PEX11 Promotes Peroxisome Division Independently of Peroxisome Metabolism



By Xiaoling Li and Stephen J. Gould


The Journal of Cell Biology, Volume 156, Number 4, Feb 18, 2002 643-651



“This web page was created as an assignment for an undergraduate course at Davidson College.”





Peroxisomes are membrane bound organelles, containing peroxidase and catalase, that import all of their protein and most of their lipid content.  Peroxisomes contribute to many lipid metabolic pathways including B-oxidation of fatty acids.  Based on evidence that peroxisomes have a direct protein import pathway, it seems that peroxisomes undergo growth and division.


It had been previously established by Li and Gould that defects in peroxisomal fatty acid B-oxidation reduce peroxisome abundance in mammalian cells.  Hence, there was the possibility of metabolic control of peroxisome division.  Furthermore, PEX11 proteins are implicated in the regulation of peroxisome abundance.  Studies in yeast, human, rodent, and protozoan forms of PEX11 have supported the notion that loss of PEX11 causes a reduction in peroxisome abundance and PEX11 overexpression causes an increase in peroxisome abundance.  These previous results would point to a direct role of PEX11 in peroxisome division.  It has recently been proposed by Roermund et al. that the role of PEX11 in peroxisome division is only a secondary function while its role in oxidation of medium chain fatty acids (MCFAs) is its primary function.


In this paper, Li and Gould test the response of PEX11 under various conditions in mammalian, yeast, and mice cells.  The authors claim that their data suggest a revised role of PEX11—PEX11 plays a direct role in peroxisome division and the loss of this protein inhibits peroxisome metabolism indirectly. 


Figure 1 illustrates that overexpression of human PEX11B induces peroxisome abundance in a multistep process.  PEX11 has two forms, PEX11B and PEX11a, that are both integral peroxisomal membrane proteins (PMPs).  In this experiment wild type human skin fibroblasts were injected with a PEX11Bmyc-containing plasmids.  The cells were processed at varying time points, 1.5, 4.5, and 48 hours, in order to determine the proliferating activity over time.  At each of the three time points, the cells were visualized by immunofluorescence with antibodies to the myc epitope or PEX14.  PEX14 is an endogenously expressed PMP.  From the results, it appears that after 1.5 hours PEX11B is detected.  After 4.5 hours, the peroxisomes elongate and after 48 hours peroxisome number increases noticeably.  By using two different antibodies, the myc which detects only those cells that take up the plasmid and the PEX14 antibody which is endogenously expressed, this figure is well controlled.  This figure clearly shows the presence and elongation of peroxisomes.  Also, cell formation is obvious after 48 hours.  


The authors next set out to determine the specificity and extent of PEX11B induced peroxisome division.  Normal human fibroblasts were transfected with PMP34myc or PEX11Bmyc expression vectors.  After incubation, the cells were processed for immunofluorescence using antibodies specific for the myc epitope tag and PEX14.  Figure 2 shows the results.  According to the bar graph in part A, peroxisome abundance is noticeably greater in the cells overexpressing PEX11Bmyc compared to the untransfected cells and the cells with the PMP34myc vector.  Using both untransfected cells and cells transfected with a general PMP provides two controls.  Representative cells are shown in parts B-E after immunofluorescence with the two antibodies.  Again, the cells expressing PEX11Bmyc (D and E) had a great increase in the number of peroxisomes compared to cells expressing PMP34myc.  The authors indicate the literature shows that when other cells expressing unrelated PMPs (besides PMP34) were also myc tagged and monitored for peroxisome abundace, this overexpression had no effect on peroxisome abundance.  This figure shows that the increase in peroxisome abundance induced by PEX11B expression reflects a specificity of PEX11B and is not a consequence of PMP overexpression.  The data is well controlled by using a PMP34myc in addition to PEX11B and by marking with an antibody to the myc and PEX14 (as in Figure 1).  I do wonder why the untransfected cells were not labeled with the two antibodies and the peroxisome abundance shown as in parts B-E.  I suppose in theory this would not show anything drastically different compared to B and C.  While one could argue that cells expressing other PMPs should be used in this experiment, the authors refer to other papers reporting that this would not give contradicting results.  The data does support that overexpression of specifically PEX11B increases peroxisome abundance in wild type human skin fibroblasts.   


Roermund et al. hypothesized that the primary role of PEX11 is in MCFA oxidation and peroxisome division is an indirect effect of this oxidation.  Therefore, a functional peroxisomal B-oxidation pathway is essential if PEX11 is to have a role in peroxisome abundance.  To test this hypothesis, Li and Gould determined if PEX11 mediated peroxisome division could occur in cells lacking a functional peroxisomal B oxidation pathway.  The human cell line PBD005 was chosen as this lacks all peroxisomal metabolic function but still contains peroxisomes.  PDB005 cells were transfected with PMP34myc and PEX11Bmyc.  The results are shown in figure 3.  Again the bar graph shows that cells overexpressing PEX11Bmyc have thirty times more peroxisomes as compared to untransfected cells and cells transfected with PMP34myc.  Representative cells (shown in B-E) labeled again with antibodies to the myc tag and PEX14 tag, show more abundance of peroxisomes in PEX11Bmyc cells than PMP34myc cells.  The same controls were employed as in figure 2.  From this data, it is concluded that the peroxisome proliferating activity of human PEX11B is not dependent on peroxisomal B-oxidation activities.  Since a cell line that is defective in all peroxisomal metabolic activities was used, it is concluded by the authors that the peroxisomal proliferating activity of human PEX11B is independent of all peroxisomal metabolic activities.  This refutes Roermund et al.’s hypothesis.  Since the PDB005 cell line was so appropriately chosen and multiple controls were done, it would appear that the data does support the author’s refute of Roermund’s claims.     


Peroxisomes are the only site of fatty acid B-oxidation in S. cerevisiae.  Li and Gould mention that previous studies have demonstrated peroxisome abundance increases when S. cerevisiae cells are shifted from glucose dependent growth to growth on fatty acids.  The abundance of peroxisomes was examined in several S. cerevisiae cell lines with varying inserts and on varying media.  In figure 4, parts A and B, the range of peroxisome abundance is established.  The S. cerevisiae laboratory strain BY4733 was transformed with a GFP containing plasmid and Gal promoter.  When shifting the same cells from glucose media to the fatty acid media oleic acid, there was a marked increase in peroxisome number (A and B).  In parts C-G, the cell line XLY1 is used.  This cell line contains the GFP plasmid with Gal promoter.  When shifting cells from glucose to galactose, there is an immediate increase in peroxisome abundance (C and D).  The galactose promoter is no longer repressed.  Galactose induced PEX13 (E) and Ypr128c (F) plasmids did not increase peroxisome abundace compared to the empty vector (D).  However, galactose induced expression of PEX11 (G) did increase peroxisome abundance up to the level observed on oleate media.  I find it curious that the plasmid contained PEX11 and not PEX11B.  I am assuming that since this has shifted from human cells to yeast, there are no alpha and beta forms.  But it would be nice if the authors clarified this point.  Finally, in figure 4, a different strain of S. cerevisiae cells, XLY2, were transfected with similar plasmids containing a gal promoter.  The galactose induced expression of PEX13 had no effect on peroxisome abundance.  Again, the galactose induced expression of PEX11 increased the peroxisome abundance to levels similar in oleate grown BY4733 cells and XLY1 cells with PEX11. 

            From these results, the authors make the claim that in the absence of fatty acids from the growth media, it is extremely unlikely that the peroxisomal fatty acid B-oxidation pathway was engaged.  This then leads to the conclusion that peroxisomal proliferation in S. cerevisiae is independent of the B-oxidation pathway.  From reviewing this data, I do believe that the author’s have a justified reason to make these claims.  There were multiple controls regarding media (both with and without fatty acids), cell line, and plasmid insert.  The results consistently showed that peroxisome abundance increased with the PEX11 insert, regardless of cell line, up to the extreme level established with growth on the oleate media.  Consistent results and appropriate controls lead me to believe the author’s conclusions. 


Once studying the role of PEX11 in human and yeast cells, the authors then focus on mouse cells.  Li and Gould test the effect of the loss of PEX11B on peroxisome metabolism.  Mouse embryonic fibroblasts from PEX11B+/+ and PEX11B-/- cell lines were cultured on serum free media and then processed for immunofluorescence using antibodies for peroxisomal enzyme catalase (a matrix marker) and PEX14. Figure 5, part A, shows that there is a decrease in peroxisome abundance in PEX11B-/- mouse fibroblasts compared to wild type cells.  The results for culturing the fibroblasts in normal conditions and in serum free media are shown for both the wild type and PEX11B-/-.  This is well controlled, yet when taking into account the error bars, I do wonder if there is a significant difference between the wild type and PEX11B-/- cells.  In the next part, wild type and PEX11B-/- mouse fibroblasts were cultured in serum free media and then visualized each with PEX14 antibody and matrix marker enzyme catalase.  The results show that the abundance of peroxisomes in wild type cells was twice that in PEX11B-/- grown under identical conditions.

            The authors reason that if PEX11B functions primarily in fatty acid oxidation, then peroxisome abundance should the same in wild type and PEX11B-/- cells when these cells are grown on serum free media—devoid of lipids and substrates of fatty acid oxidation.  On the other hand, if PEX11B’s primary function is peroxisome division, peroxisome abundance should be reduced in PEX11B-/- compared to wild type cells.  The data in figure 5 indicate just this.  The loss of PEX11B appears to affect peroxisome abundance independently of peroxisomal metabolism.  I find the data in this figure to, yet again, support the author’s claims.  My only question is with the bar graph and the idea of statistical significance.  But since Molecular Biologists often do not worry with this, the immunofluorescence data that is well controlled seems sufficient to support the author’s claims.                  


I found this paper to be well written with sufficient data to support the author’s claims.  The data was presented in a straightforward manner with multiple controls in each experiment.  My only concerns are that, first, the authors often refer to the role of PEX11 when drawing conclusions from a set of data while the data only refers to the role of PEX11B.  Second, the paper never addressed the role of PEX11a.  I wonder if this is unique to human cells and if the role of PEX11a could be exclusive to peroxisomal metabolism.  The paper provides evidence that

à overexpression of PEX11B increases peroxisome abundance in wild type human skin fibroblasts

à overexpression of PEX11B increases peroxisome abundance in a human cell line defective in all peroxisomal metabolic functions

à overexpression of PEX11 in S. cerevisiae increases peroxisome abundance in lipid free medium 

à PEX11B-/- cells have only half the peroxisome abundance as wild type cells when grown in serum free medium

I found the data adequate to support the claim that PEX11 (PEX11B in humans and mice) proteins promote peroxisome division regardless of the metabolic pathway in peroxisomes. 


Future Studies


The role of PEX11 needs to be characterized both in the metabolic pathway and in peroxisome division. 


The authors mention that a possible way in which multiple peroxisomal metabolic pathways could be damaged in PEX11 mutant cells is if the loss of PEX11 proteins alters the physical properties of the peroxisome membrane.  Hence, I propose that an experiment be performed to characterize the membrane of a wild type peroxisome, a PEX11 mutant, and another mutant such as PEX13.  This would have to be done in whichever organism the appropriate mutants exist.  FRAP (Fluorescence Recovery After Photobleaching) is one such method that can be employed to determine if these proteins are able to move within the peroxisome membrane. 


It is known that PEX11 is a membrane bound protein encoded by the PEX11 gene.  The question is if the protein is an integral membrane one and if so, which part of the protein interacts with the molecules involved in the fatty acid metabolic pathway.  First, after sequencing the gene by the Sanger Method, the protein sequence can be deduced.  Then a Kyte-Doolittle analysis can be performed to predict the hydrophobic and hydrophilic regions of the protein.  If it is an integral membrane protein, then the graph should span the X-axis.  To determine if the cytoplasm or peroxisome portion of the PEX11 protein is where initiation of metabolism occurs, the yeast two hybrid system could be employed.  The initiator protein of the fatty acid B-oxidation cycle should interact with either the part of the PEX11 that faces the cytoplasm of the cell or the part of the PEX11 that faces the lumen of the peroxisome.  If Beta gal is used as a reporter, then the cells that turn blue will contain the part of the PEX11 that interacts with the metabolic pathway.  I suppose that ideally, it would be best to crystallize the protein interacting with the molecule in the metabolic pathway.  But this is a bit more arduous.            


The authors also ponder the role of PEX11 proteins in peroxisome division.  Specifically, Li and Gould mention that peroxisome division may be sensitive to PEX11 concentrations in the peroxisome membrane.  An experiment could be performed to take concentrations of the PEX11 membrane protein at different points in the peroxisome division.   


Perhaps during peroxisome division, a part of the peroxisome membrane dissolves such as when a bud pinches off.  To investigate membrane continuity during division, FLIP (Fluorescence Loss in Photobleaching) can be employed.  The investigator would have to isolate cells at varying stages of peroxisome division in order to understand the changes in the membrane and the proteins within the membrane.   





Molecular Front Page


Davidson College


For questions or comments contact