Journal of Undergraduate Research
Volume 5, Issue 1 - October 2003

Expression and Preliminary Purification of Mouse CLN2 Serine Protease

Theresa Caridi

ABSTRACT

A hereditary neurodegenerative disorder known as classical late-infantile neuronal ceroid lipofuscinosis (LINCL) is associated with a mutation of a lysosomal protein, CLN2. LINCL is rare but eventually fatal, and no effective treatment exists. This investigation describes subcloning CLN2 into a bacterial expression cell line so that the optimal conditions for expressing protein can be determined. The project also involves purifying CLN2 inclusion bodies from the bacteria and efforts to denature and refold this insoluble protein into a soluble, properly folded form. The final goal involves attempts to purify the protein to homogeneity utilizing ion exchange chromatography and then to test for enzymatic activity through kinetic assays. Eventually this study along with further investigation can be utilized to design reduced peptide and organic peptidomemitic inhibitors. These inhibitors could then be tested against CLN2 in in vitro assays and possibly in an in vivo transgenic mouse model.

INTRODUCTION

Recently, a new class of serine proteinases that utilizes an unusual catalytic mechanism has been isolated from bacterial sources. These enzymes use a Serine-Glutamic Acid-Aspartic Acid catalytic triad, where the Histidine residue of classic serine proteinases has been replaced by Glutamic Acid as the general base. This new catalytic triad results in these enzymes having a pH profile with its maximum in the acidic region. One enzyme in this family, CLN2, has been identified to be involved in a rare but fatal human disease known as late-infantile neuronal ceroid lipofuscinosis (LINCL)1.   In 1997, Sleat et al. identified a single lysosomal protein from normal patients (CLN2) that was absent in samples from patients with LINCL. Additionally, patients with this disease had numerous mutations identified in the gene that encodes for the CLN2 protein2.

In 1999, it was determined that Tripeptidyl-peptidase I (TPP I) is apparently the CLN2 protein that was found to be absent in LINCL patients1. Sequence analysis from the mouse (Mus musculus) genome has revealed a CLN2 homologue that is almost identical to the human protein involved in the previously mentioned storage disease in children, LINCL. The mouse protein shares an 88% amino acid sequence identity to the human enzyme and was therefore cloned for use in studying expression and protein purification. The goal is to provide protein samples for eventual structure determination. Since the structures of two CLN2 related enzymes [Pseudomonas Serine Carboxyl Proteinase (PSCP) and Kumamolysin from Bacillus novosp. MN-32] have been solved by collaborating laboratories3-5, molecular replacement method can eventually be used to determine the structure of CLN2. This crystal structure, when supplemented with preexisting kinetic studies, will advance our knowledge of what interactions are important in the binding of substrates and inhibitors.

METHODS

Previously, the gene encoding the Mus musculus CLN2 protein was amplified from a mouse cDNA library. This amplification product was cloned into a pCR2.1 TOPO vector, and subsequently ligated into a pET3a expression vector. A summary of experiments performed can be seen in Table 1.

Table 1
Summary of Experiments Performed
Clone
Vector
Mutations
Bacterial Cell Line
Expression Inoculation
Refolding
Column Chromatography
Activity Assay
8vaa
pET3a
1
BL21Gold(DE)pLysS
4%
Yes
Yes
No
5%
Yes
Yes
Yes
G101S
pET3a
none
BL21Star(DE3)pLysS
3.5%
No
No
No
G101G
pET3a
none
BL21Gold(DE3)pLysS
5%
No
No
No

Site-Directed Mutagenesis


The procedure was performed on CLN2 clone 8vaa, which contained one amino acid mutation from the reference sequence. The Quickchange Mutagenesis kit (QCM) protocol was followed for primer design and for preparing and running PCR reactions. Upper and lower strand primers were designed in order to change the isoleucine in the clone to a valine as in the reference strand. The upper primer (CLNI2VQCUP) was given the following sequence: 5’-GCC AAT ATC TCC ACT TGG GTC TAC AGT AGC CCT G-3’. The lower primer (CLNI2VQCLO) was given the following sequence: 5’-CAG GGC TAC TGT AGA CCC AAG TGG AGA TAT TGG C-3’. The PCR products were digested with DpnI and then transformed into Epicurian Coli XL1-Blue Supercompetent Cells according to QCM protocol.


DNA Purification and Identification of Positive Clone


Cell stocks were prepared and then mini-prepped using QIAprep spin kit protocol. CLN2 DNA was digested with ScaI. Specific enzyme incubation time and temperature was found by referencing a NEB catalog. DNA was analyzed on a 1.0% Agarose gel before sequencing. The sequenced DNA was aligned using the Genestream and CLUSTALW Alignment software programs.


Transformation


The vectors were subcloned into either or both BL21Gold(DE3)pLysS and BL21Star(DE3)pLysS bacterial expression cell lines according to standard protocol.

Overnight Cell Cultures and Protein Expression


LB media was inoculated with cells containing plasmid DNA and then incubated at 37ÖC with shaking overnight. To isolate the plasmid, cells were spun at 4800 RPM and 4ÖC for 10min. Expression of CLN2 was done on both a small scale (1 Liter) and large scale (4 Liter) basis in LB media. Varying percent inoculations (ranging from 3.5-5%) from overnight cell cultures were attempted for comparison. Protein expression was induced with 1mM final concentration IPTG when OD600 was close to 0.5 AU. The induction was continued for an additional 3hrs and cells were harvested by centrifugation (spin for 10min at 10,000 RPM). CLN2 expression was monitored on 12% Tris-Glycine SDS-PAGE gels.


Table 2
Cell Lysis and Inclusion Body Purification Buffers
Buffer
Composition
Resuspension Amount
1 (cold)
0.01M Tris pH8.0
0.02M MgCl2
0.005M CaCl
4.2 mL/g cells
2
0.01M Tris pH 8.0
0.001M EDTA
0.002M β-mercaptoethanol
0.1 M NaCl
5mL
3
0.05M Tris pH 8.0
0.005M EDTA
0.005M β-mercaptoethanol
0.5% Triton X-100
15mL
4
0.05M Tris pH 8.0
0.005M EDTA
0.005M β-mercaptoethanol
40mL
TE
-----------------
50 mg/mL



Inclusion Body Purification, Denaturing, and Refolding


Cell lysis and inclusion body purification buffers are listed in Table 2. After resuspension with Buffer 1, Dnase I was added to a final concentration of 80 units/mL, cells were lysed through the French Press, and incubated at 37ÖC for 15min. The cell suspension was gently layered over a 10mL 27% sucrose cushion and then spun at 4ÖC, 12000 x g for 30min. Following resuspension with Buffer 2, the cell suspension was layered over the sucrose cushion and spun in the same manner as with Buffer 1. After resuspension with Buffer 3, the solution was spun at 12,000 x g and 4ÖC for 15min (followed by the same process with Buffer 4). After purification, inclusion bodies were denatured in 8M urea at 1mg/mL. Solubilization techniques were attempted at room temperature and 37ÖC for 45min. The dialysis procedure is outlined in Table 3. Purification was monitored on 12% Tris-Glycine SDS-PAGE gels.

 


Table 3
Buffer Changes and Conditions for Dialysis
Buffer
Time (hrs)
Temperature
50mM Tris pH 11.0
4
Room
50mM Tris pH 11.0
16
4°C
50mM Tris pH 9.5
6
4°C
50mM Tris pH 9.5
16
4°C
50mM Tris pH 9.5
6
4°C

 

 

Column Chromatography


An anion exchange HiTrap Q Column was equilibrated with 20mM Tris pH 9.5 (Buffer A). A program was designed to provide a linear gradient of 20mM Tris pH 9.5/1M NaCl (Buffer B), where the concentration was increased from 0 to 1M over 60min. Aliquots of 1mL were collected using a fraction collector.


Post Dialysis Activity Assays


The substrate used for enzymatic assays was a hexapeptide of the form A-X-P1-F’-X-L where P1 was one of the following amino acids: Y, F, W, L, or nL (nor leucine). X equals a mixture of the 20 naturally occurring amino acids, minus M and C, and with the addition of nL. The stock (1250μM) substrate solution prepared contained 90% water and 10% DMSO. The final substrate solution concentration was 100μM after dilution. CLN2 activity assays were performed (1) with 0.5M Na Formate pH 3.5 and (2) with buffers from Li Lin et al. The absorbance reading of the post dialysis sample at 280nm allowed for calculation of the concentration of the sample using the Beer-Lambert Law (A=εlc). The estimated molar absorbtivity coefficient was taken to be 85,870 Lmol-1cm-1 from Expasy (www.expasy.com), and the path length was taken as 1cm for the cuvette containing the sample. Therefore, the amount of each buffer used was the following: 10μL of 10% TX-100, 30μL of 5M NaCl, and 960μL 5M Na Formate pH 3.5. Kinetic data was collected on a HP8452A spectrophotometer by measuring the decrease in absorbance over the 284-324nm wavelength range.

DISCUSSION


The mutation present in CLN2 clone 8vaa was effectively eliminated with the site-directed mutagenesis procedure creating a 100% match to reference DNA. The restriction digest of the vector with ScaI allowed for detection of bands around the expected 4.5kb and 1.7kb.


The CLN2 clones with one mutation (8vaa) as well as the corrected clone (G101) were successfully transformed into bacterial expression cell lines. CLN2 8vaa was inoculated at both 4% and 5% and expressed. After 1hr of cell cultivation, cells began to die. CLN2 G101 was inoculated at both 3.5% and 5% and expressed. The 5% inoculation similarly resulted in death of cells after 1hr. However, the 3.5% inoculation allowed for steady growth of cells over a 3hr cultivation period. The progressions of cell growth for all expressions are listed in Table 4 along with the weight of each inclusion body pellet before purification. It is important to note that the 5% inoculation of clone 8vaa was a small scale expression (1L), whereas the remaining expressions were all large scale (4L). SDS gels were used to identify expressed CLN2 and to monitor inclusion body purification (Figures 1 and 2). Clones 8vaa and G101 appeared to express CLN2 at the expected 61.3 kDa.


Figure 1. 12% Tris-Glycine SDS-PAGE gel for CLN2 8vaa (5% inoculation)
Figure 1. 12% Tris-Glycine SDS-PAGE gel for CLN2 8vaa (5% inoculation). MW, molecular weight standard; T, time in hrs; S, supernatants from inclusion body purification; PD, post dialysis either at room temperature (RT) or 37ÖC.

Figure 2. 12% Tris-Glycine SDS-PAGE gel for CLN2 G101G (5% inoculation).
Figure 2. 12% Tris-Glycine SDS-PAGE gel for CLN2 G101G (5% inoculation).

Clone 8vaa inclusion bodies were further purified and refolded during dialysis, eliminating several other proteins from impure CLN2 (Figure 1). CLN2 fractions from column chromatography of 8vaa can be seen in Figures 3, 4, 5, and 6.


Figures 3-6. CLN2 Fractions 35-60 from Column Chromatography of 8vaa Inoculated at 5%.
Figures 3-6. CLN2 Fractions 35-60 from Column Chromatography of 8vaa Inoculated at 5%.
Figure 3

Figures 3-6. CLN2 Fractions 35-60 from Column Chromatography of 8vaa Inoculated at 5%.
Figure 4

Figures 3-6. CLN2 Fractions 35-60 from Column Chromatography of 8vaa Inoculated at 5%.
Figure 5

Figures 3-6. CLN2 Fractions 35-60 from Column Chromatography of 8vaa Inoculated at 5%.
Figure 6

 

These pictures of SDS gels helped to determine where CLN2 elution was most evident. Clone 8vaa was also tested for enzyme activity. However, enzymatic tests with two different buffers showed no activity. It is important to note that three of the five substrates used for these assays were precipitated.

CONCLUSION


Data in Table 4 suggests that the optimal conditions for expression of CLN2 (of the options investigated) are a 3.5% inoculation in a BL21Star(DE3)pLysS bacterial cell line. All other cases resulted in death of cells during a period of expected growth. However, it is not clear without further investigation whether the inoculation at 3.5% or the bacterial cell line used was the primary factor for a successful growth of cells. The data suggests that that the positive cell growth was achieved by inoculating the expression at 3.5%. This is likely the case since both bacterial cell lines originally allowed for rapid cell growth, and it is only over time that the higher inoculation percentages moved toward cell death. This protein may be toxic to the cells and therefore causes a decrease in cell growth.


Table 4
The Progression of Cell Growth and Inclusion Body Pellet Amount from Various Expressions
Clone
OD6DD(AU)
Weight of Pallet (g)
T=0
T=1
T=2
T=3
8vaa (4%)
0.55524
0.98254
0.88733
0.77556
4.7707
8vva (5%)
0.65254
1.08202
1.09087
0.97539
1.7468
G101 (3.5%)
0.44308
1.62247
2.18391
2.29099
13.7981
G101 (5%)
0.42804
0.87007
0.80974
0.82314
3.5976

 


Further investigation was done on clone 8vaa (with one mutation). The results from the purification and refolding process (Figure 1) suggest that the buffers used were adequate in eliminating several other proteins from the sample of CLN2. Likewise, the buffers used during column chromatography allowed for visible elution of CLN2 without other protein. However, the kinetic study to test activity of CLN2 gave negative results. It is believed that the underlying reason for these results was the precipitation of substrate. Although two of the five substrates were not visibly precipitated, they may still have been affected. More substrate is needed to conduct further studies on the enzymatic activity of CLN2.

REFERENCES

  1. Rawlings, N. D. and Barrett, A. J. (1999) Tripeptidyl-peptidase I is apparently the CLN2 protein absent in classical late-infantile neuronal ceroid lipofuscinosis. B.B.A., 1429: 496-500.

  2. Sleat, D. E., Donnelly, R. J., Lackland, H., Lui, C. G., Sohar, I., Pullarkat, R. K., and Lobel, P. (1997) Association of Mutations in a Lysosomal Protein with Classical Late-Infantile Neuronal Ceroid Lipofuscinosis. Science, 277: 1802-1805.

  3. Wlodawer, A., Li, M., Dauter, Z., Gustchina, A., Uchida, K., Oyama, H., Dunn, Ben M., and Oda, K. (2001a) Carboxyl Proteinase from Pseudomonas Defines a Novel Family of Subtilisin-like Enzymes. Nature Structural Biology, 8: 442-446.

  4. Wlodawer, A., Li, M., Gustchina, A., Dauter, Z., Uchida, K., Oyama, H., Goldfarb, N.E., Dunn, B.M. and Oda, K. (2001b) Inhibitor Complexes of the Pseudomonas Serine-Carboxyl Proteinase. Biochemistry, 51: 15602-15611.

  5. Comellas-Bigler, M., Fuentes-Prior, P., Maskos, K., Huber, R., Oyama, H., Uchida, K., Dunn, B.M., Oda, K. and Bode, W. (2002) The 1.4 Å Crystal Structure of Kumamolysin: A Thermostable Serine-Carboxyl-Type Proteinase. Structure, 10: 865-876.

  6. Lin, L., Sohar, I., Lackland, H., and Lobel, P. (2000) The Human CLN2 Protein/Tripeptidyl-Peptidase I Is a Serine Protease That Autoactivates at Acidic pH. Journal of Biological Chemistry, 276(3): 2249-2255.

--top--

Back to the Journal of Undergraduate Research