Biochem Blogs

Biochemistry blog, science writing

Cooking Up Membrane Proteins with BAM!

Graduate student Sophia Yang

Graduate student Sophia Yang

Gatzeva-topalova, P. Z., Warner, L. R., Pardi, A. & Sousa, M. C. Structure and Flexibility of the Complete Periplasmic Domain of BamA : The Protein Insertion Machine of the Outer Membrane. Struct. Des. 18, 1492–1501 (2010).

The outer membrane of gram-negative bacteria are characterized as one that has outer membrane proteins (OMPs). The correct folding and insertion of these OMPs allow for the beta-barrel structure to be embedded into the outer membrane; however, the mechanism mediating this process is still unclear.


Figure 1: Outer Membrane Protein Biogenesis Overview- The outer membrane precursors is originally synthesized in the cytosol before being processed and translocated across the inner membrane and into the periplasm. Chaperone proteins will then bind to and shuttle the protein to the BAM complex where it will be directionally inserted into the outer membrane.

There are several key players that are involved in OMP biosynthesis. The OMP precursor is synthesized in the cytosol and is tagged with a signal sequence. This for it to be post-translationally targeted to the SecYEG translocase. Following translocation into the periplasm and cleavage of the signal sequence, chaperone proteins will bind to the nascent OMP and shuttle them to the BAM complex. This multiprotein complex allows for directional insertion of the OMPs into the outer membrane.

The Beta-barrel Assembly Machine (BAM) is anchored by the beta-barrel BamA plus four lipoproteins (Bam B/C/D/E). BamA is found in all gram-negative bacteria and consists of a C-terminal beta barrel and five N-terminal polypeptide transport associated (PORTA) repeats. The structure of the first four PORTA’s have been reported, but PORTA five has not.

The goal of this paper is to determine the structure of PORTA 5 with respect with the other PORTA domains of BamA. The authors solved a crystal structure of PORTA 4-5 from E. coli BamA at 2.7 Å. PORTA5 displays the characteristic PORTA structure of two alpha helices packaged against a mixed three-strand beta sheet. The interface of PORTA4 and 5 is bridged by a 3 amino acid linker (G344 N345 R346) which forms an L-shaped conformation in the crystal between the two domains (Fig. 2A). There are several interactions that are conserved within the interface of the domains (Fig. 2B).


Figure 2- E. coli BamA PORTA4-5 crystal structure. A) L-shaped conformation in the crystal. Superposition of the two molecules in the asymmetric unit (Chain A and B). PORTA4 chain A is in salmon; PORTA5 chain A is green; grey is chain B. B) Interface of PORTA 4 and 5. Guanidinium group of R314 makes a salt bridge with D383 and a hydrogen bond with S379. Hydrogen bonds between main chain atoms are shown with red dotted lines. Interactions of R346 in the linker stabilize interface by HB to Y315 in 4 and forming a cation pi- interaction with W376 in 5. All conserved interactions in chain B except for R314.

The authors validated their crystal structure through solution NMR and SAXS experiments as independent methods for determining the orientation of the two PORTA domains. By looking at a pool of 100 structures with randomized orientations – varying the torsion angles between the three amino acid linker- the lowest energy structures were found to be in agreement with the solved crystal structure.

The crystal structure of BamA POTRA1-4 has been solved in two conformations. The extended and bent models differ in bending at linker between POTRA 2 and 3. When POTRA4 from BamA POTRA4-5 was superimposed with the same domain in POTRA1-4, this made a spliced model of the entire periplasmic domain of BamA. Subsequent analysis of the structure suggested two rigid arms- PORTA 1-2 and PORTA3-4 with a hinge point between 2 and 3. This was also validated by SAXS.

Effects of CURT1 on Grana Stacking

Graduate student Eric Waddell

Graduate student Eric Waddell

Plants, cyanobacteria, and green algae rely on photosynthesis for energy and sugar production. Chloroplast, the organelle where photosynthesis occurs, is specifically organized in higher plantae to allow for efficient photosynthesis. This structure consists of an outer and inner envelope, stroma, thylakoid membrane, and lumen. The Thylakoid membrane is further organized into grana, round disk shaped membrane structures that stack 5-20 disks high. Inside the grana and thylakoid membrane of a chloroplast, is a continuous lumen. The grana and thylakoid membranes is connected by stroma lamellae. The grana and thylakoid membrane structure is important for separation of photosynthetic proteins. Photosystem I (PSI) and Cytochrome b6/f (Cyt b6/f) complex are located along the stacked grana membranes, while Photosystem II (PSII) and ATP synthase (ATPase) are localized to the stroma-facing thylakoid membranes. The structure of the the stacked grana and physical separation of the photosynthesis complexes allows for a higher degree of control in the photosynthetic process, for example separation of PSI and PSII allows for the plant to control transfer excess excitation energy.

CURT1b, as identified in this paper, has previously been described as TMP14, thylakoid membrane phosphoprotein 14kD, and was thought to be a subunit of PSI. Previous research also found homologous genes across many plant, cyanobacteria, and green algae species. This new class of genes was named PSI-P, photosystem I phosphoproteins. Armbruster et al discovered a family of genes in Arabidopsis thaliana using co-expression of the known genes involved in photosynthesis. The unknown genes pulled from the co-expression data where then grouped hierarchically to determine their association with known photosynthetic genes. CURT1a, b, and c all grouped with other photosynthetic genes involved in PSI. However, the fourth gene (CURT1d) did not cluster with photosynthesis related genes. The predicted structure of the CURT1 gene family was determined by the sequence of the 4 homologs. They consist of an N-terminal amphipathic helix, two transmembrane alpha-helices, and a c-terminal coiled coil helix. The four helix domains are relatively conserved across different species, with the exception of the amphipathic helix. Intriguingly, the 4 domains of the CURT1 gene family are also found to be incorporated into aminoacyl-tRNA synthetases, aaRSs, in certain cyanobacteria.

Localization of the CURT1 proteins, as examined by staining of cell cultures, showed localization of CURT1s to the chloroplasts. Isolations of thylakoid, inner and outer envelope, and stroma from chloroplasts were run on gel blots and antibodies for three control proteins with known localizations, and antibodies to curt1a b and c where shown; all three localize to the thylakoid membrane. Endogenous curt1d expression is too low for western blot assays from tissue isolates, so they looked at a HA tagged over expressing line for CURT1d and found that it to is located in the thylakoid membrane. Membrane association of the CURT1s was studied using chaotropic salts to disassociate membrane bound proteins. Curt1b and CURT1d behaved similarly to PetC, a control with 1 transmembrane domain, while CURT1a and c behaved similarly to Lhcb1 and Cyt b6, which have 3 and 4 transmembrane domains.

Using a fusion of maltose binding protein for extraction, the researchers studied the endogenous protein concentrations of CURT1s. They found CURT1a to be the most abundant, and CURT1d least abundant. From previously generated insertion mutants (knockouts), they generated double, triple, and quadruple knockouts of the CURT1 genes. Interestingly, the knockout of CURT1a had a negative effect on the presence of CURT1b and CURT1c. The double mutants had a superadditive effect on the remaining CURT1 protein. And the triple knockouts showed no presence of CURT1 protein. Seven CURT1 dimer complexes where discovered using native blue page assay and sds page to separate out the interacting subunits. Homo and hetero dimers/trimers of CURT1a and b were found using a western blot assay with antibodies for each.

Photosynthetic efficiency is reduced in knockout mutants, with quantum yield and non-photochemical quenching reduced in CURT1a knockouts and drastically reduced in double, triple, and quadruple knockouts. Initial fluorescence from the PSI is increased in the initial time after light exposure from the resting state and WT can better incorporate light energy 100-200ms after exposure.  Compared to WT, the knockout mutants are impaired in cyclic electron transport.

There is a striking physiological change in knockout and overexpressing mutants. In knockout mutants we see a reduction of grana stack formation, which increases in severity with the inclusion of multiple CURT1 knockouts. Alternatively, when we overexpress CURT1 we see increased numbers of grana stacks. In triple and quadruple knockouts we see the introduction of unique vacuole-like structures between the layers of the thylakoid membrane. SEM, of chloroplasts with the outer protective membrane stripped away, showed a top-down version of the physiological changes seen previously. It also, provided a better representation of the elongated thylakoid membrane in the quadruple CURT1 knockout.

SEM and gold immunoblotting (above) for CURT1a was used to detect the areas of the thylakoid membrane that contained CURT1 proteins. The results show that the margins, or periphery, of the grana stacks contained the highest proportion of CURT1 protein. This suggested that CURT1s were responsible for the induction of thylakoid membrane curvature. To verify that the CURT1 proteins could cause curvature in a membrane, in vitro studies were conducted. The results showed that there was tubule formation in artificial lipid membranes. It also showed oligomerization of the CURT1 proteins.

The researchers provided a hypothetical model for how CURT1 proteins lead to grana stack formation in vivo. They suggest that the process is regulated by phosphorylation and subsequent transcriptional control of tap38/pph1. Phosphorylation, increases grana stacking, whereas dephosphorylation reduces grana stack formation.

Eric blog1


Armbruster, U., Labs, M., Pribil, M., Viola, S., Xu, W., Scharfenberg, M., … Leister, D. (2013). Arabidopsis CURVATURE THYLAKOID1 Proteins Modify Thylakoid Architecture by Inducing Membrane Curvature. The Plant Cell, 25(7), 2661–2678.

Influenza Virus Matrix Protein Is the Major Driving Force in Virus Budding

Graduate student Sayan Chakraborty

Graduate student Sayan Chakraborty


Centro Nacional de Biologı´a Fundamental and Centro Nacional de Microbiologia, Instituto de Salud Carlos III, Majadahonda 28220, Madrid, Spain


Influenza virus that infects humans, birds and other animals falls under the category of Orthomyxoviridae. These are enveloped viruses differing in sizes and shapes. It can be spherical or filamentous in shape. The envelope is made up of lipid bilayer, derived from the host cell in which the virus replicates, assemble the viral proteins. The viral genome consists of 8 single stranded short segmented RNA molecules. It translates into 8 different viral proteins. Nucleocapsid protein covers the RNA and polymerase required for viral replication in host cells. Polymerase has 3 components- PA, PB1, PB2. RNA and the polymerase components together form RNP.  There are three transmembrane proteins- Neuraminidase(NA), Hemaglutinin(HA) and M2.  Matrix protein (M1) and NS2 are the structural components of the virus particle. M1 has the capability to bind with lipids and interacts strongly with plasma membrane. This paper focuses on the assembly of viral proteins, major driving force behind virus particle formation and tries to find the mechanism behind Influenza A virus morphogenesis.

To solve the mechanism of virus protein assembly and to characterize the virus formation of virus like particles, they infected COS-1 cells with vTF7-3 and transfected with DNA mixtures containing the nine plasmids coding for all structural proteins  or mixtures that lacked one of the plasmids or that lacked the plasmids encoding the three polymerase subunits. To determine the presence of VLPs in the transfected cultures, the supernatants of the COS-1 cells clarified by 33% sucrose cushion centrifugation. The pellet and supernatant was analyzed for the presence of the NP, M1, HA, and M2 proteins by immunoblotting, which shows that M1 and NA are essential for biochemical detection of VLPs whereas removal of other proteins does not significantly affect the production of VLPs. Later, EM studies were conducted to check the size and morphology of these VLPs. The particles were identical in shape and size with the true virion. VLPs were observed in cultures without RNA, NA, MS2, HA, NP but were not observed in cultures not expressing M1 protein. This result inferred that all viral structural proteins, except M1, can be removed individually without compromising the formation and budding of VLPs which morphologically resemble wild-type virion. They also proved that RNPs are not required for viral assembly of the VLPs. Lipid floatation assay confirmed the interaction between the plasma membrane and the M1 protein.


Figure 1: Visualization of VLPs in infected COS-1 cells

In this paper it was shown that M1 protein, in the absence of other viral polypeptides, can assemble into virus-like budding particles which are released into the culture medium. Coexpression of the HA glycoprotein modulates the self-association and membrane binding properties of the M1 polypeptide. This paper clearly demonstrated the key role played by the structural protein M1 in the formation, assembly and budding process of influenza virus particles.

Plants, 3D printing and a recycling bin

Graduate student William (Brad) O’Dell

Primary Source:  Lei, L., et al., CELLULOSE SYNTHASE INTERACTIVE1 Is Required for Fast Recycling of Cellulose Synthase Complexes to the Plasma Membrane in Arabidopsis. Plant Cell, 2015. 27(10): p. 2926-2940.

Cellulose is a vital component of plant cell walls that confers physical strength and biological resilience to the plant.  Within plant tissue cellulose occurs in both amorphous and highly crystalline forms that are both produced by the processive addition of UDP–glucose molecules to long homo-polymeric chains.  Polymerization is catalyzed by large multi-enzyme cellulose synthase complexes (CSCs) that span the plasma membrane.  Intracellular domains catalyze the addition of each UDP–glucose unit which forces the growing cellulose chain through a membrane channel formed by interactions among transmembrane domains and into the extracellular space where crystallization and/or interactions with other cell wall polymers (hemicellulose, lignin, etc.) can occur.  Cellulose ‘extrusion’ is believed to drive the observed translational mobility of actively polymerizing CSCs within the plane of the cell membrane.  In addition, this translational motion is guided by linking CSCs to the apical microtubules lying just below the plasma membrane with an adaptor protein known as CELLULOSE SYNTHASE INTERACTIVE1 (CSI1).  The microtubule-to-CSC coupling ensures that the synthesized cellulose is drawn into elongated chains that are favorable for forming cell wall structures.  This microtubule–CSI1–CSC system is functionally analogous to contemporary 3D printing in which a head (CSCs) extrude a structural polymeric material (cellulose) guided by a track system (CSI1 and apical microtubules).  Plants were 3D printing before it was cool!  The figure below shows schematically the components of cellulose biosynthesis.


Figure Source:  Bashline, L., S.D. Li, and Y. Gu, The trafficking of the cellulose synthase complex in higher plants. Annals of Botany, 2014. 114(6): p. 1059-1067.  ©The Authors.  Reprinted with permission.

While cellulose synthesis is an elegant biological process, the “on/off” switch for polymerization is a bit, well, clunky.  CSCs synthesize cellulose whenever they are located at the plasma membrane and require translocation from the membrane to stop the synthesis.  Often when transmembrane proteins are removed from the membrane the protein is targeted to a degradation or turnover process, and new protein chains must be synthesized to replace the degraded protein.  In the case of the CSC, turnover of the complex would be an expensive way to achieve control of cellulose synthesis and create lag in the cell’s ability to resume cellulose synthesis under favorable conditions.  Lei et al. have recently shown that the connection between CSCs and apical microtubules formed by CSI1 plays an important role in a membrane translocation process by which functional CSCs can be removed from the membrane but not targeted for turnover.  CSCs that are removed from the plasma membrane form protein–lipid vesicles known as small microtubule-associated cellulose synthase complexes (SMaCCs).  By examining CSI1 and cellulose synthase cellular localization in Arabidopsis thaliana plants expressing different CSI1 domain deletion mutants, Lei et al. revealed that CSI is vital to the formation of SMaCCs and that SMaCCs localize with apical microtubules just beneath the plasma membrane by a CSI1-dependent interaction. Furthermore, the authors demonstrated by watching the recovery of CSCs in the plasma membrane that the ability to form SMaCCs speeds the recovery of cellulose synthesis after release from polymerization-inhibiting stress.  SMaCCs enabled by CSI1 act like little recycling bins that instead of allowing CSCs to go to the “shredder” of the turnover machinery keep the CSCs safe and poised to be returned to use for cellulose synthesis at the plasma membrane.  Plants were also recycling before it was cool!


Perhaps nature should inspire us with respect to 3D printing and to recycling and sustainability?

Interaction between the Linker, Pre-S1, and TRP Domains Determines Folding, Assembly, and Trafficking of TRPV Channels

Graduate student Dmitry Grinevich

Primary Source: Valverde et al. “Interaction between the Linker, Pre-S1, and TRP Domains Determines Folding, Assembly, and Trafficking of TRPV Channels.” Cell Structure. August 4, 2015.

The thermo-TRP (transient receptor potential) ion channel family is an extremely interesting group of proteins to me because of their involvement in temperature sensing in a variety of different organisms. I think this is an exciting area to study because temperature is such a crucial part of every aspect of life, all the way down to the smallest molecular mechanisms. Temperature is clearly extremely important to understand because it controls every single chemical reaction that organisms depend on to function properly. The reason these TRP channels are so exciting is because they are the mechanism which many mammals use in sensing a range of temperatures. These channels are a family categorized by a few recurring features: N-terminal Ankyrin repeats, six transmembrane helices with a pore loop, and a C-terminal TRP domain.  Additionally, they are all membrane channels which open and close to allow the flow of calcium ions into cells which then leads to downstream responses to specific temperatures. There exist a variety of slightly different TRP channels which together are able to cover the full range of temperatures an organism needs to deal with, going as low as -20 degrees Celsius all the way up to over 50 degrees Celsius. Additionally, many of these channels are able to be activated by agonists which interact with them. For example, TRPV1, which is a heat sensing channel, can also be activated by both capsaicin (a vanilloid from hot peppers) and ethanol, both of which are chemicals that we can familiarly associate from food or drink with a feeling of heat. Because these channels are the molecular mechanism that our bodies use in order to sense temperature, I think it is exceedingly important for us to understand their structure and function in our cells. There is a great interest in these channels as a target for pain therapies and developing new analgesics. Some of the heat channels are known to be involved with pain responses which makes them potential candidates for targets for new pain relieving substances.

Valverde et al. explored a conserved amino acid motif common to all TRPV channels with the belief that it had a critical function. Site directed mutagenesis to disrupt this motif revealed that it is crucial to the cellular localization of TRPV4. On top of not properly localizing to membranes, the mutant TRPV4 channels also were unresponsive to activators, including heat, capsaicin, and exposure to a hypotonic solution. They were also able to replicate the same effects when this mutation is carried out in other TRPV channels. Next, they used confocal microscopy in order to study the subcellular localization of the mutant channels. What they found is that TRPV4 wild type colocalizes with cell membrane marker concanavalin-A, while the TRPV4 mutant colocalizes with the endoplasmic reticulum (ER) marker calreticulin. Next, they pinpointed more specific amino acid locations in various parts of the TRP channel domains to figure out which domain interactions are crucial for proper channel assembly. They were able to confirm through various site directed mutants that D425, E745, and K462 are all important resideus for interactions which lead to correct channel trafficking to the membrane. In the absence of any of these residues, channels are most likely not able to tetramerize properly and then cannot be exported from the ER to the plasma membrane to carry out normal functions. Finally, some molecular modeling allowed the authors to support their findings about the importance of these three residues.


Role of Phosducin Like Protein 1 in G protein assembly in retinal rod photoreceptors

Graduate student Dipti Paudel

Lai, Chun Wan J., et al., Phosducin-like protein 1 is essential for G protein assembly and signaling in retinal rod photoreceptors. The Journal of neuroscience. May 1, 2013. 33(18):7941-7951.

Rod and cone cells are photoreceptors present in the retina of the eye. Cone cells primarily function in photopic (well-lit condition) vision whereas rods function in scotopic (intermediate light conditions) vision. Rods are more sensitive than cones, requiring only one photon for activation of intracellular signaling. G protein is a heterotrimeric protein associated with seven transmembrane G protein coupled receptors (GPCRs) and is signaling protein in rod and cone cells. When a photon binds to the GPCR, exchange of GDP to GTP activates Gα subunit separating it from the Gβy obligate dimer. The dimer is then able to perform downstream signaling through second messenger system. The original assembly of the Gβy dimer requires at least two co-chaperones: Chaperonin-Containing T complex polypeptide 1 (CCT1) and Phosducin Like Protein 1 (PhLP1). To test the role of PhLP1 in dimer formation, in vivo knockouts of PhLP1 from mice retinal rod photoreceptors were introduced and monitored for G protein expression.

First, knockout mice were created using Lox-Cre recombinase system. After two generations, mice with deleted PhLP1 (homozygous knockouts) were formed along with heterozygous and wild type mice.  Using one month old mice, immunohistochemistry was used to confirm the deletion of PhLP1 in rod cell outer segments. Using immunoblot of retinal extracts cre mediated knockouts were determined to have lost 80% of PhLP1 gene expression after 35 days. Second, the researchers looked at the expression of various G protein subunits in the absence of PhLP1. Using immunohistochemistry on retinal tissue, a disappearance of Gαt1, 1, Gγ1, and RGS9 was observed. Since Gβ4 and Gb5 are usually present in the inner nuclear layer and not the outer rod segments, their expression remained unchanged. Rhodopsin levels were also unchanged, indicating that only the G proteins were affected. To further confirm the expressions, extracts from whole retinal and retinal outer segments were immunoblotted for G proteins as well as varied cone and rod proteins. The levels of most cone and rod proteins remained unchanged but Phosducin (Pdc), a protein used for stabilization of the Gβγ complex, contained decreased levels in PhLP1 knockouts. A co-immunoprecipitation study of Pdc indicated no dimer formation in the homozygous knockout. To test the visual acuity and sensitivity of mice in the different study groups, the optomeric head turning response of wildtype and homozygous knockout mice to scotopic and photopic conditions were measured. The results indicated low response for homozygous knockout mice in scotopic conditions and an unchanged response in photopic conditions further confirming that only rod cells were affected.

In conclusion, this study showed that deletion of the co-chaperon PhLP1 decreased the expression of G proteins in the outer rod segment of the retina and inhibited assembly of the Gβγ complex.


Figure 1: The proposed mechanism for G protein heterotrimer assembly and the effects of PhLP1deletion on the assembly (shown in red).


Heat shock and light in circadian regulated plant growth

Graduate student Eric Waddell

Graduate student Jiaqi Duan

Usually, moderately warm constant ambient temperatures tend to oppose light signals in the control of plant growth. However, in this study, Karayekov and co-workers showed that brief heat shocks enhanced the inhibition of hypocotyl growth induced by light in deetiolation Arabidopsis thaliana seedlings, and that the light was perceived by phytochrome B, a circadian clock input component of Arabidopsis. In the experiments with just the wild-type seedlings, the authors discovered a synergism between high temperature and light signals in hypocotyl responsiveness. In order to further study this synergistic mechanism, the authors tested heat shock and light combination conditions in many PhyA and PhyB single or double mutant because PhyA and PhyB are red light perceiving components in the clock network. They have found that any mutants of PhyB showed absence of the synergy between heat shock and light. They then test this synergy condition with other major clock mutants that were either the main clock genes or genes in PhyB singling pathway. As for the subsequent experiments, the authors used red light instead. The experimental transgenic plants were either single or double mutants of cca1 and/or lhy, single or triple mutants of prr3/5/7/9, hy5, cop1, Pif3/4/5, elf3/4, gi and toc1. Among all the mutants and wild-type plants, they have observed that the hypocotyl length was synergistically inhibited by the combination of heat shock and red light in wild-type, but this synergy was absent in phyB and phyAphyB mutants and reduced in some mutants like pif3/4/5, elf3/4 , cop1, hy5, gi, toc1, prr7/9 and cca1 lhy. They then used luciferase reporter assay and found out that heat shock could generate a rhythm of hypocotyl growth sensitive to red light. They then started to test the affect of heat shock and red light on all the mutants that had either absence or reduced synergy between heat shock and light. They found out through the experiments that heat shock reduced the PhyB nuclear body formation and nuclear abundance of COP1 under red light, but enhanced the stability of HY5. They also found out that heat shock treatments also generated circadian rhythms of CCA1 and LHY gene expression in darkness and that temperature gating of the hypocotyl growth response to red light required CCA1, LHY, PRR7 and PRR9, and that the plants had impaired PIF4 and PIF5 expression in cca1 lhy double mutant. Since all these circadian clock genes were connected in light perceiving signaling pathway, the authors have proposed a mechanism in which how plants response to heat shock and light signals according to the figure shown above. With the presence of heat shock, there are two signaling pathways that converge to enhance the sensitivity of hypocotyl growth to light perceived by PhyB.  First, the heat decreases abundance of COP1 and causes an increase in HY5, which is a positive regulator of photomorphogenesis. The second pathway is that heat shock induces transient oscillation in the PRR7 and PRR9 expression, which leads to the oscillation of CCA1 and LHY expressions, and this causes an oscillation of PIF4 and PIF5 expressions, which are the negative regulator of photomorphogenesis. The authors from this paper also indicated that the convergence between heat shock and light signaling could affect a wide range of processes as revealed by RNA-seq analysis. As a result, the high temperature signals enhance the sensitivity to light through PhyB mediated signaling pathway in preparing the seedlings for deetiolation upon light exposure.


Reference: Karayekov, E., Sellaro, R., Legris, M., Yanovsky, M., & Casal, J. (2013). Heat Shock-Induced Fluctuations in Clock and Light Signaling Enhance Phytochrome B-Mediated Arabidopsis Deetiolation. The Plant Cell, 2892-2906.

Proteolysis Keeps the Plant Clock Ticking

Graduate student Jigar DesaiIn general, biology involves a lot of cyclical process. The water cycle, nitrogen cycle, carbon cycles, cell cycles, photosynthesis/respiration, Glycolysis, Calvin cycle; these are all cycles that that need to be renewed for life to function. The circadian clock is no different. For an organism to keep pace with 24hr cycles, it must be able to adapted to the changing time of day and be able to reset the cycle for the next day.

The Circadian Clock is a robust mechanism that helps regulate all daily activities and is mainly composed of negative feedback loops. In plants, there is morning loop, an evening loop, and a core loop. Genes in the loop are translated into proteins, which then go on to repress transcription of other genes in the opposite phase. The proteins that repressor gene transcription must degrade before the cycle can continue, making protein degradation an important factor in the renewal of the clock cycle.

Fujiwara et al. investigated the degradation of the Pseudo-response Regulator Protein (PRRs) class with their relationship with ZTL (ZEITLUPE) and time of day. The PRR proteins are known components of the core clock. If a particular PRR is knocked out, there is usually an increase or decrease in the period depending on the time of day the PRR is expressed. ZTL was found to be interesting because in the ztl knock out mutant some PRR proteins become stable (they do not degrade at the time of day they should). The researchers hypothesized that ZTL was responsible for the degradation for some of the evening PRR proteins.   ZTL is also interesting because it is a U3 ligase. Another motivating component of ZTL is that it also contains a blue light receptor. So the investigators checked the rate at which PRR proteins are degraded under red, blue, and no light. They saw that under blue light the PRR proteins remained stable and under red or no light they degraded. This is then a mechanism that protects PRR proteins degradation at the wrong time of day. Next they looked for an interaction between the PRRs and ZTL. They first used a pull down assay and checked for an interaction between all PRR proteins and ZTL. They saw interaction with TOC1 (PRR1) and PRR5, but not PRR3. Something they noticed was that PRR3 phosphorylation was different throughout the day. They then checked whether PRR3 phosphorylation affects it binding of TOC1 to ZTL. What they saw was that phosphorylated PRR3 would out compete ZTL binding to TOC1; however, unphosphorylated PRR3 could not out compete ZTL binding to PRR1. With this information they gathered they proposed the model in figure 1.


Figure 1.

Citation: Fujiwara, S., L. Wang, L. Han, S.-S. Suh, P. A. Salome, C. R. Mcclung, and D. E. Somers. “Post-translational Regulation of the Arabidopsis Circadian Clock through Selective Proteolysis and Phosphorylation of Pseudo-response Regulator Proteins.” Journal of Biological Chemistry 283.34 (2008)

Switching Fold States to Switch the Circadian Rhythm in Cyanobacteria

Graduate student Ryan Schuchman

Graduate student Ryan Schuchman

Circadian rhythms were once thought to be characteristic only to creatures that didn’t divide multiple times in a 24-hour period. These circadian rhythms synchronize many physiological processes to the constant day/night cycle of the Earth. It wasn’t until the discovery of a species of nitrogen-fixing cyanobacteria that it became clear that the circadian clock could be applied to any organism that has any sort of dependence on sunlight. Because oxygen is a potent inhibitor of nitrogenase, the cyanobacteria needed to devise an elaborate way of conducting both photosynthesis and nitrogen fixation. The solution was the evolutionary offset of activities of these metabolic processes by 180 degrees—or shifting the expression of nitrogen-fixation genes to be transcribed during the night. In cyanobacteria, the KaiABC proteins are responsible for this circadian regulation. In this pathway, KaiC is the main scaffold to which the other proteins of this pathway will bind. The expression of the ensemble of genes that are controlled by KaiC are dependent on the phosphorylation state of KaiC. Autophosphorylation of KaiC is promoted by KaiA binding, while autodephosphorylation of KaiC is favored by KaiB binding. This competitive binding also alters the binding of several other accessory proteins whose functions are activated upon interaction with KaiC. Two of these accessory proteins are SasA and CikA and they are responsible for the phosphorylation and dephosphorylation of RpaA, respectively; where RpaA is the master regulator of transcription. All of these details are summarized in the image below. In the pathway, it can be seen that all of these changes in phosphorylation state and binding interactions are depended on the fold state of KaiB. In the ground state of KaiB (gsKaiB), a non-functional homo-tetramer is formed and is the majorly favored state of KaiB. Every once in a while, KaiB will undergo a transition from its ground state fold to a thioredoxin-like fold in the fold-shifted state. Though the concentration of fold-shifted KaiB (fsKaiB) is very low, its Kd for phosphorylated KaiC is strong enough that it is able to displace the binding of KaiA and SasA. This brings up a term, in a biochemical context, that many of us probably haven’t used since General Chemistry 2: Le Chatelier’s Principle. The sequestration of fsKaiB to KaiC is what drives the transition from gsKaiB to fsKaiB. Not only does the binding of KaiB to KaiC promote autodephosphorylation of KaiC, it also recruits CikA and activates it to inhibit the phosphorylation of RpaA. The cycle then resets and this very interesting biochemical mechanism continues its function in the cell.


Figure 1: Overview of the circadian rhythem in cyanbacteria [1]

[1] Chang et al. (2015) A protein fold switch joins the circadian oscillator to clock output in cyanobacteria. Science 349: 324-328

Crystal Structure of the Heterodimeric CLOCK:BMAL1 Transcriptional Activator Complex

Graduate student Sayan Chakraborty

This paper [1] from the Takahashi lab at the University of Texas Southwestern Medical Center discusses about the structure of a transcriptional activator complex CLOCK:BMAL1 that plays a pivotal role in mammalian circadian rhythm. This paper highlights the basic structural features of BMAL1 and CLOCK and how these features are facilitating the formation of the transcriptional complex.  BMAL1 and CLOCK both the proteins are made up of three domains- bHLH, PAS A and PASB (TAD domain is not considered in the paper).The bHLH (recognizes the E-BOX DNA), PAS A, PAS B domain organizations and interactions are vividly described. Although PAS domain is widely conserved but it can adapt or change its structural conformation based on the ligand or the interacting partner. Takahashi group also performed functional studies, mutational analysis of the heterodimer.  Binding affinity of the heterodimer with E BOX part of mper/mcry DNA was calculated to be Kd~10 nM by Fluorescence Anisotropy. This paper explains the regulation of the transcriptional feedback loop and how mutations in conserved amino acid residues of the heterodimer can alter the circadian rhythm in mammals. It is known that the PER:CRY repressor complex binds with BMAL1:CLOCK heterodimer and inhibits the transcription of mper/mcry.  Based on this knowledge, the authors have provided valid justifications regarding the positioning of the binding of the repressor complex to the BMAL1:CLOCK heterodimer. Overall, this is a pioneer paper in circadian biology with detailed structural and functional explanations.


Figure1. 3D structure of CLOCK:BMAL1 reveals a tightly intertwined heterodimer with all three domains—the N-terminal bHLH domain and two tandem PAS domains (PAS-A and PAS-B)—involved in dimerization interactions.

[1] Huang et al., SCIENCE VOL 337 13 JULY 2012 (DOI: 10.1126/science.1222804)