**3.2 Effect of pH on x12 Fluc activity**

Detailed investigation on the pH-dependence of luciferase mutant activity revealed a significant further improvement in pH-tolerant profile from that of 5 Fluc (Law *et al.*, 2006). The normalised pH-dependence of activity was shown to facilitate comparison of activity across the range of pH values (Fig. 3A). The non-normalised results for 12 Fluc and UG emphasize the increase in activity for the 12 Fluc relative to UG (Fig. 3B). The high level of activity ( 80 % of maximum activity) exhibited by 12 Fluc across a range of physiologically relevant pH values (6.6 – 8.6) is likely to offer greater sensitivity and reliability when used in place of existing luciferase mutants in many applications, particularly those requiring a lower pH than the optimal for for Fluc or those that experience pH fluctuations such as whole cell or animal imaging (Frullano *et al.*, 2010).

#### **3.3 Bioluminescence spectra and kinetic properties of WT Fluc, x12 Fluc and UG with LH2 and ALH2**

The bioluminescence spectrum of WT Fluc is known to undergo a classic red bathochromic shift with LH2 at low pH, whereas x12 Fluc and UG maintained consistent yellow-green colours emission maximum of 557 nm and 560 nm respectively over the

Development of a pH-Tolerant Thermostable

Fluc with ALH2.

**Mutants** 

12 s. PMT voltage 550 mV.

**Mutants** 

at pH 6.2 not 6.5. Standard error +2nm.

**Km, µM Kcat x 108,** 

**RLU/s**

Table 3. Kinetic parameters of WT, Fluc mutants and UG with LH2

*Photinus pyralis* Luciferase for Brighter *In Vivo* Imaging 127

the emission maximum of 601 nm, 592 nm and 585 nm, respectively (Table 4). UG and x12 Fluc also had narrower half-bandwidth with ALH2 than WT Fluc, which may indicate that more rigid active sites of UG and x12 Fluc confer better protection to the emitter than WT

**WT** 592 559 558 91 66 66 **x12** 565 557 558 66 66 67 **UG** 562 560 559 73 68 70 **x2** 594 568 568 86 84 85 **x4** 560 558 561 69 66 70 **x5** 559 559 558 68 63 62

0.31 nmol of each mutant was added to 150 µM LH2 and 1 mM ATP in chilled 0.1 M TEM buffer, adjusted to a varying pH and spectra were measured after 30 s at ca. 25ºC. Half-bandwidths are widths of spectra at half maximum intensity. Data were corrected for PMT spectral response. UG was assayed

Table 2. Bioluminescent spectra of the WT, Fluc mutants and UG with LH2 at different pH

**WT** 18+1 1129+215 668+157 24.25+0.14 100 **x12** 6.8+0.2 50+3 74+2 3.56+0.04 15 **UG** 3.7+0.08 19+2 53+8 0.98+0.02 4 **x2** 12+2 1260+181 1149+110 11.37+0.03 47 **x4** 28+3 574+84 201+7 16.59+1.53 68 **x5** 9.5+1.5 1202+142 1288+51 21.33+0.04 88 Kinetic parameters were derived from flash heights (Hanes, 1932) by varying LH2 concentration in the range of 0.2 -500 M in the presence of saturating 1 mM ATP (see Materials and Methods). Specific activity was determined in the presence of 20.1 pmol enzyme by integrating light in 20 ms pulses over

In terms of kinetic parameters significant differences were observed in Km and Kcat values for LH2 and ALH2 between the WT and thermostable luciferases. In the WT both Km and Kcat for ALH2 were significantly lower than for LH2, suggesting that WT Fluc has higher affinity for ALH2 than for LH2 (Shinde *et al.*, 2006). In thermostable x12 Fluc both Km and Kcat for ALH2 were very close to those of LH2, while in UG Km for ALH2 was higher and Kcat was lower than for LH2. The mutations that invoke thermostability increase the affinity of Fluc for LH2, but reduce it for ALH2. WT Fluc had nearly 10-fold higher catalytic

**Michaelis-Menten kinetic parameters Specific activity**

**Kcat/ Km x 1013, RLU/ s x µM** 

**x 107, RLU/mg**  **% of WT at pH 7.8**

**Bioluminescent spectra λmax (nm) Half-bandwidth (nm) pH pH 6.5 7.8 8.8 6.5 7.8 8.8** 

investigated pH range of 6.2-8.8, (Table 2). Neither x12 nor UG showed any significant widening of the spectra at lower pH. Overall, bioluminescence spectra of x12 were even less pH-dependant than those of UG. Both thermostable mutants would be expected to retain their colour under physiological conditions, which may be pertinent to multispectral imaging. The tolerance of x12 Fluc to low pH may give it further advantage in terms of signal strength.

Luciferase activity (20 l at 0.42 M) was assayed by the manual mixing with 180 l of TEM, 1.11 mM ATP, 222 M LH2, 300 M CoA over a range of pH values between 6.0 and 9.5. Bioluminescence was integrated over 5 s using the luminometer at a PMT voltage of 550 mV: normalised to each luciferase enzyme total activity data (A) and non-normalised data (B). Measurements for UG were carried out using a PMT voltage of 700 mV to obtain good signal-to-noise ratio readings; data presented for the non-normalised curve have been corrected for the different PMT voltage used. The lag-time between initiation of the reaction and recording of light emission was ~ 5s. Measurements at each pH were carried out in triplicate. Error bars represent one S.E.M. within triplicate measurements.

Fig. 3. Effect of pH on the activity of WT, UG, x5 and x12 Fluc.

Kinetic parameters including Michaelis-Menten constants (Km) and catalytic constant (Kcat) were measured for the WT, x12 and UG luciferases with LH2 and ALH2. The Kms of x12 Fluc (6.8 M) and UG (3.7 M) for LH2 were lower than that of WT Fluc (18M) (Table 3), indicating an increase in affinity of the thermostable enzymes for LH2 relative to the WT. Even stabilising mutations far from the active site can affect tertiary structure and lead to increased substrate affinities and altered enzyme kinetics (Squirrell *et al.*, 1999). Catalytic efficiency characterised by Kcat/Km ratio was significantly compromised in both thermostable luciferases as compared to the WT, with UG being more thermostable but less catalytically efficient than x12. In terms of specific activity x12 Fluc retained only 15% of the specific activity of the WT, exceeding that of UG by four-fold.

For *in viv*o applications involving imaging of protease activity where ALH2 is to be detected instead of LH2. Therefore, bioluminescence properties of these mutants with ALH2 were tested. WT Fluc is known to produce pH independent orange-red emission with ALH2 (White *et al.*, 1966) and here it was seen that WT Fluc, x12 Fluc and UG were red-shifted to

investigated pH range of 6.2-8.8, (Table 2). Neither x12 nor UG showed any significant widening of the spectra at lower pH. Overall, bioluminescence spectra of x12 were even less pH-dependant than those of UG. Both thermostable mutants would be expected to retain their colour under physiological conditions, which may be pertinent to multispectral imaging. The tolerance of x12 Fluc to low pH may give it further advantage

**A B**  Luciferase activity (20 l at 0.42 M) was assayed by the manual mixing with 180 l of TEM, 1.11 mM ATP, 222 M LH2, 300 M CoA over a range of pH values between 6.0 and 9.5. Bioluminescence was integrated over 5 s using the luminometer at a PMT voltage of 550 mV: normalised to each luciferase enzyme total activity data (A) and non-normalised data (B). Measurements for UG were carried out using a PMT voltage of 700 mV to obtain good signal-to-noise ratio readings; data presented for the non-normalised curve have been corrected for the different PMT voltage used. The lag-time between initiation of the reaction and recording of light emission was ~ 5s. Measurements at each pH were

Kinetic parameters including Michaelis-Menten constants (Km) and catalytic constant (Kcat) were measured for the WT, x12 and UG luciferases with LH2 and ALH2. The Kms of x12 Fluc (6.8 M) and UG (3.7 M) for LH2 were lower than that of WT Fluc (18M) (Table 3), indicating an increase in affinity of the thermostable enzymes for LH2 relative to the WT. Even stabilising mutations far from the active site can affect tertiary structure and lead to increased substrate affinities and altered enzyme kinetics (Squirrell *et al.*, 1999). Catalytic efficiency characterised by Kcat/Km ratio was significantly compromised in both thermostable luciferases as compared to the WT, with UG being more thermostable but less catalytically efficient than x12. In terms of specific activity x12 Fluc retained only 15% of the

For *in viv*o applications involving imaging of protease activity where ALH2 is to be detected instead of LH2. Therefore, bioluminescence properties of these mutants with ALH2 were tested. WT Fluc is known to produce pH independent orange-red emission with ALH2 (White *et al.*, 1966) and here it was seen that WT Fluc, x12 Fluc and UG were red-shifted to

carried out in triplicate. Error bars represent one S.E.M. within triplicate measurements.

Fig. 3. Effect of pH on the activity of WT, UG, x5 and x12 Fluc.

specific activity of the WT, exceeding that of UG by four-fold.

in terms of signal strength.

the emission maximum of 601 nm, 592 nm and 585 nm, respectively (Table 4). UG and x12 Fluc also had narrower half-bandwidth with ALH2 than WT Fluc, which may indicate that more rigid active sites of UG and x12 Fluc confer better protection to the emitter than WT Fluc with ALH2.


0.31 nmol of each mutant was added to 150 µM LH2 and 1 mM ATP in chilled 0.1 M TEM buffer, adjusted to a varying pH and spectra were measured after 30 s at ca. 25ºC. Half-bandwidths are widths of spectra at half maximum intensity. Data were corrected for PMT spectral response. UG was assayed at pH 6.2 not 6.5. Standard error +2nm.


Table 2. Bioluminescent spectra of the WT, Fluc mutants and UG with LH2 at different pH

Kinetic parameters were derived from flash heights (Hanes, 1932) by varying LH2 concentration in the range of 0.2 -500 M in the presence of saturating 1 mM ATP (see Materials and Methods). Specific activity was determined in the presence of 20.1 pmol enzyme by integrating light in 20 ms pulses over 12 s. PMT voltage 550 mV.

Table 3. Kinetic parameters of WT, Fluc mutants and UG with LH2

In terms of kinetic parameters significant differences were observed in Km and Kcat values for LH2 and ALH2 between the WT and thermostable luciferases. In the WT both Km and Kcat for ALH2 were significantly lower than for LH2, suggesting that WT Fluc has higher affinity for ALH2 than for LH2 (Shinde *et al.*, 2006). In thermostable x12 Fluc both Km and Kcat for ALH2 were very close to those of LH2, while in UG Km for ALH2 was higher and Kcat was lower than for LH2. The mutations that invoke thermostability increase the affinity of Fluc for LH2, but reduce it for ALH2. WT Fluc had nearly 10-fold higher catalytic

Development of a pH-Tolerant Thermostable

**Km, µM Kcat x 108,** 

**RLU/s**

Table 5. Kinetic parameters of WT, Fluc mutants and UG with ALH2

displayed a sharp flash that strongly resembled the WT.

**3.5 Properties of T214 and F295 revertant mutants of x12 Fluc** 

F295L might enhance its catalytic properties.

with this substrate (Table 4).

**Mutants** 

voltage 550 mV.

*Photinus pyralis* Luciferase for Brighter *In Vivo* Imaging 129

observed in the WT (Table 2). x4 and x5 Flucs resisted red-shift at low pH in line with x12. With ALH2 none of the subset mutants showed any significant shift in the bioluminescence across the investigated pH range, which was similar to the behaviour of x12, WT and UG

**WT** 2.4+0.4 174+13 849+158 2.06+0.01 8 **x12** 6.5+0.2 38+4 59+9 1.46+0.01 6 **UG** 8.1+2.19 1.4+0.04 2.3+0.7 0.60+0.01 4 **x2** 2.3+0.1 226+24 956+68 1.99+0.03 8 **x4** 9+1 144+14 166+14 2.51+0.03 10 **x5** 1.6+0.1 215+32 1343+184 2.07+0.01 9 Kinetic parameters were derived from flash heights (Hanes, 1932) by varying ALH2 concentration in the range of 0.1-600 M used with 3mM ATP to saturate (see Materials and Methods). Specific activity was determined in the presence of 20.1 pmol enzyme by integrating light in 20 ms pulses over 12 s. PMT

In terms of kinetic parameters, if compared to the WT, the effect of mutations in x4 on Km and Kcat with LH2 differed significantly from mutations in both x2 and x5 subsets (Table 3). Mutations in x4 increased the Km, while in two other subsets decreased it, and reduced the Kcat, while in the other subsets hardly any change was observed. These results correlated well with the previously reported in literature (Law *et al.*, 2006; Tisi *et al.*, 2002; Prebble *et al.*, 2001; Branchini *et al.*, 2007). Similar trends were observed in the kinetic parameters for ALH2 (Table 5). Mutations in x4 significantly increased its Km for this substrate (as in x12), while two other mutants slightly decreased it. Kinetics of the x4 mutant with ALH2 exhibited slower rise and slower decay, reminiscent of the kinetics of x12, whereas x2 and x5

The differences in kinetic parameters and flash kinetics suggested that an undesirable effect of mutations was common to x4 and x12 mutants, but not to x2 or x5, on catalysis with both LH2 and ALH2. It was hypothesised that reversion of buried x12 Fluc mutations T214C and

x12 Fluc mutations T214C and F295L were individually reverted and their bioluminescent properties, catalytic parameters and thermal stability investigated. Bioluminescence spectra of both revertants were measured and found to match those of x12 Fluc with both substrates (Table 6). F295 revertant showed no bathochromic shift with either substrate and retain the pH-tolerance previously observed in x12 Fluc. T214 revertant had a flash kinetics similar to

**Michaelis-Menten kinetic parameters Specific activity**

**Kcat/ Km x 1013, RLU/ s x µM** 

**x 107, RLU/mg** 

**% of WT with LH2 at pH 7.8**



0.31 nmol of each mutant was added to 100 µM ALH2 and 3 mM ATP in chilled 0.1 M TEM buffer, adjusted to a varying pH and spectra were measured after 30 s at ca. 25ºC. Half bandwidths are widths of spectra at half maximum intensity. Data were corrected for PMT spectral response. UG was assayed at pH 6.2 not 6.5. Standard error +2nm.

Table 4. Bioluminescent spectra of the WT, Fluc mutants and UG with ALH2 at different pH

WT Fluc appears to provide stringent interactions dictating reaction kinetics with ALH2, but not those that govern the emitting species. This may also be indicated by larger halfbandwidths of bioluminescence spectra with ALH2 than with LH2. Low Km values were proposed to correlate with more blue-shifted emission (Kutuzova *et al.*, 1997), but this is not always true and it is possible for Km to be more linked to bioluminescence spectra halfbandwidth, i.e. the extent of vibrational freedom of oxyluciferin (Viviani *et al.*, 2001). However, here, WT Fluc has wider spectral half-bandwidth with ALH2 as compared to LH2, but lower Km. WT Fluc had only 10% of the specific activity with ALH2 as compared to LH2 at pH 7.8 (Table 5). On the contrary, x12 Fluc's activity with ALH2 (6%) is only 2.5-fold lower than that with LH2 (15%), while UG had as low as 4% with both substrates. Reasons for lower Vm and red-shifted emission of WT Fluc with ALH2 have been postulated (Shinde *et al.*, 2006; McCapra and Perring, 1985; Wada *et al.*, 2007).

#### **3.4 Properties of thermostable mutants containing subsets of x12 Fluc mutations**

Elimination of bathochromic shift at low pH, increase in thermostability and pH-tolerance were associated with the significant reduction in specific activity and changes in the essential kinetic parameters. In an attempt to identify mutations responsible for the undesirable associated changes and to improve the understanding of kinetics and activity differences between WT Fluc and x12 Fluc with LH2 and ALH2 a number of thermostable enzymes with subsets of x12 Fluc mutations were investigated (Table 1). These were x2 Fluc (E354R/ D357Y) (Baggett *et al.*, 2004), x4 Fluc (T214A/ I232A/ F295L/ E354K) (Tisi *et al.*, 2002) and x5 Fluc (F14R/ L35Q/ V182K/ I232K/ F465R) (Law *et al.*, 2006). Among the subset mutants only x2 showed bathochromic shift with LH2 at low pH, similar to that

efficiency (Kcat/Km) with LH2 and ALH2 than x12 Fluc. x12 demonstrated comparable Kcat/Km ratio for both substrates, favouring its use with ALH2 while UG had a similar catalytic efficiency to x12 Fluc for LH2, but 23-times lower Kcat/Km for ALH2 . Overall, x12

**WT** 599 601 601 74 79 79 **x12** 596 592 591 72 72 71 **UG** *588* 585 584 *71* 72 71 **x2** 596 596 595 76 76 74 **x4** 596 597 593 74 74 74 **x5** 599 600 599 74 76 73

0.31 nmol of each mutant was added to 100 µM ALH2 and 3 mM ATP in chilled 0.1 M TEM buffer, adjusted to a varying pH and spectra were measured after 30 s at ca. 25ºC. Half bandwidths are widths of spectra at half maximum intensity. Data were corrected for PMT spectral response. UG was assayed

Table 4. Bioluminescent spectra of the WT, Fluc mutants and UG with ALH2 at different pH

WT Fluc appears to provide stringent interactions dictating reaction kinetics with ALH2, but not those that govern the emitting species. This may also be indicated by larger halfbandwidths of bioluminescence spectra with ALH2 than with LH2. Low Km values were proposed to correlate with more blue-shifted emission (Kutuzova *et al.*, 1997), but this is not always true and it is possible for Km to be more linked to bioluminescence spectra halfbandwidth, i.e. the extent of vibrational freedom of oxyluciferin (Viviani *et al.*, 2001). However, here, WT Fluc has wider spectral half-bandwidth with ALH2 as compared to LH2, but lower Km. WT Fluc had only 10% of the specific activity with ALH2 as compared to LH2 at pH 7.8 (Table 5). On the contrary, x12 Fluc's activity with ALH2 (6%) is only 2.5-fold lower than that with LH2 (15%), while UG had as low as 4% with both substrates. Reasons for lower Vm and red-shifted emission of WT Fluc with ALH2 have been postulated (Shinde

**3.4 Properties of thermostable mutants containing subsets of x12 Fluc mutations** 

Elimination of bathochromic shift at low pH, increase in thermostability and pH-tolerance were associated with the significant reduction in specific activity and changes in the essential kinetic parameters. In an attempt to identify mutations responsible for the undesirable associated changes and to improve the understanding of kinetics and activity differences between WT Fluc and x12 Fluc with LH2 and ALH2 a number of thermostable enzymes with subsets of x12 Fluc mutations were investigated (Table 1). These were x2 Fluc (E354R/ D357Y) (Baggett *et al.*, 2004), x4 Fluc (T214A/ I232A/ F295L/ E354K) (Tisi *et al.*, 2002) and x5 Fluc (F14R/ L35Q/ V182K/ I232K/ F465R) (Law *et al.*, 2006). Among the subset mutants only x2 showed bathochromic shift with LH2 at low pH, similar to that

**Bioluminescent spectra λmax (nm) Half-bandwidth (nm) pH pH 6.5 7.8 8.8 6.5 7.8 8.8** 

Fluc seems to be a mutant of choice to be used with ALH2.

*et al.*, 2006; McCapra and Perring, 1985; Wada *et al.*, 2007).

**Mutants** 

at pH 6.2 not 6.5. Standard error +2nm.

observed in the WT (Table 2). x4 and x5 Flucs resisted red-shift at low pH in line with x12. With ALH2 none of the subset mutants showed any significant shift in the bioluminescence across the investigated pH range, which was similar to the behaviour of x12, WT and UG with this substrate (Table 4).


Kinetic parameters were derived from flash heights (Hanes, 1932) by varying ALH2 concentration in the range of 0.1-600 M used with 3mM ATP to saturate (see Materials and Methods). Specific activity was determined in the presence of 20.1 pmol enzyme by integrating light in 20 ms pulses over 12 s. PMT voltage 550 mV.

Table 5. Kinetic parameters of WT, Fluc mutants and UG with ALH2

In terms of kinetic parameters, if compared to the WT, the effect of mutations in x4 on Km and Kcat with LH2 differed significantly from mutations in both x2 and x5 subsets (Table 3). Mutations in x4 increased the Km, while in two other subsets decreased it, and reduced the Kcat, while in the other subsets hardly any change was observed. These results correlated well with the previously reported in literature (Law *et al.*, 2006; Tisi *et al.*, 2002; Prebble *et al.*, 2001; Branchini *et al.*, 2007). Similar trends were observed in the kinetic parameters for ALH2 (Table 5). Mutations in x4 significantly increased its Km for this substrate (as in x12), while two other mutants slightly decreased it. Kinetics of the x4 mutant with ALH2 exhibited slower rise and slower decay, reminiscent of the kinetics of x12, whereas x2 and x5 displayed a sharp flash that strongly resembled the WT.

The differences in kinetic parameters and flash kinetics suggested that an undesirable effect of mutations was common to x4 and x12 mutants, but not to x2 or x5, on catalysis with both LH2 and ALH2. It was hypothesised that reversion of buried x12 Fluc mutations T214C and F295L might enhance its catalytic properties.

#### **3.5 Properties of T214 and F295 revertant mutants of x12 Fluc**

x12 Fluc mutations T214C and F295L were individually reverted and their bioluminescent properties, catalytic parameters and thermal stability investigated. Bioluminescence spectra of both revertants were measured and found to match those of x12 Fluc with both substrates (Table 6). F295 revertant showed no bathochromic shift with either substrate and retain the pH-tolerance previously observed in x12 Fluc. T214 revertant had a flash kinetics similar to

Development of a pH-Tolerant Thermostable

properties.

50ºC.

**3.6** *In vivo* **imaging** 

**4. Conclusion** 

*Photinus pyralis* Luciferase for Brighter *In Vivo* Imaging 131

thermostability properties of x12 Fluc, but in the framework of this enzyme contributes to reduced activity and catalytic efficiency with both LH2 and ALH2. F295 revertant was hereafter named x11 Fluc. Cumulative addition of multiple phenotype-inducing mutations may enhance desired mutant properties, but some additions can clash and unduly disrupt

Flash-based activity with LH2 was compared in aliquots of 0.5 M enzyme incubated at set

mM ATP solution in TEM buffer (pH 7.8) onto 40 µl luciferase mutants. Solid lines: 40º

Fig. 4. Thermal inactivation of x12 Fluc and revertants at 40ºC and 50ºC.

mutant in mammalian cells is underway and will be published shortly.

temperatures over time. Samples were equilibrated to RT before dispensing 260 l of 70 M LH2 and 1

Human retrovirus encoding *Ppy* Fluc or x11 Fluc genes bearing a myc tag was used to transduce Raji cells, which were sorted to the same expression levels (Fig. 5A) by flow cytometric sorting for myc tag staining and cultured at 37ºC. As an example of *in vivo* imaging using x11 Fluc, one million Raji cells expressing similar levels of WT or x11 Flucs were injected into the tail veins of immunocompromised Beta2m-mice to induce systemic lymphoma (Chao *et al.*, 2011) and imaged after i.p. administration of LH2. Images revealed light signals predominantly from brain, spine and hips. x11 Fluc appeared very bright *in vivo* (Fig. 5B) because of its high thermostability and pH tolerance along with favourable kinetic parameters demonstrated in characterisation. It is expected to perform equally well under the changing physiological pH conditions and in combination with aminoluciferin used for imaging of protease assays *in vivo*. Further work to codon-optimise and test this

In the present study, we describe the construction and characterisation of the firefly luciferase mutant 12 Fluc based on the mutations previously identified as increasing thermostability of the enzyme. Detailed characterization of its bioluminescent and biochemical properties revealed that it is the only luciferase mutant reported to exhibit 80 % of total activity across a wide pH range of 6.6 - 8.8 covering physiologically

C, dashed lines:


x12 Fluc with both substrates, but F295 had a faster and brighter flash with LH2, more resembling that of the WT (not shown).

Details are as in Table 2 and 4.

Table 6. Bioluminescent spectra of x12 Fluc and its revertants at different pH with LH2 and ALH2

T214 revertant exhibited Kms and Kcats similar to x12 Fluc for LH2 and its Kcat/Km ratio was not significantly different from that of x12 Fluc (Table 7). The Km of T214 for ALH2 was slightly lower than x12 Fluc and Kcat elevated, resulting in the higher Kcat/Km ratio. T214 revertant had similar specific activities to x12 Fluc with both substrates. However, the Km of F295 revertant for both LH2 and ALH2 was lower and Kcat was higher than that of x12 Fluc resulting in a much higher Kcat/Km ratio. Specific activity of F295 revertant with both substrates was two-fold brighter than that of x12 Fluc. As a result of these findings F295 revertant afforded higher sensitivity to the flash-based detection of LH2 and ALH2 than x12 Fluc. It is possible to conclude that in x12 Fluc residue F295 contributes much more to catalysis than T214 and its reverse-mutation may significantly improve the performance of x12 Fluc providing it has no negative effect on the thermostability.


Details are as in Table 3 and 5.Specific activity was measured at pH 7.7 with LH2 and at pH 8.2 with ALH2. Table 7. Kinetic parameters of x12 Fluc revertants with LH2 and ALH2

Although T214C and F295L are thermostabilising mutations (Law *et al.*, 2006; Tisi *et al.*, 2002), revertants displayed resistance to thermal inactivation similar to that of x12 Fluc at 40ºC and 50ºC (Fig. 4). Therefore, the mutation F295L is not essential for practically useful

x12 Fluc with both substrates, but F295 had a faster and brighter flash with LH2, more

**x12 LH2** 565 557 558 66 66 67

**T214 LH2** - 560 - - 66 -

**F295 LH2** 563 559 560 71 66 70

Table 6. Bioluminescent spectra of x12 Fluc and its revertants at different pH with LH2 and

T214 revertant exhibited Kms and Kcats similar to x12 Fluc for LH2 and its Kcat/Km ratio was not significantly different from that of x12 Fluc (Table 7). The Km of T214 for ALH2 was slightly lower than x12 Fluc and Kcat elevated, resulting in the higher Kcat/Km ratio. T214 revertant had similar specific activities to x12 Fluc with both substrates. However, the Km of F295 revertant for both LH2 and ALH2 was lower and Kcat was higher than that of x12 Fluc resulting in a much higher Kcat/Km ratio. Specific activity of F295 revertant with both substrates was two-fold brighter than that of x12 Fluc. As a result of these findings F295 revertant afforded higher sensitivity to the flash-based detection of LH2 and ALH2 than x12 Fluc. It is possible to conclude that in x12 Fluc residue F295 contributes much more to catalysis than T214 and its reverse-mutation may significantly improve the performance of

x12 Fluc providing it has no negative effect on the thermostability.

**Km, µM Kcat x 108,** 

**RLU/s**

Table 7. Kinetic parameters of x12 Fluc revertants with LH2 and ALH2

**x12 LH2** 6.8+0.2 50+3 74+2 3.56+0.04 <sup>100</sup> **ALH2** 6.5+0.2 38+4 59+9 1.46+0.01 40

**T214 LH2** 6.2+0.5 49+3 83+10 3.22+0.03 <sup>79</sup>

**F295 LH2** 3.7+0.2 133+6 366+22 8.03+0.09 <sup>197</sup>

**ALH2** 4.9+0.5 66+2 138+8 2.37+0.05 58

**ALH2** 3.6+0.2 82+2 231+9 3.23+0.02 79 Details are as in Table 3 and 5.Specific activity was measured at pH 7.7 with LH2 and at pH 8.2 with ALH2.

Although T214C and F295L are thermostabilising mutations (Law *et al.*, 2006; Tisi *et al.*, 2002), revertants displayed resistance to thermal inactivation similar to that of x12 Fluc at 40ºC and 50ºC (Fig. 4). Therefore, the mutation F295L is not essential for practically useful

**ALH2** 596 592 591 72 72 71

**ALH2** - 591 - - 72 -

**ALH2** 593 592 591 70 72 73

**Michaelis-Menten kinetic Specific activity**

**Kcat/ Km x 1013, RLU/ s x µM** 

**x 107, RLU/mg**  **% of x12 with LH2 at pH 7.8**

**Bioluminescent spectra λmax (nm) Half-bandwidth (nm) pH pH 6.5 7.8 8.8 6.5 7.8 8.8** 

resembling that of the WT (not shown).

**Mutants Substrate** 

Details are as in Table 2 and 4.

**Mutants Substrate**

ALH2

thermostability properties of x12 Fluc, but in the framework of this enzyme contributes to reduced activity and catalytic efficiency with both LH2 and ALH2. F295 revertant was hereafter named x11 Fluc. Cumulative addition of multiple phenotype-inducing mutations may enhance desired mutant properties, but some additions can clash and unduly disrupt properties.

Flash-based activity with LH2 was compared in aliquots of 0.5 M enzyme incubated at set temperatures over time. Samples were equilibrated to RT before dispensing 260 l of 70 M LH2 and 1 mM ATP solution in TEM buffer (pH 7.8) onto 40 µl luciferase mutants. Solid lines: 40º C, dashed lines: 50ºC.

Fig. 4. Thermal inactivation of x12 Fluc and revertants at 40ºC and 50ºC.
