Segmentation and patterning of mesoderm

1
Lecture 6

• Segmentation and patterning of mesoderm
• Neural induction

2
Somites

• Transient blocks of embryonic mesoderm
• Give rise to axial skeleton (vertebrae, ribs)
  also muscles, dermis

The Elephant Bird (Aepyornis)
3
Somite fate map

• Nicole le Douarin
• Chick-quail chimeras

Quail cells chick
(skin)
(bone)
Signals from notochord (Shh) induce sclerotome
Figs 4.6, 4.7
4
Somitogenesis
best studied in chick embryo

• somites form from A to P as Hensens node
  regresses
• Presomitic mesoderm (PSM) buds off 1 block
  every 90 minutes

Fig 4.3
5
Two processes

• Segmentation
• How is presomitic mesoderm (PSM) divided into
  regular blocks?
• Regional specification
• How do somites know their position along
  anteroposterior (AP) axis?
• segmentation and specification must be coupled so
  that a segment has a single identity

6
how are periodic structures made?

• simultaneous division of a field
• Drosophila segments (W 5.13)
• digits in vertebrate limb (W 10.10)
• pigmentation stripes (Box 10A)
• progressive budding
• segmentation in other insects
• vertebrate somitogenesis
• (speculation is there a connection with
  periodicity in time? Circadian, ultradian
  rhythms)

7
Segmentation models

• The clock and wavefront model (Cooke Zeeman
  1976)
• PSM cells have internal cycle of gene expression
  (clock)--an ultradian rhythm
• As PSM grows, older cells move away from node,
  signals from node fall below a threshold
  (wavefront)
• Stage in cycle at which cells encounter wavefront
  determines whether they form a somite boundary
• a way to convert oscillation in time to periodic
  pattern in space

8
Molecular evidence

• Clock oscillations in expression levels of
  c-hairy1 (and other genes)
• Wavefront boundary of FGF8 domain

FGF8 secreted by node
when cells in phase I encounter low FGF8 they
make a somite boundary
9
oscillating gene expression can be generated by
negative feedback with delay
Current Biology, Vol 13, 1398-1408, 19 August
2003 Autoinhibition with Transcriptional Delay A
Simple Mechanism for the Zebrafish Somitogenesis
Oscillator , by Julian Lewis
10
PSM develops an intrinsic polarity
Fig 4.2
11
PSM knows where it is in AP axis
remember no regulation, therefore determined
Fig 4.5
12
How is AP identity specified?

• Model time spent in PSM (as measured by cycles
  of the clock) tells cells where they are along
  axis
• Clock stops once somite has formed
• anterior somites experience fewer cycles than
  posterior
• readout is expression of homeobox (Hox) genes

13
Hox genes and AP identity

• Drosophila mutants with homeotic transformations
• e.g. bithorax mutant makes extra set of wings
  instead of halteres
• homeobox (Hox) genes, encode transcription
  factors with DNA-binding homeodomain
• Found in all animals
• Many Hox genes, but only small subset involved in
  patterning AP axis
• See Box 4A

14
Colinearity in fly
Anatomy anterior to posterior DNA 3 to 5
molecular preformation?
15
Colinearity in mouse
a paralog group
16
The Hox code for AP identity
Genes at 3 end of Hox clusters expressed in
anterior Genes at 5 end expressed in
posterior a combinatorial code (roughly)
5
3
Fig 4.10
17
Evidence (1)

• Anatomical differences correlate with Hox
  expression domains in different animals

e.g. Hoxc6 and T1
Fig 4.13
18
Loss and gain of function

• Prediction
• loss of Hox function should cause anterior
  transformation
• Knock-out
• gain of Hox function should cause posterior
  transformation
• Overexpression

19
Evidence (2)loss of function

• e.g. Hoxc8 knockout causes L1 to T13
  transformation
• Hox KOs usually cause anterior transformation
• Mild defects reflect redundancy--need to KO all
  members of same paralog group

Fig 4.14
20
Evidence (3)gain of function

• HoxA7 normally expressed with anterior boundary
  in neck
• overexpress HoxA7 under control of actin promoter
• transformation of proatlas to atlas (1st cervical
  vertebra)

21
what controls the expression of Hox genes?
Early (3)
Late (5)

• WHY are gene clusters colinear with anatomy?
• Progressive opening of chromatin from 3 ends?
• older PSM has longer to open chromatin, turns
  on more genes?

22
retinoic acid may be a morphogen

• RA treatment causes complex spatial
  transformations
• both posteriorizing and anteriorizing...
• potent teratogen (used in acne cream)
• RA probably in gradient high anterior-low
  posterior
• RA itself not viasualized but can see
  biosynthetic enzyme in gradient
• Hox gene enhancers have binding sites for RA
  receptor...
• Countergradients of RA and FGF may link
  segmentation to AP identity

23
The ectoderm

• Neural tube CNS
• Brain and spinal cord
• Neural crest
• Neurons, cartilage, pigment cells, endocrine
  cells
• Ectodermal placodes
• Parts of eye, ear, nose
• Epidermis

24
Neurulation 2 steps
ectoderm
Involuting axial mesoderm
Neural plate

• Neural induction
• Signals from mesoderm (organizer) to ectoderm
• Neural tube closure (morphogenesis)

25
Neural induction the Spemann organizer graft

• Hans Spemann (1916)
• Graft dorsal blastopore lip of newt into ventral
  side
• Is second axis derived from graft or from host?
• Need to distinguish host from graft tissue

26
Hilde Mangold (1921)

• Used heterospecific transplant technique--allows
  marking of host vs graft
• Donor cells from Triturus cristatus (no pigment)
• Host Triturus taeniatus (pigmented)
• Result both kinds of cells in 2 axis

27
3 conclusions

• Dorsal blastopore lip is determined
• Grafts can dorsalize mesoderm
• Grafts can neuralize ectoderm
• the organizer

28
When does neural induction happen?
Fig 4.19

• Ectoderm is not determined until after
  gastrulation

29
Competence

• Is all ectoderm equally competent to respond?
• No--dorsal ectoderm is predisposed (maybe by
  Nieuwkoop center signals)
• organizer signals may not be essential, but help

30
activity of the organizer changes over time
Graft EARLY dorsal blastopore lip (Otto
Mangold) Duplicates head, tail
Graft LATE (Spemann, Proescholdt) Duplicates
tail, trunk
Fig 4.15
31
The organizer has structure

• First cells to move in are head organizer
  (future anterior endoderm)
• Then trunk organizer (notochord)
• Last cells to move inside are tail organizer

Fig 4.17
32
Regional specificity of neural induction
Fig 4.20

• Otto Mangold (1933)

33
Are inductive signals planar or vertical

• i,e., can organizer signal before involution, or
  only after?
• Mangold expts show that vertical signaling can
  work
• also
• Gastrulation in high salt (Holtfreter, 1933)
• Mesoderm moves outside, not inside
  (exogastrula)
• No nervous tissue made

34
Evidence for planar signals in Xenopus
Keller sandwich
Fig 4.23

• Doniach et al.,1992

35
Both signals may be important

• Planar signaling can happen in early gastrula
• Vertical signaling in later gastrula (required
  for anterior-most fates)
• Xenopus develops faster, may use more planar
  signals than newts.

36
How many distinct signals?
Fig 4.21
Antr
Postr

• Qualitatively different signals for each AP
  level? (i.e., multiple signals)

37
A two-signal model

• Signal 1 induce anterior neural fates
  (activation)
• Signal 2 progressively convert into posterior
  fates (transformation)

38
What are the signals?

• Noggin (et al.)
• Potent neural inducer
• Same mechanism as dorsalizing mesoderm inhibit
  BMP signaling
• Dominant negative BMP receptors neuralize
• Explains autoneuralization (Godsave Slack )

39
Autoneuralization

• Dissociated animal cap cells become neurons, not
  epidermis
• A community effect
• Explanation local (paracrine) BMP signaling
  required for epidermal fate

explant
dissociated
epidermis
neurons
40
Signals in the two-step model

• Signal 1 anterior neural fates (activation)
• organizer signals that inhibit BMPs (noggin,
  chordin)
• Signal 2 convert into posterior fates
  (transformation)
• FGFs or Wnts?
• (also third signal required for anteriormost
  fates)
• Cerberus

41
Neural induction signals may be conserved
chick node can induce neural tissue in Xenopus
animal caps
Fig 4.22
42
Later patterning of nervous system

• The hindbrain is segmented into rhombomeres
• each rhombomere expresses distinct set of Hox
  genes

r3 and r5 express Hoxb2 (blue stain) r4
expresses HoxB1 (brown stain) Is there a Hox
code for rhombomere identity?
Figs 4.24, 4.28
43
Each rhombomere is distinct
Rhombomeres (neural tube)
Neural crest (migrating cells)
Branchial arches (surface ectoderm)
Fig 4.27
Each rhombomere can be distinguished by the
projections of its axons and by where its neural
crest migrates to
44
Test the Hox code in the ectoderm

• correlation Hoxb1, a2 normally expressed in r4
• r4 neurons project to second branchial arch
• r4 neural crest makes cartilage of second arch,
  stapes
• loss of function Knock out Hoxa2
• prediction anterior transformation
• result stapes absent, ectopic first arch
  structures, ie. r2-to-r1
• gain of function Express Hoxb1 in r2
• Prediction posterior transformation
• Result r2 neurons project into 2nd arch, i.e.
  r2-to-r4

45
Conclusions

• Anteroposterior axis develops in mesoderm at
  gastrulation
• By neurula stage, all germ layers are patterned
  along body axes
• Organ-forming regions specified, but organs not
  yet made regulation possible

46
neurulation

• Neural Tube Closure neural plate folds in on
  itself
• Edges of neural plate become neural folds
• These become the migratory neural crest cells

Fig. 11.10
47
Neural tube closure

• Closure initiates at various points depending on
  species
• Anterior and posterior neuropores
• folding of plate primary neurulation
• lumbosacral tube forms by secondary neurulation

Mouse, day 8
48
Neural tube defects (NTDs)

• second most common birth defect (4000/yr in US)
• spina bifida cystica (myelomeningocele)
• 1 in 2000 live births
• surgically treatable
• more severe defects are lethal
• 1 in 10000
• anencephaly etc

49
most NTDs are preventable

• NTDs are associated with lack of folic acid
  (folate, vitamin B9) in diet
• estimated that 70 of NTDs preventable by folate
  supplements
• folate supplements can suppress NTDs in some
  mouse mutants
• why is folate critical?