BIOTECHNOLOGICALLY RELEVANT ENZYMES AND PROTEINS

Temperature influence on fluorescence intensity and enzymeactivity of the fusion protein of GFP and hyperthermophilicxylanase

Chong Zhang & Min-Sheng Liu & Xin-Hui Xing

Received: 25 February 2009 /Revised: 23 March 2009 /Accepted: 7 April 2009 /Published online: 24 April 2009# Springer-Verlag 2009

Abstract By constructing the expression system for fusionprotein of GFPmut1 (a green fluorescent protein mutant) withthe hyperthermophilic xylanase obtained from Dictyoglomusthermophilum Rt46B.1, the effects of temperature on thefluorescence of GFP and its relationship with the activities ofGFP-fused xylanase have been studied. The fluorescenceintensities of both GFP and GFP-xylanase have proved to bethermally sensitive, with the thermal sensitivity of thefluorescence intensity of GFP-xylanase being 15% higherthan that of GFP. The lost fluorescence intensity of GFPinactivated at high temperature of below 60°C in eithersingle or fusion form can be completely recovered bytreatment at 0°C. By the fluorescence recovery of GFPdomain at low temperature, the ratios of fluorescenceintensity to xylanase activity (Rgfp/Axyl) at 15°C and 37°Chave been compared. Even though the numbers of moleculesof GFP and xylanase are equivalent, the Rgfp/Axyl ratio at15°C is ten times of that at 37°C. This is mainly due to thefact that lower temperature is more conducive to the correctfolding of GFP than the hyperthermophilic xylanase duringthe expression. This study has indicated that the ratio of GFPfluorescence to the thermophilic enzyme activity for thefusion proteins expressed at different temperatures could behelpful in understanding the folding properties of the twofusion partners and in design of the fusion proteins.

Keywords Temperature . Green fluorescent protein .

Fusion protein . Hyperthermophilic xylanase .

Fluorescence intensity

Introduction

Green fluorescent protein (GFP), a 27-kDa protein, hasbeen widely used as a reporter in microbes, invertebrates,vertebrates, and plants (Chudakov et al. 2005). GFP and itsvariants have some unique benefits such as real-timedetection, no disruption of or toxicity to the host cells, norequirement for cofactors, and feasibility of fusion with thetarget proteins (Chalfie et al. 1994). These features makeGFP applicable as a quantitative marker for quantifying cellconcentration or the activity of a target enzyme in abioprocess. Poppenborg et al. demonstrated the potentialuse of GFP as a fusion partner for monitoring foreignprotein expression and localization in Escherichia coli(Poppenborg et al. 1997), and then, many researchers foundthat GFP fluorescence could be used to track in vivo proteinactivity when expressed by an operon fusion (two inde-pendent proteins), including the fusion with chloramphenicolacetyl transferase, organophosphorus hydrolase, heparinaseI, etc. (Albano et al. 1998; Wu et al. 2000; Chen et al. 2007).These studies indicate that GFP or other fluorescent proteinscan be used to rapidly monitor bioprocesses in terms ofenzyme activities.

Many factors affecting the quantification of fluorescenceintensity of GFP have been extensively studied, includingpH (Enoki et al. 2004), dissolved oxygen (Zhang et al.2005), cell density (DeLisa et al. 1999), and so on.Temperature can be expected to influence the fluorescenceintensity of GFP by changing its structure or its foldingprocesses. Pure GFP that has matured properly at lowtemperature can be stable even at temperatures up to 65°C(Gibbs et al. 1995). However, there has been little researchconcerning thermal effects on the fluorescence character-istics of GFP in its fused proteins, including fusion proteinsof GFP with thermophilic proteins. Also, since the folding

Appl Microbiol Biotechnol (2009) 84:511–517DOI 10.1007/s00253-009-2006-8

C. Zhang :M.-S. Liu :X.-H. Xing (*)Department of Chemical Engineering, Tsinghua University,Tsinghua Yuan,Beijing 100084, People’s Republic of Chinae-mail: xhxing@tsinghua.edu.cn

processes of wild GFP and its fusion proteins in E. coli,yeast, and mammalian cells are temperature-sensitive (Tsien1998), temperature would be a key issue for the quantifi-cation of a target enzyme by a GFP-fusion strategy,especially for the thermophilic enzymes. Moreover, for athermophilic enzyme, due to the different effects oftemperature on the folding processes of GFP and the targetprotein, a ratio of the GFP fluorescence intensity to theenzyme activity during expression of the fusion protein inbacterial cells may probably help to understand the foldingproperty of the thermophilic protein.

Hyperthermophilic xylanase, obtained from Dictyoglomusthermophilum Rt46B.1, can catalyze the endohydrolysisof 1,4-β-D-xylosidic linkages in xylan at an optimaltemperature of 90°C (Gibbs et al. 1995). Due to thedifficulties in the cultivation of thermophiles and lowenzyme productivity, expression of thermophilic enzymesin mesophilic bacterial hosts such as E. coli is a usefulstrategy. From the viewpoint of bioprocess optimizationand control, GFP is expected to be applicable formonitoring the activity of thermophilic enzymes duringthe cultivation of recombinant bacteria and/or the bio-catalysis processes.

In the work described in the present paper, byconstructing the expression system for the fusion proteinof GFPmut1 (a GFP mutant capable of well expressing at37°C) with hyperthermophilic xylanase, the effect oftemperature on the fluorescence of GFP and its relation-ship with the activity of GFP-fused xylanase has beenelucidated. In particular, the thermal characteristics of thefluorescence of GFP-xylanase and the influence oftemperature on the fluorescence of GFP in the fusedxylanase during the course of cultivation have beenstudied.

Materials and methods

Strains and plasmids

E. coli TB1 was used for fusion protein expression.Commercial plasmid pMAL-p2x was purchased fromNEB Inc. Plasmid pET21-DBc harbored hyperthermophilicxylanase gene (xynB) of D. thermophilum Rt46B.1 was akind gift from Prof. Nakamura of the Tokyo Institute ofTechnology (Morris et al. 1998). Plasmid pSG1729harbored GFPmut1 gene, which has a double substitution(F64L, S65T) in wild type (wt) GFP and its fluorescence is3.3-fold stronger than that of the wt GFP at 37°C (Cormacket al. 1996), was donated by the Bacillus Genetic StockCenter (Zeigler 2002).

The plasmid of pGFP-Xyn was constructed in this workto express the GFP and xylanase fusion protein in E. coli

TB1. The primers for clone GFPmut1 gene without the stopcodon from plasmid pSG1729 had the restriction sitesof NdeI and SacI at each end (forward primer: 5′-AAAGGAGATTCGACATATGGGTACCCTGCATATGAGTAAAGGA-3′; reverse primer: 5′-CATCGGAGCTCGAGGTACCTTTGTATAGTT CATCCATGCCATGTG-3′).The primers for clone xynB gene from plasmid pET21-DBchad the restriction sites of BamHI and HindIII at each end(forward primer: 5′-GCGGATCCATGCAAACGTCTATAACACTAACA-3′; reverse primer: 5′-GGCAAGCTTCTAACTACTTCCACTA CTGCTACTTTGA-3 ′) . TheGFPmut1 gene was inserted into pMAL-p2x at the sitesof NdeI and SacI to achieve a new plasmid pGFP-p2x.The xynB gene was then inserted into the downstream ofGFPmut1 of pGFP-p2x at the sites of BamHI and HindIIIto achieve the plasmid of pGFP-Xyn, which was able toexpress GFP and xylanase fusion protein. A linker with 24amino acids containing restriction enzyme sites waspresent between the xylanase and GFP in the fusionprotein. Bioinformatic simulation implied that this linkerpreserved the flexibilities of the two partners (data notshown). The plasmid of pGFP-Xyn was then introducedinto E. coli TB1, which was the best host for the originalplasmid of pMAL-p2x.

Expression of GFP-xylanase fusion protein

E. coli TB1 (pGFP-Xyn) was grown in Luria broth (LB;per liter: 10 g tryptone, 5 g yeast extract, and 10 gsodium chloride). The pre-cultured E. coli (OD600=1.5)was inoculated into a 500-mL flask containing 100 mLof LB medium and cultivated on a 250 rpm shaker foreach experiment. For fusion protein expression, therecombinant cells were cultivated and induced at 37°Cand 15°C, respectively, and IPTG with a final concen-tration of 1 mM was added to induce the expression offusion protein. As comparison, GFP and xylanase werealso independently expressed by the strain E. coli BL21(DE3) (pET21-DBc) and the strain E. coli BL21 (DE3)(pET28-GFP), respectively.

Harvesting of the cells and preparation of the crudeenzymes

Cells were harvested by centrifugation (9,000 × g for1 min), washed three times with phosphate bufferedsaline (PBS; pH 7.4), and then resuspended in PBS forthe subsequent experiments. Cell extracts containingcrude soluble proteins were obtained by disrupting therecovered cells with an ultra-sonicator (Ningbo Xinzhi,JY88-II) followed by centrifugation (9,000 × g for5 min) to remove the cell debris. All these operationswere carried out at 4°C.

512 Appl Microbiol Biotechnol (2009) 84:511–517

Effects of temperatures on fluorescence intensity of GFPor GFP-xylanase

To examine the decrease of the fluorescence intensity ofGFP or GFP-xylanase at high temperature, the crudeenzymes or harvested cells were incubated at a predeter-mined temperature (50°C, 60°C, and 70°C) inside thefluorescence spectrophotometer connected with atemperature-controlled water bath, and the changes influorescence intensity with the heat-treatment time weremonitored. After the fluorescence intensities of the sampleswere reduced to a stable level, the temperature was thentransferred to 0°C quickly within 1 min. Then, changes inthe fluorescence intensities of the GFP or GFP-xylanasewere recorded to examine the recovery of the fluorescenceintensity. According to the necessity, the heat treatment of asample at high temperature followed by the fluorescencerecovery at the low temperature was repeated for twocycles.

Measurement methods

The optical density at 600 nm (OD600) for evaluating thecell concentration was measured with a spectrophotometer(Shimadzu UV-1,206, Japan) with an error of less than 1%.For the measurement of fluorescence intensity of the cellextract or whole cells, a fluorescence spectrophotometer(Hitachi F-2,500, Japan) was used with the parameters ofexcitation at 488 nm and emission at 510 nm, and thetemperature of the fluorescence spectrophotometer wascontrolled by a water bath with a precise temperaturecontrol. The error for fluorescence detection was within3%. Bergquist’s assay method was used to measurethe thermostable xylanase activity (Gibbs et al. 1995). Theenzyme reaction was controlled at 90°C, which is theoptimal temperature for the thermostable xylanase. Toexamine the GFP and xylanase fusion protein expression,SDS-PAGE was performed according to the standardmethod (SamBrook 2001).

Results

Influence of temperature on the fluorescenceof GFP-xylanase during the course of cultivation

E. coli TB1 (pGFP-Xyn) was cultivated at 37°C and theninduced at either 15°C or 37°C. The expression of GFP-xylanase was confirmed by SDS-PAGE (data not shown);most of the expressed GFP-xylanase was soluble at bothtemperatures. Since almost no fluorescence was detected inthe cell debris (less than 1% of that of the soluble fraction),the GFP in the insoluble fraction did not have fluorescence

compared with that in the soluble fraction. Correspondingly,the xylanase activity in the insoluble fraction was lessthan 2% of that in the soluble fraction.

Induction temperature was found to have a significantinfluence on the fluorescence intensity and xylanaseactivity of GFP-xylanase during the cultivation. As shownin Table 1, the fluorescence intensity of GFP-xylanase inthe cells cultivated at 15°C was higher than that at 37°C.Meanwhile, the xylanase activity of the fusion protein at15°C was significantly lower than that at 37°C. The resultsthus indicated that higher induction temperature wasbeneficial for the xylanase activity but decreased thefluorescence of the GFP. Interestingly, the ratio offluorescence intensity to xylanase activity (Rgfp/Axyl) wasnot identical at the different induction temperatures. AsGFP and xylanase are transcripted and translated under thesame promoter, the number of molecules of GFP andxylanase should be equivalent ([GFP]/[xyl]=1), which isthe theoretical basis for the quantification of xylanaseactivity by monitoring the fluorescence of GFP. Thedifferences in Rgfp/Axyl may be attributed to differencesin the folding processes and/or the thermal characteristicsof the two partners in GFP-xylanase at the differentculture temperatures.

The thermal characteristics of the fluorescence of GFPin GFP-xylanase fusion protein

To better understand the influence of temperature on thefluorescence of GFP-xylanase during the course of cultiva-tion, the thermal characteristics of GFP-xylanase in the cellextract were studied at different temperatures.

As regards the activity of xylanase, the enzyme wasstable at temperatures of less than 50°C; 90% of the activityof xylanase remained even after incubation at 90°C for20 min, reflecting the high thermal stability of this enzyme.The thermal stability of xylanase-GFP was 17% lower thanthat of the single xylanase itself (75% of the activity ofxylanase remained even after incubation at 90°C for20 min).

On the contrary, the fluorescence intensities of both GFPand GFP-xylanase were found to be thermally sensitive. As

Table 1 Effects of different temperatures on expression of GFP-xylanase fusion protein in Escherichia coli after 4 h induction(induction was conducted at 0.3–0.4 of OD)

Temperature (°C) 15 37

OD600 0.53 2.12

Fluorescence intensity (RFU/OD600) 275.66 128.54

Enzyme activity (IU/mL/OD600) 1.20 5.57

Ratio of fluorescence to enzyme activity 230.08 23.09

Appl Microbiol Biotechnol (2009) 84:511–517 513

shown in Fig. 1, the fluorescence intensities of GFP andGFP-xylanase decreased linearly with increasing tempera-ture as the sample was heated from 15°C to 80°C. Theslope for the single GFP was −0.01301 (r2>0.998), whilethat for the GFP-xylanase was −0.01496 (r2>0.998), whichindicated the thermal sensitivity of the fluorescenceintensity of GFP-xylanase was 15% higher than that ofGFP.

Interestingly, the heat-induced decrease in the fluores-cence intensity of GFP in the fusion protein is a reversibleprocess. The kinetics of the heat-induced decrease influorescence intensity of GFP of GFP-xylanase andrecovery of the lost fluorescence by low-temperaturetreatment was studied. As shown in Fig. 2, on increasingthe temperature (from 0°C to 50°C, 60°C, and 70°C,respectively), the fluorescence intensities of GFP and GFP-xylanase decreased linearly with time and then attained astable level after 10 min. In the present study, the lostfluorescence of GFPmut1 in either fusion or single formcould be reversely recovered to the maximal level within5 min by shifting the temperature back to 0°C. Thefluorescence intensity of GFP lost at 50°C, 60°C, and70°C was recovered to 100%, 92%, and 20% of the initialfluorescence, respectively. Meanwhile, the recovery ratio ofthe fluorescence intensity of GFP-xylanase was slightlylower than that of GFP itself. Also, the study for thefluorescence intensity recovery by low-temperature treat-ment of the whole cells harboring GFP-xylanase exhibitedthe same trend with that of the crude enzymes (data notshown).

10 20 30 40 50 60 70 80

0.0

0.2

0.4

0.6

0.8

1.0

Rel

ativ

e F

luor

esce

nce

Inte

nsity

Temperature (0C)

GFP GFP-Xylanase

Fig. 1 Influence of temperature on the fluorescence intensities of GFPand GFP-xylanase. Escherichia coli TB1 (pGFP-Xyn) were cultivatedat 37°C to OD600=0.3–0.4 and induced at 15°C for 16 h. The cellextracts were then incubated at each temperature for 10 min

0 20 40 60 80 100

0 10 20 30 40 50 60 70

0.6

0.7

0.8

0.9

1.0

1.1

Re

lativ

e flu

ores

cenc

e in

tens

ity (

100%

)

Time (min)

GFP-Xylanase GFP

0.4

0.5

0.6

0.7

0.8

0.9

1.0

Rel

ativ

e flu

ores

cenc

e in

tens

ity (

100%

)

Time (min)

GFP-Xylanase GFP

0 10 20 30 40 50 60 70 80 900.0

0.2

0.4

0.6

0.8

1.0 GFP-Xylanase GFP

Rel

at iv

e flu

ores

cenc

e in

tens

ity (

100%

)

Time (min)

a

b

c

Fig. 2 Changes in the fluorescence intensities of GFP-xylanase andGFP by treatment at 50°C (a), 60°C (b), and 70°C (c) and followed bythe fluorescence recovery at 0°C. The cell extract was used for theseexperiments. Down arrow indicates the time to shift the temperature to0°C; Up arrow represents the time to shift the temperature from 0°Cback to each temperature

514 Appl Microbiol Biotechnol (2009) 84:511–517

Influence of cultivation temperature on the quantificationof xylanase activity by the fluorescence of GFP-xylanase

The fluorescence intensity and xylanase activity werequantitatively measured during the course of cultivation ofE. coli TB1 (pGFP-Xyn). The temperature for cultivation/induction was 15°C or 37°C. As shown in Fig. 2, with theincrease of xylanase activity, the fluorescence intensity ofGFP also increased. Indeed, the activity of xylanase showeda good linear relationship with the fluorescence intensity ofGFP at both temperatures, which means that the hyper-thermophilic xylanase activity could be rapidly quantifiedby measuring the fluorescence intensity of GFP at therelevant temperatures.

For the ease of comparison, the fluorescence intensity ofGFP-xylanase in the cells cultivated at 37°C was alsorecovered by cooling the sample to 0°C for 10 min. Eventhough the slope for the cultivated cells with fluorescence

recovery at 0°C was improved compared with the data forthe cells directly measured at 37°C during the cultivation(Fig. 3), it was still much smaller than that for thecultivation/induction at 15°C. As the fluorescence intensityfor the respective cells cultivated at 15°C and 37°C but withfluorescence recovery at 0°C was detected at the lowtemperature within 1 min, the influence of temperature onthe thermal characteristics of GFP-xylanase could benegligible during the measurement. The differences in theslopes (Rgfp/Axyl) can thus be attributed to the differences inthe folding processes of GFP and xylanase at the differenttemperatures during the expression of the fusion protein.

Discussion

By constructing the expression system for the fusionprotein of GFPmut1 with hyperthermophilic xylanase, the

0 2 4 6 8 100.0

0.5

1.0

1.5

2.0

2.5

3.0

0

2

4

6

8

10

12

0

20

40

60

OD

600

(-)

OD

600 (-

)

Time (h)

OD600

Xylanase activity Fluorescence intensity

IPTG RF

U(-

)

Enz

ym e

act

ivity

(IU

/ml)

0 2 4 6 8 10 12 14 16 180.0

0.5

1.0

1.5

2.0

0

1

2

0

20

40

60

80

100

120

140

Time (h)

OD600

Xylanase activity Fluorescence intensity

IPTG

RF

U(-

)

Enz

yme

activ

ity (

IU/m

l)

a

b

Fig. 3 Time course profilesof fluorescence intensity andxylanase activity during thecultivation at 37°C (a) and 15°C(b). The whole cells were usedfor measurement of the GFPfluorescence. Down arrowindicated the time when IPTGwas added for induction

Appl Microbiol Biotechnol (2009) 84:511–517 515

effect of temperature on the fluorescence of GFP and itsrelationship with the activity of GFP-fused xylanase hasbeen elucidated. In particular, the thermal characteristics ofthe fluorescence of GFP-xylanase and the influence oftemperature on the fluorescence of GFP in the GFP-xylanase fusion protein during the cultivation have beenstudied.

Previous studies on the thermal characteristics of thefusion protein of enhanced green fluorescent protein(EGFP) and the bacterial surface layers (rSbpA) found thatthe fluorescence intensities of EGFP and EGFP-rSbpAwerethermal sensitive (Toca-Herrera et al. 2006), and theyexplained that an increase in temperature increased theprobability of unfolding of GFP as well as of the fusionpartner from their native states; also, collisions betweenmolecules would probably increase the fluorescencequenching. These results were also consistent with ourfindings that the fluorescence intensities of both GFP andGFP-xylanases were thermally sensitive.

Another interesting finding of the present paper is thatthe lost fluorescence intensity of GFP domain inactivated athigh temperature of below 60°C in either single or fusionform can be reversely recovered by cooling the sample at0°C (Fig. 2). Previous researchers reported that GFPexpressed and matured properly at low temperature (below25°C) can be stable at temperatures up to 65°C (Gibbs et al.1995), which means that the structure of GFP can be stableat high temperatures below 65°C. As shown in the presentpaper, on reverting to low temperature, the lost fluorescenceintensity of GFP could be recovered almost completely.Particularly, it takes only 10 min for the fluorescencerecovery, which is very quick. While when the temperaturewas increased to 70°C, as the structure of the GFP domainwas probably destroyed, the fluorescence of GFP in eithersingle or fusion form could be no longer recovered. Whatshould be concerned was that the recovery ratio of thefluorescence intensity of GFP-xylanase was slightly lowerthan that of GFP itself.

Both the thermal stability and the recovery capability of thefluorescence intensity of GFP-xylanase are a little bit poorerthan that of GFP. Although the structure of xylanase used inthe present study showed less pronounced temperaturedependence than that of GFP, the presence of xylanase wouldstill probably influence the structure/chromophore of GFP tosome extent, which would ultimately alter the thermal stabilityand the recovery capability of the fluorescence intensity ofGFP-xylanase. On the contrary, previous work on the thermalcharacteristics EGFP-rSbpA found that the fluorescenceof EGFP was independent of the fusion partner of rSbpA(Toca-Herrera et al. 2006). This would be attributed to thedifferent structures of the fusion partners (xylanase andrSbpA). As suggested by the previous work (Waldo et al.1999) and the present study, productive folding of the

downstream GFP protein domain and consequent formationof the GFP chromophore may be directly related to thefolding property of the upstream protein.

By eliminating the influence of temperature on thethermal characteristics of the GFP domain through fluores-cence recovery at low temperature, the Rgfp/Axyl ratios forthe respective bacterial cells cultivated at 15°C and 37°Chave been compared (Figs. 3 and 4). Even though thenumber of molecules of GFP and xylanase are equivalent,the Rgfp/Axyl ratio at 15°C is much larger than that at 37°Ceven after the fluorescence recovery at 0°C.

Formation of the chromophore of GFP depends on thecorrect folding of the protein. GFPmut1 used in the presentstudy folds fairly efficiently at or below 37°C but tends tomisfold and form mostly nonfluorescent aggregates athigher temperatures. Accordingly, with the increase intemperature, the chromophore of GFP domain of GFP-xylanase would probably misfold, ultimately leading to adecrease of fluorescence intensity. On the contrary, lowertemperature is beneficial for the correct folding of GFP,even in the fusion protein. For this reason, the Rgfp/Axyl

ratio at 15°C is much larger than that at 37°C.Based on the findings of the present study, the

quantitative Rgfp/Axyl ratio obtained at an appropriatetemperature for a fusion protein of GFP with a thermophilicenzyme can be used to rapidly monitor the enzyme activityand optimize the bioprocess for the production of thermo-philic enzymes. As is well known, most proteins fromhyperthermophiles are intrinsically thermally stable and canfold properly at high temperatures above 60°C (Morris etal. 1998), but the influence of temperature on the foldingefficiency of hypertheromophilic or thermophilic enzymesis always not well known. The present work is possible to

0 2 4 6 8 10 120

100

200

300

400

500

600

700

150C 370C 370C after recovery at 0 0C

Xylanase Activity (IU/ml)

RF

U (

-)

Fig. 4 The linear relationships between fluorescence intensity andenzyme activity of GFP-xylanase at different cultivation/inductiontemperatures of 15°C (circle) and 37°C (square) based on the datashown in Fig. 3. The fluorescence of the cells cultivated and inducedat 37°C was measured after the fluorescence recovery at 0°C (triangle)

516 Appl Microbiol Biotechnol (2009) 84:511–517

be applied to study the influence of temperature on thefolding efficiency of the target protein:

Axyl ¼ xyl½ � � hxyl � kxyl ð1Þ

RGFP ¼ GFP½ � � hGFP � KGFP: ð2ÞIn which, kxyl is the xylanase activity per molar xylanase,

kGFP is the fluorescent intensity of GFP per molar GFP, andηGFP are the folding efficiency of xylanase and GFP,respectively.

As a fusion protein, the molecular number of xylanaseand GFP should be equal:

xyl½ � ¼ GFP½ �: ð3ÞThe above equations can be deduced as

hxyl ¼Axyl

RGFP� hGFP � kGFP

kxylð4Þ

where Axyl

RGFP, kxyl, and kGFP can be estimated directly.

The properly folded GFP should be fluorescent, whilethe wrong-folded one should not. What should be ofconcern is that the fluorescence of the wrong-folded GFPcan be recovered by shifting to low temperature, thus,

hGFP ¼ foldedGFP½ �total GFP½ � � RGFP

RrecovedGFP: ð5Þ

Thus, ηxyl can be estimated with the help of the detectionof the fluorescent intensity of fused GFP.

In fact, GFP, as an indicator, has been applied in rapidprotein-folding assay; the fluorescence of the GFP-fusionprotein is related to the productive folding of the upstreamfused protein domains (Waldo et al. 1999). In this study, byeliminating the influence of temperature on the thermalinstability of the GFP domain through fluorescence recoveryat low temperature, the Rgfp/Axyl ratio reflects only thefolding processes of the two fused domains during the courseof cultivation and induction. The ratio of GFP fluorescenceto enzyme activity for the fusion protein expressed atdifferent temperatures could be helpful in understanding thefolding properties of the two fusion partners and in thedesign of the fusion protein especially thermophilic enzymes.

What should be of concern is that we used hyperthermo-philic xylanase as a model system to study the temperatureinfluence on fluorescence intensity and enzyme activity of thefusion protein of GFP and the target fusion partner, but themethod used in this work can also be applied for othercombinations of GFP and different type of enzymes. This ismainly due to the fact that the correct folding of thechromophore of GFP is a temperature-sensitive process. Onthe contrary, the folding processes of hyperthermophilicenzymes or most of other type of enzymes are not thatthermo-sensitive as the chromophore of GFP. Taking the easy

detection of the change of GFP fluorescence intensity intoconsideration, the folding property relevant to temperature ofthe fusion partners can be easily characterized by the alterationof the fluorescence intensity of the fused GFP.

Acknowledgements This study was supported in part by theprojects of the National Natural Science Foundation of China(20836004) and 863 plan of the Ministry of Science and Technologyof China (No. 2006AA02Z203).

References

Albano CR, Randers-Eichhorn L, Bentley WE, Rao G (1998) Greenfluorescent protein as a real time quantitative reporter ofheterologous protein production. Biotechnol Prog 14:351–354

Chalfie M, Tu Y, Euskirchen G, Ward WW, Prasher DC (1994) Greenfluorescent protein as a marker for gene-expression. Science263:802–805

Chen Y, Xing XH, Ye FC, Kuang Y (2007) Soluble expression andrapid quantification of GFP-hepA fusion protein in recombinantEscherichia coli. Chin J Chem Eng 15:122–126

Chudakov DM, Lukyanov S, Lukyanov KA (2005) Fluorescent proteinsas a toolkit for in vivo imaging. Trends Biotechnol 23:605–613

Cormack BP, Valdivia RH, Falkow S (1996) FACS-optimized mutantsof the green fluorescent protein (GFP). Gene 173:33–38

DeLisa MP, Li JC, Rao G, Weigand WA, Bentley WE (1999)Monitoring GFP-operon fusion protein expression during highcell density cultivation of Escherichia coli using an on-lineoptical sensor. Biotechnol Bioeng 65:54–64

Enoki S, Saeki K, Maki K, Kuwajima K (2004) Acid denaturation andrefolding of green fluorescent protein. Biochemistry 43:14238–14248

Gibbs MD, Reeves RA, Bergquist PL (1995) Cloning, sequencing,and expression of a xylanase gene from the extreme thermophileDictyoglomus thermophilum Rt46B.1 and activity of the enzymeon fiber-bound substrate. Appl Environ Microb 61:4403–4408

Morris DD, Gibbs MD, Chin CWJ, Koh MH, Wong KKY, Allison RW,Nelson PJ, Bergquist PL (1998) Cloning of the xynB gene fromDictyoglomus thermophilum Rt46B.1 and action of the geneproduct on kraft pulp. Appl Environ Microbiol 64:1759–1765

Poppenborg L, Friehs K, Flaschel E (1997) The green fluorescent protein isa versatile reporter for bioprocess monitoring. J Biotechnol 58:79–88

SamBrook J (2001) Molecular cloning: a laboratory manual, 5th edn.Cold Spring Harbor Laboratory Press, New York

Toca-Herrera JL, Kupcu S, Diederichs V, Moncayo G, Pum D, SleytrUB (2006) Fluorescence emission properties of S-layer enhancedgreen fluorescent fusion protein as a function of temperature, pHconditions, and guanidine hydrochloride concentration. Bioma-cromolecules 7:3298–3301

Tsien RY (1998) The green fluorescent protein. Annu Rev Biochem67:509–544

Waldo GS, Standish BM, Berendzen J, Terwilliger TC (1999) Rapidprotein-folding assay using green fluorescent protein. NatBiotechnol 17:691–695

Wu CF, Cha HJ, Rao G, Valdes JJ, Bentley WE (2000) A greenfluorescent protein fusion strategy for monitoring the expression,cellular location, and separation of biologically active organo-phosphorus hydrolase. Appl Microbiol Biotechnol 54:78–83

Zeigler D (2002) Bacillus genetic stock center catalog of strains:integration vectors for gram-positive organisms. The BacillusGenetic Stock Center, Department of Biochemistry, The OhioState University, Ohio, USA

Zhang C, Xing XH, Lou K (2005) Rapid detection of a GFP-markedEnterobacter aerogenes under anaerobic conditions by aerobicfluorescence recovery. FEMS Microbiol Lett 249:211–218

Appl Microbiol Biotechnol (2009) 84:511–517 517