This web page was produced as an assignment for an undergraduate course at Davidson College.


PEX11 promotes peroxisome division independently of peroxisome metabolism


Xiaoling Li and Stephen J. Gould

Department of Biological Chemistry, The Johns Hopkins University School of Medicine, Baltimore, MD 21205

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


PEROXISOMES are cellular organelles that are small in size, but very important to cellular function. They are involved in a variety of cellular metabolic processes, including the oxidation of long and very long chain fatty acids. Peroxisome biogenesis has been a subject of interest because defects in this process cause certain lethal human diseases. Previous studies have shown that peroxisomes are formed by existing peroxisomes' growth and division. Researchers investigated peroxisome division, and two competing theories arose to explain how peroxisome division is controlled. The first theory states that there is metabolic control of peroxisome division and abundance. This is supported by evidence showing that defects in the oxidation process of peroxisomes lead to reduced peroxisome abundance. The competeing theory states that a single peroxisome membrane protein, the PEX11 protein, controls peroxisome division. This theory has been supported by studies in yeast showing that loss of PEX11 led to a reduction in peroxisome abundance, while overexpression of PEX11 led to an increase in peroxisome abundance. In this paper, Li and Gould seek to prove this second theory, that PEX11 proteins are a direct cause of peroxisome division.

Fig. 1. This figure uses immunofluorescence microscopy to show how overexpression of the human PEX11-beta gene affects the level of peroxisome abundance. The researchers injected into human skin fibrobast cells a plasmid, pcDNA3-PEX11-beta-myc, that causes overexpression of the human PEX11-beta gene. They measured the level and appearance of the peroxisomes (peroxisomes are visible via either myc or PEX14[another peroxisome membrane protein produced in peroxisomes] antibodies---they are the white spots on the dark background) after certain time periods--1.5 h, 4.5 h, and 48 h. This showed the researchers the various steps in peroxisome proliferation. They saw three dstinct steps in this process--appearance of normal peroxisomes at 1.5 h (fig. 1a), elongation of the peroxisomes at 4.5 h (fig. 1c), and a dramatic increase in peroxisome abundance at 24-48 h (fig. 1e). These distinct steps are observed in the small boxes located in Fig. 1a, 1c, and 1e.

Fig. 2. The figure displays the specificity of PEX 11-beta in increasing peroxisome abundance. The researchers tested the effects of overexpression of PEX11-beta as well as PMP34, another protein required in the peroxisome oxidation process. They transfected human skin fibroblasts with the PEX11-betamyc or PMP34myc expression vectors. They measured the peroxisome abundance of untransfected cells as a control. They found that the cells expressing PMP34myc were not significantly different from the untransfected cells in peroxisome abundance. The cells expressing PEX11-betamyc, however, showed an increase of 1000% over the PMP34myc and untransfected cells. The difference in peroxisome abundance levels between PMP34myc and PEX1-betamyc is seen by comparing immunofluorescence micrographs of PMP34myc (B and C) with those of PEX11-betamyc (D and E). The bar graph in part A quantifies the results. These results show that overexpression of PEX11-beta increases peroxisome abundance while overexpression of another peroxisome membrane protein has no effect.

Fig. 3. This figure shows the PEX11-beta overexpression increases peroxisome abundance even when peroxisome metabolism (ie the peroxisomal oxidation pathway) is not functional. To counter the other hypothesis of peroxisome division being controlled metabolically, the researchers needed to show that overexpression of PEX11-beta increases peroxisome abundance when the peroxisome metabolic pathway is non-functional. The researchers used a Zellweger syndrome cell line, PBD005, which contains peroxisomes but has a defect in PEX5 that makes fatty acid oxidation and other metabolic processes defective. Once again, the researchers transfected the PBD005 cells with the PEX11-betamyc or PMP34myc expression vectors and measured the peroxisome abundance of untransfected cells as a control. Their results were similar to Figure 2 in that overexpression of PMP34 had no effect on peroxisome abundance. Overexpression of PEX11-beta increased peroxisome abundance by 30 times compared to the control. Comparing the immunofluorescence micrographs of PMP34myc (B and C) with PEX11-betamyc (D and E) shows this difference visually, while the bar graph in part A quantifies the difference. This experiment and figure are huge pieces of evidence in the researchers' case to prove their theory and discredit the competing theory. If PEX11 proteins only increased peroxisome abundance because of their roles in a metabolic pathway, the increase of peroxisome abndance in PBD005 cells overexpressing PEX11-beta would not be seen. Because it is seen, it is evidence for a direct role of PEX11 proteins on peroxisome division and abundance.

Fig. 4. This figure shows that the increase in peroxisome abundance due to PEX11 overexpression in yeast is not affected by lack of fatty acid beta-oxidation, a peroxisomal metabolic process. The researchers moved to yeast for this experiment. They also used a specific type of green fluorescent protein that is imported into the peroxisome lumen. This will be used to measure the peroxisome abundance level instead of the fluorescent antibodies that were used before. Previous studies had shown that peroxisome abundance in yeast is increased when the cells are shifted from a glucose medium to a fatty acid medium, such as oleic acid. This increase is seen by comparing A (glucose medium) with B (oleic acid medium). The researchers then created a strain called XLY1 by deleting the PEX11 gene from the original strain ad adding pPGK1-GFP/PTS1. The XLY1 strains were also injected with a high copy GAL1 promoter-containing plasmid (pRS425/GAL1). They injected derivatives of this plasmid so that the strains would express certain peroxisome proteins--- Ypr128Cp, Pex13p, or Pex11p--- in the presence of galactose. Figure 4c show s the XLY1 strain grown on glucose; figure 4d shows the same strain grown on galactose, which increases peroxisome abundance slightly. Figure 4d was injected with an empty vector and thereforeserves as the control for the follwing modifications. Figures 4e and 4f show strains containing GAL1-PEX13 and GAL1-Ypr128c plasmids. As shown by the graphs, the overexpression of these two genes did not differ significantly from the empty vector. Figure 4g shows the strain containing GAL1-PEX11. This strain had a marked increase in peroxisome abundance as compared to teh empty vector. In fact, this strain showed comparable peroxisome abundance levels to the original strain grown on fatty acid medium (Fig. 4b). Figures 4h and 4i use a different strain, XLY2, which has a defective gene, pox1, that causes a non-functional beta-oxidation pathway for fatty acids. This serves as a semi-control to make sure that beta-oxidation of fatty acids is not occurring in these strains. It shouldn't be because the strains in Fig. 4d-g were grown on galactose media, but this further modification ensures this. Similar results are seen in that PEX13 overexpression has no effect on peroxisome abundance levels while PEX11 overexpression causes a marked increase in peroxisome abundance levels. This figure shows that fatty acid beta-oxidation is not required for PEX11 cause peroxisome abundance level increase in yeast either.

Fig. 5. This figure shows that loss of PEX11-beta in mice reduces the level of peroxisome abundance. The researchers had previously shown that overexpression of PEX11-beta causes an increase in peroxisome abundance levels. They had yet to investigate what the loss of this gene would cause. To examine this, they created knock-out mice that were homozygous mutant for the PEX11-beta gene. They generated fibroblasts from these mice and allowed them to incubate in a serum-free medium for 24 h. This serum-free medium ensured that no lipids or substrates of the fatty acid oxidation pathwa were present. They subjected homozygous wild-tpye PEX11-beta mouse fibroblasts to the same incubation. Peroxisome abundance was measured using immunofluorescence microscopy. The peroxisomes are visible as white spots on a dark background because fluorescent antibodies for PEX14 (B and D) and catalase, a peroxisomal enzme, (C and E) were used. Comparing B and C, the wild-type fibroblasts, with D and E, the mutant fibroblasts, shows the visual difference in the peroxisome abundance level. The mutant cells contain about half of the peroxisomes per slide as the wild type cells. The graph in part A quanifies the results and also includes a comparison of mutant versus wild-type cultured in normal conditions (black bars) instead of in the serum-free condition. The results are almost identical to the serum-free condition. This experiment shows that PEX11-beta's main function is peroxisome division. If PEX11-beta's main function was fatty acid oxidaton, the peroxisome abundance levels would be the same between mutant and wild-type fibroblasts grown in serum-free media (because no fatty acid oxidation could take place). Since the peroxisome abndance levels are higher in wild-type cells than in mutants, PEX11-beta has a direct effect on peroxisome division.

THIS paper does a commendable job in supporting the researchers' hypothesis that PEX11 is a direct promoter of peroxisome division. The paper is extremely well-organized, and the flow of their experiments is in logical order. They start by showing the basic process of peroxisome proliferation, then show the basic effects of overexpression of PEX11-beta, then finish by modifying the conditions of cells with PEX11-beta overexpression. This makes the paper easy to read and distinguishes it from the majority of other scientific papers. The experiments performed supported the authors' arguments, especially figures 3 and 4. These figures showed dramatic proof that PEX-11 overexpression acts independently from peroxisome metabolism/ fatty acid oxidation to increase peroxisome abundance levels. Figure 5 shows this further and also shows the effect of the loss of the PEX11 gene. This figure was a strong closing argument for their hypothesis. one problem I have with their data is in comparing the immunofluorographs with the accompanying bar graphs. Judging from the immunoflurographs, the difference in peroxisome abundance does not apear to be 1000% greater in PEX-11beta overexpressing cells in figure 2, as the bar graph shows. Another problem I have with their data is the large standard error in the PEX11-beta overexpressing cells in figures 2 and 3. The control and PMP34 cells have small standard errors, but the PEX11 cells have a very high standard error (+/- 341 pps and +/- 388 pps). This makes me partially doubt the validity of their results for the PEX11 cells, which are the most important cells for their argument. Beyond these two things, their data is believable, and, overall, I believe their argument and their proposed role of PEX11 as a direct promoter of peroxisome division.

FUTURE EXPERIMENTS. Now that there is strong evidence that PEX11 is a direct promoter of peroxisome division, it would be interesting to investigate how the PEX11 proteins function in this peroxisome division. The first experiment I propose seeks to determine whether PEX11 concentration alone influences peroxisome division or if there is post-translational modification of PEX11 that activates it and causes peroxisome division. To investigate this, I would study mutant PEX11 proteins that did not cause increased peroxisome proliferation. I would run these mutants on a western blot and compare their protein sizes with wild-type PEX11. If the sizes were equal, there would be no evidence for post-translational modification. If the mutants differed in size from the wild-tpe PEX11, there would be evidence for post-translational modifications and those mutants would need to be studied further to determine what the post-translational modification was.

Researchers have recently shown that there is a dynamin-related protein, VPS-1, involved in peroxisome division. It would be intriguing to see whether PEX11 proteins are involved with VPS-1 in any way, or whether they are involved in a separate pathway for peroxisome division. To determine this, I propose an immunoprecipitation experiment. Use antibody beads that bind to the VPS-1 protein. If PEX11 is in complex with VPS-1, you will see a band of the proper molecular weight for PEX11 on the gel when it is run after the beads are removed. IF you see this band, this is evidence that PEX11 may be involved with VPS-1 in peroxisome division. If you do not see the PEX11 band, it does not mean that PEX11 is not involved with VPS-1; the immunoprecipitation experiment just did not support this interaction. The two proteins could be interacting even though they are not in complex with one another. These two experiments will put researchers on the track to discovering how PEX11 directs peroxisome division.




If you have any questions or comments about this web page, please e-mail me, Chris Norbet

Back to Main Page

Link to Molecular Biology Page