Team:Gothenburg-Sweden/Lab Note/pre


Chalmers University of Technology

Fluorescent Proteins

Bioluminescence is used as a defense, offense and communication mechanism by various organisms in nature including insects, fish, squid, sea cacti, sea pansies, clam, shrimps and jellyfish(1). After the discovery of the Green Fluorescent Protein (GFP) by Shimomura et al. from Aequorea jellyfish(2), an unprecedented tool for visualizing living organisms was also being introduced to molecular biology, leading to a Nobel Prize in 2008(3). The Aequorea GFP is a 238 aa, 30 kDa monomer that emits green fluorescence maximum at 509 nm wavelength(4). It has a unique structure with an 11 stranded β - barrel like shape involving α- helixes running through the center of the can; to which the chromophore is attached(5). Autocatalytic formation of the chromophore of the GFP without the neccessity of any substrates or cofactors is a rather useful property of the GFPs allowing them to express in fusion with different proteins. Remarkably, the fusion of GFP to a protein does not have any vital effect on the activity or mobility of the protein in addition to its nontoxic nature(1).


Figure 1: Structure of the GFP taken from the PDB, code 1EMA



  • (1) Zimmer, M. (2002) “Green Fluorescent Protein (GFP): Applications, Structure and Related Photophysical Behavior.” Chemical Reviews. v. 102 (3) pp. 759-781.
  • (2) Shimomura, O., Johnson, F.H. & Saiga, Y. (1962) “Extraction, purification and properties of aequorin, a bioluminescent protein from the luminous hydromedusan.” Aequorea. J. Cell. Comp. Physiol. v. 59, pp. 223–239.
  • (3) Shimomura, O., Chalfie, M. & Tsien, R. Y. (2008) “The discovery and development of the green fluorescent protein, GFP.” The Nobel Prize in Chemistry. Retrieved from the website
  • (4) Prasher, D., C., Eckenrode, V., K., Ward,W.,W., Prendergast, F., G. & Cormier, M., J. (1992) “Primary structure of the Aequorea victoria green-fluorescent protein” Gene., v. 111, pp. 229-233.
  • (5) Tsien, R., Y. (1998) “The Green Fluorescent Protein” Annual Review of Biochemistry. v. 67, pp. 509–544
Plasmid backbone


SNF1 (modified subunits)



SNF1 belongs to a family of protein kinases which is highly conserved in eukaryotes (Hedbacker 2009). The SNF1 homologue in mammalian cells is called AMP- activated protein kinase (AMPK) and it regulates energy balance in mammals. It is therefore an attractive target for drug discovery against for example diabetes and obesity (Amodeo 2007). SNF1 was identified in Saccaromyces cerevisiae in 1981 when the snf1 mutation was found (Carlson 1981).
The main function of SNF1 is to help yeast cells adapt to glucose limitation and it is required for growth on other carbon sources than glucose such as sucrose (Hedbacker 2009). This is how SNF got its name, since SNF is an abbreviation for sucrose-nonfermenting (Hong 2007). SNF1 is consequently important in response to metabolic stress and it also regulates protective mechanisms against different stresses in response to glucose depletion (Hong 2007). Besides the primary role in nutrient stress, it is also important in environmental stresses including sodium ion stress, heat shock, alkaline pH, oxidative stress and genotoxic stress (Hedbacker 2009).  SNF1 regulates transcription of many genes and controls the activity of metabolic enzymes involved in fatty acid metabolism and carbohydrate storage (Hong 2007). SNF1 also controls nutrient-responsive cellular developmental processes such as meiosis, sporulation and aging (Hedbacker 2009).
SNF1 is activated when phosphorylated by the upstream kinases Sak 1, Tos 3 and Elm 1 (Hedbacker 2009). Each of all three kinases is enough to activate SNF1 in response to stress but Sak1 has the major role (Hong 2007). SNF1 is activated by glucose limitation but the glucose signal regulating activation of SNF1 is not known (Hong 2007). Hong et al. studied mutants lacking all three kinases and their result was that SNF1 still was activated in response to sodium ion and alkaline stress which according to Hong et al. suggests that stress signals might activate SNF1 by a mechanism that is independent of the upstream kinases. SNF1 is inactivated by the Ref1-GLc7 protein phosphatase 1 (Hedbacker 2009). The SNF1 protein kinase pathway is robust and low levels of SNF1 activity is mostly enough for growth under many different stress conditions (Hong 2007). Apart from activation and deactivation, regulation of SNF1 is also achieved through autoinhibition and through control of the subcellular localization (Hedbacker 2009).


Subunit structure and function:


Like its ortholog AMPK, SNF1 is heterotrimeric, which means that it consists of three subunits. The yeast genome encodes the catalytic alpha subunit Snf1, three different beta subunits, Sip1, Sip2 and Gal83, and the gamma subunit Snf4 (Hedbacker 2009). The heterotrimeric interface is located at residue 531-586 on Snf1, residues 375-412 of Sip2 and residues 38-45 of Snf4 (Amodeo, Rudolph, & Tong, 2007).


Snf1- alpha subunit:


This catalytic subunit spans 633 amino acids and contains 3 major features, namely the kinase domain, the activation site and a regulatory domain (Hedbacker 2009). The regulatory domain consists of the AIS-domain (Auto Inhibitory Sequence) which functions by binding to the kinase domain and thereby silencing the catalytic activity. The kinase domain also contains the activation site of the full protein complex, Thr210, which is phosphorylated by upstream kinases when the cells are exposed to external stress (McCartney & Schmidt 2001).


Beta- subunits:


There are three known types of the beta subunit, Sip1, Sip2 and Gal83 (Hedbacker 2009). All three beta subunits in yeast share a conserved GBD (glucogen binding domain) with their common mammalian ortholog AMPK (Wiatrowski et al 2004). The three different subunits differ in their cellular localization when glucose is limited. Sip1 localizes to the vacuolar membrane, Sip2 remains cytoplasmic while Gal83 localizes to the nucleus (Vincent et al 2001). Due to this the subunits contain a few differing elements, for example the Gal83-subunit contains a nuclear localization signal on the c-terminus (Hedbacker and Carlson 2006)


Snf4 – gamma subunit:


The Snf4-subunit consits of 2 bateman domains made up of CBS (cystathionine-beta synthase) repeats that enable them to bind to adenosine derivatives (Hedbacker 2009).  This subunit is also a part of the SNF1 activity regulating machinery through binding to the AIS -region of the Snf1 subunit when the glucose supply is low. This binding inhibits the autoinhibition of the kinase activity performed by the AIS sequence (Celenza and Carlson 1989).


  • Carlson, M., Osmond, B. C., Botstein, D. (1981) Mutants of yeast defective in sucrose utilization. Genetics, 98:25-40.
  • Celenza, J. L., Carson, M. (1989) Mutational Analysis of the Saccharomyces cerevisiae SNF1 Protein
    Kinase and Evidence for Functional Interaction with the SNF4 Protein. Molecular and cellular biology 9: 5034-5044
  • Hedbacker, K., Carlson, M. (2009) SNF1/AMPK pathways in yeast. Frontiers in Bioscience, 13:2408‑2420. 
  • Hedbacker, K., Carlson, M. (2006) Regulation of the Nucleocytoplasmic Distribution of Snf1-Gal83 Protein Kinase. Eukaryotic Cell, 5: 1950-1956. 
  • Amodeo, G. A., Rudolph, M. J., Tong, L. (2007) Crystal structure of the heterotrimer core of Saccaromyces cerevisiae AMPK homologue SNF1. Nature, 449:492-495.
  • Hong, S-P., Carlson, M. (2007) Regulation of Snf1 Protein Kinase in Response to Environmental Stress. The Journal of Biological Chemistry, 282:16838-16845.
  • McCartney, R. R , Schmidt, C. S. (2001) Regulation of Snf1 Kinase Activation requires phosphorylation of threonine 210 by an upstream kinase as well as a distinct step mediated by the Snf4 subunit. The Journal of Biological Chemistry, 276:36460-36466
  • Vincent, O., Townley, R., Kuchin, S. and Carlson, M. (2001) Subcellular localization of the Snf1 kinase is regulated by specific β subunits and a novel glucose signaling mechanism.  Genes and Development 15; 1104-1114
  • Wiatrowski, H. A., van Denderen, B. J. W, Berkey, C. D., Kemp, B. E., Stapleton, D. and Carlson, M. (2004) Mutations in the Gal83 Glycogen-Binding Domain Activate the Snf1/Gal83 Kinase Pathway by a Glycogen-Independent Mechanism. Molecular and Cellular Biology 24; 352-361


FP positions

Fusion protein design through molecular modelling

It was necessary to determine a satisfactory position for the fluorescent proteins in the SNF1 complex. The resulting fusion protein would have to give a measurable change in invivo FRET signal when the protein changes conformation due to activation.

The position was to be determined through static modelling on the SNF1 structure. This involves visualization of the crystal structure provided by Amodeo (Amodeo, G. A. et al. 2007) in RasMol. When positioning the fluorescent proteins in the complex it was impossible to predict if the conformational change due to activation of SNF1 would lead to small or large changes in the distance between the fluorophores. Small changes in distance can be measured only when the distance between the fluorophores is comparable to the Förster distance characteristic of the fluorescent pair. Otherwise the change in signal is small and can be drowned in measurement noise. Thus it was decided to find positions in the complex which would give the distances between the fluorophores close to the Förster distance.

Another consideration was aimed keeping the complex functional and possible to assemble. SNF1 consists of three subunits: subunit Snf1 composed of 633 residues, subunit β and subunit Snf4 composed of 322 residues. There are three β subunits: Sip1, Sip2 and Gal83. Sip1 consists of 863 residues, Sip2 of 415 and Gal83 of 417. Thus there exist three different versions of the SNF1 complex. All three β subunits are localised in the cytoplasm when glucose is abundant. When glucose becomes less abundant, Sip1 relocalizes to the vacuolar membrane, Gal83 relocalizes to the nucleus, and Sip2 remains cytoplasmic (Vincent, O. et al. 2001).

Gal83 is most abundant during growth on glucose, and levels of Sip2 increase during shifts to nonfermentable carbon sources. Sip1 is less abundant than either Gal83 or Sip2, and its level remains constant (Vincent, O. et al. 2001, Ghaemmaghami, S. et al. 2003). Due to some aspects of the phenotype being the result of the expression level and localization of the different β subunits it was decided not to tag Sip1, Sip2 or Gal83. The crystal structure was determined only for a part of a Sip2 containing SNF1 complex: residues 398-633 of the Snf1 subunit, residues 154-415 of the Sip2 subunit and residues 1-322 of the Snf4 subunit. Some residues were not visible in the structure due to them being part of flexible loops (Amodeo, G. A. et al. 2007).

SNF1 contains several highly conserved functional domains. Residues 41-315 in Snf1 belong to the kinase domain (Celenza, J. L. et al. 1986). Gal83 contains a glycogen-binding domain (GBD) at residues 161-243 and Sip2 contains a glycogen domain at residues 163–245 (Wiatrowski, H. A. et al. 2004). Snf4 contains two pairs of cystathionine-beta-synthase (CBS) repeats called Bateman domains which are highly conserved and occupy residues 30-322 (Bateman, A. 1997).Thus it was decided not to insert fluorescent proteins into residues 41-315 of Snf1, residues 161-243 of Gal83, residues 163-245 of Sip2 and residues 30-322 of Snf4 in order to maintain correct function of the Snf4 complex.

Residues 392-518 in subunit Snf1 belong to the conserved autoinhibitory sequence or regulatory sequence. When the complex is inactive the regulatory sequence is docked to the kinase domain. Otherwise the regulatory sequence, more specifically residues 460-495 in the Snf1 subunit, interacts with the Snf4 subunit. A component of this interaction is an anti-parallel β-sheet between residues 467-469 of Snf1 and 270-275 of Snf4, which is a part of CBS4 (Amodeo, G. A. et al. 2007). According to deletion studies residues 392-495 in Snf1 are necessary for the interaction with Snf4 and residues 392-518 for the interaction with the kinase domain (Jiang, R. et al. 1996). Thus it was decided not to insert fluorescent proteins into residues 392-518 of Snf1 in order to maintain correct autoinhibitory function and regulation by Snf4.

The heterotrimer is held together by eight-stranded β-sheet interface, four strands given by Snf1 corresponding to residues 531-586, three strands given by Sip2 corresponding to residues 375-412 and one strand given by Snf4 corresponding to residues 38-45. The hydrophobic core of this interface is enclosed by two Snf1-helices from Snf1 corresponding to residues 515-529 and 612-630. Residues 504-511 of Snf1, which are part of the regulatory sequence, form two small Sip2-strands which interact with the heterotrimer interface further stabilising it (Amodeo, G. A. et al. 2007). According to deletion studies the residues 515-633 of Snf1 are necessary for interaction with Sip2, residues 198-350 of Gal83 are necessary for interaction with Snf1 and residues 154-335 of Sip2 are necessary for interaction with Snf1, where also the glycogen-binding domain is included (Jiang, R. et al. 1997).

Residues 771-863 of Sip1, residues 332–415 of Sip2 and residues 278-417 of Gal83 are necessary for interaction with Snf4 (Yang, X. et al. 1994; Jiang, R. et al. 1997; Schmidt, M. C. et al. 2000). Thus it was decided not to insert fluorescent proteins into residues 504-633 of Snf1, residues 154-415 of Sip2, residues 771-863 of Sip1 and residues 198-417 of Gal83 in order to allow for the assembly of the SNF1 complex.
Having in mind the difficulty of creating internal fusion proteins with molecular biological techniques and predicting if such fusion protein will be correctly folded, assembled and functional, and also considering the conservation and functionality of the different subunits, it was decided that the only reasonable positions for the fluorescent proteins would be the C- and N-termini of Snf1 and Snf4.

Enhanced yellow fluorescent protein and enhanced cyan fluorescent protein were to be used as the FRET fluorophore pair. The Förster distance of this FRET pair is 49 Å. This means that distances between fluorophores shorter than 65 Å would be acceptable, giving an energy transfer efficiency of about 15 %, around which small changes in distance would still be detectable. Both fluorescent proteins are barrel-shaped with a height of 40 Å and a diameter of 20 Å. The actual fluorophore is approximately at the centre of the barrel. This information was taken into consideration when deciding which combination of the positions above would be optimal for FRET measurements.

One possibility was to position one fluorescent protein at N-terminus Snf4 and the other fluorescent protein at C-terminus Snf4. This situation was examined by measuring the distance between residues to which the fluorescent proteins would be linked. The distance between residue 7 and residue 321 in Amodeos structure is approximately 20 Å. As can be seen from Figure 2, due to the bulkiness of the SNF4 complex the fluorescent proteins would be positioned approximately parallel, thus giving a suitable distance of about 50 Å between the fluorophores. Snf4 changes conformation when activated which means a change in distance between the fluorophores could reasonably be expected. Thus this option was accepted.

N gama C gama 1

N gama C gama 2

Figure 2 : Spacefill model depicting the SNF1 complex, with a possible orientation of the two fluorescent proteins fused to N- and C-terminus of Snf4, from two different sides. The Snf1 subunit is depicted in red, Sip2 in blue, Snf4 in green, enhanced cyan fluorescent protein in cyan and enhanced yellow fluorescent protein in yellow. The N-terminal and C-terminal residues on Snf4 are depicted in white. The residues at the ends of the fluorescent proteins by which they are attached to Snf4 are also depicted in white. The residues from which Snf1 and Sip2 extend towards their N-terminal portions, which are not present in the structure, are depicted in yellow.

Another possibility was to position one fluorescent protein on the C-terminus of Snf1 and the other fluorescent protein on the C-terminus of Snf4. The distance between residue 630 in Snf1 and residue 321 in Snf4 in Amodeos structure is approximately 65 Å. As can be seen from Figure 3 , the two fluorescent proteins would be positioned on opposite sides of the SNF4 complex. Having in mind that in this case the fluorescent proteins would probably point away from each other, this would lead to a distance of up to 100 Å or more between the two fluorophores. Thus this option was rejected.

SNF1 complex

Figure 3 : Ribbon model depicting the SNF1 complex. The Snf1 subunit is depicted in red, Sip2 in blue and Snf4 in green. The C-terminal residue on Snf1 and the C-terminal residue on Snf4 are depicted in white. The residues from which Snf1 and Sip2 extend towards their N-terminal portions, which are not present in the structure, are depicted in yellow.

The last possibility would be to position one fluorescent protein on the N-terminus of Snf1 and the other fluorescent protein on N-terminus of Snf4. It was impossible to visualise this arrangement using Amodeos structure because it lacks the N-terminal part of Snf1. Instead this possibility was investigated by using a predicted structure of the SNF4 complex by (Kemp, B. E. et al. 2007). The two positions are in close proximity according to the predicted structure.  Thus this option was accepted.


  • Amodeo, G. A., Rudolph, M. J. and Tong L. (2007) Crystal structure of the heterotrimer core of Saccharomyces cerevisiae AMPK homologue SNF1. Nature, 449, 492-496.
  • Bateman, A. (1997) The structure of a domain common to archaebacteria and the homocystinuria disease protein. Trends in Biochemical Sciences, 22, 12–13. [PubMed: 9020585]
  • Celenza, J. L. and Carlson, M. (1986) A yeast gene that is essential for release from glucose repression encodes a protein kinase. Science, 233, 1175–1180. [PubMed: 3526554]
  • Ghaemmaghami, S. et al. (2003) Global analysis of protein expression in yeast. Nature, 425, 737–741. [PubMed: 14562106]
  • Jiang, R. and Carlson, M. (1996) Glucose regulates protein interactions within the yeast SNF1 protein kinase complex. Genes & Development, 10, 3105–3115. [PubMed: 8985180]
  • Jiang, R. and Carlson, M. (1997) The Snf1 protein kinase and its activating subunit, Snf4, interact with distinct domains of the Sip1/Sip2/Gal83 component in the kinase complex. Molecular and Cellular Biology, 17, 2099-2106. [PubMed: 9121458]
  • Schmidt, M. C. and McCartney, R. R. (2000) Beta-subunits of Snf1 kinase are required for kinase function and substrate definition. European Molecular Biology Organization Journal, 19, 4936–4943. [PubMed: 10990457]
  • Kemp, B. E., Oakhill, J. S. and Scott, J. W. (2007) AMPK Structure and Regulation from Three Angles. Structure, 15, 1161-1163.
  • Vincent, O. et al. (2001) Subcellular localization of the Snf1 kinase is regulated by specific β subunits and a novel glucose signalling mechanism. Genes & Development,15, 1104–1114. [PubMed: 11331606]
  • Yang, X., Jiang, R. and Carlson, M. (1994) A family of proteins containing a conserved domain that mediates interaction with the yeast SNF1 protein kinase complex. European Molecular Biology Organization Journal, 13, 5878–5886. [PubMed: 7813428]
  • Wiatrowski, H. A. et al. (2004) Mutations in the Gal83 glycogen-binding domain activate the Snf1/Gal83 kinase pathway by a glycogenindependent mechanism. Molecular and Cellular Biology, 1, 352-361.


Primer design

To see our primer design document, please follow this link here: Primer design document

Home      |      About us       |      Chalmers      |      Sponsors      |      Contact Us
Design in Chalmers, Gothenburg, Sweden