蛋白质与蛋白组学

Introduction





De novo is Latin for, "over again", or "anew".  A popular definition for "de novo peptide sequencing" is, peptide sequencing performed without prior knowledge of the amino acid sequence. Usually this rule is imposed by Edman degradation practitioners who perform de novo sequencing day in and day out, and perhaps feel a little bit threatened by that half million dollar mass spectrometer sitting down the hall, that can supposedly sequence peptides in a matter of seconds, and not days.  Actually, any research project should be started with as much information as possible, there should never be a need to restrict your starting knowledge, unless of course you are performing a clinical trial or some other highly controlled experiment.  Mass spectrometers do have the advantage when it comes to generating sequence data for peptides in low femtomole quantities.  However, Edman degradation will always enjoy the advantage when pmol quantities of a peptide are available.  At higher pmol quantities (2-10 pmol), Edman will often provide the exact amino acid sequence without ambiguity for a limited run of amino acids, 6-30 amino acids, usually taking 30-50 min per cycle of the sequencer.  However, at lower quantities, gaps and uncertainties are often encountered, even with Edman sequencing.  MS/MS enjoys sensitivity, and speed, and does not require an external standard for each amino acid or amino acid variant. MS/MS sequencing does have difficulty with isobaric or near isobaric masses, for example telling K from Q on low resolution, low mass accuracy mass spectrometers.  Another advantage is that MS/MS sequencing is never stopped by a blocked amino terminus, as is the case for Edman degradation.

Edman practitioners will often blast MS/MS sequencing on it deficiencies. We do need to approach de novo sequencing with our eyes wide open to all of its challenges and also to all of it's advantages.  As scientists we need to have faith in the derived de novo sequence without knowing the sequence ahead of time, I guess this is at the heart of de novo.  Especially when software is involved, we need to be confident enough to point to the top ranked output sequence and say, "Yes, this is the most correct sequence!"  It is appropriate to test your skills or the skills of  a software package with blinded but know sequences.  Throughout the tutorial we will look at some known and some blinded sequences to demonstrate some of the de novo sequencing principles and also to test your newly learned de novo sequencing skills.

This tutorial leans heavily on a de novo sequencing course that was presented in 1992 at the University of Virginia, taught by Professor Donald F. Hunt.  Dr. Hunt and his colleagues have generously taught this course for many years, educating generations of mass spectroscopists.  It is impossible to calculate the enormity of the contribution that Dr. Hunt and his teaching efforts have made to countless research projects, both influencing basic, and medical, and drug research.  One of the most notable early applications was the sequencing of peptides bound to MHC molecules.  This was truly ground breaking work by the Hunt lab at the University of Virginia.



Let's Start:

A common question when one begins to talk about peptide fragmentation is, "What are b and y ions?"  First we will look at the classical nomenclature, and then we will look at our first example peptide.

Peptide Fragmentation Nomenclature

b, y and a ions

The most common peptide fragments observed in low energy collisions are a, b and y ions, as described in the figure above.  The b ions appear to extend from the amino terminus, sometimes called the N-terminus, and y ions appear to extend from the carboxyl terminus, or C-terminus.  While readily observed and diagnostic for b ions, a ions occur at a lower frequency and abundance in relation to b ions. The a ions are often used as a diagnostic for b ions, such that a-b pairs are often observed in fragment spectra.  The a-b pairs are separated by 28u, the mass for the carbonyl, C=O.  
The fragment types listed above are the most common fragments observed with ion trap, triple quadrupole, and q-TOF mass spectrometers.  Follow the link to see the fully annotated fragmentation nomenclature as proposed by Biemann.  An important note: an earlier nomenclature was proposed by Roepstorff and Fohlman and later modified by Biemann.  The Biemann adaptation has been widely accepted.

Peptides do not fragment sequentially, that is to say, the first fragmentation event does not start at the amino terminus and proceed sequentially one residue at a time down the amino acid chain.  The fragmentation events are somewhat random and definitely not sequential.  In addition, some fragmentations are preferred over others as noted by the variation in the abundance of observed peaks in the spectrum below. Most of us can recognize a peptide fragment spectrum just by glancing at it.  The peaks will appear to differ by the approximate mass of an amino acid residue as shown below.





Figure 1. This is an MS/MS spectrum of the tryptic peptide GLSDGEWQQVLNVWGK. This data was collected on an ion trap mass spectrometer.  This spectrum will be the subject of our first unblinded de novo sequencing example.   


The mess of peaks normally observed in a fragment spectrum are a reflection of the population of fragment ions produced in the collision cell of a mass spectrometer.  The sequence of the peptide is determined by the mass difference between these peaks.  To complicate matters there will be y and b ions intermixed that may allow you to to establish a sequence, both forward and backward.

Those fragment peaks that appear to extend from the amino terminus are termed "b ions".  Figure 2. below demonstrates the ladder or family of "b ions" that may be observed in the fragment mass spectrum for this tryptic peptide. The b fragment peaks are labeled from the amino to the carboxyl terminus.  The fragment containing only the amino terminal amino acid is termed b1. The fragment containing the first two amino terminal amino acids is termed the b2 ion, and so forth.  The nomenclature is very simple to follow.

  

Figure 2.








Below is a closer look at the generic structure of the first six amino terminal b ions.  You can calculate the mass of any b ion,  basically it is the mass of the shortened peptide (M)-17 (OH) = b ion m/z  or or more simply M-17 = b ion m/z. To keep it simple this is the calculation for a singly charged b ion.  
  
Figure 3.




Figure 3.  Shows the first six b ions in a little bit more detail.  The b ion m/z value is basically the mass of the peptide minus OH, or -17u.  
  




Similarly, groups of peptide fragment ions appear to extend from the C-terminus, these peaks are termed, "y ions". The y ion series for our example peptide GLSDGEWQQVLNVWGK is illustrated below in Figure 4.
  


Figure 4.





  
Below in Figure 5 the first six y ions are illustrated in some detail.  To calculate the m/z value for the y ions just calculate the (M+H)1+  for the shortened peptide.

  

Figure 5




  

Figure 5. The first six y ions are illustrated.  The calculated masses are shown above each y ion in bold numbers.  
  
-  
  
y and b ion mnemonic:
-

To remember which are y ions and which are b ions you can remember that b ions are the series that extend from the amino terminus, or the front of the peptide.  To us, it would make more sense if the b ions extended from the back or C-terminus, but just the opposite is true, b ions extend from the front of the peptide, the amino terminus.  
  
The screen shot below is the output of a free on-line calculator provided by the Institute of Systems Biology.  All you need to do is paste in the sequence of your peptide and it will output the expected y and b ions. The URL for this resource is cuturl('http://db.systemsbiology.net:8080/proteomicsToolkit/index.html')  You can use these masses to casually match up the masses to the peaks in Figure 1, at the top of the page.

Figure 1. This is an MS/MS spectrum of the tryptic peptide GLSDGEWQQVLNVWGK. This data was collected on an ion trap mass spectrometer.  This spectrum will be the subject of our first unblinded de novo sequencing example

Figure 3.  Shows the first six b ions in a little bit more detail.  The b ion m/z value is basically the mass of the peptide minus OH, or -17u.

Figure 5. The first six y ions are illustrated.  The calculated masses are shown above each y ion in bold numbers.

The Rules

(the observations)


Before we go though our first MS/MS example we should take a look at some of the rules that are generally applied to de novo sequencing. These rules or observations were adapted from a 1991 de novo sequencing course taught at the University of Virginia by Professor Donald F. Hunt and Dr. Jeffrey Shabanowitz.  Here are a few of the rules and observations that were introduced in that course.  




The Rules



Loss of Ammonia and Water

y and b ion fragments containing the amino acid residues R, K, Q, and N may appear to lose ammonia, -17.
-
y and b ion fragments containing the amino acid residues S, T, and E may appear to lose water, -18.  In the case of glutamic acid, E must be at the N-terminus of the fragment for this observation to be made.

Spectral Intensity Rules

b ion intensity will drop when the next residue is P, G or also H, K, and R.
-
Internal cleavages can occur at P and H residues.  An internal cleavage fragment is a fragment that appears to be a shortened peptide with P and or H at its amino terminus, for example the peptide EFGLPGLQNK may display the b ions PGLQNK, PGLQN, PGLQ, etc.  These are the result of a double cleavage event.  The y ion intensity will often be the most prominent peak in the spectrum.
-
It is common for b and y ions or y and b ions to swap intensity when a P is encountered in a sequence. This can also be true when the basic residues H, K, or R are encountered in the sequence.
-
When a cleavage appears before or after R, the -17 (loss of ammonia) peak can be more prominent than the corresponding y or b ion.
-
When encountering aspartic acid in a sequence, the ion series can die out.

Amino Acid Composition

It is possible to observe immonium ions at the low end of the spectrum that can give a clue to the amino acid composition of a peptide.  One caveat is that if you do not see an immonium ion for a particular amino acid, this does not mean that that amino acid is absent from the sequence.  You can follow this link to learn more about immonium ions.

Isobaric Mass

Leucine and Isolucine have isobaric masses and cannot be differentiated in a low energy collision.
When we see this mass difference in a spectrum we will label it X or Lxx, adopting the Hunt nomenclature.
-
Lysine and Glutamine have near isobaric masses, 128.09496 and 128.05858 respectively. The delta mass is 0.03638 this difference can be used to differentiate K from Q on a mass spectroneter capable of higher mass accuracy and resolution, such as a  q-TOF mass spectrometer. Usually triple quadrupole or ion trap mass spectrometers are incapable of this feat.  On a lower mass accuracy mass spectrometers an acetylation can be performed to shift the mass of lysine by 42u.  If you like to live dangerously, and we do not, one can assume that a 128 mass shift internally on a tryptic peptide is a glutamine unless followed by a proline or sometimes aspartic acid.  Other instances of internal lysines left standing after a tryptic digest (this is our personal observation) is when double lysines occur in a sequence, so be careful.
-
There are instances where two residues will nearly equal the mass of a single residue, or a modified residue will nearly equal the mass of another amino acid. For more examples, see the following table.  
-

More Rules

When starting a de novo sequencing project, start at the high mass end of the spectrum; the lower number of peaks at this end often makes it easier to start sequencing.
-
The region 60 u below the parent mass can be confounded by multiple water and ammonia losses, be careful.  Realize that glycine may be your first amino acid and may fall in this region.
-
Do you want to know if your tryptic peptide ends in a K or an R?  Look for the diagnostic y1 ions at the low end of the spectrum, you may observe 147 for K or 175 for R.
-
The b1 fragment is seldom observed making it difficult to determine the order of the first two N-terminal amino acids in a peptide sequence.  Solutions for this problem can include a one step Edman degradation or an acetylation.
-
Once you know the mass of a b or y ion the corresponding y or b ion can be calculated using the following formulas.
-
y = (M+H)1+ - b +1
-
b = (M+H)1+ - y +1
-
Once you observe a y or b ion, calculate the mass of the corresponding b or y ion and go look for it in the spectrum!

谢谢楼主，正是我想要的