PEDS Advance Access originally published online on September 28, 2008
Protein Engineering Design and Selection 2008 21(11):681-687; doi:10.1093/protein/gzn049
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Facile, reagentless and in situ release of Escherichia coli intracellular enzymes by heat-inducible autolytic vector for high-throughput screening
1Department of Chemical Engineering 2Department of Chemistry, Tsinghua University, 1 Tsinghua Garden Road, Beijing 100084, China
3 To whom correspondence should be addressed. E-mail: zhanglinlin{at}mail.tsinghua.edu.cn
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and methodsResultsDiscussionFundingReferences |
|---|
In an effect to broaden the application of the heat-inducible autolytic vector pUC18-cI857/pR-SRRz-rrnB previously developed, a new vector pUC18-cI857/pR(T41C)-SRRz-rrnB (pEAS-1b) was quantitatively characterized under various growth temperatures, heat induction temperatures and durations, and IPTG (isopropyl β-D-thiogalactoside) induction times, after resolving its erratic lysis profile found previously. Escherichia coli BL21 cells harboring this vector grew well at temperatures <36°C, and lysed efficiently (97.0 ± 0.8%) just 0.5 h after heat induction at 42°C for 30 min when cell growth was performed at 35°C. Application of this autolytic vector either in 96-well plates, or on nitrocellulose membranes, or on agar plates led to facile, efficient and consistent release of intracellular recombinant enzymes (e.g., a lysis efficiency of 91.8 ± 1.1% was obtained in 96-well plates). Further application in directed evolution was illustrated by improving the thermostability of amadoriase using this vector. This reagentless and in situ cell lysis method has the potentials for lysis of miniaturized samples in clinical diagnosis and bioanalytical detection, and even for lysis of cells in the microarray format.
Keywords: cell lysis/E.coli/heat-inducible autolytic vector/high-throughput screening
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionFundingReferences |
|---|
As a widely used laboratory bacterium, Escherichia coli (E.coli) is a favorite host for use in protein expression and enzyme screening, for its fast growth rate, low cost and ease in genetic manipulation. However, the impermeable cell wall significantly limits its application in high-throughput screening and large-scale recovery of intracellular products. Therefore, cell lysis is the first and important step for releasing intracellular enzymes, for which a facile, consistent, and cost-effective method is much preferred (Aharoni et al., 2005
; Madou et al., 2006; Turner, 2006). Traditional cell disruption methods, including mechanical (e.g., sonication, French press, bead milling), physical (e.g., freeze-thawing, osmotic shock), chemical (e.g., alkali, detergents), and enzymatic (e.g., lysozyme) methods (Schein, 1989; Asenjo and Andrews, 1990; Cull and McHenry, 1990; Foster, 1992) can be tedious and cost-inefficient. Recently developed mechanical and physical cell lysis methods, using nanoscale barbs in microchannels (Di Carlo et al., 2003), laser (Dhawan et al., 2002), and single electrical pulse (Han et al., 2003), are clean, fast and reagentless, but are most suitable for miniaturized samples in clinical diagnosis and bioanalytical detection. E.coli cells expressing enzymes with intracellular lytic activities, such as D-amino acid oxidase coupled with T7 lysozyme (Chien and Lee, 2006), Lactobacillus phage LL-H muramidase (Vasala et al., 1999), bacterial phage lysis proteins [e.g., SRRz of phage 
(Kloos et al., 1994), E of Phi X174 (Witte and Lubitz, 1989; Jechlinger et al., 1999), e and t of phage T4 (Tanji et al., 1998; Morita et al., 2001)] can be disrupted, after either additional osmotic shock (Tanji et al., 1998), or freeze-thawing treatment (Chien and Lee, 2006), or addition of a chemical inducer (Kloos et al., 1994; Vasala et al., 1999; Morita et al., 2001), but which involves a significant amount of liquid handling.
We recently developed a facile, efficient, and consistent cell lysis method by combining the autolytic gene cassette SRRz from bacteriophage with a series of tightly controlled UV- and heat-inducible promoters to construct E.coli autolytic vectors (Xu et al., 2006; Li et al., 2007). In the test tube, over 60–90% of lysis efficiencies can be achieved for E.coli BL21 cells harboring these two types of vectors, respectively.
A deficiency with these autolytic vectors is that the cell growth temperature needs to be restricted to <30°C in order to stringently repress the lysis gene expression and to obtain consistent lysis. The cI857/pR(T41C) promoter, a variant of cI857/pR promoter with a T41C mutation which has been reported allowing cell growth at temperatures of up to 36°C (Jechlinger et al., 1999) was tested in our previous study, but it showed an erratic lysis pattern and thus was not explored further (Xu et al., 2006). Our subsequent exhausted trial and error experiments revealed that newly transformed cells with this vector eliminated the inconsistency. In this study, the autolytic characteristics of this vector under various growth temperatures, heat induction temperatures and durations, and isopropyl β-D-thiogalactoside (IPTG) induction times were analyzed systematically and quantitatively. Applications of this autolytic vector in high-throughput screening were illustrated by release of recombinant proteins in tubes, in 96-well plates, on nitrocellulose membranes, and on agar plates. As an example for use in directed evolution, the amadoriase (AMA) thermostability was improved using this autolytic vector.
|
Materials and methods |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionFundingReferences |
|---|
Materials
E.coli BL21 was purchased from Novagen® Merck. Pfu DNA polymerase was obtained from Tiangen (Beijing, China). All restriction enzymes were from New England BioLabs (Beverly, MA, USA). IPTG, DNase I, and pUC18 plasmid were from TaKaRa (Dalian, China). Lysozyme, o-nitrophenyl-β-D-galactopyranoside (ONPG), N-ethyl-N-(2-hydroxy-3-sulfopropyl)-m-toluidine (TOOS), and rhodamine B were obtained from Sigma-Aldrich (St Louis, MO, USA). Peroxidase and 4-aminoantipyrine were purchased from Sangon (Shanghai, China). Glycated lysine was synthesized according to the published procedure (Mossine et al., 1994). BioTraceTM NT nitrocellulose transfer membranes were purchased from Pall (Ann Arbor, MI, USA). All the other reagents used were of analytic grade.
The gene for Aspergillus fumigattus AMA, amplified from pET30a-AMA (J.Zheng and Z.Lin, unpublished results) by Pfu DNA polymerase, was inserted between the BamH I and Pst I multiple cloning sites of pUC18 and previously constructed pUC18-cI857/pR(T41C)-SRRz-rrnB (renamed as pEAS-1b) (Xu et al., 2006), yielding pUC18-AMA and pEAS-1b-AMA, respectively. pUC18-BSLA (Bacillus subtilis lipase A) and pEAS-1b-BSLA were constructed similarly by cloning the BSLA gene into the Sac I and Xba I sites of the vectors.
Growth curves of cells containing autolytic vector
Single colonies of newly transformed BL21/pEAS-1b and BL21/pUC18 were inoculated into 40 ml Luria-Bertani (LB) medium (Sambrook and Russell, 2001) containing 50 ng/µl ampicillin. After vigorously mixing, each culture was then divided into four aliquots (10 ml each). Cells were shaken at 30, 35, 36 and 37°C, respectively. OD600 (optical density at 600 nm) of the cultures were recorded at indicated intervals.
For heat lysis experiments in test tubes, E.coli BL21 cells newly transformed with pEAS-1b or pEAS-1b-AMA, were cultivated overnight and then inoculated into fresh LB/ampicillin medium. The expression of the β-galactosidase gene (located in the genome of E.coli BL21) or the AMA gene (inserted into the multiple cloning sites of the vector) was induced by addition of 1 or 0.5 mM IPTG, respectively. The lysis profiles and lysis efficiencies were then determined using the procedures previously described (Xu et al., 2006). Water-bath shakers were used for accurate and consistent temperature control. All experiments were done in triplicate or quadruple.
Heat lysis experiments in 96-well plates were also performed similarly as described (Li et al., 2007), using a shaking thermo incubator (Hangzhou Allsheng Instruments, Hangzhou, China) to improve cell growth and heat transfer in 96-well plates. The BL21/pEAS-1b-AMA cells were cultivated at 35°C overnight and then 5 µl of saturated culture was diluted into 200 µl of fresh LB/ampicillin medium in wells and grown at 35°C for 
1.5 h. The AMA expression was initiated by 0.5 mM IPTG and then continued at 35°C for 1.5 h. Heat lysis was achieved by 30 min shaking at 42°C and then additional 30 min at 35°C. For comparison, two conventional lysozyme lysis protocols were used to lyse BL21/pUC18-AMA cells prepared under the same growth and expression conditions as described earlier. Specifically, in the lysozyme protocol (1) (Wong et al., 2005), cells were centrifuged (2500g, 15 min at 4°C) and each cell pellet was resuspended in 5 mg/ml lysozyme in 50 mM potassium phosphate buffer, pH 7.5, followed by incubation for 1 h at 37°C. In the lysozyme protocol (2) (Glieder et al., 2002), cell pellets were first frozen at –20°C for 2 h, resuspended in 0.1 M sodium phosphate buffer, pH 8.0, containing 0.5 mg/ml lysozyme, 2 U/ml DNase I, and 10 mM MgCl2, and then incubated at 37°C for 1 h. For each lysis method, experiments were performed in two 96-well plates in parallel.
For heat lysis on nitrocellulose membranes, newly transformed BL21/pEAS-1b-AMA cells were plated on nitrocellulose membranes which were in turn placed on LB/ampicillin agar plates. After incubation at 35°C for 10 h, the membranes were removed and placed on LB/ampicillin agar plates containing 0.5 mM IPTG for another 3 h at 35°C for AMA expression. The membranes were then subjected to heat lysis (42°C for 30 min, followed by 35°C for 1 h). The released AMA activity of each colony was detected by soaking the membranes in the AMA substrate solution (discussed later) to allow for the development of purple color.
BL21/pEAS-1b-BSLA cells were used to illustrate heat lysis on LB/ampicillin agar plates containing 3.125% (v/v) olive oil and 0.001% (w/v) rhodamine B. The plates were incubated firstly at 35°C for 10 h for cell growth, then at 42°C for 30 min for heat induction, and finally at 35°C for 4 h for cell lysis and lipase activity assay. An orange fluorescence upon UV irradiation or a distinguishable red color under visible light indicated the lipase activity (Kouker and Jaeger, 1987).
Activity assays and lysis efficiency determination
The lysis efficiency was calculated as the percentage of extracellular enzyme activity in the sum of extracellular and intracellular activity (Xu et al., 2006; Li et al., 2007). Standard β-galactosidase activity was assayed spectrophotometrically using ONPG as substrate (Miller, 1972). The AMA activity towards glycated lysine was analyzed with a modification of the fructosyl-amino acid oxidase activity assay (Sakaue and Kajiyama, 2003). The substrate solution contained 100 mM potassium phosphate buffer (pH 8.0), 2.7 purpurogallin units of peroxidase, 0.5 mM 4-aminoantipyrine, 15 mM TOOS, and 150 mM glycated lysine. The formation of product was monitored by the increase in absorbance at 555 nm at 37°C. One unit of activity was defined as the amount of enzyme that produces 0.5 µmol of quinone dye (
555 = 39.2 cm2/µmol) per min, corresponding to the formation of 1.0 µmol of H2O2 per min.
The 1317 bp AMA gene was randomly mutated by a standard error-prone polymerase chain reaction (PCR) protocol with 0.2 mM Mn2+ (Joo et al., 1999) using EPFor: 5'-TTACGAATTCGAGCTCGGTACCCG-3' and EPRev: 5'-ACGGCCAGTGCCAAGCTTGCATGCCT-3' as primers, and then subcloned into the BamH I and Pst I sites of pEAS-1b. The resulting mutant library was transformed into E.coli BL21, followed by a stepwise screening for enhanced thermostability as follows. A total of 2100 transformants were picked onto fresh LB/ampicillin agar plates (typically 150–200 colonies per plate). After incubation for 3 h at 35°C, colonies were partially transferred to nitrocellulose membranes, the AMA expression and the heat-induced cell lysis were then performed as described earlier. After heat treatment at 55°C for 30 min, the membranes were soaked in the AMA substrate solution for activity detection. 82 colonies which showed significant purple color after heat treatment were picked from the corresponding agar plates for the subsequent quantitative screening in 96-well plates as follows. After heat-induced cell lysis, 80 µl of supernatant of each well containing the released AMA was incubated for 30 min at 55°C in a thin-walled 96-well PCR plate. Thirteen variants with higher residual AMA activities than the wild-type were expressed in test tubes and again analyzed to confirm the improved thermostability.
Purification and thermostability determination of AMA
A nickel-chelating HiTrapTM chelating HP column (Amersham Biosciences AB, Uppsala, Sweden) was used to purify the his6-tagged wild-type and the best AMA mutant G6 on an ÄKTAexplorerTM station (Amersham Pharmacia Biotech). The bound proteins were eluted with a linear imidazole gradient, and the purities for the wild-type and the mutant were >95% as judged by SDS-PAGE. The half-lives of the two enzymes at 56 and 57°C were determined in triplicate, by monitoring the residual activities after incubating the purified enzymes (100 µg/ml) at corresponding temperatures for different intervals.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionFundingReferences |
|---|
Growth stability of cells containing autolytic vector
Growth curves of E.coli BL21 cells containing pEAS-1b (Fig. 1A) were analyzed at 30, 35, 36 and 37°C. It was found that E.coli BL21 cells harboring pEAS-1b showed a similar growth pattern but higher saturated OD600 than BL21/pUC18 cells at temperatures from 30 to 36°C (data not shown). However, incubation at 37°C for >5 h caused decreases in OD600 for cells containing the autolytic vector, indicating a looser repression of the promoter at this temperature. Thus, pEAS-1b did not disrupt cell growth at 
36°C.
|
Lysis profiles: effects of cell incubation temperature
BL21/pEAS-1b cells, which were cultivated at different growth temperatures but heat shocked at the same 42°C for exactly 30 min, exhibited different lysis profiles as measured by the reporter β-galactosidase (Fig. 1B). The optimal growth temperature for cell lysis was found to be 35°C. Cell lysis was less obvious immediately after heat induction, but after continuous incubation for 0.5 h at 35°C, a dramatic increase in lysis efficiency (97.0 ± 0.8%) was seen, coupled with a simultaneous decrease in OD600 value (data not shown). The highest lysis efficiency (97.8 ± 0.4%) was obtained at 1 h after heat induction. Further incubation slightly decreased the nominal lysis efficiency, likely because the residual cells continued to grow and express intracellular β-galactosidase while the extracellular β-galactosidase levels remained nearly unchanged (Xu et al., 2006). The basal level lysis during incubation at 35°C, reflecting the repression of the promoter at this temperature, was only 0.5–0.6% (Fig. 1C, curve with solid square). For cells containing a vector lacking the promoter or the SRRz gene, the lysis efficiency was observed to be <2% after incubation under the same conditions including the heat treatment step. This suggests that the nonspecific cell lysis caused by the heat treatment step is negligible compared with the cell lysis observed for the functional autolytic vector.
It was surprising that the lysis efficiency increased as the cell incubation temperature increased from 30 to 35°C (Fig. 1B). The cell growth and lysis process can be divided into five stages: (i) overnight growth, (ii) cell growth to a OD600 of 0.4–0.5, (iii) IPTG-induced protein expression, (iv) heat-induced SRRz expression and (v) SRRz-mediated cell lysis, with stages 1, 2, 3, and 5 currently all conducted at a same temperature. It was of interest to see which stage was more susceptible to the temperature effect. Therefore, the temperatures at stages 1, 2, 3, and 5 were varied (stage 4 was all done at 42°C for 30 min) and lysis efficiencies at 1 h after heat induction were measured using β-galactosidase as the reporter. As shown in Fig. 2, the temperature for SRRz lysis (stage 5) had almost no effect on cell lysis (Fig. 2A/B, C/D, H/I and K/L), and in fact the higher temperature at this stage led to a lower apparent lysis efficiency (Fig. 2F/G and M/N), possibly because the unlysed residual cells grew faster at higher temperatures and thus decreased the nominal lysis efficiency. The effects of temperature at stages 1, 2, and 3 were more complex. Comparisons of Fig. 2A with H, E and C revealed that a decrease in the temperature of each single stage did not affect cell lysis. However, lower temperatures for two tandem stages decreased the lysis efficiency more prominently (Fig. 2A/K and A/F), whereas a tremendous decrease was seen for lower temperatures at all three stages (Fig. 2A/M). On the other hand, compared with the much lower cell lysis obtained by incubation at 30°C for all the three stages (Fig. 2N), elevating the temperature for even one single stage dramatically increased the lysis efficiency (Fig. 2N/G, N/J and N/L. Also see Fig. 2N/D, N/I and N/B).
|
Lysis profiles: effects of heat induction temperature and duration
As shown in Fig. 1C, faster lysis was observed at higher heat induction temperature (45 over 42°C). A lysis efficiency of 45.3 ± 3.6% was observed immediately after heating at 45°C, and the efficiency reached its maximum (98.7 ± 0.1%) after 0.5 h. On the other hand, a longer heat induction duration (45 over 30 min) at 42°C led to only a small change in the lysis efficiency.
Lysis profiles: effects of lysis onset
It was also of interest to determine the useful time intervals between IPTG-induced protein expression and onset of heat-induced SRRz expression. As shown in Fig. 1D, not only the lysis efficiency but also the consistency (as judged by standard deviations) decreased as the IPTG induction time increased. When the intervals were <7 h (corresponds to OD600 values <2.1), cell lysis was satisfactory in terms of both lysis efficiency and consistency, with efficiencies >89.4 ± 1.6%.
Applications: release of recombinant protein
As an example for expression and release of heterologous protein in this autolytic system, the Aspergillus fumigattus AMA gene was cloned into the autolytic vector, yielding pEAS-1b-AMA. BL21/pEAS-1b-AMA cells were heated at 42°C for 30 min after cell growth and IPTG-induced AMA expression (1.5 h) at 35°C. 95.7 ± 0.9% and 95.9 ± 0.3% AMA activities were found to be released into the culture medium at 0.5 and 1 h after heat induction, respectively, while only 3.7 ± 1.2% and 1.4 ± 0.1% of AMAs were detected extracellularly without heat treatment. These ratios were consistent with the lysis efficiencies determined using β-galactosidase as the reporter under the same conditions (Fig. 1C, curves with open and filled square), which indicated that this autolytic vector was well suited for almost complete release of intracellular enzymes.
Interestingly, the inserted heat lysis unit enhanced the expression level of AMA (located in the multiple cloning sites). After expression for 3 h at 35°C, 14.3 ± 0.6 and 41.8 ± 1.4 milli-unit of intracellular AMA activities were detected for one OD600 of BL21/pUC18-AMA and BL21/pEAS-1b-AMA cells, respectively. A similar pattern was observed for expression time of 1.5 and 4.5 h (data not shown).
Applications: heat-inducible autolysis in high-throughput screening
The utility of this autolytic vector for high-throughput screening was further evaluated in three formats: 96-well plates, nitrocellulose membranes, and agar plates. In 96-well plates, heat lysis of E.coli BL21 cells harboring pEAS-1b-AMA was carried out. For comparison, lysis of cells containing pUC18-AMA using two conventional lysozyme protocols (Glieder et al., 2002; Wong et al., 2005) was also performed. Cell lysis mediated by the autolytic vector pEAS-1b-AMA exhibited higher lysis efficiency and consistency (91.8 ± 1.1%) and lower basal level lysis (2.1 ± 1.1%), and thus was superior over the lysis mediated by lysozyme (with lysis efficiencies of 42.3 ± 9.1% and 55.7 ± 6.2%, respectively) (Table I). The extracellular AMA activity released by heat-inducible autolysis was 4–7-folds higher than those of the two lysozyme methods, which was the consequence of both higher lysis efficiency and increased AMA expression using the autolytic vector.
|
The autolytic vector was also applied to release intracellular enzymes from BL21/pEAS-1b-AMA cells plated out on nitrocellulose membranes. After heat lysis of the cells on nitrocellulose membrane for 30 min at 42°C and then further incubation for 1 h at 35°C, almost all the colonies of BL21/pEAS-1b-AMA showed uniformed purple color instantly, within 1 min after contacting with the substrate solution (Fig. 3A), which indicated that efficient and consistent cell lysis had occurred on the membrane. However, no AMA activity was detected for BL21/pUC18-AMA colonies on the membrane (Fig. 3B), suggesting that neither the expressed AMA could secrete out of the cell nor the substrate could permeate through the cell.
|
The autolytic vector with the BSLA gene inserted (pEAS-1b-BSLA) was used to evaluate the lysis performance of the vector on agar plates, as the lipase shows a characteristic orange fluorescence upon UV irradiation on plates containing olive oil and rhodamine B, which can also be identified by a distinguishable red color under visible light (Kouker and Jaeger, 1987
). In BL21/pEAS-1b-BSLA and BL21/pUC18-BSLA cells, the lipase gene showed a remarkable leaky expression level, and thus IPTG was omitted in the experiments. Cells on agar plates were heated for 30 min at 42°C, followed by incubation for 4 h at 35°C for cell lysis and color development. As shown in Fig. 3C, almost all the colonies of BL21/pEAS-1b-BSLA showed red color (under visible light) in a uniformed manner. As lipases often have lethal effects on E.coli cells, and the activity can be detected on agar plates containing olive oil and rhodamine B after prolonged incubation, lipase-producing microorganisms have been routinely screened on agar plates without any lysis method (Dartois et al., 1992; Prim et al., 2000). Similarly, in our study, BL21/pUC18-BSLA cells showed the characteristic red color after prolonged incubation (1–2 days), but in an uneven manner (picture not shown). However, a uniformed release of lipase by the autolytic vector should be much preferred in high-throughput screening. Applications: directed evolution of AMA
AMAs, which catalyze the oxidative deglycation of glycated amino acids to yield corresponding amino acids, glucosone, and H2O2, are of potential importance as tools for uncoupling hyperglycemia and glycation reactions (Takahashi et al., 1997; Wu et al., 2000). Using the autolytic vector pEAS-1b-AMA, 2142 Aspergillus fumigattus AMA variants created by error-prone PCR were stepwisely screened on nitrocellulose membranes and in 96-well plates as described in detail in Materials and methods. The best mutant (G6), with four nucleotide substitutions (A184G, A630C, A752T and T773A) and thus three amino acid changes (N62D, Y251F and F258Y), showed 3.0- and 2.6-fold increased half-lives than the wild-type at 56 and 57°C, respectively (Table II).
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionFundingReferences |
|---|
In this study, we first re-examined and quantitatively characterized the previously constructed autolytic vector pEAS-1b, with the aim to broaden the cell growth temperature for the E.coli autolytic system (Xu et al., 2006
; Li et al., 2007). In test tubes, E.coli BL21 cells harboring this vector grew well at temperatures <36°C, and lysed efficiently (>90%) when cell growth was performed at temperatures from 33 to 36°C, followed by IPTG induction for <7 h, and then heat induction at 42 or 45°C for 30 or 45 min. A lysis efficiency of 97.0 ± 0.8%, which is suitable for most applications, was obtained under these conditions: cell growth at 35°C, IPTG induction for 1.5 h, heat induction at 42°C for 30 min, and further incubation at 35°C for just 0.5 h. Cell lysis by this vector was found to be tightly controlled, with a basal level lysis of 0.5–0.6% at 35°C. Compared with the previously reported autolytic vector employing the cI857/pR promoter, which requires cell growth at 28°C, heat induction at 38°C for 2 h, and further 6 h post-heat induction to reach its highest lysis efficiency (96.3 ± 0.4%) (Xu et al., 2006), this autolytic vector broadens the growth temperature and shortens the time required for maximal cell lysis, and thus complements and improves the previous low temperature one.
Releases of AMA by sequential expression of AMA and lysis genes located on the autolytic vector pEAS-1b-AMA both in test tubes and in 96-well plates were quite efficient, consistent and tightly controlled, with lysis efficiencies of 95.7 ± 0.9% and 91.8 ± 1.1%, and basal level lysis of 3.7 ± 1.2% and 2.1 ± 1.1%, respectively. Compared with the two conventional lysozyme-mediated lysis methods, which are widely used for high-throughput screening in 96-well plates, this heat-inducible autolysis method has obvious advantages for its reagentless nature, simpler procedure, shorter screening time, higher efficiency, and better consistency (Fig. 4). In the cases of high-throughput screening on nitrocellulose membranes and on agar plates, consistent in situ releases of intracellular enzymes were also observed simply by heating the membrane or agar plate at 42°C for 30 min. Finally, an AMA mutant with half-lives at 56 and 57°C increased by 3.0- and 2.6-fold than the wild-type were identified just by pre-screening 2100 AMA colonies on nitrocellulose membranes and then re-screening 80 colonies in 96-well plates. Thus this heat-inducible autolysis method should be useful in protein evolution, given that the target protein is stable after 30 min of heat induction at 42°C. Furthermore, this facile, efficient, reagentless, and in situ cell lysis method should also be applicable to lysis of miniaturized samples (e.g., picoliter to nanoliter cells or even single cell) in clinical diagnosis and bioanalytical detection (Dhawan et al., 2002; Han et al., 2003), and to lysis of cells integrated in a lab-on-chip format [e.g., the so called micro-colony array (Pohn et al., 2007)].
|
The significant inconsistent lysis efficiency of this autolytic vector observed before (Xu et al., 2006
) was found via exhausted tests to be due to permanent storage (data not shown). Interestingly, newly transformed cells with this vector exhibit consistent lysis patterns, and thus are useful in high-throughput screening, as in this case cells are routinely newly prepared with libraries of enzyme variants. However, The autolytic vector using the wild-type cI857/pR promoter, which permits cell growth at 28°C and lysis at 38°C, is much more stable and shows resistance to freeze-thawing cycles. It is noteworthy that permanent storage has no negative effect on cell growth, plasmid extraction, or recombinant protein expression for cells containing either autolytic vector.
The incubation temperature prior to heat induction seems to affect cell lysis mediated by the phage lysis genes (Figs 1B and 2), which is undocumented in the literature to our best knowledge. Interestingly, a similar phenomenon was also observed for another two similar autolytic vectors with replacement of the cI857/pR(T41C) promoter by two other E.coli heat shock promoters, htpG and clpB, respectively (Z.Cai and Z.Lin, unpublished results). It is noteworthy that the mechanism of heat induction of these two heat shock promoters (Nonaka et al., 2006) is quite different from that of cI857/pR(T41C) promoter (Ptashne et al., 1980; Jechlinger et al., 2005). Further investigation reveals that at least a short incubation at higher temperatures such as 35°C prior to heat induction is required to achieve efficient cell lysis (Fig. 2G, J, L and N), and the stages closer to heat induction are more susceptible to the temperature effect than the stages further away (efficiency difference: Fig. 2A/F > A/K, N/L > N/G). Moreover, both the positive effects of higher temperature (i.e., 35°C) and negative effects of lower temperature (i.e., 30°C) on cell lysis efficiency appear to be additive (efficiency: Fig. 2A > H > K > M, N< G < D < B). These results suggest that the SRRz-mediated cell lysis is dependent on the state of cells before heat induction, and cells cultivated at 35°C seems to be better responsive to heat induction than those cultivated at lower temperatures. But the intrinsic reasons are unknown.
Finally, IPTG induction time prior to heat induction also exhibits obvious effects on cell lysis efficiency. It has been reported that E.coli cell disruption mediated by T4 phage lysis gene t or e was efficient when the two genes were expressed in the logarithmic growth phase, while much less or almost no disruption occurred in the stationary phase (Tanji et al., 1998; Morita et al., 2001). This also holds true for cell lysis mediated by SRRz lysis genes from phage , as the lysis efficiency declined sharply when the IPTG induction time was increased form 7 h (OD600 = 2.1) to 9 h (OD600 = 2.2) (Fig. 1D), which indicated the entry of stationary phase as measured for the growth curve (data not shown).
|
Funding |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionFundingReferences |
|---|
National Basic Research Program of China (2003CB716002); High-tech Research and Development Program of China (2003AA214061).
Conflict of interest: none declared.
|
Footnotes |
|---|
Edited by Alan Berry
|
References |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionFundingReferences |
|---|
Aharoni A., Griffiths A.D., Tawfik D.S. Curr. Opin. Chem. Biol. (2005) 9:210–216.[CrossRef][Web of Science][Medline]
Asenjo J.A., Andrews B.A. Separation Processes in Biotechnology—Asenjo J.A., ed. (1990) New York: CRC Press. 177–208.
Chien L.J., Lee C.K. Biochem. Eng. J. (2006) 28:17–22.[CrossRef]
Cull M., McHenry C. Methods Enzymol. (1990) 182:147–153.[Web of Science][Medline]
Dartois V., Baulard A., Schanck K., Colson C. Biochim. Biophys. Acta (1992) 1131:253–260.[Medline]
Dhawan M.D., Wise F., Baeumner A.J. Anal. Bioanal. Chem. (2002) 374:421–426.[CrossRef][Web of Science][Medline]
Di Carlo D., Jeong K.H., Lee L.P. Lab. Chip (2003) 3:287–291.[CrossRef][Web of Science][Medline]
Foster D. Biotechnology (1992) 10:1539–1541.[CrossRef][Medline]
Glieder A., Farinas E.T., Arnold F.H. Nat. Biotechnol. (2002) 20:1135–1139.[CrossRef][Web of Science][Medline]
Han F.T., Wang Y., Sims C.E., Bachman M., Chang R.S., Li G.P., Allbritton N.L. Anal. Chem. (2003) 75:3688–3696.[Medline]
Jechlinger W., Szostak M.P., Witte A., Lubitz W. FEMS Microbiol. Lett. (1999) 173:347–352.[CrossRef][Web of Science][Medline]
Jechlinger W., Glocker J., Haidinger W., Matis A., Szostak M.P., Lubitz W. J. Biotechnol. (2005) 116:11–20.[CrossRef][Web of Science][Medline]
Joo H., Lin Z., Arnold F.H. Nature (1999) 399:670–673.[CrossRef][Web of Science][Medline]
Kloos D.U., Stratz M., Guttler A., Steffan R.J., Timmis K.N. J. Bacteriol. (1994) 176:7352–7361.
Kouker G., Jaeger K.-E. Appl. Environ. Microbiol. (1987) 53:211–213.
Li S., Xu L.H., Hua H., Ren C.A., Lin Z.L. J. Biotechnol. (2007) 127:647–652.[CrossRef][Web of Science][Medline]
Madou M., Zoval J., Jia G.Y., Kido H., Kim J., Kim N. Annu. Rev. Biomed. Eng. (2006) 8:601–628.[CrossRef][Web of Science][Medline]
Miller J.H. Experiments in Molecular Genetics (1972) New York: Cold Spring Harbor Press.
Morita M., Asami K., Tanji Y., Unno H. Biotechnol. Prog. (2001) 17:573–576.[CrossRef][Medline]
Mossine V.V., Glinsky G.V., Feather M.S. Carbohydr. Res. (1994) 262:257–270.[CrossRef][Web of Science][Medline]
Nonaka G., Blankschien M., Herman C., Gross C.A., Rhodius V.A. Genes Dev. (2006) 20:1776–1789.
Pohn B., Gerlach J., Scheideler M., Katz H., Uray M., Bischof H., Klimant I., Schwab H. J. Biotechnol. (2007) 129:162–170.[CrossRef][Web of Science][Medline]
Prim N., Blanco A., Martinez J., Pastor F.I.J., Diaz P. Res. Microbiol. (2000) 151:303–312.[Medline]
Ptashne M., Jeffrey A., Johnson A.D., Maurer R., Meyer B.J., Pabo C.O., Roberts T.M., Sauer R.T. Cell (1980) 19:1–11.[CrossRef][Web of Science][Medline]
Sakaue R., Kajiyama N. Appl. Environ. Microbiol. (2003) 69:139–145.
Sambrook J., Russell D.W. Molecular Cloning: A Laboratory Manual (2001) 3rd edn. New York: Cold Spring Harbor Laboratory Press.
Schein C. BioTechnology (1989) 7:1141–1149.
Takahashi M., Pischetsrieder M., Monnier V.M. J. Biol. Chem. (1997) 272:12505–12507.
Tanji Y., Asami K., Xing X.H., Unno H. J. Ferment. Bioeng. (1998) 85:74–78.[CrossRef]
Turner N.J. Enzyme Assays: High-Throughput Screening, Genetic Selection and Fingerprinting—Reymond J.L., ed. (2006) Weinheim: WILEY-VCH Press. 142.
Vasala A., Isomaki R., Myllykoski L., Alatossava T. J. Ind. Microbiol. Biotechnol. (1999) 22:39–43.[CrossRef][Web of Science]
Witte A., Lubitz W. Eur. J. Biochem. (1989) 180:393–398.[Web of Science][Medline]
Wong T.S., Wu N., Roccatano D., Zacharias M., Schwaneberg U. J. Biomol. Screen. (2005) 10:246–252.
Wu X.L., Takahashi M., Chen S.G., Monnier V.M. Biochemistry (2000) 39:1515–1521.[CrossRef][Web of Science][Medline]
Xu L.H., Li S., Ren C., Cai Z., Lin Z.L. BioTechniques (2006) 41:319–322.[Medline]
Received April 27, 2008; revised August 21, 2008; accepted August 22, 2008.
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||