23 January 2009. Whether you’re a job seeker or a postmitotic cell, it’s not a good idea to burn your bridges. So suggests a new study that describes how accumulating damage to irreplaceable nuclear pore complex (NPC) proteins may contribute to age-related disease. The work appears in today’s issue of Cell.
NPCs are elaborate transport channels that span the nuclear membrane and haul cargo between the nucleus and cytoplasm. Like a slew of other protein complexes in the cell, NPCs do break down and reassemble on occasion—for instance, during mitosis. But what about in neurons and other mature cells that are no longer dividing? These cells treat most protein structures as we would an old car—“you just trash it and buy a new one,” said Martin Hetzer of the Salk Institute for Biological Studies in La Jolla, California, who was lead investigator on the new study. But renewal of NPCs isn’t so simple. Unlike other transport channels that are embedded in single lipid bilayers, NPCs span a double membrane and have indispensable roles in both structure and transport. Plus, at 90 million daltons, they are some of the largest protein complexes in the cell. “The NPC is more like a bridge. You can't simply replace a bridge while traffic is happening on it,” Hetzer told ARF. Likewise, differentiated cells have “no chance to completely take this protein complex apart without affecting the integrity of the nuclear membrane.”
From a series of functional assays in worms and mammalian cells, Hetzer’s team concludes that certain structural components of the NPC do not turn over in postmitotic cells. Instead, those proteins have extremely long half-lives, leaving them susceptible to oxidative damage and age-related wear and tear that compromises nuclear pore integrity and allows cytoplasmic proteins to aggregate in the nucleus. In light of other studies on abnormal protein trafficking across the nuclear membrane, (for review, see Chu et al., 2007), the new data suggests faulty nuclear pores as another potential culprit in neurodegenerative disease.
“I think it's a pretty unique finding,” said George Perry, University of Texas at San Antonio, who was not involved with the new study. “It could open up a whole new avenue to look at a range of diseases—especially age-related disease—because the issue of why cells are different as people get older has never been really well addressed.”
To begin addressing how NPCs are maintained in differentiated cells, first author Maximiliano D’Angelo and colleagues looked at whether a key structural protein in the NPC (the scaffold nucleoporin Nup160) is differentially expressed in dividing versus postmitotic cells. They reasoned that expression of new Nup160, but not of a peripheral nucleoporin (Nup153), would be repressed in differentiated cells if in fact nuclear pores only get assembled during mitosis. Sure enough, this is what they saw. In the nematode worm Caenorhabditis elegans, Nup160 promoter activity—read out as expression of green fluorescent protein (GFP)—was restricted to dividing embryonic cells that coexpressed cyclin B, and did not appear in postmitotic cells of adult worms. This pattern seemed to apply in mammalian cells, too. Using quantitative PCR of muscle cell mRNA, the researchers found high expression of a variety of nucleoporins in dividing C2C12 mouse myoblasts, whereas in terminally differentiated myotubes, expression of only scaffold nucleoporins took a nosedive.
If non-dividing cells repress expression of new scaffold nucleoporins, one would expect that the low levels remaining after the final round of mitosis could sustain the cell for the long haul. To test this idea, D’Angelo and colleagues used RNA interference (RNAi) to knock down expression of various nucleoporins in wild-type worms and daf-2 worms (these have longer life spans due to an insulin/IGF-1 receptor mutation). For most nucleoporins, a 60 to 90 percent reduction in mRNA levels was embryonic lethal, consistent with these proteins’ essential role in pore assembly in dividing cells. By contrast, when the researchers performed RNAi in postmitotic adult worms, only depletion of the dynamic nucleoporins reduced life span, whereas knocking down expression of the scaffold nucleoporins had no effect, even in long-lived worms. These findings suggest that differentiated cells do not need to replenish their NPC scaffold during the worm’s lifetime. “Inside the cell, I think this is one of—if not the most—stable protein complex known so far,” Hetzer said.
As further support for the resilience of scaffold nucleoporins, his team showed that these proteins are not exchanged once incorporated into the nuclear envelope. They mixed wild-type mouse myoblasts with myoblasts expressing GFP-tagged versions of a scaffold nucleoporin (Nup107), a structural transmembrane nucleoporin (Pom121), or a nuclear lamina component (Lamin). All of these proteins readily incorporated into the NPCs and lamina of dividing cells. The researchers then serum-starved the cells to induce differentiation into multinucleated myotubes. After two days, they could not detect any GFP-Nup107 incorporating into non-fluorescent nuclei from wild-type cells, suggesting that the endogenous Nup107 was not getting replaced. Pulse-chase experiments in C2C12 myoblasts confirmed that Nup107 sticks around for a long time. Whereas other proteins—including structural proteins (tubulin and lamin A) and dynamic nucleoporins (Nup62 and Nup153) showed measurable turnover within days, Nup107 hardly degraded over several weeks.
The NPC scaffold seems hardy, but does it remain in tip-top shape into an organism’s golden years? Normally, NPCs only allow molecules 60 kDa or smaller into the nucleus. The researchers put aging nuclear pores to the test by seeing how effectively they excluded larger molecules. Whereas nuclei from young worms successfully blocked entry of 70-kDa fluorescent dextran, nearly a third of old worm nuclei failed to restrict these molecules. Similarly, 70-kDa dextran was largely excluded from nuclei of non-dividing cells from young (three months) but not old (28 months) rat brains, suggesting that NPC function sags with age in mammals as well.
Taking a closer look at NPCs in leaky nuclei from old rats, the researchers found decreased protein levels of certain scaffold nucleoporins (e.g., Nup93) while expression of others (e.g., Nup107) remained unchanged, relative to intact nuclei. Location within the NPC may explain this difference. Nup93’s position near the NPC’s central channel makes it especially vulnerable to oxidative damage from reactive oxygen species, whereas Nup107 appears protected from such harm. Consistent with this idea, the researchers found carbonyl groups (indicators of oxidative protein damage) on Nup93 but not on Nup107. Furthermore, young adult worms treated with an oxidative stress inducer (paraquat) accumulated leaky nuclei more quickly and at higher percentages than did untreated worms.
Whether induced by aging or oxidative stress, one implication of the progressive loss of nuclear integrity would be increased leakage of cytoplasmic proteins into the nucleus of aging cells. This is in fact what the researchers saw. The old rat nuclei that allowed influx of 70-kDa dextran also contained tubulin βIII, a protein normally restricted to the cytoplasm. Intriguingly, the nuclear tubulin often clumped together in filaments that seemed to disrupt chromatin structure.
John Woulfe, a neuropathologist at Ottawa Hospital in Canada, has reported the presence of rod-shaped intranuclear bodies staining with β-tubulin in normal human brain (Woulfe and Munoz, 2000). These structures increase with aging, and seem to disappear in Alzheimer disease (Woulfe et al., 2002). “If the structures that [Hetzer] is showing are the same as our structures, and if indeed these form as a response to nuclear leakiness, then the implications for AD are potentially profound,” Woulfe said in an interview with ARF. “The big chunks of tubulin he's showing in the nuclei—could they be disrupting nuclear bodies and thereby disrupting nuclear function? Could they be mechanically disrupting the structure of the nucleus?”
Nuclear inclusions appear in other neurodegenerative diseases, too. Such structures lie at the root of Huntington disease, for instance, and some studies have suggested that these inclusions cause mayhem in the nucleus by sequestering proteins needed for transcription (see ARF related news story). However, more recent investigations (Arrasate et al., 2004; Mitra et al., 2008) seem to indicate that nuclear bodies in HD are neuroprotective.
Based on his team’s new findings, Hetzer proposes that cells may form these sorts of nuclear aggregates to protect themselves from toxic cytoplasmic proteins that leak into the nucleus through damaged pores. His group is now teaming up with Eliezer Masliah at the University of California, San Diego, to assess NPC function and look for signs of oxidative damage in brain tissue from old and young people, compared with samples from Alzheimer’s and Parkinson’s patients. The newly published data already seem to “point to the deterioration of [NPCs] becoming a very important aspect of these diseases,” Hetzer said.
In AD, where concepts of pathogenesis have focused heavily on events directly linked to β amyloid deposition, “critical events in the cell nucleus have garnered well-deserved attention,” Woulfe noted in a post-interview e-mail to ARF (See full comment below and upcoming Live Discussion). “Histone deacetylases, transcriptional regulation, and DNA repair have taken center stage. The study by Hetzer’s group brings the nuclear pores, the gateway to the nucleus, into the mix.”—Esther Landhuis.
D’Angelo MA, Raices M, Panowski SH, Hetzer MW. Age-Dependent Deterioration of Nuclear Pore Complexes Causes a Loss of Nuclear Integrity in Postmitotic Cells. Cell 2009 January 136:284-295.