The Whole Erk Catalogue Tales of a Geek Abroad

I'm working on writing up a report on my research from the last few months, and I'm getting tired of FTPing it to myself back and forth. Therefore, I'm going to save it here along with some commentary, as a little insight into what my life has been about for the last while. This is a work-in-progress.

Introduction
Planaria – free living flatworms of class Turbellaria – have been known for their incredible regenerative capacity since Abraham Trembley and later Peter Simon Pallas first considered them “immortal under the edge of the knife” nearly 250 years ago (Newmark and Sanchèz Alvarado, 2002). Their nearly limitless regenerative plasticity allows a full organism to regenerate from a fragment as small as 1/279 of the original organism. In addition to this fascinating ability, planaria are very interesting as a model organism due to their extreme simplicity yet resemblance to upper organisms: they are bilaterally symmetrical, cephalised, have cute eyespots which function almost identically to more advanced eyes, and possess a central nervous system consisting of paired ganglia and longitudinal nerve cords. However, their eyes are very nearly as simple as possible using the light-detection rhodopsin system standard to most visual organisms, and the planarian contains only three distinct tissue layers. Its circulatory, digestive, and respirtatory systems are virtually nonexistent: as a flatworm it accomplishes these functions via diffusion. The planarian represents a fascinating bridge between extremely complex and extremely simple organisms, making it ideal as a research topic. [This paragraph is quoted with minor modification from my research proposal "Exploring gene function and expression in Planaria using RNA interference knockouts."]

The purpose of this project was to initiate RNA interference in planarian of species Dugesia rotolytica (black planarians) and Girardia tigrina (brown planarians). RNA interference is a method in which double stranded RNA matched to the sequence (or a part of the sequence) of a gene's messenger RNA is introduced into an organism. The double stranded RNA is used as a targetting mechanism to specifically silence expression of the messenger RNA, leading to gene knockout (Sanchèz Alvarado and Newmark, 1999).

The gene targeted is the sine oculis gene, a conserved gene important to eye development in Drosophila. SO has been shown to function similarly in planarians: RNA interference knockouts of SO do not exhibit eye growth (Pineda et al., 2000). Interference will be initiated using a "soaking method" whereby the planarian is cut and exposed to a dsRNA-laced environment (Orii et al, 2003). Sine oculis has been chosen as a target because eye knockout causes an easily visualised change in the planarian's phenotype: the lack of eyespots can be readily observed under a dissecting microscope. The soaking method was chosen as an interference technique because of its simplicity of use compared to other techniques, namely microinjection (Sanchèz Alvarado and Newmark, 1999) and consumption of dsRNA-expressing bacteria (Newmark et al., 2003). It is hoped that using the premade materials from this experiment (bacteria containing plasmids expressing the single-stranded RNAs which can be used to initiate interference) can be used to relatively quickly generate double stranded RNA's for interference in the future, and that the procedure from this experiment can be used as a third year level Developmental Biology (Biol330 at Thompson Rivers University) lab.

Procedure
Step 1: Annealing single-stranded DNA to form double-stranded inserts with "sticky" ends
Twelve units of single-stranded DNA were ordered from Promega Inc. These each consisted of half of a fragment of the sine oculis gene from a planarian species, along with a portion of the sticky end from EcoR1 at the 5' end and HindIII at the 3' end.

The gene sequences ordered included a 68bp fragment of the Girardia tigrina sine oculis (GtSO) gene (Genbank accession number AJ251660), a 87 bp fragment of Dugesia japonica sine oculis (DjSO) gene (Accession No. AJ312218), and a scrambled sequence (Scram) comprised of the same sequence as GtSO with several nucleotide “chunks” transposed across the sequence. A comparison of the nucleotide and protein sequences of GtSO and DjSO can be found at http://www.rmxp.org/erk/sequence.html.

TABLE 1: Single stranded DNAs used to construct double stranded DNA inserts

Note: The terms "forward" and "reverse" used in the final names (i.e. "GtSO forward sequence (GtSO1)") refer to the 'direction' of the single-stranded RNA that will later be generated from the T7 promoter on the pBlueScript vector. The "forward" GtSO sequence exactly matches the sequence selected from the genbank entry for GtSO, while the "reverse" sequence is its base-pair match, mirrored 5' to 3'.

The single stranded DNAs were suspended in STE, combined in a single 1.5mL microfuge tube, and heated to 95 degrees in a water bath in a heating block, with a sheet of tinfoil overtop to help insulate. The tubes were left to incubate for 5 minutes, then the heat was turned off and the bath left to drop to room temperature overnight. (Later annealing work showed that satisfactory results could be achieved by removing the tubes from heat and letting them cool in the open air, a much faster process). The annealed DNA sequences were run on a 5% acrylamide gel to test the annealing (see Figure 1) At this point it was found that the original GtSO2 ordered did not arrive as expected and/or was not successfully resuspended in STE; a new batch was ordered and annealed at a later date.

Figure 1: 5% acrylamide gel showing single stranded and annealed DNA. Lanes contain, from left to right: [1] 50 bp ladder marker, [2] DjSO2 forward, [3] DjSO2 annealed, [4] DjSO1 reverse, [5] DjSO1 forward, [6] DjSO1 annealed, [7] Scram1 reverse, [8] Scram1 forward, [9] Scram1 annealed, [10]GtSO2 reverse, [11] GtSO2 forward, [12] GtSO2 annealed, [13]GtSO1 reverse, [14]GtSO1 forward, [15] GtSO1 annealed.

2. Linearisation of pBlueScript
The efficacy of the EcoR1 and HinD3 stocks used was tested by restricting samples of pBlueScript with each restriction enzyme independantly and testing the linearisation of the circular plasmid – the supercoiled, uncut plasmid moves further on the gel than the cut linearised plasmid. Restrictions were carried out according to standard protocols (Sambrook and Russel, 2001). Incubation was carried out at 37 degrees for at least 1 hour; later samples were left overnight in a 37 degree bath to ensure no unrestricted material remained.

Once the efficacy of both enzyme stocks was confirmed (see Figure 1) a double restriction was performed to excise the multiple-restriction site and provide a location in which to insert the target gene.

Figure 2: 0.7% Agarose gel showing effectiveness of EcoR1 and HinD3 restriction enzyme stocks. From left to right, the lanes contain: [1] a 25bp ladder marker, [2] circular pBlueScript (the two bands seen are circular 'nicked' pBlueScript and supercoiled plasmid, which migrates further), [3] pBlueScript linearised with EcoR1, [4] pBlueScript linearised with HinD3, [5] pBlueScript linearised with both enzymes simultaneously, and [6] 50bp ladder marker.

3. Insertion of DNA into pBlueScript vector and ligation
The ligation was carried out using standard protocols (Sambrook and Russel, 2001). Seven samples were ligated: GtSO1 and 2, Scram 1 and 2, DjSO 1 and 2, and linear pBlueScript (cut with only one restriction enzyme) with no added insert to demonstrate the efficacy of the ligase and ligase buffer used.

Evidence of successful ligation was somewhat difficult to determine (see Figure 3). However, comparing linear and circular pBlueScript banding to the patterns yielded from ligation did confirm that the ligase did something; to test the integrity of the ligated genes more accurately the simplest method was to proceed to transformation. If transformed bacteria were observed, clearly the ligations were effective.

Figure 3: Gel showing effectiveness of ligase reaction. Lanes contain, from left to right: [1] 25 bp ladder marker, [2] circular pBlueScript, [3] pBlueScript linearised with EcoR1 and HinD3, [4] Ligated pBlueScript + GtSO1 insert, [5] Ligated pBlueScript + Scram1 insert, [6] Ligated pBlueScript + DjSO1 insert, [7] Ligated pBlueScript + DjSO2 insert, and [8] 50 bp ladder marker.

4. Transformation of ligated vectors into competent E. coli cells
Ligated pBlueScript vectors were transformed into competent cells according to the Inoue protocol (Sambrook and Russel, 2004)

5. Plating and selection of transformed cells
Cells were plated on LB-agar (see Table 4) containing ampicillin, X-galactose, and IPTG. Each transformation was plated onto two plates, labelled “A” and “B” (eg. GtSO1a, DjSO2b, etc.). The conrols were only plated onto one plate each. Transformed cells were cells capable of growing on the medium due to the ampicillin resistance coded on pBlueScript, but incapable of digesting X-galactose due to the interrupt caused by the inserted DNA, and therefore not stained blue. Plates were incubated at 37 degrees for approximately 18 hours.

It was observed that DjSO transformants still had some ability to digest pBlueScript, appearing as mostly white colonies with a very small amount of blue stain. This may be because, unlike GtSO and Scram, the DjSO inserts do not frameshift the gene they are inserted into, as they possess a number of base-pairs divisible by three. Therefore, the general protein coded by the recombinant gene on the plasmid is still the same, but with about 29 extraneous amino acids added to the middle.

6. Inoculations using transformants
Successful transformant colonies were taken from the growth plates and inoculated into 5 mL of LB-ampicillin broth and grown in a shaking incubator overnight. As controls, tubes containing uninoculated LB-amp broth and untransformed E. coli were grown as well.

7. Miniprep isolation of DNA from grown transformants
Minipreps (Sambrook and Russel, 2001) were carried out using 4.2 mL of the incoculum; the remaining 0.8 mL were mixed with 0.2 mL of glycerol and frozen for future use.

8. Characterisation of miniprep DNA
The plasmids isolated by miniprep were run on a gel to determine the approximate size of the inserts in the plasmid - it was assumed if the inserts were in the expected size range, the identity was the fragment expected. On the gel, miniprep plasmids digested by only one restriction enzyme (EcoR1) were run next to double-digested minipreps (EcoR1+HinD3). It was expected that the double-digested plasmids would have the same retardation factor as linear pBlueScript, as the insert had been fully excised, while single-digested plasmids would migrate less far than the linear pBlueScript as they possessed the 50-75bp insert added by ligation earlier. This additional retardation would be very slight, as the insert is so small relative to the 1900bp pBlueScript vector. It was also hoped that the excised inserts would be visible on the double-digested gel; however, insufficient DNA was used to visualise the very small fragment (see Figure 4).

Figure 4 (click to enlarge): gel showing efficacy of minipreps, quantifying miniprep DNA, and demonstrating identity/size of inserts in miniprep plasmids. Please note that on the original gel all lanes contained DNA; the information loss from first photographing with a polaroid camera, then photographing that photograph to digitally store the information has cost resolution in some lanes. Lanes contain, from left to right:
[1] 10 uL of linear pBlueScript	[11] 5uL of EcoR1 digested Scram1.2a culture
[2] 5uL of EcoR1 digested GtSO1.1a culture	[12] 5uL of EcoR1/HinD3 digested Scram1.2a culture
[3] 5uL of EcoR1/HinD3 digested GtSO1.1a culture	[13] 5uL of EcoR1 digested Scram1.1b culture
[4] 5uL of EcoR1 digested GtSO1.2a culture	[14] 5uL of EcoR1/HinD3 digested Scram1.1b culture
[5] 5uL of EcoR1/HinD3 digested GtSO1.2a culture	[15] 5uL of EcoR1 digested Scram2.2b culture
[6] 5uL of EcoR1 digested GtSO2.1a culture	[16] 5uL of EcoR1/HinD3 digested Scram2.2b culture
[7] 5uL of EcoR1/HinD3 digested GtSO2.1a culture	[17] 5uL of EcoR1 digested Scram2.3b culture
[8] 5uL of EcoR1 digested GtSO2.2a culture	[18] 5uL of EcoR1/HinD3 digested Scram2.3b culture
[9] 5uL of EcoR1/HinD3 digested GtSO2.2a culture	[19] 5 uL of linear pBlueScript
[10] 50 bp ladder marker	[20] 0.2 ug of circular pBlueScript.

9. Synthesis of single-stranded RNA using T7 RiboMAX kit
RNA was synthesised from pBlueScript's T7 promoter using a RiboMAX kit by Promega. As there was no information available on the specific strain of pBlueScript used, it was unsure if the HinD3 site was closer to the T7 promoter (pBlueScript SK-, Figure 5a) or further away (pBlueScript SK-, Figure 5b). Therefore, two samples of each miniprep were prepared, one digested with EcoR1, which would produce RNA from T7 using SK+, and one digested with HinD3, which would produce RNA from SK-.

A)
B)
Figure 5: A) EcoR1 and HinD3 cut sites on pBlueScript SK-. Arrow shows the direction of RNA synthesis from the T7 promoter. B) EcoR1 and HinD3 cut sites on pBlueScript SK-. Arrow shows the direction of RNA synthesis from the T7 promoter.

Linearised samples were prepared for RNA synthesis using the RNAse cleanup protocol described in step 10. RiboMAX RNA preparations were visualised on a gel (RNA samples were incubated at 65 degrees for 7 minutes in a denaturing buffer containing formaldehyde and other RNA denaturing factors before loading onto the gel); the gel indicates that single-stranded RNA was successfully produced. See Figure 6. The GtSO1 sample used to produce RNA was at a concentration of about 4 ng/uL, and the GtSO2 sample was at 40 ng/uL. No Scram samples were synthesised at this time, as the main objective was to demonstrate the possibility of producing RNA with the samples, and to anneal single-stranded RNA if enough was produced. The concentration of GtSO2 was about half the minimum recommended by the RiboMAX kit, and RNA synthesis went forward quite well; the concentration of GtSO1 was apparently insufficient to generate noticeable quantities of RNA.

Figure 6: The very bright bands on this gel show the RNA produced using the RiboMAX kit. Lanes contain, from left to right: [1] 50 bp ladder marker, [2] RNA produced from GtSO2.1a miniprep digested with EcoR1, [3] RNA produced from GtSO2.1a miniprep digested with HinD3, [4] RNA produced from GtSO1.1a miniprep digested with EcoR1, [5] RNA produced from GtSO1.1a miniprep digested with HinD3, [6] Linear pBlueScript, [7] RNA produced from the luciferase control gene in non-nuclease-free TE buffer, and [8] RNA produced from the luciferase control gene in nuclease-free water. Previous gels showed more RNA from the TE buffer control, indicating that the RNAse inhibitors work to preserve RNA for a while, but it does eventually degrade in non-nuclease-free buffer.

10. RNAse cleanup and removal from isolated plasmids.
Miniprep plasmids were incubated in the presence of RNAse, then treated with proteinase K to denature the RNAse (Sambrook and Russel, 2001). After proteinase K treatment each sample (100uL total volume) was washed with a mixture of 50uL phenol and 50uL chloroform (actually 24:1 chloroform:amyl alcohol). From this point, all work was carried out in as nuclease-free an environment as possible. Fresh PCR-safe barrier pipette tips and nuclease-free microfuge tubes were the primary precaution taken. The TE layer (top) was removed and saved, and the phenol/chloroform layer was back-extracted with 100uL of TE. The TE layers for each sample were combined to make a 200uL mixture to which 2.5 mL of -20C 95% ethanol and 500 uL of 3M acetic anhydride were added. The ethanol mixtures were centrifuged for 10 minutes, the liquid decanted, and the pellet washed with 70% ethanol before being centrifuged for an additional 5 minutes. The supernatant was discarded and the flasks left to air dry thoroughly. Pellets were resuspended in TE; in the future it would be wise to suspend them in nuclease-free water instead.

Conclusion and future work
Due to time constraints, RNA interference has not yet been initiated in the planarian stocks. However, single-stranded RNA has been successfully synthesised; all that remains is to anneal the single-stranded RNA fragments and introduce the resultant double-stranded RNA into planarians. The high yield of single-stranded RNA from the kit and relative simplicity of the steps leading up to RNA production suggest that this may still be a viable third-year developmental biology lab. The main forseeable problem is the limited number of uses available in the riboMAX kit; a less expensive way to synthesise RNA might be desirable for an undergraduate level laboratory.

References
Newark, P. and Sanchèz Alvarado, A. 2002. "Not your father's planarian: A classic model enters the era of functional genomics". Nature Rev Genet, 3:210-220.

Newark, P.; Reddien, P.; Cebria, F.; and Sanchèz Alvarado, A. 2003. "Ingestion of bacterially expressed double-stranded RNA inhibits gene expression in planarians". Proc Natl Acad Sci, 100:11861-11865.

Orii, H.; Mochii, M.; and Watanabi, K. 2003. "A simple 'soaking method' for RNA interference in the planarian Dugesia japonica." Dev. Genes. Evol. 213:138-141

Pineda, G.; Gonazlez, J.; Callaerts, P.; Ikeo, K.; Gehring, W.; and Saló, E. 2000. “Searching for the prototypic eye genetic network: Sine oculis is essential for eye regeneration in planarians.” Proc. Natl. Acad. Sci. 97:4525-4529.

Sambrook, J. and Russel, D. 2001. Molecular Cloning: A Laboratory Manual (Third Edition). Cold Spring Harbor Laboratory Press.

Sanchèz Alvarado, A. and Newmark, P. 1999 "Double-stranded RNA specifically disrupts gene expression during planarian regeneration". Dev. Biol. 96:5049-5054.