Chapter 4: Cell Membrane and Cell Surface

1
Chapter 4 Cell Membrane and Cell Surface
I. Cell Membrane II. Cell Junctions III. Cell
Adhesion IV. Extracellular Matrix
http//www.cbi.pku.edu.cn/chinese/documents/chenjg
/
2
(No Transcript)
3

4
I. Biomembranes Their Structure, Chemistry and
Functions

• Learning objectives
• A brief history of studies on the structrure of
  the plasma membrane
• Model of membrane structure an experimental
  perspective
• The chemical composition of membranes
• Characteristics of biomembrane
• An overview of the functions of biomembranes

5

• 1. A brief history of studies
• on the structrure of the
• plasmic membrane

A. Conception Plasma membrane(cell membrane),
Intracellular membrane, Biomembrane. B. The
history of study ?Overton(1890s) Lipid
nature of PM
6

• ? J.D.Robertson(1959)
• The TEM showingthe trilaminar appearance of
  PM
• Unit membrane model
• ? S.J.Singer and G.Nicolson(1972)
• fluid-mosaic model
• ? K.Simons et al(1997)
• lipid rafts model
• Functional rafts in Cell
• membranes.
• Nature 387569-572

7
2. Singer and Nicolsons Model of membrane
structure The fluid-mosaic model is the central
dogma of membrane biology.

• The core lipid bilayer exists in a fluid state,
  capable of dynamic movement.
• Membrane proteins form a mosaic of particles
  penetrating the lipid to varying degrees.

The Fluid Mosaic Model, proposed in 1972 by
Singer and Nicolson, had two key features, both
implied in its name.
8
3. The chemical composition of membranes
A. Membrane Lipids The Fluid Part of the Model

• Membrane lipids are amphipathic.
• There are three major classes of lipids

Phospholipids Phosphoglyceride and
sphingolipids Glycolipids Sterols ( is only
found in animals)
9
Figure 10-2. The parts of a phospholipid
molecule. Phosphatidylcholine, represented
schematically (A), in formula (B), as a
space-filling model (C), and as a symbol (D). The
kink due to the cis-double bond is exaggerated in
these drawings for emphasis.
10
Figure 10-3. A lipid micelle and a lipid bilayer
seen in cross-section. Lipid molecules form such
structures spontaneously in water. The shape of
the lipid molecule determines which of these
structures is formed. Wedge-shaped lipid
molecules (above) form micelles, whereas
cylinder-shaped phospholipid molecules (below)
form bilayers.
11
Figure 10-4. Liposomes. (A) An electron
micrograph of unfixed, unstained phospholipid
vesicles (liposomes) in water. The bilayer
structure of the vesicles is readily apparent.
(B) A drawing of a small spherical liposome seen
in cross-section. Liposomes are commonly used as
model membranes in experimental studies. (A,
courtesy of Jean Lepault.)
12
Figure 10-5. A cross-sectional view of a
synthetic lipid bilayer, called a black membrane.
This planar bilayer is formed across a small hole
in a partition separating two aqueous
compartments. Black membranes are used to measure
the permeability properties of synthetic
membranes.
13
Figure 10-6. Phospholipid mobility. The types of
movement possible for phospholipid molecules in a
lipid bilayer.
14
Figure 10-7. Influence of cis-double bonds in
hydrocarbon chains. The double bonds make it more
difficult to pack the chains together and
therefore make the lipid bilayer more difficult
to freeze.
15
Figure 10-8. The structure of cholesterol.
Cholesterol is represented by a formula in (A),
by a schematic drawing in (B), and as a
space-filling model in (C).
16
Figure 10-9. Cholesterol in a lipid bilayer.
Schematic drawing of a cholesterol molecule
interacting with two phospholipid molecules in
one leaflet of a lipid bilayer.
17
Figure 10-10. Four major phospholipids in
mammalian plasma membranes. Note that different
head groups are represented by different symbols
in this figure and the next. All of the lipid
molecules shown are derived from glycerol except
for sphingomyelin, which is derived from serine.
18
Figure 10-11. The asymmetrical distribution of
phospholipids and glycolipids in the lipid
bilayer of human red blood cells. The symbols
used for the phospholipids are those introduced
in Figure 10-10. In addition, glycolipids are
drawn with hexagonal polar head groups (blue).
Cholesterol (not shown) is thought to be
distributed about equally in both monolayers.
19
Figure 10-12. Glycolipid molecules.
Galactocerebroside (A) is called a neutral
glycolipid because the sugar that forms its head
group is uncharged. A ganglioside (B) always
contains one or more negatively charged sialic
acid residues (also called N-acetylneuraminic
acid, or NANA), whose structure is shown in (C).
Whereas in bacteria and plants almost all
glycolipids are derived from glycerol, as are
most phospholipids, in animal cells they are
almost always produced from sphingosine, an amino
alcohol derived from serine, as is the case for
the phospholipid sphingomyelin. Gal galactose
Glc glucose, GalNAc N-acetylgalactos-amine
these three sugars are uncharged.
20
Membrane proteins
Figure 10-13. Six ways in which membrane proteins
associate with the lipid bilayer. Most
trans-membrane proteins are thought to extend
across the bilayer as a single a helix (1) or as
multiple a helices (2) some of these
"single-pass" and "multipass" proteins have a
covalently attached fatty acid chain inserted in
the cytoplasmic monolayer (1). Other membrane
proteins are attached to the bilayer solely by a
covalently attached lipid - either a fatty acid
chain or prenyl group - in the cytoplasmic
monolayer (3) or, less often, via an
oligosaccharide, to a minor phospholipid,
phosphatidylinositol, in the noncytoplasmic
monolayer (4). Finally, many proteins are
attached to the membrane only by noncovalent
interactions with other membrane proteins (5) and
(6). How the structure in (3) is formed is
illustrated in Figure10-14.
21
Figure 10-14. The covalent attachment of either
of two types of lipid groups can help localize a
water-soluble protein to a membrane after its
synthesis in the cytosol. (A) A fatty acid chain
(either myristic or palmitic acid) is attached
via an amide linkage to an amino-terminal
glycine. (B) A prenyl group (either farnesyl or a
longer geranylgeranyl group - both related to
cholesterol) is attached via a thioether linkage
to a cysteine residue that is four residues from
the carboxyl terminus. Following this
prenylation, the terminal three amino acids are
cleaved off and the new carboxyl terminus is
methylated before insertion into the membrane.
The structures of two lipid anchors are shown
underneath (C) a myristyl anchor (a 14-carbon
saturated fatty acid chain), and (D) a farnesyl
anchor (a 15-carbon unsaturated hydrocarbon
chain).
22
Figure 10-15. A segment of a transmembrane
polypeptide chain crossing the lipid bilayer as
an a helix. Only the a-carbon backbone of the
polypeptide chain is shown, with the hydrophobic
amino acids in green and yellow. (J. Deisenhofer
et al., Nature 318618-624 and H. Michel et al.,
EMBO J. 51149-1158)
23
Figure 10-17. A typical single-pass transmembrane
protein. Note that the polypeptide chain
traverses the lipid bilayer as a right-handed a
helix and that the oligosaccharide chains and
disulfide bonds are all on the noncytosolic
surface of the membrane. Disulfide bonds do not
form between the sulfhydryl groups in the
cytoplasmic domain of the protein because the
reducing environment in the cytosol maintains
these groups in their reduced (-SH) form.
24
Figure 10-18. A detergent micelle in water, shown
in cross-section. Because they have both polar
and nonpolar ends, detergent molecules are
amphipathic.
25
Figure 10-19. Solubilizing membrane proteins with
a mild detergent. The detergent disrupts the
lipid bilayer and brings the proteins into
solution as protein-lipid-detergent complexes.
The phospholipids in the membrane are also
solubilized by the detergent.
26
Figure 10-20. The structures of two commonly used
detergents. Sodium dodecyl sulfate (SDS) is an
anionic detergent, and Triton X-100 is a nonionic
detergent. The hydrophobic portion of each
detergent is shown in green, and the hydrophilic
portion is shown in blue. Note that the bracketed
portion of Triton X-100 is repeated about eight
times.
27

Figure 10-21. The use of mild detergents for
solubilizing, purifying, and reconstituting
functional membrane protein systems. In this
example functional Na-K ATPase molecules are
purified and incorporated into phospholipid
vesicles. The Na-K ATPase is an ion pump that
is present in the plasma membrane of most animal
cells it uses the energy of ATP hydrolysis to
pump Na out of the cell and K in, as discussed
in Chapter 11.
28
Figure 10-22. A scanning electron micrograph of
human red blood cells. The cells have a biconcave
shape and lack nuclei. (Courtesy of Bernadette
Chailley.)
29
Figure 10-24. SDS polyacrylamide-gel
electrophoresis pattern of the proteins in the
human red blood cell membrane. The gel in (A) is
stained with Coomassie blue. The positions of
some of the major proteins in the gel are
indicated in the drawing in (B) glycophorin is
shown in red to distinguish it from band 3. Other
bands in the gel are omitted from the drawing.
The large amount of carbohydrate in glycophorin
molecules slows their migration so that they run
almost as slowly as the much larger band 3
molecules. (A, courtesy of Ted Steck.)
30
Figure 10-25. Spectrin molecules from human red
blood cells. The protein is shown schematically
in (A) and in electron micrographs in (B). Each
spectrin heterodimer consists of two
antiparallel, loosely intertwined, flexible
polypeptide chains called a and b these are
attached noncovalently to each other at multiple
points, including both ends. The phosphorylated
"head" end, where two dimers associate to form a
tetramer, is on the left. Both the a and b chains
are composed largely of repeating domains 106
amino acids long. In (B) the spectrin molecules
have been shadowed with platinum. (D.W. Speicher
and V.T. Marchesi, Nature 311177-180 B, D.M.
Shotton et al., J. Mol. Biol. 131303-329)
31
Figure 10-26. The spectrin-based cytoskeleton on
the cytoplasmic side of the human red blood cell
membrane. The structure is shown schematically in
(A) and in an electron micrograph in (B). The
arrangement shown in (A) has been deduced mainly
from studies on the interactions of purified
proteins in vitro. Spectrin dimers associate
head-to-head to form tetramers that are linked
together into a netlike meshwork by junctional
complexes composed of short actin filaments
(containing 13 actin monomers), tropomyosin,
which probably determines the length of the actin
filaments, band 4.1, and adducin. The
cytoskeleton is linked to the membrane by the
indirect binding of spectrin tetramers to some
band 3 proteins via ankyrin molecules, as well as
by the binding of band 4.1 proteins to both band
3 and glycophorin (not shown). The electron
micrograph in (B) shows the cytoskeleton on the
cytoplasmic side of a red blood cell membrane
after fixation and negative staining. (B,
courtesy of T. Byers and D. Branton, PNSA.
826153-6157)
32
Figure 10-31. The three-dimensional structure of
a bacteriorhodopsin molecule. The polypeptide
chain crosses the lipid bilayer as seven a
helices. The location of the chromophore and the
probable pathway taken by protons during the
light-activated pumping cycle are shown. When
activated by a photon, the chromophore is thought
to pass an H to the side chain of aspartic acid
85. Subsequently, three other H transfers are
thought to complete the cyclefrom aspartic acid
85 to the extra-cellular space, from aspartic
acid 96 to the chromophore, and from the cytosol
to aspartic acid 96. (R. Henderson et al. J. Mol.
Biol.213899-929)
33
Figure 10-32. The three-dimensional structure of
a porin trimer of Rhodobacter capsulatus
determined by x-ray crystallography. (A) Each
monomer consists of a 16-stranded antiparallel b
barrel that forms a transmembrane water-filled
channel. (B) The monomers tightly associate to
form trimers, which have three separate channels
for the diffusion of small solutes through the
bacterial outer membrane. A long loop of
polypeptide chain (shown in red), which connects
two b strands, protrudes into the lumen of each
channel, narrowing it to a cross-section of 0.6 x
1 nm. (Adapted from M.S. Weiss et al., FEBS
Lett.280 379-382)
34
Figure 10-33. The three-dimensional structure of
the photosynthetic reaction center of the
bacterium Rhodopseudomonas viridis. The structure
was determined by x-ray diffraction analysis of
crystals of this transmembrane protein complex.
The complex consists of four subunits, L, M, H,
and a cytochrome. The L and M subunits form the
core of the reaction center, and each contains
five a helices that span the lipid bilayer. The
locations of the various electron carrier
coenzymes are shown in black. (Adapted from a
drawing by J. Richardson based on data from J.
Deisenhofer et al., Nature 318618-624)
35
4. Characteristics of biomembrane
A. Dynamic nature of biomembrane

• Fluidity of membrane lipid. It give membranes
  the ability to fuse, form networks, and separate
  charge
• Motility of membrane protein.

The lateral diffusion of membrane lipids can
demonstrated experimentally by a technique called
Fluorescence Recovery After Photobleaching
(FRAP).
36
Figure 10-34. Experiment demonstrating the mixing
of plasma membrane proteins on mouse-human hybrid
cells. The mouse and human proteins are initially
confined to their own halves of the newly formed
heterocaryon plasma membrane, but they intermix
with time. The two antibodies used to visualize
the proteins can be distinguished in a
fluorescence microscope because fluorescein is
green whereas rhodamine is red. (Based on
observations of L.D. Frye and M. Edidin, J. Cell
Sci. 7319-335)
37
Figure 10-35. Antibody-induced patching and
capping of a cell-surface protein on a white
blood cell. The bivalent antibodies cross-link
the protein molecules to which they bind. This
causes them to cluster into large patches, which
are actively swept to the tail end of the cell to
form a "cap." The centrosome, which governs the
head-tail polarity of the cell, is shown in
orange.
38
Figure 10-37. Diagram of an epithelial cell
showing how a plasma membrane protein is
restricted to a particular domain of the
membrane. Protein A (in the apical membrane) and
protein B (in the basal and lateral membranes)
can diffuse laterally in their own domains but
are prevented from entering the other domain, at
least partly by the specialized cell junction
called a tight junction. Lipid molecules in the
outer (noncytoplasmic) monolayer of the plasma
membrane are likewise unable to diffuse between
the two domains lipids in the inner
(cytoplasmic) monolayer, however, are able to do
so.
39
Figure 10-38. Three domains in the plasma
membrane of guinea pig sperm defined with
monoclonal antibodies. A guinea pig sperm is
shown schematically in (A), while each of the
three pairs of micrographs shown in (B), (C), and
(D) shows cell-surface immunofluorescence
staining with a different monoclonal antibody (on
the right) next to a phase-contrast micrograph
(on the left) of the same cell. The antibody
shown in (B) labels only the anterior head, that
in (C) only the posterior head, whereas that in
(D) labels only the tail. (Courtesy of Selena
Carroll and Diana Myles.)
40
Figure 10-39. Four ways in which the lateral
mobility of specific plasma membrane proteins can
be restricted. The proteins can self-assemble
into large aggregates (such as bacteriorhodopsin
in the purple membrane of Halobacterium) (A)
they can be tethered by interactions with
assemblies of macromolecules outside (B) or
inside (C) the cell or they can interact with
proteins on the surface of another cell (D).
41
cell coat
Figure 10-41. Simplified diagram of the cell coat
(glycocalyx). The cell coat is made up of the
oligosaccharide side chains of glycolipids and
integral membrane glycoproteins and the
polysaccharide chains on integral membrane
proteoglycans. In addition, adsorbed
glycoproteins and adsorbed proteoglycans (not
shown) contribute to the glycocalyx in many
cells. Note that all of the carbohydrate is on
the noncytoplasmic surface of the membrane.
42
Figure 10-42. The protein-carbohydrate
interaction that initiates the transient adhesion
of neutrophils to endothelial cells at sites of
inflammation. (A) The lectin domain of P-selectin
binds to the specific oligosaccharide shown in
(B), which is present on both cell-surface
glycoprotein and glycolipid molecules. The lectin
domain of the selectins is homologous to lectin
domains found on many other carbohydrate-binding
proteins in animals because the binding to their
specific sugar ligand requires extracellular
Ca2, they are called C-type lectins. A
three-dimensional structure of one of these
lectin domains, determined by x-ray
crystallography, is shown in (C) its bound sugar
is colored blue. Gal galactose GlcNAc
N-acetylglucosamine Fuc fucose NANA sialic
acid.
43
5. An Overview of membrane functions
1. Define the boundaries of the cell and its
organelles. 2. Serve as loci for specific
functions. 3. provide for and regulate transport
processes. 4. contain the receptors needed to
detect external signals. 5. provide mechanisms
for cell-to-cell contact, communication and
adhesion

44
II. Cell junction, Cell adhension Extracellular
matrix
Learning Objectives 1. Integrating Cells into
Tissues 2. Cell junctons Cell-cell adhension and
communication 3. Cell-Matrix adhension 4.
Extracellular matrix Components and
Functions 5. Cell Walls
45
(No Transcript)
46
Figure 19-1 Simplified drawing of a cross-section
through part of the wall of the intestine. This
long, tubelike organ is constructed from
epithelial tissues (red), connective tissues
(green), and muscle tissues (yellow). Each tissue
is an organized assembly of cells held together
by cell-cell adhesions, extracellular matrix, or
both.
47
Tight junctions
Figure 19-2 The role of tight junctions in
transcellular transport.    Transport proteins
are confined to different regions of the plasma
membrane in epithelial cells of the small
intestine. This segregation permits a vectorial
transfer of nutrients across the epithelial sheet
from the gut lumen to the blood. In the example
shown, glucose is actively transported into the
cell by Na-driven glucose symports at the apical
surface, and it diffuses out of the cell by
facilitated diffusion mediated by glucose
carriers in the basolateral membrane. Tight
junctions are thought to confine the transport
proteins to their appropriate membrane domains by
acting as diffusion barriers within the lipid
bilayer of the plasma membrane these junctions
also block the backflow of glucose from the basal
side of the epithelium into the gut lumen.
48
Figure 19-3 Tight junctions allow cell sheets to
serve as barriers to solute diffusion.    (A)
Schematic drawing showing how a small
extracellular tracer molecule added on one side
of an epithelial cell sheet cannot traverse the
tight junctions that seal adjacent cells
together. (B) Electron micrographs of cells in an
epithelium where a small, extracellular,
electron-dense tracer molecule has been added to
either the apical side (on the left) or the
basolateral side (on the right) in both cases
the tracer is stopped by the tight junction. (B,
courtesy of Daniel Friend.)
49
Figure 19-4 Structure of a tight junction between
epithelial cells of the small intestine.    The
junctions are shown schematically in (A) and in
freeze-fracture (B) and conventional (C) electron
micrographs. Note that the cells are oriented
with their apical ends down. In (B) the plane of
the micrograph is parallel to the plane of the
membrane, and the tight junction appears as a
beltlike band of anastomosing sealing strands
that encircle each cell in the sheet. The sealing
strands are seen as ridges of intramembrane
particles on the cytoplasmic fracture face of the
membrane (the P face) or as complementary grooves
on the external face of the membrane (the E face)
(see Figure 19-5). In (C) the junction is seen as
a series of focal connections between the outer
leaflets of the two interacting plasma membranes,
each connection corresponding to a sealing strand
in cross-section. (B and C, from N.B. Gilula, in
Cell Communication R.P. Cox, ed., pp. 1-29)
50
Figure 19-5 A current model of a tight
junction.    It is postulated that the sealing
strands that hold adjacent plasma membranes
together are formed by continuous strands of
transmembrane junctional proteins, which make
contact across the intercellular space and create
a seal. In this schematic the cytoplasmic half of
one membrane has been peeled back by the artist
to expose the protein strands. Two peripheral
proteins associated with the cytoplasmic side of
tight junctions have been characterized, but the
putative transmembrane protein has not yet been
identified. In freeze-fracture electron
microscopy the tight-junction proteins would
remain with the cytoplasmic (P face) half of the
lipid bilayer to give the pattern of
intramembrane particles seen in Figure 19-4B,
instead of staying in the other half as shown
here.
51
Anchoring junctions

52
Figure 19-7 Construction of an anchoring
junction.    Highly schematized drawing showing
the two classes of proteins that constitute such
a junction intracellular attachment proteins and
transmembrane linker proteins.
53
Adhesion belts
Figure 19-8 Adhesion belts between epithelial
cells in the small intestine. This beltlike
anchoring junction encircles each of the
interacting cells. Its most obvious feature is a
contractile bundle of actin filaments running
along the cytoplasmic surface of the junctional
plasma membrane. The actin filaments are joined
from cell to cell by transmembrane linker
proteins (cadherins), whose extracellular domain
binds to the extracellular domain of an identical
cadherin molecule on the adjacent cell.
54
Figure 19-9 The folding of an epithelial sheet to
form an epithelial tube.    It is thought that
the oriented contraction of the bundle of actin
filaments running along adhesion belts causes the
epithelial cells to narrow at their apex and that
this plays an important part in the rolling up of
the epithelial sheet into a tube (although
cellular rearrangements are also thought to play
an important part). An example is the formation
of the neural tube in early vertebrate development
55
Septate junction
Figure 19-11 A septate junction.    Electron
micrograph of a septate junction between two
epithelial cells of a mollusk. The interacting
plasma membranes, seen in cross-section, are
connected by parallel rows of junctional
proteins. The rows, which have a regular
periodicity, are seen as dense bars or septa.
(From N.B. Gilula, in Cell Communication R.P.
Cox, ed., pp. 1-29)
56
Desmosomes
Figure 19-12 Desmosomes.    (A) An electron
micrograph of three desmosomes between two
epithelial cells in the intestine of a rat. (B)
An electron micrograph of a single desmosome
between two epidermal cells in a developing newt,
showing clearly the attachment of intermediate
filaments. (C) A schematic drawing of a
desmosome. On the cytoplasmic surface of each
interacting plasma membrane is a dense plaque
composed of a mixture of intracellular attachment
proteins (including plakoglobin and
desmoplakins). Each plaque is associated with a
thick network of keratin filaments, which are
attached to the surface of the plaque.
Transmembrane linker proteins, which belong to
the cadherin family of cell-cell adhesion
molecules, bind to the plaques and interact
through their extracellular domains to hold the
adjacent membranes together by a Ca2-dependent
mechanism. (A, from N.B. Gilula, in Cell
Communication, pp. 1-29 B, from D.E. Kelly, JCB.
2851-59)
57

Figure 19-13 The distribution of desmosomes and
hemidesmosomes in epithelial cells of the small
intestine.    The keratin filament networks of
adjacent cells are indirectly connected to one
another through desmosomes and to the basal
lamina through hemidesmosomes.
58
(No Transcript)
59
Communicating junctions
gap-junction channel

Figure 19-14 Determining the size of a
gap-junction channel.    When fluorescent
molecules of various sizes are injected into one
of two cells coupled by gap junctions, molecules
smaller than about 1000 daltons can pass into the
other cell but larger molecules cannot.
60
Figure 19-15 A model of a gap junction.  The
drawing shows the interacting plasma membranes of
two adjacent cells. The apposed lipid bilayers
(red) are penetrated by protein assemblies called
connexons (green), each of which is thought to be
formed by six identical protein subunits (called
connexins). Two connexons join across the
intercellular gap to form a continuous aqueous
channel connecting the two cells.
61
Figure 19-16 Gap junctions as seen in the
electron microscope.    Thin-section (A) and
freeze-fracture (B) electron micrographs of a
large and a small gap junction between
fibroblasts in culture. In (B) each gap junction
is seen as a cluster of homogeneous intramembrane
particles associated exclusively with the
cytoplasmic fracture face (P face) of the plasma
membrane. (From N.B. Gilula, in Cell
Communication R.P. Cox, ed., pp. 1-29)
62
Figure 11-33 Three classes of channel
proteins.    The postulated relationship between
the number of protein subunits and pore diameter.
(Adapted from B. Hille, Ionic Channels of
Excitable Membranes, 2nd ed. Sunderland, MA
Sinauer, 1992.)
63
Figure 19-17 A proposed model for how
gap-junction channels may close in response to a
rise in Ca2 or a fall in pH in the cytosol.    A
small rotation of each subunit closes the
channel. The model is based on an image analysis
of electron micrographs of rapidly frozen tissue
in which the structure of gap junction channels
in their presumed open state was compared with
their structure in a Ca2-induced closed state.
It is possible that a similar mechanism operates
in the opening and closing of the gated ion
channels discussed in Chapter 11. (After P.N.T.
Unwin and P.D. Ennis, Nature 307609-613)
64
Figure 19-18 Summary of the various cell
junctions found in animal cell epithelia.    This
drawing is based on epithelial cells of the small
intestine.
65
Figure 19-19 Plasmodesmata.    (A) The
cytoplasmic channels of plasmodesmata pierce the
plant cell wall and connect all cells in a plant
together. (B) Each plasmodesma is lined with
plasma membrane common to two connected cells. It
usually also contains a fine tubular structure
(20-40nm), the desmotubule, derived from smooth
endoplasmic reticulum.
66



Figure 19-20 Plasmodesmata as seen in the
electron microscope.    (A) Longitudinal section
of a plasmodesma from a water fern. The plasma
membrane lines the pore and is continuous from
one cell to the next. Endoplasmic reticulum and
its association with the central desmotubule can
be seen. (B) A similar plasmodesma in
cross-section. (Courtesy of R. Overall.)
67
III. Cell Adhension

There Are Two Basic Ways in Which Animal Cells
Assemble into Tissues
Figure 19-21 The simplest mechanism by which
cells assemble to form a tissue.    The progeny
of the founder cell are retained in the
epithelial sheet by the basal lamina and by
cell-cell adhesion mechanisms, including the
formation of intercellular junctions.
68

Figure 19-22 An example of a more complex
mechanism by which cells assemble to form a
tissue.    Neural crest cells escape from the
epithelium forming the upper surface of the
neural tube and migrate away to form a variety of
cell types and tissues throughout the embryo.
Here they are shown assembling and
differentiating to form two collections of nerve
cells in the peripheral nervous system. Such a
collection of nerve cells is called a ganglion.
Other neural crest cells differentiate in the
ganglion to become supporting (satellite) cells
surrounding the neurons. Although it is not
shown, the neural crest cells proliferate rapidly
as they migrate.
69
Figure 19-23 Organ-specific adhesion of
dissociated vertebrate embryo cells determined by
a radioactive cell-binding assay.    The rate of
cell adhesion can be measured by determining the
number of radioactively labeled cells bound to
the cell aggregates after various periods of
time. The rate of adhesion is greater between
cells of the same kind. In a commonly used
modification of this assay, cells labeled with a
fluorescent or radioactive marker are allowed to
bind to a monolayer of unlabeled cells in
culture.
70

Cadherin
Figure 19-24 Schematic drawing of a typical
cadherin molecule.    The extracellular part of
the protein is folded into five similar domains,
three of which contain Ca2-binding sites. The
extracellular domain farthest from the membrane
is thought to mediate cell-cell adhesion the
sequence His-Ala-Val in this domain seems to be
involved, as peptides with this sequence inhibit
cadherin-mediated adhesion. The cytoplasmic tail
interacts with the actin cytoskeleton via a
number of intracellular attachment proteins,
including three catenin proteins. a-catenin is
structurally related to vinculin. X represents
uncharacterized attachment proteins involved in
coupling cadherins to actin filaments.
71
Figure 19-25 Distribution of E- and N-cadherin in
the developing nervous system.    Immuno-fluoresce
nce micrographs of a cross-section of a chick
embryo showing the developing neural tube labeled
with antibodies against E-cadherin (A) and
N-cadherin (B). Note that the overlying ectoderm
cells express only E-cadherin, while the cells in
the neural tube have lost E-cadherin and have
acquired N-cadherin. (Courtesy of Kohei Hatta and
Masatoshi Takeichi.)

72
Selectin
Figure 10-42 The protein-carbohydrate interaction
that initiates the transient adhesion of
neutrophils to endothelial cells at sites of
inflammation.    (A) The lectin domain of
P-selectin binds to the specific oligosaccharide
shown in (B), which is present on both
cell-surface glycoprotein and glycolipid
molecules. The lectin domain of the selectins is
homologous to lectin domains found on many other
carbohydrate-binding proteins in animals because
the binding to their specific sugar ligand
requires extracellular Ca2, they are called
C-type lectins. A three-dimensional structure of
one of these lectin domains, determined by x-ray
crystallography, is shown in (C) its bound sugar
is colored blue. Gal galactose GlcNAc
N-acetylglucosamine Fuc fucose NANA sialic
acid.
73
Figure 19-26 Three mechanisms by which
cell-surface molecules can mediate cell-cell
adhesion.    Although all of these mechanisms can
operate in animals, the one that depends on an
extracellular linker molecule seems to be least
common.
74
NCAM

Figure 19-27 Schematic drawing of four forms of
NCAM.    The extracellular part of the
polypeptide chain in each case is folded into
five immunoglobulinlike domains (and one or two
other domains called fibronectin type III repeats
for reasons that will become clear later).
Disulfide bonds (shown in red) connect the ends
of each loop forming each Ig-like domain.
75
Figure 19-28 A summary of the junctional and
nonjunctional adhesive mechanisms used by animal
cells in binding to one another and to the
extracellular matrix. The junctional mechanisms
are shown in epithelial cells, while the
nonjunctional mechanisms are shown in
nonepithelial cells. A junctional interaction is
operationally defined as one that can be seen as
a specialized region of contact by conventional
and/or freeze-fracture electron microscopy. Note
that the integrins and cadherins are involved in
both nonjunctional and junctional cell-cell
(cadherins) and cell-matrix (integrins) contacts.
The cadherins generally mediate homophilic
interactions, whereas the integrins mediate
heterophilic interactions. Both the cadherins and
integrins act as transmembrane linkers and depend
on extracellular divalent cations to function
for this reason, most cell-cell and cell-matrix
contacts are divalent-cation-dependent. The
selectins and integrins can also act as
heterophilic cell-cell adhesion molecules the
selectins bind to carbohydrate, while the
cell-binding integrins bind to members of the
immunoglobulin superfamily. The integrins and
integral membrane proteoglycans that mediate
nonjunctional adhesion to the extracellular
matrix are discussed later.
76
Figure 19-29 Importance of the cytoskeleton in
cell adhesion.   This drawing illustrates why
cell-adhesion molecules must be linked to the
cytoskeleton in order to mediate robust cell-cell
or cell-matrix adhesion. In reality, many
adhesion proteins would probably be pulled from
the cell with bits of attached membrane, and the
holes left in the membrane would immediately
reseal.
77
IV. Cell Coat and The Extracellular Matrix

Figure 19-30 Cells surrounded by spaces filled
with extracellular matrix.    The particular
cells shown in this low-power electron micrograph
are those in an embryonic chick limb bud. The
cells have not yet acquired their specialized
characteristics. (Courtesy of Cheryll Tickle.)
78
(No Transcript)
79
Figure 19-31 The connective tissue underlying an
epithelial cell sheet.    It consists largely of
extracellular matrix that is secreted by the
fibroblasts.
80
Figure 19-32 Scanning electron micrograph of
fibroblasts in connective tissue. The tissue is
from the cornea of a rat. The extracellular
matrix surroun-ding the fibroblasts is composed
largely of collagen fibrils. The glycoproteins,
glycosaminoglycans, and proteoglycans, which
normally form a hydrated gel filling the
interstices of the fibrous network, have been
removed by enzyme and acid treatment. (T. Nishida
et al. Invest. Ophthalmol. Vis. Sci.
291887-1890)
81
Figure 19-57 The comparative shapes and sizes of
some of the major extracellular matrix
macromolecules.    Protein is shown in green,
glycosaminoglycan in red.
82
Figure 19-33 The repeating disaccharide sequence
of a dermatan sulfate glycosaminoglycan (GAG)
chain.    These chains are typically 70 to 200
sugar residues long. There is a high density of
negative charges along the chain resulting from
the presence of both carboxyl and sulfate groups.
83
Figure 19-35 The repeating disaccharide sequence
in hyaluronan, a relatively simple GAG.    It
consists of a single long chain of up to 25,000
sugar residues. Note the absence of sulfate
groups.
84
Figure 19-36 The linkage between a GAG chain and
its core protein in a proteoglycan molecule.    A
specific link tetrasaccharide is first assembled
on a serine residue. In most cases it is not
clear how the serine residue is selected, but it
seems to be a specific local conformation of the
polypeptide chain, rather than a specific linear
sequence of amino acids, that is recognized. The
rest of the GAG chain, consisting mainly of a
repeating disaccharide unit, is then synthesized,
with one sugar residue being added at a time. In
chondroitin sulfate the disaccharide is composed
of D-glucuronic acid and N-acetyl-D-galactosamine
in heparan sulfate it is D-glucosamine (or
-iduronic acid) and N-acetyl-D-glucosamine in
keratan sulfate it is D-galactose and
N-acetyl-D-glucosamine.
85
Figure 19-37 Examples of a large ( aggrecan ) and
a small (decorin) proteoglycan found in the
extracellular matrix.    They are compared to a
typical secreted glycoprotein molecule
(pancreatic ribonuclease B). All are drawn to
scale. The core proteins of both aggrecan and
decorin contain oligosaccharide chains as well as
the GAG chains, but these are not shown. Aggrecan
typically consists of about 100 chondroitin
sulfate chains and about 30 keratan sulfate
chains linked to a serine-rich core protein of
almost 3000 amino acids. Decorin "decorates" the
surface of collagen fibrils, hence its name.
86
Figure 19-38 An aggrecan aggregate from fetal
bovine cartilage.  (A) Electron micrograph of an
aggrecan aggregate shadowed with platinum. Many
free aggrecan molecules are also seen. (B)
Schematic drawing of the giant aggrecan aggregate
shown in (A). It consists of about 100 aggrecan
monomers (each like the one shown in Figure
19-37) noncovalently bound to a single hyaluronan
chain through two link proteins that bind to both
the core protein of the proteoglycan and to the
hyaluronan chain, thereby stabilizing the
aggregate the link proteins are members of the
hyaladherin family of hyaluronan-binding proteins
discussed previously. The molecular weight of
such a complex can be 108 or more, and it
occupies a volume equivalent to that of a
bacterium, which is about 2 x 10-12 cm. (A,
courtesy of Lawrence Rosenberg.)
87
Figure 19-39 Electron micrograph of proteoglycans
in the extracellular matrix of rat
cartilage.    The tissue was rapidly frozen at
-196C and fixed and stained while still frozen
(a process called freeze substitution) to prevent
the GAG chains from collapsing. The proteoglycan
molecules are seen to form a fine filamentous
network in which a single striated collagen
fibril is embedded. The more darkly stained parts
of the proteoglycan molecules are the core
proteins the faintly stained threads are the GAG
chains. (E.B. Hunziker et al., J. Cell Biol.
98277-282.)
88
Collagen
Figure 19-40 The structure of a typical collagen
molecule.    (A) A model of part of a single
collagen a chain in which each amino acid is
represented by a sphere. The chain contains about
1000 amino acid residues and is arranged as a
left-handed helix with three amino acid residues
per turn and with glycine as every third residue.
Therefore an a chain is composed of a series of
triplet Gly-X-Y sequences in which X and Y can be
any amino acid (although X is commonly proline
and Y is commonly hydroxyproline). (B) A model of
a part of a collagen molecule in which three
alpha chains, each shown in a different color,
are wrapped around one another to form a
triple-stranded helical rod. Glycine is the only
amino acid small enough to occupy the crowded
interior of the triple helix. Only a short length
of the molecule is shown the entire molecule is
300 nm long. (From model by B.L. Trus.)

89
Figure 19-41 Electron micrograph of fibroblasts
surrounded by collagen fibrils in the connective
tissue of embryonic chick skin.    The fibrils,
which are organized into bundles that run
approximately at right angles to one another, are
produced by the fibroblasts. These cells contain
abundant endoplasmic reticulum, where secreted
proteins such as collagen are synthesized. (C.
Ploetz et al. J. Struct. Biol. 10673-81)
90
(No Transcript)
91

Figure 19-42 Hydroxylysine and hydroxyproline
residues.  These modified amino acids are common
in collagen they are formed by enzymes that act
after the lysine and proline are incorporated
into procollagen molecules
92
Figure 19-43 The intracellular and extracellular
events involved in the formation of a collagen
fibril.    Note that collagen fibrils are shown
assembling in the extracellular space contained
within a large infolding in the plasma membrane.
As one example of how the collagen fibrils can
form ordered arrays in the extracellular space,
they are shown further assembling into large
collagen fibers, which are visible in the light
microscope.
93
Figure 19-44 How the staggered arrangement of
collagen molecules gives rise to the striated
appearance of a negatively stained fibril. (A)
Since the negative stain fills only the space
between the molecules, the stain in the gaps
between the individual molecules in each row
accounts for the dark staining bands. An electron
micrograph of a portion of a negatively stained
fibril is shown below (B). The staggered
arrangement of the collagen molecules maximizes
the tensile strength of the aggregate. (B,
courtesy of Robert Horne.)
94
Figure 19-45 The covalent intramolecular and
intermolecular cross-links formed between
modified lysine side chains within a collagen
fibril.    The cross-links are formed in several
steps. First, certain lysine and hydroxylysine
residues are deaminated by the extracellular
enzyme lysyl oxidase to yield highly reactive
aldehyde groups. The aldehydes then react
spontaneously to form covalent bonds with each
other or with other lysine or hydroxylysine
residues. Most of the cross-links form between
the short nonhelical segments at each end of the
collagen molecules.
95

Figure 19-46 Electron micrograph of a
cross-section of tadpole skin.    Note the
plywoodlike arrangement of collagen fibrils, in
which successive layers of fibrils are laid down
nearly at right angles to each other. This
arrangement is also found in mature bone and in
the cornea. (Courtesy of Jerome Gross.)
96
Figure 19-47 Type IX collagen.    (A) Schematic
drawing of type IX collagen molecules binding in
a periodic pattern to the surface of a
type-II-collagen-containing fibril. (B) Electron
micrograph of a rotary-shadowed
type-II-collagen-containing fibril in cartilage
sheathed in type IX collagen molecules an
individual type IX collagen molecule is shown in
(C). (B and C, from L. Vaughan et al., J. Cell
Biol. 106991-997.)
97
Figure 19-48 The shaping of the extracellular
matrix by cells.    This micrograph shows a
region between two pieces of embryonic chick
heart (rich in fibroblasts as well as heart
muscle cells) that has grown in culture on a
collagen gel for four days. A dense tract of
aligned collagen fibers has formed between the
explants, presumably as a result of the
fibroblasts in the explants tugging on the
collagen. (D. Stopak and A.K. Harris, Dev. Biol.
90383-398)
98
Elastin
Figure 19-49 A network of elastic
fibers.    These scanning electron micrographs
show a low-power view of a segment of a dog's
aorta (A) and a high-power view of the dense
network of longitudinally oriented elastic fibers
in the outer layer of the same blood vessel (B).
All of the other components have been digested
away with enzymes and formic acid. (K.S. Haas et
al., Anat. Rec. 23086-96.)
99
Figure 19-50 Stretching a network of elastin
molecules.    The molecules are joined together
by covalent bonds (indicated in red) to generate
a cross-linked network. In the model shown each
elastin molecule in the network can expand and
contract as a random coil, so that the entire
assembly can stretch and recoil like a rubber
band.
100
Fibronectin
Figure 19-51 The structure of a fibronectin
dimer.    As shown schematically in (A), the two
polypeptide chains are similar but generally not
identical. They are joined by two disulfide bonds
near the carboxyl terminus. Each chain is almost
2500 amino acid residues long and is folded into
five or six rodlike domains connected by flexible
polypeptide segments. Individual domains are
specialized for binding to a particular molecule
or to a cell, as indicated for three of the
domains. For simplicity, not all of the known
binding sites are shown. (B) Electron micrographs
of individual molecules shadowed with platinum
arrows mark the carboxyl termini. (C) The
three-dimensional structure of a type III
fibronectin repeat, as determined by nuclear
magnetic resonance studies. It is the main type
of repeating module in fibronectin and is also
found in many other proteins. The Arg-Gly-Asp
(RGD) sequence shown is part of the major
cell-binding site (shown in blue in A) that we
discuss in the text. (B,J. Engel et al., J. Mol.
Biol. 15097-120 C, A.L. Main et al., Cell
71671-678)
101
Figure 19-52 How type IV collagen molecules are
thought to assemble into a multilayered
network.    The model is based on electron
micrographs of rotary-shadowed preparations of
these molecules assembling in vitro. (Based on
P.D. Yurchenco et al., J. Histochem. Cytochem.
3493-102)
102
Figure 19-53 Three ways in which basal laminae
(yellow lines) are organized.    They surround
certain cells (such as muscle cells), underlie
epithelial cell sheets, and are interposed
between two cell sheets (as in the kidney
glomerulus). Note that in the kidney glomerulus
both cell sheets have gaps in them, so that the
basal lamina serves as the permeability barrier
determining which molecules will pass into the
urine from the blood.
103
Figure 19-54 Scanning electron micrograph of a
basal lamina in the cornea of a chick
embryo.    Some of the epithelial cells (E) have
been removed to expose the upper surface of the
matlike basal lamina (BL). A network of collagen
fibrils (C) in the underlying connective tissue
interacts with the lower face of the lamina.
(Courtesy of Robert Trelstad.)
104
Figure 19-55 The structure of laminin.    A
schematic drawing of a laminin molecule is shown
in (A), and electron micrographs of laminin
molecules shadowed with platinum are shown in
(B). This multidomain glycoprotein is composed of
three polypeptides (A, B1, and B2) that are
disulfide bonded into an asymmetric crosslike
structure. Each of the polypeptide chains is more
than 1500 amino acid residues long. Three types
of Alpha chains, three types of B1 chains, and
two types of B2 chains have been identified,
which in principle can associate to form 18
different laminin isoforms. Several such isoforms
have been found, each with a characteristic
tissue distribution. There are also several
isoforms of type IV collegen, each with a
distinctive tissue distribution. Thus basal
laminae are chemically diverse, which is not
surprising in view of their functional diversity.
(J. Engel et al., J. Mol. Biol. 15097-120)
105
Figure 19-56 A current model of the molecular
structure of a basal lamina. The basal lamina (A)
is formed by specific interactions between the
proteins type IV collagen, laminin, and entactin
plus the proteoglycan perlecan (B). Arrows in (B)
connect molecules that can bind directly to each
other. (Based on P.D. Yurchenco and J.C.
Schittny, FASEB J. 41577-1590.)
106
Figure 19-58 Regeneration experiments indicating
the special character of the junctional basal
lamina at a neuromuscular junction.    When the
nerve, but not the muscle, is allowed to
regenerate after both the nerve and muscle have
been damaged (upper part), the junctional lamina
directs the regenerating nerve to the original
synaptic site. When the muscle, but not the
nerve, is allowed to regenerate (lower part), the
junctional lamina causes newly made acetylcholine
receptors to accumulate at the original synaptic
site. These experiments show that the junctional
basal lamina controls the localization of other
components of the synapseon both sides of the
lamina.
107
Figure 19-59 Importance of cell-surface-receptor-b
ound protease.    In (A) human prostate cancer
cells make and secrete the serine protease UPA,
which binds to cell-surface UPA receptor
proteins. In (B) the same cells have been
transfected with DNA that encodes an excess of an
inactive form of UPA, which binds to the UPA
receptors but has no protease activity by
occupying most of the UPA receptors, the inactive
UPA prevents the active protease from binding to
the cell surface. Both types of cells secrete
active UPA, grow rapidly, and produce tumors when
injected into experimental animals. But the cells
in (A) metastasize widely, whereas the cells in
(B) do not. In order to metastasize, tumor cells
have to crawl through basal laminae and other
extracellular matrices on the way into and out of
the bloodstream. This experiment therefore
suggests that proteases must be cell-surface
bound to mediate migration through the matrix.
108
Figure 19-60 The subunit structure of an integrin
cell-surface matrix receptor.    Electron
micrographs of isolated receptors suggest that
the molecule has approximately the shape shown,
with the globular head projecting more than 20 nm
from the lipid bilayer. By binding to a matrix
protein outside the cell and to the actin
cytoskeleton inside the cell, the protein serves
as a transmembrane linker. The alpha and beta
chains are both glycosylated and are held
together by noncovalent bonds.
109
Figure 19-29 Importance of the cytoskeleton in
cell adhesion.    This drawing illustrates why
cell-adhesion molecules must be linked to the
cytoskeleton in order to mediate robust cell-cell
or cell-matrix adhesion. In reality, many
adhesion proteins would probably be pulled from
the cell with bits of attached membrane, and the
holes left in the membrane would immediately
reseal.
110
Figure 19-61 Coalignment of extracellular
fibronectin filaments and intracellular actin
filament bundles.    The fibronectin is
visualized in two rat fibroblasts in culture by
the binding of rhodamine-coupled anti-fibronectin
antibodies (A). The actin is visualized by the
binding of fluorescein-coupled anti-actin
antibodies (B). (R.O. Hynes and A.T. Destree,
Cell 15875-886)
111

Figure 19-62 How the extracellular matrix could
propagate order from cell to cell within a
tissue.    For simplicity, the figure represents
a hypothetical scheme in which one cell
influences the orientation of its neighboring
cells. It is more likely, however, that the cells
would mutually affect one another's orientation.
112
Figure 19-63 Cells can regulate the activity of
their integrins. In (A) cell activation leads to
a change in the extracellular binding site of the
integrin so that it can now mediate cell
adhesion. In (B) the tyrosine phosphorylation of
the cytoplasmic tail of the integrins impairs
their ability to bind to the actin cytoskeleton.
As integrins must bind to the cytoskeleton to
mediate robust cell-matrix adhesion, the
phosphorylation causes the integrins to relax
their grip on the extracellular matrix.
113
Cell walls

• Plant cell walls provide protection against
  abrasion, osmotic stress, and pathogens.
• Microfibrils of cellulose form the fibrous
  component of the cell wall.

C. The matrix of cell wall contains
hemicellulose, pectins, and hydroxyproline-rich,pr
oline-rich, and glycine-rich structural proteins.
114
Figure 19-64 Plant cell walls.    (A) Electron
micrograph of the root tip of a rush, showing the
organized pattern of cells that results from an
ordered sequence of cell divisions in cells with
rigid cell walls. (B) Section of a typical cell
wall separating two adjacent plant cells. The two
dark transverse bands correspond to plasmodesmata
that span the wall. (A, courtesy of Brian
Gunning B, courtesy of Jeremy Burgess.)
115
Figure 19-65 Scale model of a portion of a
primary cell wall showing the two major
polysaccharide networks.    The orthogonally
arranged layers of cellulose microfibrils (green)
are cross-linked into a network by H-bonded
hemicellulose (red). This network is coextensive
with a network of pectin polysaccharides (blue).
The cellulose and hemicellulose network provides
tensile strength, while the pectin network
resists compression. Cellulose, hemicellulose,
and pectin are typically present in roughly equal
quantities in a primary cell wall. The middle
lamella is pectin rich and cements adjacent cells
together.
116
Figure 19-66 The orientation of cellulose
microfibrils in the primary cell wall of an
elongating carrot cell.    This electron
micrograph of a shadowed replica from a rapidly
frozen and deep-etched cell wall shows the
largely parallel arrange-ments of cellulose
microfibrils, oriented perpendicular to the axis
of cell elongation. The microfibrils are
cross-linked by, and interwoven with, a complex
web of matrix molecules. (Brian Wells and Keith
Roberts.)
117
Figure 19-68 The cortical array of microtubules
in a plant cell.    (A) A grazing section of a
root-tip cell from Timothy grass, showing a
cortical array of microtubules lying just below
the plasma membrane. These microtubules are
oriented perpendicular to the long axis of the
cell. (B) An isolated onion root-tip cell. (C)
The same cell stained by immunofluorescence to
show the transverse cortical array of
microtubules. (A, courtesy of Brian Gunning B
and C, courtesy of Kim Goodbody.)
118
Figure 19-69 One model of how the orientation of
newly deposited cellulose microfibrils might be
determined by the orientation of cortical
microtubules.    The large cellulose synthase
complexes are integral membrane proteins that
continuously synthesize cellulose microfibrils on
the outer face of the plasma membrane. The distal
ends of the stiff microfibrils become integrated
into the texture of the wall, and their
elongation at the proximal end pushes the
synthase complex along in the plane of the
membrane. Because the cortical array of
microtubules is attached to the plasma membrane
in a way that confines this complex to defined
membrane channels, the microtubule orientation
determines the axis along which the microfibrils
are laid down.
119
? ?

• ?????????????????????????