Induction and patterning of mesoderm in Xenopus

1
Lecture 5

• Induction and patterning of mesoderm in Xenopus

2
Origin and specification of germ layers

• Three early embryonic tissues endoderm,
  mesoderm, ectoderm
• How are they made different?
• How does their topology arise (gastrulation)
• How are they patterned w.r.t the body axis
  asymmetries?

3
Fate mapping tells us where things come from in
normal development
Mark cell at early stageand look later to see
what it made
4
Fate map of frog blastula

• Composite, based on many cell marking experiments
• Normal cell mixing limits resolution
• Fate map does not tell us if cells specified or
  determined to these fates--need to do isolation
  or transplant experiments

Fig 3.18
5
Tissues formed from explants in culture
Fig 3.25

• animal cells make epidermis, but not neurons
• marginal cells make only most dorsal and ventral
  fates
• Vegetal explants make large yolky cells, sort
  of endoderm?

6
compare fate and specification
cf Fig 3.28
fate map--describes normal development, based on
cell marking
specification map--describes cell behavior in
isolation (in vitro)
7
How are germ layers specified?

• are determinants localized along An-Vg axis?
• VegT -- behaves as localized endodermal
  determinant
• Determinants for ectoderm not known
• Mesoderm is different

Ec
Ec
Ec
Ec
Ec
Ec
Ec
Ec
M M M M M M M
En En En En En En
8
Mesoderm induction in vitro
Fig 3.25

• Nieuwkoop (1960s)
• Dale and Slack (1980s, using fluorescent labels)
• Analyze properties of induction in vitro

9
Does induction require cell-cell contact?

• Place micropore filter between vegetal and animal
  explants
• Mesoderm induction occurs normally

10
What is the range of the inductive signal?
Animal cap
4 cells 80 mm
mesoderm
Vegetal base

• Block cell movement, division in animal cap
  (cytochalasin--inhibits actin)
• Long-range diffusion or relay of short-range
  signals?

11
Competence

• Property of receiving cells
• Animal cap cells competent 4-11 hours after
  fertilization
• Induction requires at least 2 hours contact
• 5 hours gives full induction of mesoderm
  specific genes (e.g. Brachyury)

Scale bar about 500 mm
12
Can single cells be induced?

• No need a critical mass of gt100 cells
• The community effect
• May be mediated by embryonic FGF (eFGF)
  signaling between induced cells

Fig 3.26
13
Timing of mesoderm gene expression set by
internal clock, not time of induction

• Blocking protein synthesis has no effect on
  timing of competence or response to induction

Fig 3.27
14
Tentative model

• Mesoderm fate requires an inductive signal from
  vegetal cells to animal cells
• Signal is diffusible, spreads over several cell
  diameters
• Caveat explant experiments are artificial (in
  vitro, not in vivo)
• Vegetal cells normally contact marginal zone (MZ)
  cells by the time MZ can be dissected its
  already specified
• Data show inductive signals sufficient, but are
  they necessary?
• best evidence comes from inhibition of putative
  signals in vivo...

15
Four divisions of mesoderm
Dorsal midline

• Axial mesoderm (notochord)
• Paraxial mesoderm (somites)
• bone, skeletal muscle
• Intermediate mesoderm
• kidneys, gonads
• Lateral plate mesoderm
• blood, connective tissue, cardiovasular system

Fig 2.8
ventral midline
16
At late blastula only two types of mesoderm are
specified
paraxial and intermediate mesoderm specified by
interactions between dorsal and ventral mesoderm?
Fig 3.28
17
Dorsal and ventral vegetal cells send different
signals
Notochord muscle
Blood, vascular tissue
Fig 3.30
18
The four signal model
Formal model, to explain embryological
experiments What is the molecular basis of each
signal?
Fig 3.29
19
Evidence (this time with added caveats)

• Correlation show it
• Caveat limits of detection
• Loss of function block it
• Caveats specificity, redundancy
• Gain of function move it
• Caveats specificity. need negative control.

20
Candidate mesoderm inducers

• Members of TGFb protein family
• Vg1
• activin
• Xnr1-4
• (FGF)
• will not discuss

TGFb Transforming Growth Factor b. Can convert
normal cells into cancerous cells
(transformation). Think of as signals that can
elicit any kind of developmental response.
21
The Vg1 story

• Isolated by molecular screen for localized RNAs
  (Doug Melton, 1983)
• Made by oocyte before fertilization and localized
  to vegetal cortex
• TGFb-family signal

22
Is Vg1 sufficient?

• Inject Vg1 mRNA into animal caps (for 10
  years)--no effect
• TGFb proteins require post-translational
  modification
• enzymes that process Vg1 not made by animal
  cells?
• Processed Vg1 induces dorsal mesoderm in animal
  caps

dimerization
cleavage
Active (processed) Vg1
23
Activin

• Jim Smith (1988)
• isolated in functional screen test supernatants
  from Xenopus tissue culture lines for mesoderm
  induction
• find XTC-MIF activin, another TGFb protein
• activin sufficient to induce mesoderm, BUT
• very low levels present in vivo. Maybe activin
  is mimicking another TGFb protein?

24
Loss of function techniques

• Ways to inhibit function
• Disrupt DNA
• genetic mutation--only in genetically tractable
  animals
• Disrupt RNA
• antisense RNAs (RNA interference)
• morpholinos--short antisense RNAs that disrupt
  translation
• Disrupt protein
• Blocking antibodies
• dominant negative protein

25
Dominant negative TGFb receptors
TGF beta receptors were known to be heterodimers
(AB) Truncated receptors(AC) will form
heterodimers that dont work
Inject mRNA encoding B into both cells at 2-cell
stage
AB
AB
Fig 3.36
(Actually first done with FGFRs Amaya et al.
1991. Expression of a dominant negative mutant of
the FGF receptor disrupts mesoderm formation in
Xenopus embryos. Cell 66257-70)
26
Vg1 vs. Activin

• Correlation show it
• Vg1 mRNA in right place, right time ?
• Activin mRNA not detectable ?
• Gain of function move it
• Vg1 Processed Vg1 can induce mesoderm, but not
  all types ?
• Activin potent mesoderm inducer ?
• Loss of function block it
• Vg1, Activin dominant negative TGFb receptors
  block induction ?

27
signals can be instructive or permissive

• Permissive necessary for some subsequent choice
• Instructive directly affects choice of fate

stay alive
Become mesoderm
How to tell difference
28
Activin has dose-dependent effects

• In explants
• High activin induces dorsal mesoderm
• Lower activin induces ventral mesoderm
• Argues against simple permissive role
• Vg1 also has dose dependent effects
• secreted factors with dose-dependent effects are
  called morphogens

Fig 3.42
29
Morphogens

• Long-range inductive signals
• Effects vary with concentration (gt2 possible
  fates)
• simple explanation of pattern formation
• Pure theory until late 1980s

Lewis Wolpert
Fig 1.22
30
Is activin (or Vg1) a morphogen in vivo?

• Both have concentration dependent effects in
  explant assays
• But neither expressed in gradient in vivo?
• They may be mimicking another set of TGFb
  proteins the Xnrs
• isolated in 1995, related to mouse protein nodal
• hence Xenopus nodal related (Xnr) proteins
• turn on in vegetal cells, graded expression!

31
Turning on Xnr signaling

• Low level activation by VegT in all vegetal cells
• High level activation by VegT PLUS b-catenin
• b -catenin forms dimers with TCF/LEF to activate
  transcription of Xnr5, Xnr6
• Xnrs 5 and 6 turn on early, before the MBT
• An example of combinatorial control (an AND
  operation to computer scientists)
• Other Xnrs turn on later, very dynamic

32
Current best model
Fig 3.35
33
Modeling the network
The mesendoderm network on the web http//www.not
tingham.ac.uk/biology/Genetics/staff/rogerpatient/
networks/mesendoderm/index.htm
34
Conclusion so far

• Signal 1 (ventral mesoderm) probably low level of
  Xnr (induced by maternal VgT)
• Signal 2 (dorsal mesoderm) probably high levels
  Xnr (induced by VgT b-catenin)

35
Signals 3 and 4

• Dorsalizing signals (3) from the organizer
  counteract ventralizing signals (4) from ventral
  mesoderm
• Candidates for signal 3
• Noggin
• Chordin
• Frzb (Frizbee)

36
noggin

• make cDNA library from Li-treated (dorsalized)
  embryos
• screen pools of cDNAs for dorsalizing activity
• subdivide pools to find individual cDNAs find
  noggin
• Expressed in right place, time
• can rescue axis in UV-treated eggs
• (Is it necessary? )
• protein was unfamiliar (novel), so unclear how
  it functions biochemically

Bill Smith Richard Harland, 1992. Expression
cloning of noggin, a new dorsalizing factor
localized to the Spemann organizer in Xenopus
embryos. Cell 70 829-40.
37
chordin

• another organizer signal
• similar to a Drosophila protein, SOG
• In flies, SOG and TGFb have opposing effects on
  dorsoventral axis
• see section 5.9
• could organizer proteins act by antagonizing TGFb
  proteins?

38
Signal 4 may be BMP-4

• TGFb subfamily, bone morphogenetic proteins
• BMP-4 expressed throughout blastula
• Potent ventralizing factor in explants, in vivo
• Candidate for signal 4

39
Noggin binds and inhibits BMP-4
BMP-4

• Purified noggin protein binds very tightly to
  purified BMP-4
• Noggin-BMP4 complex cannot bind receptor
• Model noggin creates a dorsal BMP-free zone
• chordin does same thing
• result is a gradient of BMP-4 activity from high
  (ventral) to low (dorsal)
• Frzb is analogous inhibitor of Wnt signals

noggin
BMP-4
V D
40
Noggin itself is not an instructive signal

• Do noggin ( chordin) do anything other than
  antagonize BMP-4 signaling?
• NO. Inhibition of BMP-4 signaling (dominant
  negative receptor) is sufficient to dorsalize
  mesoderm

41
Readout of mesoderm induction

• Low levels of Xnr activate Brachyury in all
  mesoderm
• High level of Xnr also induce transcription
  factors (goosecoid etc) in organizer region

Fig 3.41
42
mesoderm induction/patterning

• Mesoderm in Xenopus induced by TGFb signals from
  vegetal cells
• Chick Vg1 can induce primitive streak
  (mesoderm)
• Mouse TGFb nodal required for mesoderm
• mesoderm may be specified by conserved TGFb
  signals from endoderm
• dorsalization of mesoderm by inhibition of
  ventralizing BMP signals