Dear Gerwald, dear Artem,

As, in the absence of quantum mechanical calculations, we can only 
speculate about the contribution of mesomeric states it seems to me 
from the data presented that the left formula (i.e. without the 
mesomeric effect) may match the structural situation, in particular 
the shorter bond distance to the His NE2 atom.

In this case only the bridging carbon and the amino nitrogen have to 
be in the same plane with the imidazolium ring while there should be 
rotational freedom around the aminoN-Calpha bond. This is difficult 
to see from the attached picture but the precise local geometry also 
depends on the bonding state and force field that was applied during 
the previous refinement step.

The heating during preparation, together with a favourable effect of 
close contacts within the protein, may drive such a condensation 
reaction. However, I do not see from the experimental details 
disclosed how a subsequent reduction may have taken place unless it 
happened already inside the host cell.

High resolution ESI-MS should actually reveal whether a reduction 
occurred or not. At least we can often count hydrogen mass 
differences when we look for the proper formation of disulphide bonds.

Arne



At 5:11 Uhr -0400 11.10.2008, gerwald jogl wrote:
>Thank you. I was thinking along the lines of amination but was 
>concerned about the reactivity of the His nitrogen...
>
>To add more information about the system: this is an untagged 
>thermophile protein, no His column used for purification. It was 
>heated to 65 degC during purification at pH 6.8 and crystallized at 
>pH 8.0 imidazole buffer and calcium acetate present.
>I believe the two mesomeric states shown by Dr. Skerra would require 
>that the His_N - C - amino_N - Calpha are positioned in a plane, but 
>this is not the case. The tetrahedral angle is approaching 90 
>degrees instead. (I have included a small screenshot with refmac 
>density after refinement with a carbon atom as atom X. The 
>temperature factor of this carbon is similar to the surrounding 
>atoms after refinement.)
>I guess the geometry would argue for the bis-amine formation 
>requiring reduction, as suggested by Artem.
>The bis-amine formation would have to go through the amidine 
>intermediate and maybe the N-term is flexible enough to accomodate 
>the planar group before reduction.
>There is a somewhat similar reaction observed in the crosslinking of 
>collagen. There, a lysine is converted in an allysine (aldehyde 
>instead of terminal amino group). The allysine reacts with another 
>lysine to form an imine, which can be reduced with NADH to form a 
>stable crosslink. So, I guess if we assume that the His is reactive 
>enough to attack the formyl group and the N-term is flexible enough 
>to accomodate the amidine then a crosslink by this mechanism is at 
>least conceivable.
>I will submit a sample to mass-spec analysis to verify the presence 
>of an additional carbon atom.
>Thanks for the advice.
>Gerwald
>
>
>
>
>Prof. Dr. Arne Skerra wrote:
>>Dear Colleagues,
>>
>>The amidine originating from the condensation reaction between the 
>>Met N-formyl group and the His side chain should be the other way 
>>around! This would also be roughly consistent with the reported 
>>geometry. In fact, there should be two mesomeric states, provided 
>>there is N protonation.
>>
>>Arne Skerra
>>
>>
>>
>>At 5:56 Uhr -0400 10.10.2008, Artem Evdokimov wrote:
>>>If you're 100% sure that this is only one atom then amination comes to mind.
>>>I have no clue what conditions would favor such reactivity but it is
>>>possible that the formyl group on the Met was aminated with the cyclic N of
>>>the histidine, resulting in either a substituted bis-amine (requires
>>>reduction, may not be stable) or in a cyclic amidine (the latter requires
>>>four atoms - N1-C=N-Ca - to be in the same plane). This amidine should have
>>>some pretty interesting properties (probably more like a Schiff base than an
>>>amidine).
>>>
>>>Could you tell us a bit more about the system you're working with?
>>>
>>>Artem
>>>
>>>P.S. The bond lengths you describe are not typical for C-N, however at 1.9A
>>>resolution it is not very likely that the values you measured actually
>>>correspond to the bond lengths (hopefully, because otherwise you have
>>>
>>>-----Original Message-----
>>>From: CCP4 bulletin board [mailto:[log in to unmask]] On Behalf Of
>>>gerwald jogl
>>>Sent: Thursday, October 09, 2008 4:24 PM
>>>To: [log in to unmask]
>>>Subject: [ccp4bb] Potential N-terminal cyclization
>>>
>>>Hi All,
>>>
>>>I am looking for some input regarding an unusual feature in one of our
>>>structures. Maybe someone has come across something similar or has some
>>>thoughts about it...
>>>
>>>We have a 1.9 A crystal structure with well-defined density in the
>>>N-terminal region. The side chain of His3 is oriented towards the
>>>N-terminal amino group of Met1 and there is a nice difference density
>>>'ball' right in between the His and the amino group suggesting that
>>>there is an additional atom that is covalently linked. There are two
>>>molecules in the asu and both show the same feature (no ncs refinement).
>>>My problem is that I cannot come up with a reaction that would result in
>>>such a linkage and there is not much to be found in the literature.
>>>
>>>If I place a hypothetical atom in the difference density peak, I can
>>>measure distances and angles. Here are some numbers: Distance from atom
>>>X to the N-terminal nitrogen 1.72 (1.64 in chain B). Distance from atom
>>>X to His NE2 1.58 (1.45 in chain B; the temperature factors of the His
>>>side chain are slightly more consistent with this ring orientation
>>>compared to the 180 degree flip that would bring CE1 towards atom X).
>>>Atom X is coplanar with the His ring. The angle between NE2 - X - N1 is
>>>95 deg (106 in chain B). The angle between X - N1 - Calpha1 is 111
>>>degrees in both chains. As the N-terminal methionine is still present,
>>>it is possible that the methionine formyl-group was present before the
>>>hypothetical reaction. However, there is only one 'atom' in the
>>>difference density.
>>>
>>>Any comments or suggestions would be highly welcome.
>>>Gerwald
>>
>>
>
>
>Thank you. I was thinking along the lines of amination but was 
>concerned about the reactivity of the His nitrogen...
>
>To add more information about the system: this is an untagged 
>thermophile protein, no His column used for purification. It was 
>heated to 65 degC during purification at pH 6.8 and crystallized at 
>pH 8.0 imidazole buffer and calcium acetate present.
>I believe the two mesomeric states shown by Dr. Skerra would require 
>that the His_N - C - amino_N - Calpha are positioned in a plane, but 
>this is not the case. The tetrahedral angle is approaching 90 
>degrees instead. (I have included a small screenshot with refmac 
>density after refinement with a carbon atom as atom X. The 
>temperature factor of this carbon is similar to the surrounding 
>atoms after refinement.)
>I guess the geometry would argue for the bis-amine formation 
>requiring reduction, as suggested by Artem.
>The bis-amine formation would have to go through the amidine 
>intermediate and maybe the N-term is flexible enough to accomodate 
>the planar group before reduction.
>There is a somewhat similar reaction observed in the crosslinking of 
>collagen. There, a lysine is converted in an allysine (aldehyde 
>instead of terminal amino group). The allysine reacts with another 
>lysine to form an imine, which can be reduced with NADH to form a 
>stable crosslink. So, I guess if we assume that the His is reactive 
>enough to attack the formyl group and the N-term is flexible enough 
>to accomodate the amidine then a crosslink by this mechanism is at 
>least conceivable.
>I will submit a sample to mass-spec analysis to verify the presence 
>of an additional carbon atom.
>Thanks for the advice.
>Gerwald
>
>
>
>
>Prof. Dr. Arne Skerra wrote:
>
>>Dear Colleagues,
>>
>>The amidine originating from the condensation reaction between the 
>>Met N-formyl group and the His side chain should be the other way 
>>around! This would also be roughly consistent with the reported 
>>geometry. In fact, there should be two mesomeric states, provided 
>>there is N protonation.
>>
>>Arne Skerra
>>
>>
>>
>>At 5:56 Uhr -0400 10.10.2008, Artem Evdokimov wrote:
>>
>>>If you're 100% sure that this is only one atom then amination comes to mind.
>>>I have no clue what conditions would favor such reactivity but it is
>>>possible that the formyl group on the Met was aminated with the cyclic N of
>>>the histidine, resulting in either a substituted bis-amine (requires
>>>reduction, may not be stable) or in a cyclic amidine (the latter requires
>>>four atoms - N1-C=N-Ca - to be in the same plane). This amidine should have
>>>some pretty interesting properties (probably more like a Schiff base than an
>>>amidine).
>>>
>>>Could you tell us a bit more about the system you're working with?
>>>
>>>Artem
>>>
>>>P.S. The bond lengths you describe are not typical for C-N, however at 1.9A
>>>resolution it is not very likely that the values you measured actually
>>>correspond to the bond lengths (hopefully, because otherwise you have
>>>
>>>-----Original Message-----
>>>From: CCP4 bulletin board 
>>>[<mailto:[log in to unmask]>mailto:[log in to unmask]] On 
>>>Behalf Of
>>>gerwald jogl
>>>Sent: Thursday, October 09, 2008 4:24 PM
>>>To: <mailto:[log in to unmask]>[log in to unmask]
>>>Subject: [ccp4bb] Potential N-terminal cyclization
>>>
>>>Hi All,
>>>
>>>I am looking for some input regarding an unusual feature in one of our
>>>structures. Maybe someone has come across something similar or has some
>>>thoughts about it...
>>>
>>>We have a 1.9 A crystal structure with well-defined density in the
>>>N-terminal region. The side chain of His3 is oriented towards the
>>>N-terminal amino group of Met1 and there is a nice difference density
>>>'ball' right in between the His and the amino group suggesting that
>>>there is an additional atom that is covalently linked. There are two
>>>molecules in the asu and both show the same feature (no ncs refinement).
>>>My problem is that I cannot come up with a reaction that would result in
>>>such a linkage and there is not much to be found in the literature.
>>>
>>>If I place a hypothetical atom in the difference density peak, I can
>>>measure distances and angles. Here are some numbers: Distance from atom
>>>X to the N-terminal nitrogen 1.72 (1.64 in chain B). Distance from atom
>>>X to His NE2 1.58 (1.45 in chain B; the temperature factors of the His
>>>side chain are slightly more consistent with this ring orientation
>>>compared to the 180 degree flip that would bring CE1 towards atom X).
>>>Atom X is coplanar with the His ring. The angle between NE2 - X - N1 is
>>>95 deg (106 in chain B). The angle between X - N1 - Calpha1 is 111
>>>degrees in both chains. As the N-terminal methionine is still present,
>>>it is possible that the methionine formyl-group was present before the
>>>hypothetical reaction. However, there is only one 'atom' in the
>>>difference density.
>>>
>>>Any comments or suggestions would be highly welcome.
>>>Gerwald


-- 

------------------------------------------------------------------------
Prof. Dr. Arne Skerra                                  [log in to unmask]
Lehrstuhl f. Biologische Chemie               Phone: +49 (0)8161 71-4351
Technische Universitaet Muenchen                Fax:               -4352
85350 Freising-Weihenstephan
Germany                                         http://www.wzw.tum.de/bc
------------------------------------------------------------------------