Functional analysis of ER and Golgi subdomains

Lead Research Organisation: University of Leeds


Understanding how proteins are sorted to the right place in a living cell is comparable to the task of the Royal Mail to first sort letters and packages by destination and secondly to deliver them correctly at minimal cost and in a reasonable timeframe. The Golgi apparatus is a sorting station of the secretory pathway from which proteins can be sorted in at least three directions, to the cell surface (secreted proteins), to the vacuole or the endoplasmic reticulum (ER). The ER can be compared to the manufacturing site for a range of products that will be shipped first to the Golgi for subsequent sorting and delivery. To stay in business, manufacturing sites require efficient transport routes to distribution centres. This also means that vehicles should not travel when they are not properly loaded, and they should be doing something useful when they return, such as bringing some mail back. This is not an easy task, and the same is true for the sorting of proteins in living cells. However, cells are very efficient at organising cost-effective supply chains and have developed very efficient transport processes in which no step is perfect but little is left to chance due to efficient back-up mechanisms. This includes the recycling of essential machinery so that they can be engaged in multiple transport reactions during their life-span.

Numerous protein sorting signals (address labels) have been described in the last 20 years and in many cases receptor molecules (lorry drivers) have been identified that bind to the sorting signals and package them into transport vesicles (the lorries). It is however much less clear how the sorting receptors know how to find their way in the cell. In other words how exactly do the lorry drivers reach their right destination, who gives them their instructions and who sends them back for new jobs when they have delivered their cargo? In the cell, sorting receptors must not only bind to ligands in one compartment, they must also transport them to another compartment, release them there, and return back to the original compartment to select new proteins again.

We would now like to use our recent discoveries to shed light on the complete process of ER retention in plants. Many proteins exported from the ER are meant to be returned, just like the wooden pallets are not discarded in modern logistic firms but they are in fact recycled so that they can be used again to be packed with new cargo. One problem is that there are many more proteins to be recycled than receptors available. Receptors have to bind their cargo very efficiently in the Golgi and preferentially release them in the ER in a place where they will stay for a longer time, whilst the receptors move back to the Golgi using a frequent flyer ticket. This is why we find it exciting that ER import sites and ER export sites are physically separated from each other and we can now start to explore how this is achieved.

We have also discovered that plants contain two types of Golgi bodies. This is very exciting, because it suggests that instead of one basic distribution centre the plant uses at least two. It was always believed that the Golgi apparatus is a single organelle, so our finding is of major importance. The new Golgi type can be seen with fluorescently tagged receptors (lorry drivers) which represent a new class found in plants, algae and some unicellular organisms but is absent in animals and yeasts. We must first learn how this new class of receptors reaches the new class of Golgi bodies, and next we would like to see what happens if we eliminate this class of receptors and see why plants use these receptors.

Finally, we need to identify more proteins residing in the new compartments that we have discovered to gain a better understanding of protein sorting to the plant vacuoles, which host the vast majority of edible protein on earth. Plant proteins constitute a major food source and must be harnessed as well as we can.

Technical Summary

Transport of proteins between organelles of the plant secretory pathway is controlled by complex cascades of cyclic reactions involving conditional conformational changes and interactions between proteins and other bio-molecules. Using innovative in vivo protein transport assays my team has discovered two novel compartments of the secretory pathway: 1) The plant ER import site (ERIS) was found to be physically separated from ER export sites and associated Golgi bodies (daSilva et al., 2004, Plant Cell 16, 1753-1771) and may help to explain how cargo and receptors segregate in the ER. 2) A specialized subpopulation of Golgi bodies was discovered (see case for support) that specifically contains high levels of a membrane spanning protein that resembles the HDEL-receptor ERD2 but is differentiated by an additional transmembrane domain at the N-terminus (Hadlington and Denecke, Curr Opin Plant Biol 3, 461-468). These two new compartments enrich the roadmap of the known transport steps and their further analysis is essential to understand the true complexity of the pathway in plants.

This project will deliver specific knowledge on the sorting motifs and transport route required to mediate segregation of transport machinery to ERIS domains of the ER, their biogenesis and the transport cycle of ERD2. ERIS function is one of the last missing links in the retrograde transport route of the pathway and we will identify new proteins that take part in the process of COPI vesicle fusion with the ER. In addition, we will complete the knowledge base on anterograde ERD2 transport and unravel the biological meaning of the plant-unique ERD2-related protein family that specifically labels only a subset of Golgi bodies. These observations stand out internationally, and will explore completely new avenues of fundamental research and plant biotechnology allowing us to ultimately design artificial protein sorting receptors and storage organelles for biotechnology.

Planned Impact

Plant Biotechnology is in urgent need for new breakthroughs which demonstrate measurable consumer benefits in the use of plants to manufacture molecules of added value. The public is still extremely sceptical regarding any form of research leading to genetic manipulation of plants, and it is therefore of paramount importance to discover new strategies to harness the potential of plants to produce real consumer values. Constructive engineering to obtain gain of function in the secretory pathway is crucially dependent on a complete understanding of all the organelles and sub-organelles that exist in the pathway, as well as the signals/mechanisms that allow receptors to complete a full transport cycle. This project will yield a critical mass of data to design of custom made receptor molecules that can boost ER retention of plant cells without affecting the overall yield. It will pave the way towards next generation pathway engineering strategies and the beginning of true synthetic biology for bio-manufacturing procedures involving the plant secretory pathway as a green factory.

Prof. Denecke (Leeds) has extensive experience with IP and filing patents, supplemented by significant experience and passion for process technology and process innovation. The University of Leeds is priviledged to maintain knowledge/technology transfer offices to maximise the potential for exploitable IP and to facilitate industrial links.

When plant molecular biology was at its infancy J. Denecke became a pioneer in plant cell transfection and transient expression systems. He developed the first quantitative secretion assays for plant cells, introduced a variety of cargo molecules for the analysis of protein sorting signals, including transport time-courses, dose response analysis and multiple cargo assays. Leading scientists worldwide have adopted the system and the laboratory hosts regular visitors worldwide to acquire the necessary skills for their own research goals. J. Denecke was the first to demonstrate with proper proteins that secretion to the cell surface is the default pathway and that soluble proteins can be secreted without carrying active sorting signals. He spearheaded new experimental strategies to co-express cargo molecules of the secretory pathway with effector molecules to specifically interfere with house-keeping transport reactions. The main principle is to monitor early responses of a complex system by rapid transient expression systems in which one parameter is the sole variable. This allows researchers to identify primary responses and not long-term indirect effects. Since 2003, access to state of the art confocal laser scanning microscopes at Leeds has introduced live fluorescence microscopy as a key-technique which has profoundly transformed the research. Results have shed light on fundamental biological questions and were published in high impact journals at regular intervals, including 20 publications in The Plant Cell, the leading journal in the field, and three filed patents on the use of pathway engineering in biotechnology.

Whilst at York and later in Leeds, his research group has organised international training workshops on transient expression technology. Researchers leaving the Denecke lab have often published well above average and the lab has a strong track-record in launching young scientists to become research leaders themselves as over 60% of the PhD students and postdoctoral researchers from Denecke's team now work in permanent academic environments running their own research teams, including a number of assistant professors (Luis daSilva, Brazil; Peter Pimpl, Germany; Nathalie Leborgne-Castel, France). Other researchers work for the government (Belinda Phillipson, CSL York) or industry (Phil Taylor, Monsanto; Jane Hadlington, Qiagen).


10 25 50
Description The project has exceeded the expected research outcomes on the ERD2 gene family by a very large margin and met the research outcomes regarding the ERP gene family.
1) It was firmly established that ERPs do not play a complementary role to ERD2 in ER retention.
2) Subcellular localisation of ERPs was restricted mainly to the endoplasmic reticulum and specific subdomains that host Golgi membrane proteins.
3) ERPs appear to slow down both ER export and Golgi-mediated recycling, possibly due to their interaction with ERD2
4) ERPs have 8 well-defined transmembrane domains and have both N-terminus and C-terminus exposed in the cytosol
5) ERD2 by contrast has a lumenal N-terminus and a cytosolic C-terminus.
6) We have established the first robust quantitative receptor activity assay for ERD2 based on a gain-of-function cargo sorting essay in living plant cells.
7) The activity assay was used to establish that one ERD2 molecule can prevent the secretion of at least 200 HDEL- or KDEL- proteins, demonstrating an extraordinary in vivo efficiency.
8) We documented a previously unrecognised role of the cytosolic carboxy-terminus in the biological activity of ERD2.
9) The activity was also used to carry out an extensive mutagenesis analysis which revealed novel ERD2 mutants that mediate dominant secretion of HDEL proteins, a crucial finding that permits deeper understanding of receptor recycling between the ER and the Golgi apparatus.
10) A functional cross-kingdom comparison of ERD2 from plants, green algae, fungi, yeasts, protists, and animals revealed only very few non-functional variants (Saccharomyces cerevisiae, Plasmodium falciparum) which are currently been used to map and identify further functional domains of ERD2. The results clearly demonstrate that ERD2 is an ancient developmental trait of eukaryotic cells and extremely conserved.
11) We identified a novel antagonism between ERD2 and the vacuolar sorting machinery, leading to induced secretion of vacuolar proteins when ERD2 is overexpressed.
12) We established a crucial role of two conserved Leucine residues in the cytosolic C-terminus of ERD2 which is necessary for Golgi residency and its function in mediating ER retention of HDEL proteins in the plant ER.
13) Most importantly we have succeeded in generating a biologically functional fluorescent ERD2 fusion which is Golgi-resident and does not show evidence for recycling to the ER. Previously used fluorescent fusions or epitope fusions of ERD2 were shown to be present in both the Golgi and the ER but we show that these are biologically inactive due to masking of the conserved Leucine residues in the C-terminus.
14) Further work in 2018 established that even small extensions to the ERD2 C-terminus (i.e. using a c-myc tag) inhibit ERD2 function thus providing an explanation for the original observation that ERD2 re-distributes to the ER upon ligand-binding.
15) The ERD2 C-terminus was established to be a Golgi retention signal, rather than an ER export signal, and crucially is needed to prevent redistribution back to the ER. This is confirmed by demonstrating that non-functional ERD2 fusions with a masked or otherwise compromised C-terminus all re-distribute to the ER whilst functional ERD2 fusions with an unobstructed C-terminus do not redistribute.
16) Experimental evidence was obtained to show that imposing COPI-mediated retrieval signal on ERD2 renders the molecule non-functional, further refuting the original recycling model. Points 14-16 are currently been prepared for publication in high ranking journal, together with point 7 of the list of achievements that is unpublished so-far.
17) Publication from a competing group (Brauer et al., 2019 Science 363, 1103-1107) describing the crystal structure of chicken ERD2 and claiming mutually exclusive COPI and COPII binding signals modulated by ligand-binding, but ignoring our published findings and continuing localisation studies with a C-terminal fluorescent protein fusions (ERD2-GFP) which is biologically inactive, required further data from our team. Publication of points 14-16 was delayed and further research was carried out with the following key-findings:
17.1 Since Chicken ERD2 and human ERD2 are almost identical in protein sequence, and human ERD2 is functional in our in vivo cargo transport assay in plant protoplasts, we first tested C-terminal fluorescent tagging and show that this inactivates human ERD2 and results in a dual ER-Golgi localisation of the inactive fusion protein.
17.2 We then constructed a biologically active fluorescent fusion of human ERD2 by inserting an additional transmembrane domain at the N-terminus and show that it remains biologically active and is exclusively Golgi localised.
17.3 We then tested the role of lysine residues in the C-terminus of human ERD2. Despite claims by Brauer et al., 2019, none of the lysines were required for the biological function of human ERD2 and their replacement by alanine did not change the Golgi residency of ERD2.
17.4 We then showed that the conserved leucine residues at position -5 and -3 from the C-terminus are equally important for human ERD2 as they are for plant ERD2, mutation of which abolishes biological activity and causes partial ER -Golgi localisation, similar to that observed by C-terminally tagged ERD2.
17.5 We noticed that the final 5 aminoacids harbouring the conserved di-leucine motif for Golgi retention are not included in the crystal structure published by Brauer et al., 2019 Science 363, 1103-1107 and explored key-result 16 further by mutating the COPI signal. As a result, the ERD2 fusion regains biological activity. These results firmly establish that COPI-mediated recycling is not involved in ERD2 function.

Interim conclusion: Our additional work in 2019 firmly establishes that our results on permanent Golgi residency of ERD2 are not plant-specific but conserved across eukaryotes as diverse as plants and human, and further research will now require a completely new approach to explain the gate-keeper function of ERD2. This will require dedicated funding, for which a very convincing case has now been made, combining key results 1-17, in particular results obtained in 2019 with human ERD2 (17.1 to 17.5)

Progress in 2020:

18) To establish that Golgi-retention per-se is required for ERD2 function, we set out to test if an alternative Golgi retention signal can re-actvate ERD2 when the C-terminus is masked by a large fluorescent protein such as YFP. This experiment was severely interrupted due to lock-downs in 2020, but we succeeded in obtaining further key-results which really should settle any further doubt regarding our new working model:
18.1: Supplementing an inactive ERD2-YFP fusion with a recently discovered N-terminal Golgi retention signal (LPYS) and a TM domain to ensure cytosolic display (LPYS-TM-ERD2-YFP) restores biological activity in a quantitative HDEL-cargo cell retention assay, leading to a clear dose-response.
18.2: LPYS-TM-ERD2-YFP is exclusively Golgi localised, as opposed to ERD2-YFP or TM-ERD2-YFP which are also found in the ER. The combined results provide positive evidence for the essential role of Golgi-retention in ERD2-function, establishing a causal link between activity and Golgi-residency.

Overall conclusion: The work carried out over the years after completing the research grant systematically pointed in the same direction and fully justify our proposal that ERD2 acts as a gate-keeper in the Golgi, rather than a taxi-driver shuttling between Golgi and ER membranes. Due to the central role the ER-Golgi interface plays in the secretory pathway, this is not a further detail, but a fundamental discovery furthing our understanding of the secretory pathway and opening many new avenues of future research,
Exploitation Route The results are fundamentally extremely exciting because they shed light on a role of the ERP gene family in modulating ERD2 activity. The ERP gene family is found in plants, green algae, and the group of protists collectively referred to as alveolates, including Plasmodium, Paracemcium, diatoms, watermolds).
The finding that demonstrates how very efficient the ERD2 receptor cycle operates between the ER and the Golgi shows the critical importance of this step in the functioning of the secretory pathway. Researchers may be inspired by our quantitative in vivo assays and upgrade to similar methodology for the analysis of their own sorting receptors.
The results are equally exciting from an applied perspective as they can be directly used to engineer transgenic plants that can store more recombinant proteins in the endoplasmic reticulum. First experiments with crop plants are extremely promising and will impact on food security, energy security medical security. The analysis from the functional cross-kingdom comparison of ERD2 refutes recent data on the role of serine phosphorylation in ERD2 function and provided examples of ERD2 forms that are more amenable to up-regulating the ER retention capacity of plants because they do not exhibit the antagonistic effect with vacuolar sorting. Storing high value proteins in the plant ER will have a tremendous impact on affordable healthcare as well as food security on Earth.
Key-results 7 and 14-16 will form the basis for a further high impact publication and a renewed application for BBSRC-funded research projects to understand the proposed gate-keeper function of ERD2. This is a fundamental breakthrough in understanding one of the first and most conserved step in the secretory pathway of eukaryotes that has remained elusive to date and indeed fundamentally mis-interpreted by the field for the last 30 years.
Key-results 17.1-17.5 establish the need to carry out work in other than plant systems and may shed light on the origins of the ER and the Golgi in eukaryotes.
Key results 18.1 and 18.2 completely refute the old recycling model, and fully justify the gate-keeper model. The impact of this fundamental discovery should make it into Cell Biology textbooks as a landmark discovery.
Sectors Agriculture, Food and Drink,Education,Energy,Environment,Healthcare,Manufacturing, including Industrial Biotechology,Pharmaceuticals and Medical Biotechnology

Description The newly gained understanding about functioning and efficiency of plant ERD2 genes has allowed us to successfully engineer plants with drastically upregulated levels of recombinant HDEL or KDEL proteins. This result is unexpected, because receptor-levels were deemed critical and overexpression of receptors was reported to lead to lethal Brefeldin-A like effects. In our experiments, we could not observe such negative effects, but could demonstrate 1) an extra-stoichiometric recycling capacity of plant HDEL receptors and 2) the ability of plants to accommodate drastically increased receptor levels without developmental growth phenotypes. In the last year of the award we have succeeded in identifying ERD2 variants from other organisms that can up-regulate the ER retention capacity of plants even more than their endogenous ERD2. The findings have far-reaching consequences as they pave the way towards engineering starch crops to store edible proteins in their endoplasmic reticulum. This is currently the most viable strategy to increase protein security on Earth, to replace wasteful unsustainable arable land allocation for animal farming. Further data have been obtained by comparing the human ERD2 protein with the plant ERD2 protein. We could show that serine phosphorylation is not involved in human ERD2 signalling and that human ERD2 function relies on the same two conserved leucine residues that we have identified in the plant protein. Heterologous expression of human ERD2 in plants to facilitate human protein production for medical purposes in plants is currently being tested, In 2018, a key construct produced during the award was successfully used to localise ERD2 exclusively in the Golgi apparatus. In a land-mark publication, we challenge the established view that ERD2 recycles together with its ligands back to the ER and propose a new model that accomodates the extra-stoichiometric recycling capacity of plant HDEL receptors observed during the award period. We show that C-terminal tagging of ERD2, systematically used in the field by all research attempts to localise ERD2 in vivo in plants and other eukaryotic models, renders ERD2 non-functional and refutes the recycling model. A new biologically active fluorescent ERD2 fusion developed during the award is exclusively localised to the Golgi apparatus (Silva-Alvim et al., 2018) and supports a Gate-keeper function for ERD2, rather than a recycling receptor. The result is a fundamental breakthrough in our understanding of ERD2 in plants and other eukaryotes. The findings are currently being used to carry out research into refining the gate-keeper model and to develop crops with enhanced protein content in non-storage tissues. Research in 2019 shows that our data are not plant-specific but generally valid in eukaryotes, greatly enhancing the case for follow-up funding of this work. Research in 2020 completes the final test to establish a causal link between Golgi-retention and ERD2 function. Furthermore, our screen of ERD2 homologs from various eukaryotic kingdoms has established that tardigrade ERD2 has a higher activity in plants than endogenous plant ERD2, and that this can be explained by the ability to boost ER retetion of soluble cargo without toxic side-effects. The aim to use enhanced ER retention capacity in starch crops to boost protein content has therefore become much more feasible in 2020 and could really contribte to achieving food security and quality in the future. Research in 2021 demonstrated that Golgi-retention per-se is a key feature of ERD2 function, illustrated by the fact that an alternative Golgi-retention signal can re-activated ERD2-YFP despite the masking of its own Golgi-retention signal at the ERD2-C-terminus. We also managed to demonstrate that replacing the ERD2 C-terminus with the c-terminus of a p24 protein (containing established and tested COPI and COPII signals) inactivates ERD2, but if the COPI signal is mutated, the presence of a COPII signal compensates for the lack of the Golgi-retention signal. Whilst this shows again that ERD2 does not normally require fast ER export, the newly engineered ERD2 construct with just the COPII signal can enhance the retention of K/HDEL proteins in plants without negative effects on plant viability. Therefore, the construct can be overexpressed to higher levels than wild type ERD2, which enabled us to drastically boost recombinant protein retention in vegetative tissues. We are currently exploring this for the sustainable upregulation of edible proteins (as well as industrial enzymes and pharmaceutical proteins) in potato tubers, in order to contribute to 1) food quality and security, and enable 2) a sustainable approach to generate more food and more bio-energy (instead of food versus fuel). The results may lead to evidence to help public acceptance of GM technology in plants, as they offer real consumer benefits.
First Year Of Impact 2021
Sector Agriculture, Food and Drink,Chemicals,Energy,Environment,Manufacturing, including Industrial Biotechology,Pharmaceuticals and Medical Biotechnology
Impact Types Cultural,Societal,Economic,Policy & public services

Title Multiple gene knock down by hybrid antisense transcript 
Description Using the model species Nicotiana benthamiana, a transient expression assay was developed to introduce a hybrid anti-sense transcript covering a gene family for simultaneous knock-down from a single gene construct. As a feasibility study, the ERD2 gene family was targeted to test functional complementation by Arabidopsys thaliana ERD2 and a biologically active fluorescent derivative (YFP-TM-ERD2, see Silva-Alvim et al., 2018). Due to different codon usage between Arabidopsis thaliana and Nicotiana benthamiana ERD2, the complementing transcripts are resistant to the effect of the introduced antisense transcripts. 
Type Of Material Technology assay or reagent 
Year Produced 2018 
Provided To Others? Yes  
Impact Using this new technique, genetic complementation assays can be done within 48 hours rather than lengthy plant transformations during several months. It thus represents an enormous step forward for phenotypes that can be scored with cell populations biochemically, rather than whole plant tissues morphologically. The method is fast and quantitative and can be used by many peers in the field with an interest in plant biochemistry, cell biology and molecular biology.. 
Title Triple expression vector for transient expression analysis 
Description We have established plasmid vectors for transient expression which contain 1) a common cytosolic enzyme to normalise cell transfection rates, 2) a cargo molecule to monitor the functioning of the secretory pathway (either secretion, vacuolar sorting or ER retention) and 3) a third gene that can encode variants of a receptor, or any other protein that controls transport through the secretory pathway. 
Type Of Material Technology assay or reagent 
Year Produced 2017 
Provided To Others? Yes  
Impact Gain-of-function assays (or the use of dominant mutants) is often unpopular with geneticists because the effects are dose-dependent and not clear cut black versus white. However, with adequate quantification, dose-response assays can be a key to obtain far more information than from gene knockouts as they report on the action of the protein, rather than documenting the consequence of its absence. The use of in-built reference markers for normalisation of transfection, and the implementation of a constant cargo level has helped us enormously to compare receptors from different organisms and with different mutations. It will help others to do more complex gain-of-function assays too, whether it has to do with designer drugs, medicines or industrial enzymes.