Talk:Peripheral membrane protein/Archive 1

Untitled

I have made some changes here.Biophys 20:46, 25 October 2006 (UTC)

• They look good to me - thanks! – ClockworkSoul 20:59, 25 October 2006 (UTC)

Work in progress...Biophys 07:28, 5 November 2006 (UTC)

I included a list of peripheral/amphitropic protein families with known 3D structures. Please correct me if I made any errors. The list is incomplete. Basically, this is a set of examples. Biophys 02:12, 7 November 2006 (UTC)

First version of "binding" section is done. Some references are needed. Biophys 05:46, 13 November 2006 (UTC)

To do list

Right now I can see the following shortcomings in this article.

1. More can be said about biology of peripheral proteins: what they are doing, why they are so important; their involvement in signal transduction pathways, membrane targeting, etc. I tried to classify them into different groups and describe briefly each of them. This should be done better.

2. Physical chemistry of membrane-protein binding (but this could be a separate article): forces defining the binding, how deep these proteins penetrate to the lipid bilayer, etc. There are certain controversies here.

3. There are many membrane-associated proteins not mentioned at all in this article.

4. It would be good to make stubs about some groups of proteins already included in the article, such as GLA or PX domains.

5. Pictures of different types of membrane-associated proteins in membrane; I could provide some. Biophys 19:30, 14 January 2007 (UTC)

Original Research?

I am a bit concerned that parts of this article could be construed as original research? First, the beginning of the section "Binding to the lipid bilayer" is written in a passive-past tense that I would expect when reading an original journal article, this immediately makes me suspicious. Second, I notice that nearly all of the external links (86 out of 89) are to a University of Michigan site, this seems odd to me. Third, why are the external links needed for proteins that have Wiki-articles, wouldn't the external link be expected to be in that specific article? (Looking around most specific articles contain the exact same links.) I could be wrong, I am not familiar with the UMich site, but something feels a bit off here. --DO11.10 19:24, 16 January 2007 (UTC)

Thanks for comments! I made these links to UofM site simply because it was easy to do (that is why it is "a bit off" - I agree). I could not find any other database that provides 3D structures specifically for peripheral membrane proteins (it would be very difficult to decide which proteins are "peripheral membrane" in PDB). So, one can take a quick look at the pictures of these proteins using the provided links. I thought it would be more convenient to have all links here and in other relevant articles. If others disagree, some links can be deleted here and left only in specific article (if there is such article). You are very welcome to provide any additional or better links or references, organize these links differently, include more proteins, etc.! One could take a look which families of membrane-associated proteins can be found in Pfam, SMART, SCOP or other places, and include the found information here. Biophys 20:21, 16 January 2007 (UTC)

Pfam includes ~ 2000 membrane protein families [1] (but many of them are transmembrane or "DUF" - unknown function). Through the OPM database (the currently included references) one can easily reach PDBsum and then Pfam and Uniprot record for the proteins currently included in the article, if it helps. Biophys 21:09, 16 January 2007 (UTC)

No, this this not original research, and the text was not copied from any other site (I did some copying from Pfam summaries in stub about C2 domain). But you are very welcome to improve and extend the text; it is certainly far from perfect. Biophys 20:21, 16 January 2007 (UTC)

Of course, we have to make references on some reviews or key papers rather than a link to [1]. I will provide some. Biophys 20:31, 16 January 2007 (UTC) One might go to link [1], find Table 2b, pick PDB code of any protein in this Table, and then take a look at citation of experimental studies in the page of each protein. But this is not necessary. Biophys 20:48, 16 January 2007 (UTC)

Okay, just wanted to make sure that it wasn't. Here are a few other suggestions (I would be happy to help with some of this):

1. A few figures would indeed be nice.
2. On the drawing at the top, is it possible to denote the intracellular/extracellular locations of the proteins? From the text I can deduce which is which, but I cannot tell from the drawing.
3. The list "categories with known 3D structures" doesn't really convey much information to me about how important the proteins are. These sections would probably be way more informative if each section were turned into a table that included name, reaction(?) or related molecules, functions mediated, organisms found in, with a link to the Umich site.... something like that.

Let me know what you think and if I can help you.--DO11.10 21:16, 16 January 2007 (UTC)

Could you prepare an example of such Table with only two or three proteins included? There are too many related proteins. I think it would be easier to have a good individual article for each protein and keep all database links there. We also have a Category of Peripheral membrane proteins. This upper-level article should probably only include references to most notable groups and representatives of peripheral proteins. So, I will try to work in this direction. Any suggestions? Biophys 02:57, 20 January 2007 (UTC)

Lipid bilayer

I will edit first lipid bilayer article.Biophys 05:07, 17 January 2007 (UTC)

Channel-forming proteins and peptides

Generally, this page is very good, however, I am confused by the inclusion of this last section. Surely, proteins/peptides that form channels must by their very nature span the entire membrane, thus making them integral membrane proteins. Pore forming toxins (work in progress) are polymorphic proteins that exist at various points in their life as soluble proteins (monomeric state), peripheral membrane proteins (pre-pore state) & integral membrane proteins (pore state). I am happy to be disagreed with, but wanted to raise the potential point of contention. :Bassophile 13:14, 17 January 2007 (UTC)

It is great that you making the article about Pore forming toxins! Not all channel-forming peptides and proteins are toxins. Some of them are involved in apoptosis, or may even form channels in vitro as an artifact. So, the question is: can they be defined as "peripheral" if they associate with membranes irreversibly and adopt the transmembrane orientation? They were described as such in review by Goni F.M. 2002. Non-permanent proteins in membranes: when proteins come as visitors. Mol. Membr. Biol. 19: 237-245. In fact, they have two different structural states, one of which is "peripheral" (or simply water-soluble and dissociated from membrane) and another is "transmembrane". They also qualify as "amphitropic" according to definition by Johnson J.E. and Cornell R.B. 1999. (Amphitropic proteins: regulation by reversible membrane interactions. Mol. Membr. Biol. 16: 217-235.). Anyway, what would you suggest? Maybe such proteins should be described in more detail in a separate article? That could be reasonable. Biophys 17:49, 17 January 2007 (UTC)

It's a little difficult. Membrane associated proteins could be classed as transmembrane, perhipheral & perhaps transient or amphi-pathic/tropic for this last class? However, I've not seen this notation used in the literature. Proteins that exist in two such different states are going to be difficult to pidgeon-hole. I suggest a paragraph or two on the peripheral membrane protein page, and as you suggest, create a separate page for amphitropic membrane proteins??? Bassophile 20:43, 17 January 2007 (UTC)

I will try to clarify this in the text.Biophys 21:25, 17 January 2007 (UTC)

I included "Classification" paragraph to clarify this. Please feel free to edit. Biophys 23:21, 17 January 2007 (UTC)

Further development

• I will download to Wikimedia several pictures of peripheral proteins with calculated hydrophobic boundaries and layers of bulk lipid phosphates (this seems to be easy). Biophys 20:23, 22 January 2007 (UTC)

• I will also remove all excessive links to OPM database if they are present in the articles about specific proteins. Biophys 20:40, 22 January 2007 (UTC)

I can include the links in the table I have been procrastinating, I think that they really are useful, but only when mentioned in the proper context (i.e. when you know what info the links will be providing). So just leave them for now, I will work on a table and then you can see if they still need to be removed. --DO11.10 22:19, 22 January 2007 (UTC)

Agree. Thank you! Biophys 22:24, 22 January 2007 (UTC)

Table

Okay here is a table, what do you think?

Table 1 moved to article, see Peripheral membrane protein#Enzymes

Some of the info may be incorrectly stated or too simplistic (This is not my area of specialty), that is why I am pasting it here, please feel free to copyedit or whatever, and move into the article when finished.

• Table format is much easier to read - especially for non-specialists (read: me). Nice work. -- MarcoTolo 00:02, 23 January 2007 (UTC)

I think this is really great and helps a lot! I did not see any scientific reviews about peripheral proteins where this material would be organized and represented in the way you suggested (although it should be!). We probably should make an additional column with Pfam links (or may be SMART links for membrane-targeting domains) - that would be convenient for a reader. For example, [2] for alpha-toxins, such as 1olp. These links can be found by clicking: 1. OPM link (select one of proteins in the family) -> 2. PDB sum ->3. Pfam, and then click on the graphical symbol of Pfam domain (upper left corner). Alpha/beta-hydrolases will have several Pfam links. What do you think?

There are some non-trivial moments here. For example, alpha/beta-hydrolase superfamily includes bacterial, fungal, gastric and pancreatic lipases, palmitoyl protein thioesterases, cutinase, and cholinesterases. Should they be considered together? Possibly yes, because we do not want too large Table. May be some less notable examples (such as vitelline protein) should be excluded to simplify the Table? I will check the content of the Table and modify as needed. Could you prepare similar tables for other groups of peripheral proteins (probably with an additional column for Pfam links) and put them here? That would help a lot. Thank you. Very good work! Biophys 05:23, 23 January 2007 (UTC)

Table Update 1: Okay, I took a look at the Pfam information and found that they can actually be used as references in most cases, whereas the Umich site cannot, I think that is an important distinction, so I have included the Pfam links as refs. In some cases I had to pick a representative compound, but I think that the table is clearly improved by these additions. I also excluded the vitelline protein and expanded others. For the alpha/beta hydrolases, the functioning was too complicated to explain in a table, (that group needs it's own subpage) so I just used your description. What do you think?--DO11.10 19:14, 23 January 2007 (UTC)

I think it is a good idea to make Pfam links as references. I have to think about alpha/beta hydrolases and check other content. May be we should change the order of protein families in the Table, e.g. all enzimes that process lipids would be in the beginning of the Table? Looks good. The Table provides a lots of links to other Wikipedia articles. It can be easily expanded if needed. Biophys 20:02, 23 January 2007 (UTC)

This Table will be too big if we include all currently present examples. I think we can limit ourselves only by most prominent examples:

• Structural domains: only Annexins and GLA-domains
• Membrane-targeting domains - all (this is related to cell signaling)
• Transporters - all except Polyisoprenoid-binding protein and Oxysterol-binding protein(?)
• Electron carriers - none(?)
• Polypeptide ligands - none(?) since one could argue that their actual (final) targets are other proteins
• Channel-forming - seem to belong partly to Pore forming toxins and partly to antimicrobial peptides and could be described there using similar Tables; they certainly deserve to be described.

So, what do you think? Biophys 18:06, 24 January 2007 (UTC)

I was actually planning on making several tables, one for each group of proteins (I think that this makes it easier, to know what kinds of proteins you are going to be reading about), but I tend to agree, there are too many examples. I agree with your assessment of what should be included, except for a couple of things.

1. Structure, should also include spectrin.
2. The polypeptide ligands- I would like to see the venoms listed here (most can be grouped together), since they are, well really cool. Also the defensins and cytolysins should be included, maybe under a broader heading? A larger, more inclusive table could be prepared for subpages, with "details" links from this page. What do you think?--DO11.10 18:42, 24 January 2007 (UTC)

Agree. Feel free to include, exclude, or group together whatever you think is necessary. Defensins and cytolysins are polypeptide toxins against bacteria, but they have different origin (vertebrane/invertebrate and bacterial). Note: there are three different groups of defensins in this article: references 61, 83, and 84.Biophys 19:37, 24 January 2007 (UTC)

The "Polypeptide ligands" and "Channel-forming proteins and peptides" in this article could be divided into two different groups: (a) polypeptide hormones and inhibitors involved in regulation within the same organism, and (b) polypeptide toxins that work against a different organism. Unfortunately, we do not have articles or Categories, specifically about polypeptide hormones and polypeptide toxins (I said "polypeptide" because it might be difficult to distinguish "peptides" and "proteins", although this is common practice). There are categories: Category:Toxicology, Category:Neurotoxins, Category:Ion channel toxins - you may take a look. You should also consult with articles Pore forming toxins and antimicrobial peptides, although they are incomplete. Biophys 19:37, 24 January 2007 (UTC)

Table2

Here is the second table (for the Structural domains). I left the most interesting ones in. (Moved to article--DO11.10 00:31, 26 January 2007 (UTC))

I am thinking that the membrane targeting domain could be expanded very slightly without making it into a table (i.e just make each entry in the list a full sentence or two.) Also since several of the same pathways (PIPs, etc.) are discussed in the Enzyme section, maybe the membrane targeting section should precede the structure section? Also the abbreviations for the PIPs need to be consistent. For example, in the Phosphatidylinositol (3,5)-bisphosphate article they use PtdIns(3,5)P2, so I think that we should stick with that notation throughout the article. What do you think?--DO11.10 18:28, 25 January 2007 (UTC)

Looks good. Everything is reasonable. Could you just go ahead and edit the article as you think is better? If anything is wrong, we can easily make corrections. You are right, "PtdIns(3,5)P2" nomenclature is more rigorous. No objections about that. Biophys 18:40, 25 January 2007 (UTC)

I made some changes, and added a third table, hopefully you agree with my changes and the layout and content of the table, feel free to change anything you would like. I had two "leftover" peptides.

α-Helical peptide hormones [3] and Tricyclic peptide RP71935 [4] Should these be included somewhere? --DO11.10 00:31, 26 January 2007 (UTC)

human / animal / general physiology

You are doing great job here. However, I have maybe one small remark... Those tables are excelent, I am watching with anticipation their emergence, but should not be the role of the protein in physiology more specified? I mean, it looks like it it is not considered necessary to specify the type of organisms where the process/protein has a role, while it is actually in human/animal physiology part, while when it takes a part in organisms of another kingdoms/domain there it is adequately assigned. I do not mean it is intentional POV :))) , I am actually almost certain it is subconscious, but I believe it should be equally labeled/described/assigned in all cases.

I might do it, but first I wanted to ask you on your opinion Reo ^ON | +++ 23:24, 25 January 2007 (UTC)

Do you mean to include something like "Animal Lipoxygenases" instead of simply "Lipoxygenases" in first column? Or say something like "lipid signaling in multicellular organisms" in third column? Please do, if others do not mind. It is very good to include kingdoms, unless that would increase the size of the Tables too much. But this is not so easy. For example, there are also homologous "Plant Lipoxygenases" (probably also peripheral membrane proteins), although they are not included in the OPM database, (this database is rather incomplete with regard to peripheral proteins). There are also bacterial phospholipases C (included in OPM), but I am not sure what is their function in bacteria. So you have to check. And so on. Biophys 00:09, 26 January 2007 (UTC)

That was pretty much the reason that I did not include that information in the tables (I had planned to but quickly realized it would be a serious amount of work). But if you want to hunt down the information, it would be good to include. --DO11.10 00:31, 26 January 2007 (UTC)

@Biophys -> there are also homologous "Plant Lipoxygenases" .... There are also bacterial ... Yes, and that is in fact exactly the reason way I first did the post here :)). Because it is simply not so simple. On the other hand It should be somehow clarified. Because to do it everyvhere would be difficult I propose to fix at least the situations where are plainly animal related informations like "Involved in function of clotting factors in the blood coagulation cascade"

IMHO in the third column, rather then in the first, but what do you think?

But I apologize for today, because I am editing from Central Europe and here is quite a late after midnight and I am adequately fatigued and ready for bed. Have a good time on the other side of that big puddle in between :). Reo ^ON | +++ 00:45, 26 January 2007 (UTC)

O'K, just correct tomorrow what you think should be corrected. Biophys 01:24, 26 January 2007 (UTC)

Further work

I will be very busy for a few days. Please feel free to edit anything. By the end of this week, I am going to download several pictures of peripheral proteins in membrane with bound lipids.Biophys 21:10, 30 January 2007 (UTC)

I finished the draft (unforunately, I do not have much time). It would be great if someone could edit my English, ask questions, criticized, added any references, and so on Biophys 02:57, 9 February 2007 (UTC)

I'm still catching up with RNA interference, but you've done some great work on this article. I'll give it a more in-depth look this weekend. Opabinia regalis 06:00, 9 February 2007 (UTC)

Thank you! Biophys 16:07, 9 February 2007 (UTC)

OK, when I said 'this weekend' I meant 'sometime this week'. Sorry ;) You can also try putting it up for peer review - or the MCB peer review here, which ought to get more traffic than it does. Opabinia regalis 06:36, 13 February 2007 (UTC)

No need to apologize. Whenever you can is fine. Thank you.Biophys 06:58, 13 February 2007 (UTC)

OK, few ideas from a fossil brain. I made some wording changes and did some copyediting (reigning in those definite articles ;) but didn't edit the content.

• The article is very well illustrated, but your images on commons have scanty descriptions (or in one case, {{description}}), which makes them hard to reuse elsewhere, and makes people flip back and forth between the larger image and the captioned thumbnail in the article. It'd be better to copy the local captions to the image description pages.

• For the clueless, the article could use just a standard stylized lipid bilayer diagram, without any proteins in it. Oddly lipid bilayer doesn't have one, but you seem quite capable of making nice diagrams, hint hint ;) A chemical structure of a phospholipid wouldn't hurt either.

There are two pictures of that kind in Cell membrane. Would they work? First picture in this articale could be even replaced by first pictuere from Cell membrane. Is not it? Actually, picture 1 in this article is a little misleading. It does not show membrane interfacial region, for example. Biophys 05:32, 18 February 2007 (UTC)

• Are the purification properties of peripheral membrane proteins important enough for the first paragraph of the lead?

In my opinion, they are not so important. But that is how it is usually described in textbooks. Probably could be moved somewhere. Biophys

• Leads generally don't need citations for facts that are repeated in more detail in the text. The citation on the 'reversible attachment' section looks weird, because the sentence is very general and in no way surprising. It would be better to leave off the footnote and describe the reference's point in more detail later, or alternatively, expand the sentence in the lead and leave the footnote where it is.
• The level of detail in the article varies substantially from one section to the next. Parts of the 'binding' section is very detailed and specific, but other sections are too general; e.g. 'Some polypeptide hormones, antimicrobial peptides, and neurotoxins...' - which ones? How was this measured? Another example, the membrane-binding mechanisms section is rather general; c
• 'The hydrophobic interactions are important even for ... proteins of natural origin' - I don't understand that. Is 'proteins of natural origin' what you mean here?

Corrected. Biophys 05:32, 18 February 2007 (UTC)

• 'the effective concentration of water that changes exponentially from nearly zero to ~2M' - from the hydrophobic core to the exterior? Where are these measurements made?

In model lipid bilayers using spin-labeled lipid analogues. Strictly speaking, this is debatable and may depend on lipid composition, presence of proteins, etc. Biophys 05:32, 18 February 2007 (UTC)

• 'The corresponding hydrocarbon membrane boundary planes pass through the carbonyl groups of phospholipids' - may need a diagram here; I'm not too sure what this is saying. The parallel boundary of the membrane is defined by the locations of the carbonyl groups?

Good comment! No. I can see - this is not explained at all. First, we need a better article about lipid bilayer to explain how the "acyl chain boundary" was defined there (yes, they correspond locations of the lipid carbonyl groups, but not "by definition", this is a kind of experimental result). Second question is how these boundaries were defined for the protein structures shown in Figures. Both questions are described in some of the refences, but they are not explaine in this article. It might be better to write a separate article about protein-lipid or protein-membrane interactions/association, but here focus mostly on proteins themselves rather than on protein-bilayer interactions.Biophys 05:32, 18 February 2007 (UTC)

• There are a lot of parentheticals in the article used to give examples, such as '(palmitoylation, myristoylation, or prenylation)' - these would all read better if the examples were de-parenthesized and written into the text.
• 'or to the entire lipid molecules (such as in GPI or cholesterol)' - what does it mean to bind to an entire lipid molecule?

Corrected.Biophys 19:56, 18 February 2007 (UTC)

• Physiological ionic strength = 0.14M NaCl? If that's going to be stated, it should be explained how one converts ionic strength of a heterogeneous solution of ions into a molarity of sodium chloride.
• The long list of factors at the end of the binding affinity section could use a citation.

I will do it. Biophys

• All the lists and tables at the end should have their external links converted to 'proper' references, or at least given titles rather than numbers (for example, the OPM accession number). I'm not sure why the lists are lists and the tables are tables; have the lists just not been converted yet, or is there a reason to leave them un-tabled? I'm thinking these lists might better exist as a separate list article (or multiple). Opabinia regalis 04:30, 14 February 2007 (UTC)

Lists can be made as Tables. Do you suggest making each category of peripheral proteins a separate Wikipedia article? Could be done. What others think?Biophys 19:56, 18 February 2007 (UTC)

Thank you very much! I have to think about this and make some changes.Biophys 04:40, 14 February 2007 (UTC)

That will require some time. If someone wants to correct any of these flaws, please do. Biophys 17:11, 14 February 2007 (UTC)

To-do list proposed above

If time is short, we might simply exclude Hisactophilin, Lactoferricin, and Glycolate oxidase from this article. We should do Lipopeptides and Cyclotides. Lipopeptides is also a sub-Category of Peptides. We have already articles about lipopeptides Daptomycin and Surfactin. Membrane lateral pressure is important but may be a tall order (images are needed), I can give some references if someone wants to try. Biophys 19:43, 18 February 2007 (UTC)

Recent edits

You are welcome to improve anything in the article (it is fine to cite articles several times, etc.). I will be back from a trip and can take a look at the edits in the end of next week. Biophys 19:55, 28 February 2007 (UTC)

I've added a stub for Ptdins4P, and I'll tackle cyclotides over the next day or two. Pbergson 22:41, 5 March 2007 (UTC)

I have made a stub for cyclotides.Biophys 04:50, 9 March 2007 (UTC)

Lactoferricin is stubby goodness now... -- Scientizzle 18:58, 14 March 2007 (UTC)