FUNDAMENTALS 1: 11:00-12:00Scribe: LOUISA WARREN

TUESDAY, AUGUST 17TH, 2010Proof: RACHEL TUCKER

MILLERCOLLAGEN CHEMISTRY AND BIOLOGYPage1 of 6

Fiber-Forming Collagens (specifics) [S13]
Type I is essentially a heterotrimer, two 1 type I chains and one 2 chain
The 1 type 1 chains come from one kind of gene, the 2 come from another kind of gene
They are common in large connective tissue, things like bone, ligament, tendon
Major connective tissues almost exclusively type I collagen
The paradigm of that would be the collagen in bone
Type II collagen is a homotrimer with only one kind of chain, 1 type II
These are in cartilages and related tissue
Related tissue- things like the vitreous humor of the eye and the fibrous soft tissue in a vertebral column at the articulating surfaces
Type III collagen a homotrimer found largely in soft tissues
Makes heterotypic fibers with the type I collagen
Procollagen to collagen conversion is slow and incomplete, forming pN molecules
pN molecules: C terminus is ok, but at the amino terminus they retain some of the pN collagen- some of the globular domain is retained in the type III collagen
The same applies to the types V and XI collagen shown on the next slide
Fiber-Forming Collagens (specifics) cont [S14]
Type V and XI collagens form heterotypic fibers
Type I and V are together in the cornea
Type II and XI are together in the cartilage
The procollagen to collagen conversion yields pN molecules, molecules which have a small remnant of the globular domain that is still attached to the final form of the molecule
What does that do?
Heterotypic Fibers [S15]
This is an example of that situation with the cornea
It has type I and V heterotypic fibrils
You form these fibers in their quarter-staggered array we talked about
This would be the electron microscope observation- light, dark, light dark
Type V go into fibers but the amino terminal non-helical, globular domainwould not fit into the fiber and would stick out above the fiber
When you have these domains circulating around the fiber, you inhibit the addition of new molecules to make the fiber bigger
This is how you get the small, delicate fibers found in the cornea, where there is a mixture of type I and type V
Type I are shown as essentially straight lines, where type V have this large, globular domain sticking out
This essentially controls fiber size
Synthesis- Assembly of a Collagen Molecule [S16]
How do you synthesize this kind of molecule?
You synthesize it through the endoplasmic reticulum system because collagen is a secreted molecule; will be secreted to the extracellular space.
Therefore, when it is synthesized, its first home will be the lumen of the endoplasmic reticulum
After ribosome synthesizes collagen, it is hydroxylated and glycosylated
The newly synthesized collagen grows with hydroxyl groups and carbohydrates on it (such as lactose and glucose shown here)
Once chains have been completed, we have chain association
Chain association is controlled by the primary structure of each chain at the C-terminal region, which is a globular domain- a domain that does not have triple helix
C terminal globular domain is responsible for sorting all chains made and putting the right chains together
They come together due to the binding properties/ association properties of the nonhelical, primary structure at the carboxy terminal of each chain
Once the chains come together at C terminus, spontaneous folding process gives triple helix all the way to N terminal globular domain
Globular domains shown by the jagged lines as opposed to the thick, black lines
This C terminal domain very important
It is an interesting situation because certain cells like fibroblasts synthesize types I and III simultaneously
Never does a type III chain get into a type I molecule or a type I chain into a type III molecule
The molecules are all assembled – type III molecules with three identical type III chains and type I molecules with 2 identical type I 1 chains and one 2 chain
The chains never get mixed up- the sorting out and distribution of the chains is really from the C-terminal domain
C terminal domain has a certain primary structure of about 12 amino acids that signifies where the chain will go
Another interesting aspect of C terminal domain goes back to what we talked about the other day
In quaternary structure, complicated structures and molecules are nice because they exclude mutated chains
This doesn’t work that well with collagen because it is the C-terminal domain that dictates whether a chain will get into the final structure
If there is no mutation in the C-terminal domain, a very destructive chain mutated in the helical domain will get into the molecule
This gives rise to a situation called osteogenesis imperfecta- “imperfect bone formation/ generation of imperfect bone” so you cannot make normal bone if you have a defective type I collagen
If there is a mutation in this chain which prevents the triple helix from forming (all you need to do is change one amino acid out of the entire 1,000)
You can change a lysine to a valine and make it impossible for the triple helix to form in that particular position, and then the molecule is never complete
When the molecule is never complete, it is not useful as a functional molecule, and it cannot form bone
This is a lethal mutation and a bizarre example of how nature tries to improve but falls short
Keep in mind that 3 chains come together because of the virtue of the primary structure of C-terminal globular domain
Once the three chains come together, then triple helix formation is spontaneous, and the end result of this process is a collagen molecule
Prolyl Hydroxylation [S17]
Very instrumental in the stability of the collagen molecule is the formation of hydroxy proline
This (figure on slide) is proline in peptide linkage- shows peptide chain direction, C-terminus, amino terminus (demonstrated on second reaction, second reactant on the slide—C-terminus and N-terminus are shown with R groups)
That proline (usually that in the Y position residue is hydroxylated in collagen molecules to a great extent.
In Gly-x-y, the proline in the 3rd position will be hydroxylated
Hydroxylation occurs through an enzyme called prolyl 4-hydroxylase
It requires molecular oxygen, ferric iron, and ascorbic acid (vitamin C) –shows usefulness of vitamin C
Alpha ketoglutarate accompanies this whole process and is converted to succinic acid
need Alpha ketoglutarate and succinic acid because when you add OH to proline use one oxygen atom, have to do something with the other- single O very dangerous molecule
decarboxylate alphaketoglutarate, take up oxygen to make succinic acid
This is a reaction we will also see in the tricarboxylic acid cycle, but it also occurs when a proline is hydroxylated
The reason that you need to have alpha ketoglutarate and succinic acid is that when you add a hydroxyl group to proline you use one of the oxygen atoms and you have another oxygen atom that has to do something
An oxygen atom by itself is a very dangerous molecule
So if you decarboxylate alpha ketoglutarate, you will need another oxygen atom to make the carboxyl terminus of succinic acid
This side reaction is needed just as a place to put the extra oxygen atom during hydroxylation
This converts primary of collagen structure from Gly-Pro-Pro to Gly-Pro-Hydroxyproline and makes a rather hydrophobic chainbecome rather hydrophilic
You have added a polar group.
You do this 100-125 times to each column of a chain, and the structural requirement is that the molecule has to take a -turn configuration
Lysyl Hydroxylation [S18]
The same thing occurs with lysine, and the lysine is hydroxylated on the -carbon atom
This goes from a hydrophobic to hydrophilic side chain, and carbohydrate is added to the hydroxyl position
Doesn’t occur ver often with collagen; not much time spent here
This transformation from lysine to hydroxylysine is very important for cross-link formation
Specificity of Chain Association [S19]
I have already told you that that C-terminal domain is very important for specifying how the chain will come together
This picture is like type I collagen where the solid line indicates the two type 1 1 chains and the dotted line indicates the 2 chain
This is a cartoon to show that there the types of chains must fit in a way to allow the chains to come together in a particular fashion
You could not have 3 1 chains all together and you could not have 3 2 chains. You have to have 2 1 chains and 1 2 chain
That is the value and virtue of the carboxy terminal domain
Chemical Stability [S20]
The three chains wrap around each other into coiled- coil arrangement
This coiled-coil accommodates three chains rather than just two like the coiled-coil in myosin, which has two -helix type chains coming together, wrapping around each other
These have 3 polyproline chains coming together, wrapping around each other
Polyproline helix is a much wider, open helix than the  helix
The pitch of -helix is 5.5 angstroms, and the pitch of polyproline helix is about 10 angstroms, so it is like taking an  helix and jerking it to twice the size of the original
This is the reason why glycine is required for every third amino acid
It is the only amino acid that can accommodate these arrangements
Why glycine? – glycine is small, it has only a hydrogen side chain, has hardly any side chain at all, so glycine is an absolute requirement
If you take away the glycine and put another amino acid in the position where glycine should be, then you cannot form this triple helix beyond the point where that glycine should be
Space Filling Model of Collagen Molecule [S21]
This is an example of one turn- look at the yellow or the pink to see one complete turn
Each helix is polyproline wrapped around in a coiled-coil fashion
Then shows magnified version of that with the 3 chains wrapped around each other as they go along the molecule
You can see where only glycine is allowed, and where you can have proline (based on the space between the fibers of the helix)
This is a way of magnifying the structure
Extracellular Processing of Collagen [S22]
Once molecules are secreted from the cell, what do you to with them? How do you get them to go into fibril forms?
When you make collagen fibers, there is not much that needs to be done- you do not need special enzymes, special currents/ special electromagnetic waves, or any kind of process outside of ordinary chemical/physical processes
First, you remove the amino terminal propeptide in most collagens, especially type I and type II, but in type III, V, and XI you leave it on
You always removed C-terminal domain by procollagen C protease
All 3 chains are cleaved, and the molecule can be “guillotined” on the C-terminal and amino-terminal ends
You essentially get rid of 1/3 of total procollagen molecule- “wasted” so to speak, we will come back to this
Then you form fibers by precipitation in this unusual, regular, quarter-staggered array
Then the small, whisker-like extensions are acted upon by an enzyme called lysyl oxidase which converts lysine side chains to aldehyde components, which set up cross-links between the molecules
Cross-links here are different from cross-links in elastin
Realize that you have to modify the protein by removing about 1/3 of the protein before you can form fibers
Fiber Architecture [S23]
The fibers that are formed are essentially what you saw the other day, you have seen this before
This is the quarter-staggered array, although not exactly a quarter
The D distance is 680 angstroms or 68 nm
The molecule is about 4.4 D-lengths long
It is all staggered by D or a multiple of D – molecule number 1’ staggered relative to 5 by D, and that leaves a gap
This material is not really new
Extracellular Processing of Collagen [S24]
Let’s go back to fiber formation and how we stabilize the fibers
Keep in mind that structural substances such as collagen fibers must have the ability to resist tension
Imagine a ligament, which is essentially a massive collagen fiber
It can be a tiny periodontal ligament or an anterior ligament in the knee, like the one Chipper Jones tore
Basically, you put tension on the ligaments- you can tear the ligament itself or pull the ligament out of the bone
The reason the ligament doesn’t fall apart in normal functioning is that all the molecules are hooked to each other
When you put tension on the ligament or tendon, you essentially pull the whole structure because every subunit is hooked up in a chain-like link
What you do with collagen fibers is that you crosslink them by the old aldehyde-formaldehyde routine, but not as uncontrolled as formaldehyde fixing. It is a normal physiological fixing by aldehydes
Preparation for Cross-Linking [S25]
You have lysine or hydroxylysine residues in those small nonhelical domains at the ends of molecules
If it’s lysine, you will have lysyl oxidase, molecular oxygen, and copper
You oxidatively deaminate the lysine side chain and convert it to what we call allysine so instead of the carbon with the primary amine, it is now an aldehyde function
It is not formaldehyde, but it is lysine with a derived aldehyde
For a hydroxylysine, the same operation occurs and gives you hydroxyallysine
So we can convert both lysine and hydroxylysine to the aldehyde molecules
What do we do with aldehydes?
Aldehydes have a great tendency to interact with free amines to form a Schiff base
This is the loss of water from the two compounds to make a carbon-nitrogen bond
Reactions for Cross-Links [S26]
This is an example of that
If you react two aldehydes, you get an aldol condensation- not important for the collagen system
But if you react an aldehyde with a free amine, you get a Schiff base, where the carbon with hydrogen double bonds to the nitrogen
The interesting thing about Schiff base is that because it is formed by the loss of water, you can easily add back water to break it
A Schiff base is a very unstable compound and therefore not usually a very nice compound for a lasting, lifetime cross-link in a collagen molecule, which should last a lifetime
Suppose you do this reaction between a hydroxyaldehyde and a free lysine side chain: you make Schiff base, which can be internally reduced by oxidizing at a hydroxyl group and C to get two H atoms and converting this to ketone function and reducing the Schiff base to a secondary amine
That kind of internal oxidation/reduction is called an amidory rearrangement
That gives you a stable secondary amine, which is not easily hydrolyzed back to the original reactants
Consequently, by using hydroxyallysine instead of allysine to crosslink, you now have a stable cross-link that will last a lifetime
Remember when I talked about Desmosine, which involves 4 lysines for elastin
Collagen doesn’t have this many, more free lysines are used to make a compound called (audio not clear) I won’t bother you with that molecule
Most collagen molecules involve 2 lysines
Shows hydroxylysino-5-oxonoleucine
Location of Cross-Links [S27]
Where would these crosslinks be in the fiber architecture?
Most cross-links are between symmetrical sides of the collagen molecule
They will originate at the amino terminal and C-terminal and then go to three lysines from the N-terminal and one lysine from another position
So the way which the cross-links are exposed in the actin fiber are between molecules which are linked by 4D spaces relative to each other
This molecule’s (top in molecule in slide) amino terminus forms a cross-link aldehyde with lysine in the molecule’s helical domain
This molecule (bottom) forms a cross-link from it’s amino terminus to the C-terminal region of the other molecule
So adjacent molecules are cross-linked with each other by 2 positions/ 2 cross-links
On the other side, one molecule can be cross-linked by two links to another molecule, and the same for the other
So molecules in a fiber will be readily cross-linked, all of them will be attached at least partially to each other
Cross-Links in a Fiber: Physical Stability [S28]
This slide shows that particular region with the head-to-tail linking of all of the molecules in a fiber
When the fiber is curled around, one C-terminal head is linked to another N-terminal head/region
All of the molecules in the fiber are linked in this way by 2 cross-links, indicated by each of the thick lines
This shows the staggering of the molecule, each molecule attached to the next by two cross-links
That provides the physical stability of the fiber
The fiber will not fall apart when tension applied
The chemical stability provided by the fact that the helical domains are impervious to most enzymatic activity/ to proteases that we talked about (trypsin, chymotrypsin, elastase)
None of these proteases can come in and destroy the fibers at helices
There are several proteases, however, called collagenases that can destroy collagen fibers
This is a two-edged sword:
We need collagenases when we are growing and developing.