Discovery of a New Associate to Human OXR1 Protein

By DALIA, Kielce, Poland
Supervision: Dr Paul Hoff Backe & Prof Magnar Bjoras
Finalist of the Intel International Science and Engineering Fair (Intel ISEF); Issuer: Society for Science and the Public, May 2015
Finalist in The E(x)plory Scientific Competition 2015; Issuer: Fundacja Zaawansowanych Technologii, March 2015
The Talent of Swietokrzyskie, Marshal Office of the Swietokrzyskie Voivodeship, October 2014
Let's Talk about [X] Multidisciplinary Student Research Conference - University of Glasgow, February 2015

CONTACT AUTHOR

CONDENSED VERSION:

I have conducted my experiments during a two-month internship in Rikshospitalet University Hospital in Oslo, Norway. I worked there on a project characterizing protein-protein interactions with the human oxidation resistance gene 1 (OXR1). The objectives of my work were: to undertake the purification process for hOXR1A, hOXR1T1, CNN2 and TOMM20 proteins; to investigate the possible interaction between hOXR1 protein isoforms and the candidate TOMM20 and CNN2 proteins; and to form a theoretical background for my results and investigate the possible causes of obtaining such results.
The first step of my experiments was growing a bacteria culture in order to express the desired proteins. I took competent bacteria BL21-RIPL and transformed them (by heat shock) with appropriate plasmid containing human cDNA gene for the protein I wanted to obtain and additional His-tag, made of six histidins.
 
Table 1 shows plasmids used for particular proteins.
 
Transformed bacteria were grown overnight in an incubator (37 degrees) on kanamycin LB plates. After the overnight growth, I selected a single colony and grew it in overnight liquid culture in 37 degrees, at 180 rpm shaking. The next step was growing big expression culture in 6 flasks, each of capacity 1 l. I added 10 ml of overnight liquid culture to each 1l flask and started growing bacteria at 37 degrees, 200 rpm shaking. In the meantime I measured Optical Density (OD, absorbance) which indicates the density of bacteria culture, on a spectrophotometer. I grew the culture until the OD reached about 0,5 and then I changed the temperature to 18 degrees and waited until the OD reached 0,6-0,8 for inducing the expression of protein with IPTG (Isopropyl β-D-1-thiogalactopyranoside). This component binds to the lac repressor and releases the tetrameric repressor from the lac operator in an allosteric manner, thereby allowing the transcription of genes in the lac operon. The plasmid construction includes the cDNA gene of the desirable protein, therefore the IPTG presence causes the production of the protein. The flasks with IPTG remained shaking at 18 degrees overnight.

After the overnight shaking, I poured the content of the flasks into 1l centrifuge bottles and spun them on 5500 rpm at 18 degrees for 30 minutes in order to obtain cell pellet. I homogenized the cell pellet in a binding buffer (300 mM NaCl, 50 mM Tris) and then the bacteria cells were lysed by sonication. This process uses sound, vibrations and heat to destroy the bacterial cells and release the protein into the solution. I repeated the sequence of sonication three times for 30 seconds with a minute pause with incubation on ice before each repetition. After this experiment I spun the solution on 15 000 rpm at 4 degrees (to protect the protein from denaturation) for 30 minutes. I collected the supernatant containing the protein and disposed of the bacteria residues.

In order to obtain pure protein without any additional proteins from the bacteria I used nickel affinity chromatography. I used a slurry containing immobilized nickel with ability to bind the 3D structure of the His-tag. Only the human proteins present in plasmids contained a His-tag, other bacteria proteins did not have one. I mixed the supernatant from spinning with immobilized nickel and left it on shaking for binding for 0,5 hours. Then I poured the solution in the column and collected the flowthrough, in case the protein did not bind to the column. I washed the column with 100 ml of binding buffer (300 mM NaCl, 50 mM Tris) and added 20ml of elution buffer (300 mM NaCl, 50 mM Tris, 300mM Imidazole). The structure of imidazole is similar to the structure of the His-tag. It enables imidazole to replace the His-tag and in this way the protein is eluted from the column. I collected the eluted protein in 1ml fractions and run them on the SDS gel, including supernatant from spinning and flowthrough from the column. Sodium dodecyl sulfare (SDS) polyacrylamide gel electrophoresis (PAGE) is  a technique used to separate proteins according to their electrophoretic mobility. A protein, denatured by sodium dodecyl sulfare, is mixed with marker, e.g. Seeblue Plus 2, and loaded into wells in polyacrymide gel. Then the voltage is applied. The protein migrates in the porous gel and reaches the point of its size, not able to be pushed further inside pores of decreasing diameter. The size can be defined by the use of protein standard, performing a colorful ladder at the end of electrophoresis. Then the gel is stained and the stain reveals the location of the protein, to which it binds.

For further purification of hOXR1 A, protein size exclusion chromatography was used, which was Superdex 75 column. Size exclusion chromatography, such as Superdex75 column, uses porous gels in which smaller molecules have to traverse a longer way inside the column than the big molecules, as the smaller ones are caught inside the pores.  Using the ÄKTA FPLC Purifier 10 pump I purified hOXR1 A. Protein-rich fractions, judged by absorption at 280 nm, were applied to a 12 % SDS-PAGE gel, using 5 µl Seeblue Plus 2 as marker. Coomassie blue was used as gel staining dye.

For further purification of hOXR1 T1, protein ion exchange chromatography was used, specifically  the Resource Q anion exchange column. This method uses the charge of the proteins to cause their binding to the oppositely charged column, letting all the remaining proteins out. I applied the nickel affinity chromatography flowthrough fraction of hOXR1T1, as it was rich in the protein, on the ÄKTA FPLC Purifier 10 pump. Protein-rich fractions, judged by absorption at 280 nm, were applied to a 12 % SDS-PAGE gel, using 5 µl Seeblue Plus 2 as marker. Coomassie blue was used as gel staining dye.

In order to run the interaction tests Dynabeads® (Invitrogen) were used as a binding material. Dynabeads® magnetic separation technology utilizes the gentle affinity interactions between bead-bound ligands and their specific targets. During these tests one protein containing His-tag (bead-bound ligand) is mixed with magnetic beads, to which it binds. Then, after a short period of incubation, the candidate protein (specific target) without His-tag is added and the sample is incubated again. After this procedure the proteins are eluted. Two bands of appropriate size present at the SDS-PAGE gel indicate the interaction between those two proteins. Although the second protein does not have the His-tag necessary for the column binding it may be present in the eluent, which means that the interaction occurs.

CNN2 candidate protein was first cleaved from the His-tag; TEV proteis enzyme as well as β-mercaptoethanol were added to CNN2 fractions, and they were left over two nights at 4 degrees for incubation. 12 % SDS-PAGE gel confirmed the cleavage. Two fractions were then applied on Superdex 75 column according to the procedure described for hOXR1A protein. The combined elution fractions of CNN2 protein were concentrated using Amicon Ultra-4 Centrifugal Filter Device with a 10K cut-off filter.
In order to carry the interaction tests Dynabeads ® magnetic separation beads were divided into 5 tubes.
 
Table 2 presents the content of each interaction test tube.
 
The component 1 was added first to each tube; then the tubes were incubated for 10 minutes in 4°C. The component 2 was then added and incubation in the same conditions was repeated. The tubes were then placed on a magnet and the supernatant was discarded. Eluted fractions were then run on 12 % SDS-PAGE gel.

In order to confirm the results of the interaction test the Western blot test was conducted. This method is used to detect specific proteins in a given sample. It uses gel electrophoresis to separate native proteins, and then transfer them to a membrane (typically nitrocellulose), where they are stained with primary antibodies specific to the target protein. The primary antibodies are then marked with secondary antibodies which contain an active place, where a fluorescent marker can bind. This is used for chemiluminescent detection; if a signal is present, it indicates the presence of a specific protein targeted antibody, and, therefore, of the protein itself. A commercially available, polyclonal antibody against human CNN2 was ordered. This antibody has been produced by immunization of mice. 10 µl of content of each interaction tube were applied to a 12 % PAGE gel. MagicMark™ XP Western Protein Standard (Invitrogen) was used as a standard.
 
Table 3 shows the order of samples on Western blot gel/membrane.
 
The results of my experiments are presented on the figures below.

Figure 1: A 12 % PAGE gel of fractions 1 to12 from nickel affinity chromatography purification of hOXR1A. The arrow indicates hOXR1A band.

Figure 2: A 12 % PAGE gel of fractions 1 to10 from nickel affinity chromatography purification of hOXR1T1. The arrow indicates hOXR1T1 band.
 
Figure 3: A 12 % PAGE gel of fractions 1 to12 from nickel affinity chromatography purification of CNN2 expressed for 4 hours. The arrow indicates CNN2 band. 

Figure 4: A 12 % PAGE gel of fractions 1 to10 from nickel affinity chromatography purification of TOMM20. The arrow indicates TOMM20 band. The expression of the protein was not efficient enough, therefore further purification steps were suspended.

Figure 5: A 12 % PAGE gel of fractions A4 to A10 from Resource Q6 fast liquid protein chromatography purification of hOXR1T1. The arrow indicates hOXR1T1 band.

Figure 6: A 12 % PAGE gel of fractions A6 to A10 from Superdex 75 fast protein liquid chromatography purification of hOXR1A. The arrow indicates hOXR1A band.

Figure 7: A 12 % PAGE gel of CNN2 protein after cleavage from His-tag. It is smaller without the His-tag, hence further down in the gel. The arrow indicates CNN2 band.

Figure 8: A 12 % PAGE gel of CNN2 protein after fast protein liquid chromatography Superdex75 column. The arrow indicates CNN2 band.

Figure 9: A 12 % PAGE gel of Dynabeads ® interaction tests. Bands labelled as follows: 1 - hOXR1 A, 2 - hOXR1 T1, 3 - CNN2, 4 - hOXR1A + CNN2, 5 - hOXR1T1 + CNN2. The arrow indicates CNN2 band together with hOXR1 band, suggesting interaction.
 
Figure 10.  Western Blot interaction test result. Bands labelled as follows: 1 - hOXR1A, 2 - hOXR1T1, 3 - CNN2,  4 - hOXR1A+CNN2, 5 - hOXR1 T1+CNN2 The interaction between hOXR1 A and CNN2 is visible in band 4. The black stain in that band represents the fluorescent marker attached to the antibody against CNN2. Therefore CNN2 protein was present in the sample 7 with hOXR1 A protein.

Human OXR1 protein, known as oxidation resistance protein 1, does not in fact have a fully determined function, although it has been identified many times as an oxidation resistance protein. Intensive research is being conducted on this protein. One of the possibilities to investigate the function of any protein is to identify candidate proteins with known function that may interact with the target protein - either in 'protein-to-protein' reaction or by forming a protein complex with other proteins involved. In this way, if the interaction between those proteins is confirmed, the function of the target protein may be deduced.
Calponin is an actin filament-associated regulatory protein originally found in smooth muscle cells, implicated in the regulation and modulation of its contraction. Three homologous isoforms of calponin have been identified in vertebrate species. Expression of h2-calponin occurs in smooth muscle and non-muscle cells including epidermal keratinocytes, lung alveolar cells, endothelial cells, fibroblasts and myeloid blood cells. The function of h2-calponin includes stabilizing actin cytoskeleton and inhibiting cytokinesis. The conserved structures of calponin isoforms suggest conserved functional mechanisms. It is capable of binding to actin, calmodulin, troponin C and tropomyosin. The interaction of calponin with actin inhibits the actomyosin Mg-ATPase activity.  

The discovery of the interaction between those two proteins must lead to common points in which those proteins may occur to function together. Most common function of CNN2 protein is regulation on smooth muscle contraction, where calponin-2 inhibits the activity of one of the ATPases -  an enzyme that catalyzes the decomposition of ATP into ADP and a free phosphate ion. Actomyosin Mg-ATPase is a mitochondrial transmembrane ATP-ase; it is thus present in an environment where leakage of activated oxygen from mitochondria during oxidative phosphorylation, causing oxidative stress, may occur. HOXR1 protein is thought to function to protect cells from oxidative damage; it can be hypothesized that calponin-2, which inhibits ATPase activity and therefore diminishes the intensity of oxidative phosphorylation, binds to hOXR1 to give a signal about potential danger of oxidative stress and to start the repair process. The hypothesis seems possible when considering additionally that calponin-2 is mostly present in smooth muscle, where a lot of mitochondria activity occurs, and mitochondrial localization is required for OXR1 protein to prevent oxidative damage.

Several publications confirm the importance of CNN2-hOXR1 interaction in different types of cells and intracellular reactions, although it has not been discovered yet and it is not mentioned in those publications. I present some of them with explanation of possible role of the  CNN2-hOXR1 interaction in them.

 Evidence exists that diminished expression of h2-calponin in prostate cancer cells promotes cell proliferation, migration and the dependence of cell adhesion on substrate stiffness.  This leads to rapid spreading and promotion of cancer.  Considering the fact that CNN2 is capable of binding to hOXR1 protein it seems possible that a decreased level of CNN2 causes less efficient protection from oxidative damage. HOXR1 protein function appears to be crucial in cancer cells, as oxidants produced in oxidative stress can cause direct damage to the DNA, causing mutations. Oxidative species may also suppress apoptosis and promote proliferation and invasiveness. Functions of unsuppressed oxidative species are very similar to the effect of diminished h2-calponin expression in prostate cancer, which may indicate the direct concurrence of CNN2 and hOXR1 in protecting the cells from developing cancer.

Recently it was also discovered that exposure to oxidative species significantly increased the expression of calponin, and other thin filament-associated proteins in gastric smooth muscle. An increased expression of calponin can perhaps enhance hOXR1 protection from oxidative damage. In this case two scenarios are possible: 1.) due to oxidative stress hOXR1 protein expression is increased and it leads to enhanced expression of CNN2 protein, or 2.) increased CNN2 expression affects the level of expression of hOXR1 protein. The second possibility seems more probable because of the fact that together with CNN2 other thin filament-associated proteins are also highly expressed when exposed to oxidative stress.

Lately, calponin-2 has also been identified in CD90+ lymphocyte T-cells in mouse model of Alzheimer Disease (AD). In those cells, increased oxidative stress levels also occurred, which may indicate the importance of calponin-2 and its interaction with hOXR1 in protection from oxidative damage, which is one of the main causes of AD.  In another study Sirt2 protein knock down in primary human umbilical vein endothelial cells under oxidative stress was tested. Sirt2 is a protein which mediates cellular stress responses and is highly expressed in vascular endothelial cells. Its knock down changed expression of 340 genes, including calponin-2. This is also a lead indicating the possible impact of CNN2 protein on hOXR1 protein and confirming the hypothesis of the role of hOXR1 in protecting against oxidative damage.
Those examples clearly confirm that hOXR1-CNN2 interaction is crucial in numerous cell processes and that it can affect correct development and growth of human cells among different organs. The discovery that I have made gives many further research opportunities and is essential for treatment of such diseases as Alzheimer Disease. My discovery provides a tool for researchers around the world for better understanding of the processes and reactions that they test, leading to a better understanding of the human body and metabolism.
 
​FULL PAPER: