Signal Transduction Education
Signal Module - Obesity and Regulation of Food Intake - Biocomputing Practical

Dr IJsbrand Kramer
Signal Transduction - Course SVI632 - 6 ECTS

Below you find the screenprints of databases that we are going to use in the search for information about biomolecules involved in the regulation of food intake.

For the search of scientific publications: PubMed

For the search of information about proteins (amino acid sequence, domains, function, essential articles and links to other sites : SwissProt

To find out about chromosome location, genomic region (sequence, exons, introns) , transcripts (cDNA): Entrez Nucleotide

For information about protein structure : PDB

For information about pathologies linked to gene mutations: OMIM

Work in pairs and search SwissProt for the following proteins

As you scroll down the Insulin page, answer the following questions:

What is the accession code?

How many amino acids?

Why do they call it “precursor”?

Why are there two chains, A and B, whereas it is just one gene? Explain.

What is a signal peptide and why does the insulin precursor have one? (If you have forgotten, go to resource 8 “acheminement des proteins à travers le reticulum endoplasmique et le Golgi” (page 3/35).

In what year was human insulin amino-acid sequence published?

What is the sequence of the signal peptide of human insulin? Why does it start with an M?

Using the amino acid scheme in the” casier” of Ulysse, write down what the amino-acid abbreviations stand for and write down the nature of the amino acid: non-polar, aromatic, polar uncharged or polar charged

Would you qualify this peptide sequence as “non-polar” or “highly polar” and what information can you deduce from its presumed level of (a)polarity?

As you scroll down the Leptin page, answer the following questions:

What is the accession code?

The synonyms are OB, OBS. Why?

How many amino acids?

Is Leptin implicated in a Mendelian inheritable disease? If so, which one and what is the code?

How many chains make up Leptin?

How many disulphide bonds?

What are the consequences of a disulphide bond?

Are their natural variants in the leptin amino acid sequence (due to mutations of the coding region of the gene)?

With respect to aa94, exactly what signal nucleotide (SNP) change in the gene sequence (codon) has lead to a change of methionin for the original valine? NB In order to help, type “code génétique” in Google and search for images. Select the image of www.cellbiol.net and discover about codons.

If you were to cite a paper in which the mRNA sequence of human Leptin was first published, which one would you choose?

As you scroll down the Ghrelin page, answer the following questions:

What is the access code?

How many amino acids has the precursor protein?

Ghrelin comes in different isoforms; how many how are they obtained?

The ghrelin precursor is processes; what are the products and what is their function?

Where in the cell are secreted proteins generally cleaved ? What is the common name for the enzymes that process pro-peptides (other than the enzyme that removes the signal peptide)? (If you have forgotten, go to resource 8 “acheminement des proteins à travers le reticulum endoplasmique et le Golgi” (apres page 28/35 (vers la voie de la sécrétion contrôlee).

Ghrelins are modified after translation: how? Make a drawing of the chemical composition of the modifying component.

Was does “des” signify in des-acyl ghrelin and in des-Gln-ghrelin?

Find the position of the lacking amino acid in the precursor form of ghrelin.

Go to family and domain databases, Pfam “Graphical view”; what is represented by the cryptic words “motilin_ghrelin”?

Finally, find a “free article”in Pubmed that reports about the enzyme involved in the octanoylation of ghrelin, write down the authors and the title below:

In this TD we are going to explore the molecular composition of the insulin receptor. First we will explore the domain architecture of its ectodomain with the use of SwissProt and Pfam (protein families based on seed alignment and hidden Markov models ). Next we are going to employ the Protein Data Bank (PDB) to find the coordinates of the amino acids of the kinase domain of the insulin receptor both in its inactive and active state. Using the instructions provided in the exercise you will discover the highly conserved amino acid residues of protein kinases and you will learn how phosphorylation of the activation segment causes a shift in the conformation that renders the protein kinase competent for phosphorylating its substrates.

Write a little report (either alone or in groups of two) of this activity including a description of:

1. the processing of the insulin receptor precursor
2. the domain architecture of the receptor
3. the mode of activation of the intracellular kinase domain

Home-made images of your PyMol exercise are essential.

Send the report in the form of a “ .doc “ file to i.kramer@iecb.u-bordeaux.fr.

Start with Expasy SwissProt (http://expasy.org/) and search for the insulin receptor homo sapiens

As you scroll down answer the following questions

What is the accession code and the abbreviation for this receptor?

How many chains are there and what are their recommended names?

What is catalytic activity is carried by the insulin receptor?

How many membrane passages has the receptor?

Are defects of the receptor implicated in disease(s), if so, which are the OMIM accession codes?

Which are the amino-acids (numbers) that constitute the subunit-alpha and which comprise the subunit-beta?

Which amino acids cross the membrane? Write down the sequence in single letter code and verify their hydrophobicity (see standard amino acid file in your casier)

Scroll all the way down until you reach “Family and domain databases” and then select “Pfam [Graphical view]”. What are the names of the domains that are recognized (by the sequence comparison programs) in the extracellular segment (subunit alpha + beta)?

Suggestion: Cut and paste the figure into this document and make a short list underneath to explain the abbreviated codes. Add the amino-acids numbers behind each domain.

1 2 3 4 4 5

Would you be able to indicate the position of the membrane passage in this domain representation?

We now are going to have a look at the insulin receptor ectodomain (extracellular component) Scroll up until you come across “Cross-references” and 3D structure databases; then select, at Entry, the protein segment which the largest amino acid sequence of the ectodomain (remember amino acids 28-957 are at the outside of the cell, the rest goes inside).

Click on the appropriate PDB (protein data bank) “Entry” code

You know have entered the PDB, an information portal to biological macromolecular structures

Figure 1

You are first going to download the coordinates of the insulin receptor on your server, click on the telephone shown in the red-lined square above (figure 1). Call it “ InsR-ecto- 2dtg “

In order to find out what a PDB file is, click on the text icon just right of the telephone symbol. What you get is a long list of: COMPND (compound), SOURCE, JRNL (journal) , REMARK, SEQADV modifications in the amino acid sequence of tested protein compared to the sequence of the “standard” protein), SEQRES (“standard” amino acids of protein in question), HELIX (helix configuration), SHEET (beta sheet configuration), SSBOND (disulphide bond), CISPEP (prolines and other amino acids found in cis conformation), CRYST1 (unit cell parameters, space group and Z value), ORIGX ( ), SCALE (transformation from the orthogonal coordinates as contained in the entry to fractional crystallographic coordinates), and finally, and most importantly, ATOM (showing the coordinates of the atoms (N, C, Calpha, Cbeta, Cgamma, O etc) in a three dimensional space (X,Y,Z).

Scroll down from the title and you find Molecular Description Asymmetric Unit with a list of “polymers” (meaning here polypeptide chains or segments of proteins)

1. Polymer 1 FAB 83-7 HEAVY CHAIN, chains: A
2. Polymer 2 FAB 83-7 LIGHT CHAIN, Chains: B
3. ......
4. Polymer 5 Insulin receptor Fragment INSULIN RECEPTOR ECTODOMAIN

Heavy chains and light chains are parts of antibodies, not of the insulin receptor, why have they been included in the structure analysis?

Open PyMol (on the desktop of your computer) and import the 2dtg file

In the PyMol tool bar (“The PyMol molecular graphics system”) you go to display and select “Sequence”, then click, and than select “sequence mode” and finally “chains”

Figure 2

From the introductory page you have learned that chains A, B, C and D are part of an antibody so we like to remove them from the screen.

Click, in PyMol Viewer, on A, B, C and D (the left hand corner of the screen). The selected chains become red.

1) change colour of the antibody Fab fragments by going to the right hand corner toolbar, at the level of “sele” (for select) and click on grey (at the bottom of the list) and then click on “grey80).

Figure 3

2) remove the Fab fragments from the screen by going to “sele” (for selected items) and click on H (for hide), then select “everything”.

What is left on the screen is the ectodomain of the insulin receptor in green. You will see the chain indications A, B, C and D in grey20 wherease E is green (A B C D E)

Figure 4

Select E and go to (sele) , select S (for show) and then click on “cartoon”, the ribbons should appear on your screen. To remove the lines you next click on H (for hide) and you select “lines”.

You now see a long V-shaped protein, folded with lots of beta-sheets and some helices.

Try all other forms of S (show) and do not forget to remove (Hide) the previous ones. This helps you to get a feel of how proteins can be represented. Also by playing around you familiarize with the action buttons of PyMol. Take your time for this exercise. Every time you have a new representation, you turn the protein around with the mouse in order to get a three dimensional feel of the ectodomain of the insulin receptor.

Where is the membrane proximal side of the protein?

Figure 5

Remember, the insulin receptor precursor is cut into two chains. Go back to your SwissProt entry (see above) and find out where exactly the protein is cut. Then go the display and select “sequence mode” and then “residue codes” . What you see are soft grey (grey80) letters, which are the amino-acids of the Fab fragment of the antibodies (which we ignore) and green letters, which are amino acids of the insulin ectodomain. You now search for the cleavage site. The numbers of the amino acids on top of PyMol viewer should correspond to the numbers in the SwissProt database but always verify the identity of the amino acids as a “backup”.

Select the alpha-chain amino acids and colour them “orange”. Then select the amino acids of the beta-chain and colour them “yellow-orange”.

Now find the representation that according to you shows best that the protein is made up of two chains. You may change colours if you like. You may also change the black background. You can this this by going to the Display in the PyMol Molecular Graphics System (see figure 2) and click on background, then choose another colour).

How the two chains are hold together (apart from an extensive surface contact)? Find the relevant disulphide bridges with the help of SwissProt.

Go to the casier in Ulysse and bring up the article of “Insulin Receptor Structure McKern” and learn how the receptor is made of two subunits with only one operational insulin binding site.

In this practical you are going to explore the molecular composition of the intracellular kinase domain of the insulin receptor and in particular how phosphorylation changes its structure and as a consequence its activity. You will finish the TD by showing how the tyrosine phosphatase PTP1B exerts a negative control on the insulin signalling pathway. You are going to employ the Protein Data Bank (PDB) to study the positioning (coordinates) of the amino acids of the kinase domain both in its inactive and active state. Using the instructions provided in the exercise you will discover where highly conserved amino acids in kinase domains are situated and you will get a rough impression how they are involved in the the regulation of kinase activity.

Write a little report (either alone or in groups of two) of this activity including a description of:

1. the characteristic structure of the kinase domain
2. the mode of activation of the intracellular kinase domain
3. the mode of de-activation by PTP1B

Home-made images of your PyMol exercise are essential.

Send the report in the form of a “ .doc “ file to i.kramer@iecb.u-bordeaux.fr.

Start with Expasy SwissProt (http://expasy.org/) and search for the insulin receptor homo sapiens

As you scroll down answer the following questions

WWhat is the accession code and the abbreviation for this receptor?

What are the amino acids (numbers) that constitute the kinase domain?

How many structure analyses have been reported for the kinase domain or proteins that interact with the kinase domain (like IRS-2)? Write down their accession codes and why you think is their so much interest for the structure of the kinase domain of the insulin receptor?

Choose 1IRK ( 1irk) and click on it to be redirected to the PDB-pages concerned. Download the PDF file.

Open the file in PyMol

In “PyMol Molecular Graphics System” (menu)
Go to “Display” -> sequence
Go to “Display” -> sequence mode -> Residue Codes
Go to “Display” -> Background -> White
Go to “Display” -> Cartoon -> fancy helices

In “PyMol Viewer”
Go to “All” and select S (show :) -> cartoon
Go to “All” and select H (hide :) -> lines as well as -> waters
Go to “All” and select C (color:) -> grays -> gray 60

Then start to select the following amino-acids and present them as “sticks” with “colours by element”
Go to “(sele)”, select S (show:) -> sticks
Go to “(sele)”, select C (color:) -> by element CHNOS (carbon green, third from top in list)

K1030 (lysine), E1047 (glutamate), D1150 (aspartate) and D1132 (aspartate catalytic residue). D1132 is located in the VHRD, valine-histidine-arganine-asparate sequence which is part of the protein kinase sequence signature

Next you select the following amino-acids and present them as sticks with red colour

Y1158, Y1162, Y1163 (all tyrosines)

Next you select the following amino acids and present them as “cartoon” with “yellow “ colour

G1003, Q1004, G1005, S1006, F1007, G1008, M1009

This is a glycine rich loop, what is its function?

Next you select the following amino acids and present them as “cartoon” with “yellow “ colour

D1150 -> P1172

What is the name of this segment?

In this segment you find a highly conserved sequence

DFG, Aspartate-Phenylalanine-Glycine, which is part of the protein kinase sequence signature. After this segment you find an APE (arginine-proline-glutamate sequence which you often find in protein kinases (sometimes ALE)

Finally, you select the proline in the sequence LPVRWMAPE, just after the activation segment, present it as stick and colour it blue. This proline is another highly conserved residue but only for tyrosine protein kinases, the serine/threonine have a threonine instead and this plays, in part, a role in substrate selectivity (tyrosine or serine/threonine)

Get an orientation that shows nicely all the amino-acids you have selected.

Saving your work

In “PyMol Molecular Graphics System” (menu)
Go to “File” -> Save Session -> InsR inactive 1irk, it will be saved as a .pse file (printer page segment bitmap). Whatever goes wrong, you can reinitialize PyMol and open the .pse file (you have a back-up)
Go to “Ray” and wait until the image is ready (everything gets very smooth once 100% has been reached)
Go to “File”-> save image as -> png (portable network graphics) ->InsR inactive 1irk
When you are ready, turn the protein around, play with it for a while, present the entire protein in “spheres” and in “surface” and in the end, save an image, in “cartoon”, with “white background”. If you have forgotten the names of the residues you have selected, you can label them (do this with a “black background”
Go to “Display” -> Background -> Black
In order to get a label, bring the cursor on the residue (sticks representation) and click with the right mouse button (then follow instructions below)

Pay special attention to tyrosine Y1162, proline P1172 and aspartate D1132

Go to “File” -> reinitialize

Go to “Plugin” -> PDB loader service and enter 1ir3 (this gives the same result as 1IR3)

Repeat exactly what you have done before but notice that the tyrosines in the activation segment are now indicated as PTR, meaning phosphotyrosine

This structure has to components that were not present in the previous one

1. Go to “Display” -> Sequence Mode -> Chains
   How many chains are there? What is B?
1. For the sequence you can also return to the” Display” -> Sequence Mode -> Residue Codes and search for the short sequence
2. For more information you can go to the PDB page (using 1ir3 to find it and check in “sequence details” what sequence is represented by B, write down the amino-acid sequence
   Select “B” by clicking on “B” in PyMol Viewer then go to “S” (show :) as -> sticks. Go to “C” (colour:) -> blues -> marine. You now SEE a short peptide bound to the protein kinase (again verify the tyrosine in the substrate and its relation to P1172 and D1132
2. Go to “Display” -> sequence mode -> residue codes and find ANP & Mg Mg

Represent ANP as “Sticks” and colour by “Element” ( CHNOS ) with carbon en rose (fourth from top of list)

Why is it named ANP and not ATP?

Represent Mg Mg as “Dots”and colour -> Cyans -> Pale Cyan

Turn around and try to find the differences between 1ir3 and 1irk (use the image in the casier to get some help)

Save your image both as a .pse file and, after having applied “Ray”, as a .png file

You can again play around with representations to get a feel of what the active protein kinase and its substrate peptide looks like

Further reading: find a short article describing the insulin receptor kinase and its mode of activation in the casier (structure of the insulin receptor tyrosine kinase - SR Hubbard)

Start with PDB and enter 1g1f (1G1F)

Find out which proteins are presented.
This time few instructions, apply the tricks that you have learned in the sections

In order to find which phosphotyrosine residue is the first target of the tyrosine protein phosphatase, take into account that cysteine 215 normally is the catalytic residue (which, importantly, is replaced by an alanine in this structure in order to render the phosphatase inactive, otherwise you would never be able to fix the tyrosine-phosphate substrate!)

Save your file as a .pse and, after having RAYed, save it as a .png for your report

In this TD you are going to be confronted with the type of body measurements that are performed around the world in order to get an impression of the general (as well as abdominal) adiposity state of humanity. The tests are simple because they have to be employed in large populations (thousands of individuals). The measurements therefore only provide a rough and indirect estimate of fat content, but numerous comparisons have shown that the outcome of these tests are comparable to more sophisticated methods (density measurements, and others). Such measurement serve to a better understanding the relationship between longevity and adiposity, essential for insurance companies (to calculate premium rates) as well as for public health authorities (to estimate budgets). One distinguishes two types of adiposity; gluteofemoral (buttocks and thighs, also referred to as lower-body fat), which is poorly associated with chronic diseases that may reduce lifespan, and abdominal (waist, also referred to as upper-body fat), which is associated with chronic diseases (Pischon T, Boeing H, Hoffmann K, Bergmann M , Schultze MB et al. General and abdominal adiposity and risk of death in Europe. N Engl J Med 2008;359:2105-2119, Haslam DW, James WP. Obesity. Lancet 2005;366:1197-1209; Wang Y, Rimm EB, Stampfer MJ, Willet WC, Hu FB. Comparison of abdominal adiposity and overall obesity in predicting risk of type 2 diabetes among men. Am J Clin Nutr 2005;81:555-563).

This afternoon you are going to estimate your body mass index (BMI) to get an impression of your relative adiposity (do you belong to the skinny or the not-so-skinny group), your waist-to-hip ratio, to get an estimate of what type of adiposity it concerns (affecting your life expectation or not) and finally you are going to measure two representative skin-folds in order to estimate your fat- and your lean-body mass.

It should be stressed that the association between adiposity and public health only applies to large populations and not to individuals, meaning that people with a high BMI may outlive those with an “ideal BMI” (estimated around 25) and that the skinny ones may “cost” a great deal more in hospital charges than the not-so-skinny ones. It should also be stressed that you do not necessarily live healthy and happily (and then instantly die with a smile on your face at the age of 85) when you follow the instructions of health-care authorities to the letter. Firstly there is more to life than BMI or waist-to-hip ratios and secondly, with respect to physiology and pathology, it is insane to consider men all equal in view of the enormous heterogeneity of its genetic composition (certain statements of health-politics have the sole purpose to enforce hegemony of the ruling majority). However, it is certainly not unwise to follow health-care instructions; it may improve the quality of life in the five years that precede your last heart beat. In particular, whilst acknowledging a certain bias, it has been proven that coming to the SVI632 Signal Transduction course on bicycle extends a healthy and happy life span with a considerable number of years.

Measure weight (kilogram) and hight (meters) and measure BMI = weight/height2 =

Where do you stand?

Measure circumference at largest point of buttocks (hip) and half way in between lower ribs and iliac crest (waist) Waist circumference(cm) Hip circumference (cm) Ratio waist/hip =

From Pischon et al. N Engl J Med 2008;359:2105

subscapular

triceps

T = triceps skinfold thickness in mm, S = subscapular skinfold thickness in mm, M = midarm circumference in cm, and A = age in months

Fat content is: % of fat x body weight

Lean body mass equals: body weight minus fat content (this measure is sometimes used to estimate doses of medicaments or doses of general anesthesia)

1. Classification de l´état nutrionnel chez l´adulte en function de l´indice de masse corporelle (IMC) (selon l´international obesity task force (1998))
2. Calculatrice universelle de l´indice de la masse corporelle et du poids idéal
3. Outil de calcul et de classement de l´IMC (BMI)
4. Similar measurements on halls.md health calculators and charts
5. A series of “health-based calculators” including body fat calculation, body mass index, calorie intake and expenditure during exercise (note: 1 kg is 2.2 pounds, 1 inch is 2.54 cm and 1 foot is 30.48 cm and 1m70 equals 5 feet 6.1 inch tall)
6. Body fat percentage Wikipedia article