Building 1, Floor 3, Room 375 | (816) 926-4326

Research Highlights


How do pores in the nuclear envelope form?

Figure 6-FCCS-finalOne of the outstanding questions in cell biology centers around how fenestrated regions in the nuclear envelope are created. By combining super-resolution imaging with bimolecular fluorescence complementation (BiFC), we show that spindle pole body (SPB) fenestra are composed of two domains that are regulated by Mps3. By mapping physical interactions at high resolution, our data suggests a non-canonical SUN-KASH interaction occurs at SPB fenestra, where the proteins fold back on themselves to form a hairpin connected by both luminal and extra-luminal binding. This unique interaction forms during SPB insertion, raising the possibility that similar SUN-KASH interactions could play a general role in nuclear envelope fenestration.

An atlas of the fission yeast centrosome

Figure 6-FCCS-finalStructured illumination microscopy with single-particle averaging (SPA-SIM) is a general strategy to increase resolution and decrease noise. We applied this methodology to study fission yeast centrosome assembly, building the first comprehensive molecular model of the S. pombe SPB. This approach led to important structural and functional insights not ascertained through investigations of individual subunits, including functional similarities between Ppc89 and the budding yeast SPB scaffold Spc42, distribution of Sad1 to a ring-like structure and multiple modes of Mto1 recruitment.


This paper was included in a special collection of papers on the cell biology of cilia and centrosomes in The Journal of Cell Biology.

New tools to study nuclear envelope composition

Figure 6-FCCS-finalThe nucleus envelope is one of the defining features of eukaryotic cells, yet its composition and regulation are poorly understood. We developed a novel split-GFP system to study the composition of the inner nuclear membrane (INM) and systematically screened all known and predicted integral membrane proteins in yeast for their ability to access the INM.


We used our library to determine the role that protein quality control systems play in INM composition, comparing protein localization in mutants lacking components of ER-associated degradation (ERAD), inner nuclear membrane associated degradation (INMAD) and vacuolar degradation pathways. The INMAD has been proposed to remove mistargeted proteins from the INM through ubiquitin-mediated proteolysis. Our data indicates that the INMAD system not only targets ‘foreign’ proteins from the INM, but it also regulates ‘resident’ INM components. Thus, INMAD plays a role INM homeostasis. INMAD (Asi1)-dependent ubiquitination of the nucleoporin Pom33 controls its distribution in the INM, suggesting that INMAD has roles in nuclear function beyond protein degradation.

SPA-SIM: super-resolution imaging of yeast centrosome duplication

We developed a novel two-color structured illumination microscopy with single-particle averaging (SPA-SIM) approach to study the localization of all eighteen components of the yeast centrosome during duplication using endogenously expressed fluorescent protein derivatives. The increased resolution and quantitative intensity information obtained using this method allowed us to demonstrate that centrosome assembly is coupled with nuclear envelope insertion and it provided clues regarding the spatiotemporal control of centrosome duplication. SPA-SIM is a general strategy to increase resolution and decrease noise and it can be used to investigate other intracellular structures.

Licensing of yeast centrosome duplication by phosphoregulation

AvenaStrikingFigure_Figure2A_300dpiTwo collaborative papers with Mark Winey’s lab at the University of Colorado-Boulder illustrate the importance of phosphorylation in ensuring that spindle pole bodies (SPBs, the yeast equivalent of centrosomes) occur once and only once per cell cycle.


Using a combination of genetics and cytology, including super-resolution imaging and electron microscopy, to reduplicated centrosomes, we show that Cdk1 phosphorylation of Sfi1 blocks yeast centrosome duplication during S and G2/M phase.  And importantly, we show that dephosphorylation of Sfi1 by Cdc14 licenses centrosome duplication for the next cell cycle.

SUN proteins control the distribution of Ndc1

Figure 6-FCCS-finalNdc1 is an integral membrane protein that is localizes to SPBs/centrosomes and nuclear pore complexes (NPCs), making it an ideal candidate to be a shared factor involved in membrane insertion of both complexes.


To test this idea, we created two systems to study Ndc1 binding partners in vivo. Analysis of a series of NDC1 mutants in this system pointed to the existence of at least one additional, essential Ndc1 binding partner, which we showed was the SUN protein Mps3. 


Although Mps3 and Ndc1 are both components of the SPB, our data suggested their binding was SPB-independent.  Moreover, it was enhanced by deletion of nucleoporins, suggesting that the Mps3-bound Ndc1 pool is important for controlling Ndc1 levels at both the NPC and SPB.





SUN proteins are non-essential for mitotic growth

SGA hitsOne of the most surprising results of Witkin et al. 2010 was that deletion of some nucleoporins rendered MPS3 non-essential.  As a structural component of the spindle pole body, it is difficult to envision how the organelle is able to duplicate in its complete absence.


A similar SUN-like domain appears in the membrane protein Slp1, and we find that slp1∆ exacerbates the growth defect of certain mps3 mutants. However, further characterization of Slp1 and its binding partner Emp65 show no role in SPB duplication. Because Mps3 is the sole SUN protein, SUN proteins are not essential for mitotic growth.



Nuclear envelope lipid composition and SPB duplication

Figure for Chen 2014Duplication of the SPB occurs by the formation of a soluble precusor in the cytoplasm, then this ~0.5 GDa protein complex is inserted into the double lipid bilayer of the nuclear membrane in a poorly understood process. In a trilogy of papers, we further invested nuclear membrane insertion of the SPB, showing that it requires the function of Mps3 

and ER membrane bending proteins Rtn1 and Yop1/DP1. In collaboration with Orna Cohen Fix’s lab, we also found that the growth and SPB duplication defect of mps3 mutants was exacerbated by deletion of SPO7, which affects the lipid composition of the nuclear membrane. This work provided the first evidence that lipid composition of the nuclear membrane affects SPB duplication.

Divergent mitotic strategies affect SPB composition and function

NEW-Figure1 mutants

Our unexpected finding that Mps3 and other SPB components are non-essential in certain genetic backgrounds made us reconsider the overall function of SPB proteins and their role in duplication and microtubule nucleation. We wondered if the organelle might be more plastic than originally anticipated and if small changes in the primary sequence of cytoskeletal components could translate into large changes in the organization of the microtubule cytoskeleton, resulting in differences in morphology, nuclear movements, cell division and growth.


To address this issue we turned to Ashyba gossypii, a multinucleated filamentous fungus that is closely related to the budding yeast Saccharomyces cerevisiae. We have extensively characterized its microtubule cytoskeleton by fluorescence microscopy, electron tomography and genetics.


Our results suggest that the multinucleate growth mode of A. gossypii has resulted in different demands on the microtubule cytoskeleton that have driven evolution of SPB components and microtubule regulators to fit its unique life-style.

Mps3 is targeted to the inner nuclear membrane by H2A.Z

Htz1 image 2Localization of SUN proteins to the inner nuclear membrane is essential but how this occurs is poorly understood. In this paper, we show that Mps3 localized to the inner nuclear membrane through a non-canonical pathway that does not appear to require a nuclear localization sequence or other features anticipated to be involved in inner nuclear membrane localization of other proteins including Sun1 and Sun2 in humans/mouse and UNC-84 in C. elegans. Rather, Mps3 is localized via binding to the histone variant H2A.Z in a chromatin-independent manner.





SUN proteins organize chromosomes in the mitotic nucleus

Bupp 2007Mutants in the N-terminal region of Mps3 have no defects in mitotic growth.  However, cells lacking this region of Mps3 displayed altered patterns of telomere distribution within the nucleus and were unable to silence of subtelomeric genes, suggesting that Mps3 functions as a nuclear membrane receptor for telomeres during S phase and for certain types of double-stranded DNA breaks. Mps3 binding to chromosome is partially dependent on its acetylation by Eco1. Together with Ali Shilatifard’s lab we have studied the involvement of histone modifications on telomeric and mating type silencing and cell cycle control.