres is harder than degeneration of halteres from wings. But, how this evolutionary event happened in history? Scientists have at least three hypotheses to explain the transition from four-winged to two-winged insects.

About 60 years ago, the first hypothesis is that the four-winged common ancestor of flies and dragonflies lacked the Ubx gene. Somewhere in the evolutionary lineage leading to flies, Ubx appeared and resulted in the transformation of the third thoracic wings into halteres. But after the development of molecular biology, it is discovered that Ubx gene is present throughout the insects, this hypothesis could be rejected.

Then, it is common to think that the Ubx gene is in all insects but expressed different between two-winged and four-winged insects. So the second hypothesis was that the evolution of two-winged from four-winged insects is caused by the mutations that affected the regulation of Ubx. But when the Ubx expression pattern was examined in a number of different insects, it was found to be highly conserved, even in wingless insects. So, the second hypothesis could also be rejected.

The third hypothesis, currently favored, focus on the downstream targets of Ubx gene. Evidence shown that in flies, various genes involved in processes such as vein formation, sensory hair growth, and cell proliferation came under the control of Ubx. Considering that the wings have veins and sensory hairs, whereas the halteres do not; the wing is far larger and contains a greater number of cells than does a haltere, it is very likely the expression of Ubx in flies’ third thoracic segment cause the formation of halteres instead of wings. But these same genes in butterflies are not regulated by Ubx, at least not in the same way.

Combining these three hypotheses and combining the basic rule of Darwinian Theory, these three kinds of mutations must be happened in history (and even nowadays), but most of them caused lethal larva, like the null mutation of Ubx gene. Some of these mutations lower the fitness of insects that might have less offspring than others, like wings are not developed in flies’ third thoracic segment nor are halteres, which make flies have trouble on flying. After one or more generation, these mutates are completely die out and we cannot see such population in wild nowadays. What we can see today is the only few of the collection of such mutations, which is rare events, but the result is the dominance of such mutates. This is the most attractive parts of evolution for me.


Nicholas H. Barton, Derek E.G. Briggs, Jonathan A. Eisen, David B. Goldstein, Nipam H. Patel. 2007. Evolution. Cold Spring Harbor Laboratory Press

Why Enucleation in Mammals Erythrocyte? (BIOL 5104 blog 8)

The sentence “…a small number of fetal red blood cells are seen in the maternal blood circulation.” in the text book P309 triggered my thinking weather that is one of the reasons vivipara do not have nucleated red blood cell in their circulatory system. If large amount of nucleated blood cell in the mother’s blood occasionally contact with fetus’ blood, strong immune response will kill the fetus, so under selection, only individuals with enucleated erythrocytes can survive, I guess.

In order to test my hypothesis, I started searching for information. Firstly, I found camel and giraffe’ erythrocytes have nuclear by Google. I thought if there is evidence showing their placenta structures are different from other mammals to have less possibility of maternal and fetal blood mixing, which should be a support of my hypothesis. But later, after searched for some academic articles, I found it is wrong. “All are unique among mammals, because camel’s erythrocytes are biscuit- or wafer-shaped ovals, with very little volume, and having little packed volume in whole blood (only 27-28 %). The hypothesis that they throw back to the ancestral reptiles and other lower vertebrates seemed superficially obvious, and perhaps led to the error often published that camel erythrocytes are nucleated. In fact, they are anucleate, but are unusual and novel in vertebrates.” (Long, 2007)

Furthermore, I found a credible source meaning that enucleated erythrocytes are dominant in blood in all mammal individuals including all marsupials, placental mammals, Prototheria and monotremes. Meanwhile all known non-mammalian vertebrates, including fish, amphibian, reptiles and birds, have nucleated erythrocytes (Baskurta and Meiselman, 2010). But in certain conditions, some animals will change the amount of nucleated erythrocytes or enucleated erythrocytes in their blood. Some marsupials have some circulating nucleated red blood cells as adults (Stonehouse, 1977). “Occasionally, reptilian RBC lacking nuclei are observed. These are called ‘erythroplastids,’ and their extruded nuclei found in the plasma are referred to as ‘hematogones’” (Mauro, 1997). In fetuses with intrauterine growth restriction, higher nucleated red blood cell counts at birth and longer persistence of nucleated red blood cell count are observed. Evidence shows that metabolic acidemia is in charge of causing that (Baschat, et al., 1999). These facts seem to be refutes to my hypothesis. So I decided to focus on the differences of erythropoiesis between mammal and non-mammals.

Then, I first realized, in early stage of human embryos, blood cells are not only generated by marrow. “In developing embryos, blood formation occurs in aggregates of blood cells in the yolk sac, called blood islands. As development progresses, blood formation occurs in the spleen, liver and lymph nodes. When bone marrow develops, it eventually assumes the task of forming most of the blood cells for the entire organism. ” (From Wikipedia)

What’s more, the hepatic erythropoiesis will occur in adults when excessive bleeding happened and a lot of blood cell is needed in emergency. What interested me is hepatic erythropoiesis will generate nucleated red blood cell (Baschat, et al., 1999), different from marrow. Will there be some similarities between non-mammals nucleated red blood cell? The answer seems to be no. Some net users said birds’ bones are empty inside, so they do not have marrow. They generate their blood cell by division of mature blood cells as the erythrocytes are nucleated. These net users must misunderstand something that some of the birds’ bones have marrow, and the red blood cells are produced in the bone marrow. This picture shows which bones of pigeon are empty inside and which are filled with marrow.



(Ornithology, Avian Circulatory System)

Finally, I found one explanation about why mammals’ erythrocytes are enucleated but non-mammalian vertebrates have nucleated erythrocytes in the aspect of evolution. But this explanation do not convince me: “Mammals, which had developed an aerobic metabolism, emerged in the Triassic, when the oxygen content in the atmosphere was by approximately 50% lower than the current level and even lower than in the Jurassic period. A drastic decrease in the total content and percentage of oxygen in the Triassic was connected with the prevalence of arid conditions on the continents. Under these conditions, mammals got rid of the nuclei in erythrocytes (having obtained enucleate and biconcave cells, where the surface area of the contained hemoglobin was larger), which led to thinner capillaries, while the biconcave shape provided a larger exchange area. Birds, which originated from more advanced reptiles, had established powerful respiratory and circulatory systems and, since they emerged at the time when the oxygen content in the Earth atmosphere approached the present level, had no need to eliminate the nuclei from their erythrocytes.” (Gavrilov, 2013)

I am wondering, in the Triassic, new coming mammals are suffering low oxygen content, why reptiles at that time are not? Why the ancestors of birds do not get rid of the nuclei in erythrocytes in Triassic? In other words, most of us have no doubt that enucleated erythrocytes are advanced than nucleated erythrocytes, but why this advanced feature only limited in mammals? I believe there should be some connection between other features of mammals and their enucleated erythrocytes. Viviparity seems not be the connection, as the evidences shown in the former part of this blog. But I will keep trying to find that. One point I am thinking is that there are two main stages of enucleated erythrocytes development in mammals; the first stage is called primitive erythropoiesis (Palis, 2014), which looks similar with the non-mammal’s erythropoiesis. Investigating their similarities and differences in molecular level, may give us some clue that how enucleated erythrocytes is evolved from nucleated erythrocytes.


Charles A. Long. Evolution of Function and Form in Camelid Erythrocytes. Conference on Cellular & Molecular Biology – Biophysics & Bioengineering, Athens, Greece, August 26-28 (2007) 18–24.

O.K. Baskurta, H.J. Meiselmanb. Lessons from comparative hemorheology studies. Clinical Hemorheology and Microcirculation 45 (2010) 101–108.

Nicholas A. Mauro, Russel E. Isaacks. Examination of Reptilian Erythrocytes as Models of the Progenitor of Mammalian Red Blood Cells. Comp. Biochem. Physiol. Vol. 116A, No. 4, pp (1997) 323–327.

Ahmet A. Baschat, Ulrich Gembruch, Irwin Reiss, Ludwig Gortner, Chris R. Harman, and Carl P. Weiner. Examination of Reptilian Erythrocytes as Models of the Progenitor of Mammalian Red Blood Cells. Am J Obstet Gynecol. Volume 181, Number 1 (1999) 190-195.

Valery M. Gavrilov. Origin and development of homoiothermy: A case study of avian energetics. Am J Obstet Gynecol. Advances in Bioscience and Biotechnology, 2013, 4, 1-17.

James Palis. Primitive and definitive erythropoiesis in mammals. Am J Obstet Gynecol. FrontiersinPhysiology. January (2014) Volume5, Article3. 1-9.


Brief Review of Pigment Generation, Part III (BIOL 5104 blog 7)

Migration of Melanoblast (continued)

In Xenopus embryo, contact inhibition of locomotion was found happens during neural crest cell migration. When two migrating neural crest cells meet, they stop, collapse their protrusions and change direction. In contrast, when a neural crest cell meets another cell type, it fails to display contact inhibition of locomotion; instead, it invades the other tissue, like metastatic cancer cells. Inhibition of non-canonical Wnt signaling abolishes both contact inhibition of locomotion and the directionality of neural crest migration. Wnt signalling members localise at the site of cell contact, leading to activation of RhoA in this region. This contact inhibition of locomotion should be present in any species and breeds which have uniformly distributed pigmentation. But the I still don’t know whether this in chicken is similar like this in Xenopus.

Recently, two distinct types of melanocyte are described in mice, dermal (non-cutaneous) melanocytes and epidermal melanocytes (Aoki, et al., 2009). Experiments have shown that ectopically expression of Endothelin 3 (EDN3) affect the dermal pigmentation but not the hair color (Garcia, et al., 2008). These non-cutaneous like dermal melanocytes are incapable of contributing to epidermal hair follicle pigmentation further highlighting the functional differences between these two melanocyte populations (Aoki, et al., 2011). Similar dermal pigmentation phenotype is seen when EDN3 is ectopically expressed in the chicken, which is called dermal hyperpigmentation or Fibromelanosis (FM), a breed character of Silkie chickens. EDN3 is up regulated in Silkie chicken during migration of melanoblast and even in the adult Silkie chicken skin tissue. But Silkie expressing FM can be white in feather (so called feather, it looks like mammals hair or chick’s down) and also can be other feather color, which indicates that FM does not affect feather pigmentation (Dorshorst, et al., 2011). Previous studies have shown that in avian, EDN3 exerts trophic, mitogenic and melanogenesis-promoting activities on early neural crest cells in culture. EDN3-responsive precursors were identified in neural crest clonal cultures and include fate-restricted glial and melanocytic cells as well as bipotent glia–melanocyte (GM) precursors. The GM progenitor is thus a main target of survival and mitogenic activities of EDN3. Experiments have shown that EDN3 appeared capable for transition between glia and melanocytes in vitro (Douarin and Dupin, 2003). But it seems like glia, or at least its function, is not affected by high expression of in EDN3 Silkie chicken. The explanation need to be further investigated.

When we are talking about FM in Silkie, we have to mention another gene, inhibitor of dermal melanin (Id). The phenotype of Id gene is the abnormal migration of melanoblast, which will invade into the ventral pathway normally reserved for neuronal and glial cell lineages. So, Id causes pigmentation of internal connective tissue and the exterior (Dorshorst, et al., 2010). Similar like inhibitors prevent glial cell entering dorsolaterally pathway, something called barrier molecules (PNA-binding molecules) can stop melanoblast to migrate ventrolaterally. But what the barrier molecules exactly are is still unknown, we can only label them by PNA (Lection Peanut Agglutinin). In this way, we can know the distribution of barrier molecules is normal in Silkie during most of the time of melanoblast migration. But in later stage, they disappear and then melanoblast migrates from the dorsolateral space to the ventral part. So, the melanoblast in the ventral part is not because of their migration in the same way as the neural/glial precursors at the beginning. Some other experiments have shown that the abnormal migration of melanoblast is not an autonomous property of Silkie neural crest, but the environment difference around the pathways, which consists with previous observation of PNA-binding molecules’ down regulation (Faraco, et al., 2001).

The identification of the barrier molecules and the detail mechanism of Id gene will give us better understanding about neural crest cells’ path finding.


Carlos Carmona-Fontaine, Helen K. Matthews, Sei Kuriyama, Mauricio Moreno, Graham A. Dunn, Maddy Parsons, Claudio D. Stern, and Roberto Mayor. Contact Inhibition of Locomotion in vivo controls neural crest directional migration, Nature. 2008 December 18; 456(7224): 957–961.

Aoki H, Yamada Y, Hara A, Kunisada T. Two distinct types of mouse melanocyte: differential signaling requirement for themaintenance of non-cutaneous and dermal versus epidermal melanocytes. Development. 2009. 136: 2511–2521.

Garcia RJ, Ittah A, Mirabal S, Figueroa J, Lopez L, et al. Endothelin 3 induces skin pigmentation in a keratin-driven inducible mouse model. J Invest Dermatol. 2008. 128: 131–142.

Aoki H, Hara A, Motohashi T, Osawa M, Kunisada T. Functionally distinct melanocyte populations revealed by reconstitution of hair follicles in mice. Pigment Cell Melanoma Res. 2011. 24: 125–135.

Ben Dorshorst, Anna-Maja Molin, Carl-Johan Rubin, Anna M. Johansson, Lina Stromstedt, Manh-Hung Pham, Chih-Feng Chen, Finn Hallbook, Chris Ashwell, Leif Andersson. A Complex Genomic Rearrangement Involving the Endothelin 3 Locus Causes Dermal Hyperpigmentation in the Chicken. PLoS Genetics. 2011. 7(12): e1002412.

Nicole M Le Douarin, Elisabeth Dupin. Multipotentiality of the neural crest. Current Opinion in Genetics & Development.  2003. 13:529–536.

Ben Dorshorst, Ron Okimoto, Chris Ashwell. Genomic Regions Associated with Dermal Hyperpigmentation, Polydactyly and Other Morphological Traits in the Silkie Chicken. Journal of Heredity. 2010. 101(3):339–350.

Cloris D. Faraco, Sonia A.S. Vaz, Maria Veronica D. Pastor, Carol A. Erickson. Hyperpigmentation in the Silkie Fowl Correlates With Abnormal Migration of Fate-Restricted Melanoblasts and Loss of Environmental Barrier Molecules. Developmental Dynamics. 2001. 220:212–225.

Brief Review of Pigment Generation, Part II (BIOL 5104 blog 6)

Migration of Melanoblast

In part I, we have reviewed how two types of neural crest cell, neural/glial precursors and melanoblast are differentiated from neural tube. After this process, the neural/glial precursors will migrate ventrolaterally and the melanoblast will migrate dorsolaterally. As we are focused on pigment generation, here we are going to talk about the migration of the melanoblast only.

There are several mechanisms that prevent nonmelanoblast to migrate dorsolaterally, one of them is ephrin. The expression of ephrin in posterior of somite and in dorsolateral pathway excludes the nonmelanoblast neural crest cells and only allows them to enter the anterior part of somite in ventrolateral way. So, why melanoblast can migrate dorsolaterally between the ectoderm and the somite in the presence of ephrin? The key receptor is EphB2 expressed only in melanoblast, which not only allow melanoblast to survive in ephrin, but also act as chemoattractant, to attract melanoblast to migrate dorsolaterally. Another inhibitor of nonmelanoblast is Slits. Similarly, the receptors of Slits, Robo1 and Robo2 in melanoblast allow them to migrate dorsolaterally. It is possible that ephrin and Slits act in conjunctly (Melissa and Carol, 2007). The last inhibiter of nonmelanoblast I want to mention here is endothelin3 (ET3), the ligand of EDNRB2. EDNRB is up regulated in neurons and glial precursors but cannot help them to survive in the presence of ET3. Then, melanoblast down regulates EDNRB and at the same time up regulates EDNRB2. Some experiments have shown that without these inhibitor, nonmelanoblast will migrate dorsolaterally even they do not have these receptors (this receptors act in the role of path finding). Interestingly, although the ligand of EDNRB2 and EphB2 are different, the overexpression of EDNRB2 can maintain normal dorsolateral migration of melanoblast in the absence of EphB2, and vice versa (Melissa, et al., 2008), suggesting they may have overlap in pathway. Other repulsive cues present in the dorsolaterally pathway that restrict the migration of neuronal precursors include spondins, chondroitin sulfate proteoglycans and PNA-binding molecules (Jia, et al., 2005).

In the view of spatial expression, while melanoblast is migrating between ectoderm and dermamyotome, ET3 is expressed in both ectoderm and dermamyotome (Nagy, et al., 2006). But Steel Factor (SLF) only up regulated in dermamyotome after melanoblast are well into the dorsolaterally pathway, at the same time of the expressing of c-KIT in melanoblast (Reedy, 2003). In the mouse, Kit is necessary to maintain the survival of melanoblast, and for their consequent dispersal onto the dorsolaterally pathway (Wehrle-Haller, 2001). But it is different in chickens. c-KIT knocking out does not affect the migration of melanoblast, just affect the maintenance of melanoblast function.

Here I have present the main part of melanoblast migration. Besides, there are some other interesting processes happened during migration, which I will introduce in the following part.


Melissa L. Harris, Ronelle Hall, Carol A. Erickson. Directing pathfinding along the dorsolateral path – the role of EDNRB2 and EphB2 in overcoming inhibition. Development. 2008. 135, 4113-4122

Melissa L. Harris, Carol A. Erickson. Lineage Specification in Neural Crest Cell Pathfinding. Developmental Dynamics. 2007. 236:1–19.

Jia, L., Cheng, L. and Raper, J. Slit/Robo signaling is necessary to confine early neural crest cells to the ventral migratory pathway in the trunk. Dev. Biol. 2005. 282, 411-421.

Nagy, N. and Goldstein, A. M. Endothelin-3 regulates neural crest cell proliferation and differentiation in the hindgut enteric nervous system. Dev. Biol. 2006. 293, 203-217.

Reedy, M. V., Johnson, R. L. and Erickson, C. A. The expression patterns of c-kit and Sl in chicken embryos suggest unexpected roles for these genes in somite and limb development. Gene Expr. 2003. Patterns 3, 53-58.

Wehrle-Haller, B., Meller, M. and Weston, J. A. Analysis of melanocyte precursors in Nf1 mutants reveals that MGF/KIT signaling promotes directed cell migration independent of its function in cell survival. Dev. Biol. 2001. 232, 471-483.

Polydactyly (BIOL 5104 blog 5)

During the Christmas vacation of last year, I have driven to Key West, FL. That’s my first trip in the U.S. and really wonderful. What most impressed me are the Hemingway cats in Key West.

The Hemingway’s Cats (Fig. 1), descended from a single polydactylous cat given to the American author Ernest Hemingway by a ship’s captain in the 1930s. Around 40 descendants populate the grounds of the Ernest Hemingway Home in Key West ( of which about half have extra toes. Polydactylous cats, also called ‘mitten’ cats, are a well-known curiosity of the pet world. Valued as good luck charms by sea captains on sailing ships in the past centuries, polydactylous cats are frequently found along the northeastern coastal regions of North America.


Fig.1 Hemingway’s Cat (Photo taken by me)

Recent studies have shown that the polydactyly in cat is caused by point mutations in the 778 bp conserved noncoding element of the limb-specific cis-regulator of Shh termed the ZRS (Lettice et al. 2008). In the developing limb, Sonic Hedgehog (Shh) is normally expressed in a region of the posterior mesoderm, called the zone of polarizing activity (ZPA), and is required for proper anterior/posterior limb patterning (reviewed in Mariani and Martin, 2003). A highly conserved long-range limb-specific Shh enhancer is present within intron 5 of the LMBR1 gene (Dorshorst et al. 2010). This enhancer is called the ZPA regulatory sequence (ZRS) and is also known as mammalsfishes-conserved-sequence-1 (MFCS-1). The function of ZRS and Shh in variety of vertebrates’ digit development is quite similar. In normal birds, 4 toes is common and is thought as loss of the fifth toe from their ancestor. But evidence has shown that the polydactyly in chicken is not the reappear of the lost fifth toe, but has the same mechanism as 6 toes human, mice and cats. Chicken polydactyly is the existence of one or more extra most anterior toes developed preaxially (Fig. 2), some polydactyly cause the extra toe and also cause the loss of the normal most anterior toe. Different from other vertebrates, polydactyly can be a kind of breed character that is fixed in some breeds (e.g. Silkie, Sultan, Houdan, Dorking et al. We are not thinking Hemingway’s Cats as a breed, so I don’t think there are some breeds that fixed with polydactyly in vertebrates other than chickens). So the study of polydactyly in chicken is much earlier. In 1944, Warren has studied the genetics of polydactyly of chicken, the results is so complex that we will not talk about that here. Because of its variety of expression, it is hard to apply defined mapping. Some studies around 2000 showed the association of SNP within intron 5 of LMBR1 is strongly associated with polydactyly in Silkie chicken. In the meantime gain-of-function and loss-of-function analysis confirmed that the Lmbr1 gene is required for limb formation and reciprocal changes in levels of Lmbr1 activity can lead to either an increase or decrease in the number of digits in the vertebrate limb (Clark et al., 2001). Then researchers started to study the relationship between the expression and the function of Lmbr1 and polydactyly. But little progress is made in that area. Till recent work in the mouse has identified the ZRS region in the intron of LMBR1 which has nothing to do with the expression of Lmbr1 but can cis-regulate and repress Shh expression in the limb, (Maas and Fallon 2005), and similar mechanism found in human, researchers started to focus on the study of Shh in polydactyly chicken. Shh expression in the posterior of polydactyly chicken limbs is abnormal and demonstrates that ZRS long-range Shh control element is not an anterior specific element but general to regions of the limb that have the potential to express Shh. In ZRS mutated chickens, the size of the Shh expression domain and the range of Shh signaling increased. Interestingly, this increased range of Shh signaling played a role in induction of anterior Shh expression in the polydactyly chicken leg bud. A loss of programmed cell death in the anterior of the leg bud has been implicated in the formation of preaxial polydactyly in the Dorking chicken which is thought to lead the polydactyly (Ian et al. 2011).


Fig.2 Polydactyly in Chicken (Photo taken by me)

But two interesting things are shown by Warren that low temperature during incubation can inhibit the expression of polydactyly and when asymmetry of digit number individuals are found, the left leg are always be the side has more digit. This phenomenon was repeated later (Huang et al. 2006) but no one can give explanation till now. Mammal (like mouse) has some advantages on studying the digit development than chickens, but not in these two aspects: 1. mammal embryo developed in low temperature will cause much more problem rather than the development of digit. 2. Asymmetry of digit number in mammals is so rare. But if that happened in mammals, it will be much easier to find the clue of the mechanism by compare the development process between deferent species, and that must help us to have a better understanding on vertebrates digit development.


Laura A. Lettice, Alison E. Hill, Paul S. Devenney, Robert E. Hill. Point mutations in a distant sonic hedgehog cis-regulator generate a variable regulatory output responsible for preaxial polydactyly. Human Molecular Genetics, 2008, Vol. 17, No. 7 978–985

Mariani FV, Martin GR. Deciphering skeletal patterning: clues from the limb. Nature. 2003.423:319–325.

Ben Dorshorst, Ron Okimoto, And Chris Ashwell. Genomic Regions Associated with Dermal Hyperpigmentation, Polydactyly and Other Morphological Traits in the Silkie Chicken. Journal of Heredity. 2010. 101(3):339–350

Warren DC. Inheritance of polydactylism in the fowl. Genetics. 1944. 29:217–231.

Yan Qun Huang, Xue Mei Deng, Zhi Qiang Du, Xiangpin Qiu, Xiaohui Du,Wen Chen, Mireille Morisson, Sophie Leroux, F. Abel Ponce de Léon, Yang Da, Ning Li. Single nucleotide polymorphisms in the chicken Lmbr1 gene are associated with chicken polydactyly. Gene. 2006. 374 10–18.

Clark, R.M., et al. Reciprocal mouse and human limb phenotypes caused by gain- and loss-of-function mutations affecting Lmbr1. Genetics. 2001. 159, 715–726.

Ian C. Dunn, I. Robert Paton, Allyson K. Clelland, Sujith Sebastian, Edward J. Johnson, Lynn McTeir, Dawn Windsor, Adrian Sherman, Helen Sang, Dave W. Burt, Cheryll Tickle, Megan G. Davey. The Chicken Polydactyly (Po) Locus Causes Allelic Imbalance and Ectopic Expression of Shh During Limb Development. 2001. 159, 715–726.

Brief Review of Pigment Generation, Part I (BIOL 5104 blog 4)

Melanoblast generation

I would like to write a brief review that covers the processes of pigment generation in chicken. Because of my limit time and effort, this review will be divided into several blogs that posted at most one blog each week.

When we are talking about melanoblast generation, neural crest has to be mentioned. Neural crest is derived from neural tube after gastrula stage. Or more precisely, neural crest cells are specified at the border of the neural plate and the non-neural ectoderm. The first group of cell that differentiated into neural crest cell will migrate ventrolaterally and become neurons and glial. In chicken embryo, one day later (Reedy et al., 1998), the following groups of neural crest cells become melanoblast then migrate dorsolaterally.

So, our first question is what causes the neural tube cells becoming into neural crest cells. At least we know Wnt signaling is in charge of this process, Wnt6 induces neural crest production through the noncanonical signaling pathway and Wnt1 inhibits neural crest induction through canonical signaling pathway. One of the targets of these pathways is FOXD3 (Forkhead box transcriptional repressor, Corina et al., 2007), which is necessary for neural/glial precursors and also can inhibit MITF (Microphthalmia-associated transcription factor, key transcriptional factor for melanoblast formation) expression. So, FOXD3 acts like a switch between neural/glial precursors and melanoblast derived from neural tube, and Wnt6 active FOXD3 while Wnt1 inactive FOXD3. Bone morphogenetic proteins (BMP) also induce FOXD3 during the initial generation of neural crest (Taneyhill and Bronner-Fraser, 2005). Evidence shows that neural crest induction is underway during gastrulation (By expression of the paired box transcription factor PAX7) and well before proper neural plate appearance (when Wnt6 and BMP expressed). But there are other results indicate that the expression of PAX7 is regulated by Wnt6 and BMP (Martı´n L et al., 2006). So, the relationship between PAX7, Wnt6 and BMP in the initiation of neural crest seems unclear.

MITF, a key transcriptional factor for melanoblast formation, is mentioned above. It is activated by PAX3, SOX10 and WNT3A which are exist in neural tube before the formation of neural crest. So MITF should be depressed in other way otherwise the first group of neural crest cell should be melanoblast. That is through the binding of FOXD3 with MITF-M (the melanocyte-specific isoform of MITF) promoter which prevents the binding of PAX3 with this promoter. As FOXD3 is expressed exclusively in the neural/glial precursors, it acts like a key that regulates the lineage switch between neural crest derived glial cells and pigment cells (Aaron et al., 2009). Besides, BMP-4 is expressed in the dorsal neural tube throughout the time when neural/glial precursors are migrating, but is decreased coincident with the timing of melanoblast migration later. This expression pattern suggests that BMP-4 antagonizes melanogenesis (Jin et al., 2001) in conjunction with FOXD3. This lineage switch occurs while the neural crest precursors are still resident in the neural tube.



Reedy, M. V., Faraco, C. D. and Erickson, C. A. The delayed entry of thoracic neural crest cells into the dorsolateral path is a consequence of the late emigration of melanogenic neural crest cells from the neural tube. Dev. Biol. 1998. 200, 234-246.

Corina Schmidt, Imelda M. McGonnell, Steve Allen, Anthony Otto, and Ketan Patel. Wnt6 Controls Amniote Neural Crest Induction Through the Non-canonical Signaling Pathway. Developmental Dynamics. 2007. 236:2502–2511.

Taneyhill, L. A. and Bronner-Fraser, M. Dynamic alterations in gene expression after Wnt-mediated induction of avian neural crest. Mol. Biol. Cell. 2005. 16, 5283-5293.

Martı´n L. Basch, Marianne Bronner-Fraser, Martı´n. Garcı´a-Castro. Specification of the neural crest occurs during gastrulation and requires Pax7. Nature. 2006. Vol 441, 11, 218-222.

Aaron J. Thomas, Carol A. Erickson. FOXD3 regulates the lineage switch between neural crest derived glial cells and pigment cells by repressing MITF through a non-canonical mechanism, Development 136, 1849-1858 (2009).

Jin, E. J., Erickson, C. A., Takada, S. and Burrus, L. W. Wnt and BMP signaling govern lineage segregation of melanocytes in the avian embryo. Dev. Biol. 2001. 233, 22-37.

Why Countershading? (BIOL 5104 blog 3)

Countershading is a kind of pattern in animal coloration which means the dorsal hair or feather color is darker than the ventral hair or feather color. Most of us think it is a kind of crypsis for animals not to be found easily by others. But in fact, people are still debating about why countershading and we still have little understanding about this common phenomenon in many species. There are two aspects for the question of why countershading, the function of countershading on animals and the molecular mechanism of countershading development.

One main problem about crypsis theory is that why ventral hair or feather color is lighter? We will all agree with that when looking upward, it will be hard to find the fish because their ventral color is light and very close to the color of the sky, same thing about looking fish downward. But is that situation the same in some mammals? It seems no chance for others to look at a rodent, a felid or a Artiodactyla upward. If the hair colors of these animals are all dark in dorsal and in ventral, it seems to be a better crypsis. So, there are several alternative theories to explain the function of lighter ventral hair color, like illuminate the food under the body, protection from ultraviolet light; thermoregulation; and protection from abrasion. Other theory says the lighter color in ventral help the cubs finding the papilla easier which convinced me more.

While looking at birds, most of the birds have countershading and the main function of that may be crypsis because their situation is similar with fish rather than mammals. But another phenomenon confused me is the sexual dichromatism, which means the pattern of males and females are different.


Pic.1 Cardinal: left female, right male. (Photo from Google)

Pic.2 Penacook: left female, right male. (Photo from Google)


Pic.3 Mallard: left female, right male. (Photo from Google)


Pic.4 Red Jungle fowl (ancestor of all domestic chickens): left male, right female. (Photo from Google)

In these birds above, we can find sexual dichromatism and countershading only happened in females not in males. Here is a problem; sexual dichromatism is more common in birds rather than mammals. If the lighter color in ventral helping the cubs finding the papilla, this sexual dichromatism is more possible to happen in mammals rather than in birds as there is no lactation in birds. In other words, if the countershading only serves as crypsis, is that means male birds need less protection than females? I would like to say yes to this question but more evidence is needed.

So, the real function of countershading needs to be discovered in the future. But we have a better understanding on the molecular mechanism of countershading development. The main pigment in mammals and in birds is melanin, which can form to kinds of particles: eumelanin and pheomelanin. Eumelanin shows a black/brown color and pheomelanin shows a yellow/red color. Melanocortin 1 receptor (MC1R) plays a critical role in the synthesis of melanin. MC1R is a G protein-coupled receptor that activates the cAMP signaling pathway, after binding with α -melanocyte-stimulating hormone (α-MSH) or adrenocorticotropic hormone (ACTH), eumelanin will be synthesized. The agouti signaling protein(ASIP) has a competitive effect on bingding MC1R, when MC1R is coupled with ASIP, the synthesis of eumelanin will be repressed and pheomelanin will be synthesized. In mice, there are five kinds of ASIP mRNA variants and are controlled by two kinds of promoters: the hair cycle-specific promoter and the ventral-specific promoter. The hair cycle-specific promoter acts at the midpoint of the hair growth cycle to produce hairs with a black base. The ventral-specific promoter directs expression throughout the entire hair growth cycle of ventral but not dorsal hair follicles. In this way countershading is developed. In rabbits, only two kinds of ASIP mRNA variants are expressed by the action of two promoters to produce countershading pattern.

Scientists focused on the function of ASIP just in recent years. The chicken feather follicles express at least seven kinds of ASIP mRNA variants using three promoters. And the chicken ASIP gene is expressed in a wide variety of tissues, which contrasts with the expression of the wild-type mouse agouti (the murine ASIP) gene which is limited only to skin which suggests that ASIP in birds may play more roles than that in mammals. In 2012, scientists found that the distal ASIP promoter not only acts to produce countershading in chicks and adult females, but also plays an important role for creating sexual plumage dichromatism controlled by estrogen. What surprised me is that the countershading and sex dichromatism is associated again in molecular level! I believe this is kind of evidence that no countershading in male birds is a kind of benefit for the whole population in evolution. If we understand this benefit, we will be able to solve the problem of why countershading.

In addition, it is the melanocytes that produce the melanin, so the presence of melanocytes is also a key factor that determines the pattern of coloration. What I can tell now is that the melanoblasts (precursor of melanocytes) are derived from neural crest cells (NCCs), a multipotent population of cells emigrating from the dorsal neural tube. Considering countershading, I believe it is a benefit at least in chicken that the melanoblasts migrate from dorsal to ventral not ventral to dorsal. Because some time the migration will not complete at the time of hatching (See Pic5), so countershading is formed. In other words, the migration from dorsal to ventral gives the pigment to the chick at least in the dorsal part and somehow protects the chick (although the countershading function is still in debate). I would like to talk about more about NCCs and chicken pigmentation next week after digging deeper in the articles.


Pic.4 Chick display countershading(Photo taken by me).


H. Vrieling, D.M. Duhl, S.E. Millar, K.A. Miller, G.S. Barsh, Differences in dorsal and ventral pigmentation result from regional expression of the mouse agouti gene, Proc. Natl. Acad. Sci. USA 91 (1994) 5667–5671.

L. Fontanesi, L. Forestier, D. Allain, E. Scotti, F. Beretti, S. Deretz-Picoulet, E. Pecchioli, C. Vernesi, T.J. Robinson, J.L. Malaney, V. Russo, A. Oulmouden, Characterization of the rabbit agouti signaling protein (ASIP) gene: transcripts and phylogenetic analyses and identification of the causative mutation of the nonagouti black coat colour, Genomics 95 (2010) 166–175.

C. Yoshihara, A. Fukao, K. Ando, Y. Tashiro, S. Taniuchi, S. Takahashi, S. Takeuchi, Elaborate color patterns of individual chicken feathers may be formed by the agouti signaling protein, General and Comparative Endocrinology 175 (2012) 495–499.

Eri Oribe, Ayaka Fukao, Chihiro Yoshihara, Misa Mendori, Karen G. Rosal, Sumio Takahashi, Sakae Takeuchi, Conserved distal promoter of the agouti signaling protein (ASIP) gene controls sexual dichromatism in chickens, General and Comparative Endocrinology 177 (2012) 231–237.

Ben Dorshorst, Anna-Maja Molin, Carl-Johan Rubin, Anna M. Johansson, Lina Stromstedt, Manh-Hung Pham, Chih-Feng Chen, Finn Hallbook, Chris Ashwell, Leif Andersson, A Complex Genomic Rearrangement Involving the Endothelin 3 Locus Causes Dermal Hyperpigmentation in the Chicken, PLoS Genetics, 12, 2011, Volume 7, Issue 12, e1002412.

Wikipedia – Countershading



Spontaneity Tissue-specific Mutation Helps Curing Inborn Immune Deficiency (BIOL 5104 blog 2)

This is a true story told by Dr. Liwu Li yesterday. It is really amazing so I would like to share this with you.

Even the topic I am using is so attractive. We know genomic equivalence, which means in one organism, each cell has the same sets of genes. Mutation may happen in single cell but should not be tissue-specific except in embryo stage. But how can tissue-specific mutation just happen in a 25 years old human being?  I can’t imagine this before knowing the following story, the answer to this question is cancer. Then, more and more people are curious in curing of inborn immune deficiency. It is a big problem in medicine and in biology as this kind of deficiency is determined by genes. But if we can find a way to repair the genes even in adult cells, it must be great improvement in medicine and biology. Here comes a possible way.

Once a day, a physician encountered a boy with a very strange disease. The boy’s skin is full of ulcer everywhere because all of his neutrophil (a kind of leukocyte) do not work. Of cause the boy is very easily to be sick. After study, it is just a single mutation in the end of chromosome 13 causes this symptom. So it is a kind of inborn immune deficiency. As I mentioned above, till now we have nothing to do to cure inborn immune deficiency. That physician did not give up, he explored the articles and find one single paper in 1960s reports one patient who has the almost the same symptom as the boy. But the results are the same, no way to cure that. However, very coincidently, the physician found that the patient described in that paper is the mother of this boy! Why the physician didn’t notice this at the beginning? Because the mother does not have the symptom and she even forgot she used to have. When she was 25, the symptom suddenly and spontaneity disappeared without any medicine or treatment! And that happened several years before she gave birth to this boy. By sequencing the mothers’ different tissue cells, the physician found that the mother does have the single mutation in different tissue cells except in neutrophil. This can explain that the mother doesn’t have the symptom but it can be inherited to her son. But why this happened? It is answered by a very simple experiment, observing the neutrophil’s chromosome by microscope: the whole part of the end of chromosome 13 which contains that SNP is missing.

Chromosome breakage causes the chromosome deletion. There are three possible results of chromosome breakage, first on is been repaired. There are repair mechanisms to relink our chromosome which happened every day in out body. But there is very little possibility that we fail to repair a broken chromosome or relink chromosome in a wrong way. Within this situation, most of such cells will encounter apoptosis, very little of them will become cancer cell because cancer gene is somehow be activated by chromosome breakage or falsely relinking. What’s more, our body has variety of mechanism to recognize the cancer cell and kill them before they develop. So, back to our story, the mother’s neutrophil is the results of such an extremely rare event, suffer chromosome breakage but still alive, even functions normally. First, the region of chromosome 13 that the neutrophil lost contains many genes including the single mutation causing that syndrome, but luckily, none of them are useful to neutrophil. Second, if only one neutrophil has such broken chromosome, it will not make any difference, but it is impossible that multi neutrophils have the same broken chromosome at the same time. What’s miraculously is this chromosome breakage also triggers the cancerization of that single neutrophil. And furthermore, it didn’t develop into the kind of invasive spreading cancer but a little bit aggressive that killed other neutrophils and become dominant. As the result, most of the neutrophils are daughter cells of that single neutrophil which has the normal immune function.

That’s all about the story that I heard, but it will never come an end. It may open a whole new gate for medicine and biology. If we can find the ways to change different tissue’s single cell into a kind of cancer cell which is aggressive but not invasive spreading and still functions normally like the neutrophil in the above story, it may be a wonderful way to cure inborn genetic deficiency. On the other hand, this extremely rare event (chromosome breakage in that specific region in that specific cell) happened in that mother. Is this really a random event? By suspecting this question, we may discover other great things.

Smart way for animals to build them self (BIOL 5104 blog 1)

All of us are developed from a single cell, zygote. The information carried by the zygote acts like a commander, he commands his general (daughter cells), and each general commands their own subordinates (daughter cells). As the result, 0.2 million billion of cells (approximate number of cells in an adult man), are generate from (or controlled by) the only one zygote; no doubt, zygote is a perfect commander that he command a huge number of cells and seldom make mistakes, because we all have right number of eyes, ears, limbs, brain, heart and so on, and they are in right position. Some people will say it feels like a delicate machine that does a lot of work but no mistake. I would like to say: no, our body is much better than a machine. “One of the critical differences between you and a machine is that a machine is never required to function until after it is built. Every animal has to function even as it builds itself.”(From our text book) That is one reason that developmental biology attracts me.

When I am reading the developmental history of the leopard frog, the most amazing thing for me is the formation of the ectoderm, mesoderm and endoderm. If the embryo chooses another way to develop, that differentiates by organ at the beginning. There may be one cell; its mission is to from the head, and other single cells they have to become hand, heart, liver all by themselves. For the one “head cell”, it have to develop into variety of nerve cells to build the brain, it also should generate skull and skin to protect the brain. Considering hair, eyes, teeth and so on, that single cell acts similar as zygote. It may be too much work for that cell and may be easier to make mistake. Another problem is, if all parts of head is formed by one cell, it is hard for it to connected to the body as many vessels, nerves, muscles need to be recognized and joined together smoothly. So, forming three layers of cell is a smart way of building an animal, although it seems strange at the beginning that the shape of the embryo doesn’t like an animal at all.

Another amazing thing is the appearance of mesoderm. It is not only the precursor of the important organs (heart, blood, skeleton, gonads and kidneys), but also provide a wonderful space for the organs to stay. Food is taken in from mouth and goes through the digestive tract then move out from anus. So all the space surrounded by digestive tract (formed by endoderm) is used for digestion. Of course organs cannot be putted there as they may be digested as well. Respiratory tract (formed by endoderm) is also not suitable for organs as it is direct connected to the air, they may suffer the risks from outside. So, endoderm cannot provide a space for organs. Obviously, the space outside the ectoderm is not suitable; it is silly to have heart above the skin for the vulnerability. For the similar reason, the coelenterate cannot have complex organs as they do not have mesoderm. So the smart choice is the space between endoderm and ectoderm, and mesoderm serves for that.

When I say it is smart, I am not saying the animal itself knows these, even the smartest species human being cannot fully understand these till now. I would like to believe that, the nature is the smartest one in the world, even she cannot think at all.