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

Review Paper

The immunoglobulin superfamily protein Izumo is required for sperm to fuse with eggs

Naokazu Inoue, Masahito Ikawa, Ayako Isotani & Masaru Okabe

Nature (2005) Vol. 434: 234 - 238

Critique and Summary:

The authors of this paper claim to have found a novel protein that is required in order for sperm to fuse with eggs. They termed this protein Izumo.

Figure 1.

A) This panel compares the sequence similarity between the human and mouse Izumo sequences. The immunoglobulin like sequence is indicated as well as the sequence that indicates the transmembrane region.
B) This panel shows a schematic diagram of the transmembrane portion of Izumo as well as the loop that is formed by a disulfide bridge.
C) Panel c is a Western blot that is probed with an anti-mouse Izumo antibody. The blot shows a light band in the testis lane and a much darker band in the sperm lane. Both of these bands are 56.5kDa in length which corresponds to the molecular weight of the mouse Izumo protein. There is an undetectable amount of protein in the other tissue types. However, because a loading control is not present one cannot make any firm conclusions about the lack of a band in any of the other lanes. It would have been nice to see a control band in all of the lanes to ensure that protein was in fact loaded into that lane. Also, a similar intensity of control band across all lanes would then allow a comparison of the intensities of the bands present in the testis and sperm. The authors report that the same amount of protein was loaded in each lane but without the visual evidence it is hard to believe them.
D) Panel d is also a Western blot. Protein was extracted from human sperm and run on the gel. A human anti-Izumo antibody was used to detect Izumo. There is one band present that corresponds to the correct molecular weight (37.2kDa) of human Izumo. As in panel c there is no loading control present. Also, only one tissue type is shown in this panel. It would be nice to see the experiment performed on the mouse tissue types duplicated among different human tissue types.
E) This panel shows immunostained sperm from mice. The top image is of the sperm under normal light microscopy. The bottom two images are of the same sperm under a fluorescent microscope. The sperm in panel e that are green were from a transgenic mouse line that expressed GFP in the acrosome of the sperm. These sperm did not appear to have detectable amounts of Izumo present while the acrosome-reacted sperm (stained in red) did fluoresce red. Sperm were stained red with a polyclonal antibody to mouse Izumo protein. This panel demonstrates that sperm do not have detectable amounts of Izumo protein until they have undergone the acrosome reaction.
F) Panel f repeats the experiment performed in panel e but with human sperm instead of mouse sperm. However, the acrosome-reacted sperm were stained with anti-CD46 antibody that glows green. The sperm that were not acrosome reacted did not fluoresce. After staining with polyclonal anti-human Izumo antibody, only the sperm that had undergone the acrosome reaction were red. The non-acrosome reacted sperm did not have detectable amounts of Izumo present.

Figure 2.

A) This panel shows the full Izumo allele, targeting vector, and targeted allele. Using restriction enzymes, exons 2-10 of the Izumo allele were replaced with a neomycin-resistance gene. This splicing caused the normally 15kb allele to be shortened to a 6.9kb mutant fragment. This schematic also shows the binding site of the probe used in subsequent blotting assays. The probe binds to the 3' end of both the wt and mutant alleles.
B) This panel is a confirmation that the researchers were in fact successful at creating a mutant allele that was of the correct size (6.9kb). However, as in the previous figure, loading controls were not used. The three lanes represent the different genotypes (+/+, +/-, -/-) present. The first lane (+/+) shows only one band at 15kb. In the heterozygote, a very faint band is present at 15kb with a much stronger band present at 6.9kb. Interestingly, the band present in the homozygous recessive (-/-) lane is of a lesser intensity than that of the heterozygote. This is where a loading control would be helpful. One cannot be sure that the lesser intensity is due to less DNA loaded into the lane or because there is really a less intense band in the -/- lane.
C) A Northern Blot is shown in this panel that compares the total testis RNA of wt, heterozygote and mutant mice. A 1.6kb band is present in both the wt and heterozygote lanes but not in the mutant lane. This time GAPDH was used as a loading control.
D) The researchers then performed a Western blot of sperm protein to show that the Izumo protein was not present in mutant mice. Several other sperm proteins are also shown to indicate that only Izumo is disrupted in the process of replacing exons 2-10 with a neomycin-resistance gene. Again, loading controls are not used in the Izumo blot so the lack of a band in the mutant lane could be due to incorrect loading.

Figure 3.

A) This bar graph shows that under normal mating conditions males that had a homozygous Izumo mutation were unable to produce litters with wt females. Heterozygous males were able to produce litters even when mated with homozygous female mutants. These data suggest that Izumo is necessary in males in order for normal reproduction to occur.
B) Researchers then performed in vitro fertilization of normal female mice with either heterozygous (+/-) male mice or homozygous mutant (-/-) male mice. Of the eggs examined, only those fertilized with sperm from +/- male mice formed pronuclei.
C) This panel demonstrates the fusion capabilities of both +/- and -/- sperm. The +/- sperm were able to fuse and fertilize the egg normally while there were many -/- sperm on the surface of the zona pellucida but none that were able to effectively fertilize the egg.
D) This panel shows that many of the sperm from -/- males were able to accumulate in the perivitelline space (upper panel). The lower panel just shows that the sperm in the perivitelline had undergone the acrosome reaction (red stain). A comparison to a wt or +/- accumulation would have been helpful and would have shed more light on whether this accumulation is normal or a result of the mutation.
E) This bar graph shows the number of sperm that have fused with an egg 2 hours and 6 hours after insemination. Eggs inseminated with +/- sperm have a normal amount of sperm fusing to the egg while none of the -/- sperm fused to the eggs at either time point.
F) This panel is a pictorial representation of the information supplied in panel e. Fused sperm were stained by Hoechst 33342. Only the egg inseminated with +/- sperm show any fusion. The egg inseminated with -/- sperm does not show any sperm fusion. However, it is interesting to note that there is one sperm in the -/- Hoechst stained panel that looks very similar to the +/- sperm that the researchers say have fused to the egg. They do not mention this finding in the figure legend or anywhere else in the paper. I am not sure if it is really a sperm that has fused to the egg or not.

Table 1.

The researchers used a method called ICSI (intracytoplasmic sperm injection) to inject -/- sperm directly into eggs. Because the -/- sperm were unable to fuse to the eggs by themselves it was one way to determine if the -/- sperm were able to functional normally once inside the egg. The numbers presented in the table suggest that the development of eggs injected with -/- sperm was the same as the development of eggs injected with +/- sperm. These data lend further support to the fact that Izumo is responsible for egg fusion.

Figure 4.

A) In this experiment, the zona of hamster eggs was removed in order to allow mouse sperm to fuse with the egg. As can be seen from the staining of the sperm, only +/- mouse sperm was able to fuse to the zona-free hamster egg. Many -/- sperm are present but have not bound to the egg (swelling is not present).
B) This panel shows the effect of anti-Izumo antibody on the fusion of sperm with zona-free hamster eggs. These eggs were either treated with IgG (control) or anti-Izumo antibody. No fusion can be seen in the anti-Izumo antibody treatment but fusion can be seen in the control treatment.

Future Experiments:

Although some controls are missing from several experiments, these experiments do lend support to the fact that Izumo is responsible for sperm-egg fusion. However, more work needs to be done to determine the mechanism by which Izumo regulates this fusion. First one would need to determine if Izumo was the only protein needed in order for fusion to occur. In order to see if other proteins are involved, an immunoprecipitation could be performed. Also, the yeast two-hybrid-system could be implemented to see the more functional side of protein expression. The data presented here suggest that Izumo is the only protein involved in this fusion process so I am not sure if any other proteins would be revealed in the immunoprecipitation or yeast-two-hybrid assay.

Also, the authors state frequently that they believe Izumo to be a member of the immunoglobulin superfamily. Other than sequence similarity data, they have not performed any other confirmatory tests. One experiment they could perform would be protein crystallization. This would allow them to examine the actual 3D structure of the protein to determine if it was in fact an immunoglobulin.


Inoue, N. et. al. The immunoglobulin superfamily protein Izumo is required for sperm to fuse with eggs. Nature 434: 234-238 (2005).

Back to Homepage
Molecular Biology Homepage
Davidson College Department of Biology
Davidson College

Questions or comments? E-mail me

© Copyright 2005 Department of Biology, Davidson College, Davidson, NC 28035