US 20060183212 A1
Luciferase enzymes with greatly increased thermostability, e.g., at least half lives of 2 hours at 50° C., cDNAs encoding the novel luciferases, and hosts transformed to express the luciferases, are disclosed. Methods of producing the luciferases include recursive mutagenesis. The luciferases are used in conventional methods, some employing kits.
1. A thermostable beetle luciferase that retains at least 50% activity in aqueous solution for at least 2 hours at a temperature of about 50° C., wherein the thermostable beetle luciferase has a plurality of amino acid substitutions at residues corresponding to residue 2, 27, 91, 144, 173, 183, 204, 220, 221, 262, 294, 353, 354, 356, 386, 394, 399, 412, 413 or 499 of SEQ ID NO:37, wherein the plurality of substitutions is associated with enhanced thermostability relative to a corresponding beetle luciferase without the plurality of substitutions.
2. The thermostable luciferase of
3. The thermostable luciferase of
4. The thermostable luciferase of
5. The thermostable luciferase of
6. The thermostable luciferase of
7. The thermostable luciferase of
8. The thermostable luciferase of
9. The thermostable luciferase of
10. The thermostable luciferase of
11. The thermostable luciferase of
12. The thermostable luciferase of
13. The thermostable luciferase of
14. The thermostable luciferase of
15. The thermostable luciferase of
16. The thermostable luciferase of
17. The thermostable luciferase of
18. The thermostable luciferase of
19. The thermostable luciferase of
20. The thermostable luciferase of
21. The thermostable luciferase of
22. The thermostable luciferase of
23. The thermostable luciferase of
24. A solid substrate comprising the luciferase of
25. A fusion protein comprising the luciferase of
26. A kit comprising: a container comprising the luciferase of
27. The kit of
28. The kit of
29. The kit of
30. The kit of
31. The kit of
32. The kit of
33. The kit of
34. The kit of
35. The kit of
36. The kit of
37. The kit of
38. A fusion protein comprising the luciferase of
39. A fusion protein comprising the luciferase of
40. A thermostable beetle luciferase that retains at least 50% activity in aqueous solution for at least 2 hours at a temperature of about 50° C., wherein the thermostable beetle luciferase has a plurality of amino acid substitutions relative to a corresponding beetle luciferase which does not retain at least 50% activity in aqueous solution for at least 2 hours at a temperature of about 50° C., wherein at least three substitutions are associated with increased thermostability, and wherein at least one substitution is to a consensus amino acid identified by comparing a plurality of wild-type beetle luciferases selected from the group consisting of Luciola cruciata luciferase, Luciola lateralis luciferase, Pyrophorus plagiophthalamus luciferase, Pyrocoelia miyako luciferase, Lampyris noctiluca luciferase, Photuris pennsylvanica luciferase, Phengodes sp. luciferase and Photinus pyralis luciferase.
41. The thermostable beetle luciferase of
This application is a divisional of U.S. patent application Ser. No. 10/378,168, filed on Feb. 23, 2003; which is a continuation of U.S. patent application Ser. No. 09/396,154, filed on Sep. 15, 1999 (now U.S. Pat. No. 6,602,677); which is a continuation-in-part application of U.S. application Ser. No. 09/156,946, filed Sep. 18, 1998 (abandoned), and of PCT/US98/19494, filed Sep. 18, 1998, both of which claim the benefit of U.S. application Ser. No. 60/059,379, filed Sep. 19, 1997; the disclosures of which are incorporated by reference herein.
The invention was made with grants from the Government of the United States of America (grants 1R43 GM506 23-01 and 2R44 GM506 23-02 from the National Institutes of Health and grants ISI-9160613 and III-9301865 from the National Science Foundation). The Government may have certain rights to the invention.
The invention is directed to mutant luciferase enzymes having greatly increased thermostability compared to natural luciferases or to luciferases from which they are derived as measured, e.g., by half-lives of at least 2 hours at 50° C. in aqueous solution. The invention includes mutant luciferase enzymes that are resistant to inhibition by a substrate inhibitor, e.g., a substrate analog. The invention is also drawn to polynucleotides encoding the novel luciferases, and to hosts transformed to express the luciferases. The invention is further drawn to methods of producing luciferases with increased thermostability and the use of these luciferases in any method in which previously known luciferases are conventionally employed. Some of the uses employ kits. The invention also provides a method of producing a polynucleotide sequence encoding an enzyme that is resistant to inhibition by an inhibitor, and a method which yields a polynucleotide sequence encoding an enzyme having enhanced enzymological properties.
Luciferases are defined by their ability to produce luminescence. Beetle luciferases form a distinct class with unique evolutionary origins and chemical mechanisms (Wood, 1995).
Although the enzymes known as beetle luciferases are widely recognized for their use in highly sensitive luminescent assays, their general utility has been limited due to low thermostability. Beetle luciferases having amino acid sequences encoded by cDNA sequences cloned from luminous beetles are not stable even at moderate temperatures. For example, even the most stable of the luciferases, LucPpe2, obtained from a firefly, has very little stability at the moderate temperature of 37° C. Firefly luciferases are a sub-group of the beetle luciferases. Historically, the term “firefly luciferase” referred to the enzyme LucPpy from a single species Photinus pyralis (Luc+ is a mutant version of LucPpy, see U.S. Pat. No. 5,670,356).
Attempts have been reported to mutate natural cDNA sequences encoding luciferase and to select mutants for improved thermostability (White et al., 1994; from P. pyralis, and Kajiyama and Nekano, 1993; from Luciola lateralis.) However, there is still a need to improve the characteristics and versatility of this important class of enzymes.
The invention is drawn to novel and remarkably thermostable luciferases, including luciferase enzymes with half-lives of at least 2 hours at 50° C., or at least 5 hours at 50° C., in an aqueous solution. As described hereinbelow, after 2 hours at 50° C. in an aqueous solution, a thermostable luciferase of the invention lost less than 5% luminescence activity. The mutant luciferases of the present invention display remarkable and heretofore unrealized thermostability at 22° C. in an aqueous solution and at temperatures at least as high as 60° C. in an aqueous solution. For example, the luciferases of the invention are thermostable for at least 10 hours at 50° C.; for at least 2 hours, preferably at least 5 hours, more preferably at least 10 hours, and even more preferably at least 24 hours, at 60° C.; and/or for at least 100 days, preferably at least 200 days, more preferably at least 500 days, and even more preferably at least 800 days, at 22° C., in aqueous solution. For example, after 30 days at 22° C. in an aqueous solution, a thermostable luciferase of the invention lost less than 5% luminescence activity. Preferably, the thermostable luciferases of the invention have enhanced luminescence intensity, enhanced signal stability, enhanced substrate utilization, and/or decreased Km, relative to a reference, e.g., a native wild-type, luciferase. The invention is further directed to the mutant luciferase genes (e.g., cDNA or RNA) which encode the novel luciferase enzymes. The terminology used herein is, e.g., for the mutants isolated in experiment 90, plate number 1, well B5, the E. coli strain is 90-1B5, the mutant gene is luc90-1B5, and the mutated luciferase is Luc90-1B5.
As defined herein, a “thermostable” enzyme, e.g., a luciferase, or an enzyme which has “thermostability”, is an enzyme which under certain conditions, e.g., at certain temperature, in aqueous solution and/or for certain periods of time, has an increased retention of activity relative to a reference enzyme. For example, for a thermostable luciferase, a reference luciferase may be native wild-type luciferase or recombinant wild-type luciferase. Preferably, for beetle luciferases, the activity is luminescence under conditions of saturation with luciferin and ATP. One measure of thermostability of an enzyme is the half-life of the enzyme in an aqueous solution (the time over which 50% of the activity is lost) at a stated temperature.
The invention further encompasses expression vectors and other genetic constructs containing the mutant luciferases, as well as hosts, bacterial and otherwise, transformed to express the mutant luciferases. The invention is also drawn to compositions and kits which contain the novel luciferases, and use of these luciferases in any methodology where luciferases are employed.
Various means of random mutagenesis were applied to a luciferase gene (nucleotide sequence), most particularly gene synthesis using an error-prone polymerase, to create libraries of modified luciferase genes. This library was expressed in colonies of E. coli and visually screened for efficient luminescence to select a subset library of modified luciferases. Lysates of these E. coli strains were then made, and quantitatively measured for luciferase activity and thermostability. From this, a smaller subset of modified luciferases was chosen, and the selected mutations were combined to make composite modified luciferases. New libraries were made from the composite modified luciferases by random mutagenesis and the process was repeated. The luciferases with the best overall performance were selected after several cycles of this process.
Methods of producing improved luciferases include directed evolution using a polynucleotide sequence encoding a first beetle luciferase as a starting (parent) sequence, to produce a polynucleotide sequence encoding a second luciferase with increased thermostability, compared to the first luciferase, while maintaining other characteristics of the enzymes. A cDNA designated lucPpe2 encodes a firefly luciferase derived from Photuris pennsylvanica that displays increased thermostability as compared to the widely utilized luciferase designated LucPpy from Photinus pyralis. The cDNA encoding LucPpe2 was isolated, sequenced and cloned (see Leach et al., 1997). A mutant of this gene encodes a first luciferase LucPpe2 [T249M]. However, the methods of the invention are not limited to use with a polynucleotide sequence encoding a beetle luciferase, i.e., the methods of the invention may be employed with a polynucleotide sequence encoding other enzymes.
In an embodiment of a mutant luciferase, the amino acid sequence is that of LucPpe2 shown in
Diluted extracts of recombinant E. coli that expressed mutant luciferases made by the methods of the invention were simultaneously screened for a plurality of characteristics including light intensity, signal stability, substrate utilization (Km), and thermostability. A fully automated robotic system was used to screen large numbers of mutants in each generation of the evolution. After several cycles of mutagenesis and screening, thereby creating mutant libraries of luciferases, an increased thermostability compared to LucPpe2 [T249M] of about 35° C. was achieved for clone Luc90-1B5 which also essentially maintained enzymatic activity (there was only negligible loss in activity of 5%) when kept in aqueous solution over 2 hours at 50° C., 5 hours at 65° C., or over 6 weeks at 22° C.
Mutant luciferases of the present invention display increased thermostability for at least 2 hours at 50° C., preferably at least 5 hours at 50° C., and in the range of at least 2 hours, preferably at least 24 hours, and more preferably at least 50 hours, at temperatures including 50° C., 60° C., and/or at temperatures up to 65° C. In particular, the present invention comprises thermostable mutant luciferases which, when solubilized in a suitable aqueous solution, have a thermostability greater than about 2 hours at about 50° C., more preferably greater than about 10 hours at 50° C., and more preferably still greater than 5 hours at 50° C. The present invention also comprises mutant luciferases which, when solubilized in a suitable aqueous solution, have a thermostability greater than about 2 hours, more preferably at least 5 hours, even more preferably greater than about 10 hours, and even more preferably still greater than about 24 hours, at about 60° C. The present invention further comprises mutant luciferases which when solubilized in a suitable aqueous solution have a thermostability greater than about 3 months at about 22° C., and more preferably a thermostability of at least 6 months at 22° C. An embodiment of the invention is a luciferase mutant having thermostability at 65° C., wherein a loss of activity of about 5-6% was found after 6 hours (equivalent to a half-life of 2 days). The half-lives of enzymes from the most stable clones of the present invention, extrapolated from data showing small relative changes, is greater than 2 days at 65° C. (corresponding to 6% loss over 6 hours), and about 2 years at 22° C. (corresponding to 5% loss over 9 weeks).
In particular, the invention comprises luciferase enzymes with embodiments of amino acid sequences disclosed herein (e.g., mutant luciferases designated Luc49-7C6, Luc78-0B10; Luc90-1B5, Luc133-1B2, and Luc146-1H2, as well as all other beetle luciferases that have thermostability as measured in half-lives of at least 2 hours at 50° C. The invention also comprises mutated polynucleotide sequences encoding luciferase enzymes containing any single mutation or any combination of mutations of the type which convert an amino acid of the reference beetle luciferase into a consensus amino acid. Conserved amino acids are defined as those that occur at a particular position in all sequences in a given set of related enzymes. Consensus amino acids are defined as those that occur at a particular position in more than 50% of the sequences in a given set of enzymes. An example is the set of beetle luciferase sequences shown in
Nucleotide sequences encoding beetle luciferases are aligned in
The invention also provides methods to prepare enzymes having one or more desired properties, e.g., resistance to inhibition by a substrate analog of the enzyme or enhanced enzymological properties. The method comprises selecting at least one isolated polynucleotide sequence encoding an enzyme with the desired property, e.g., an enzymological property, from a first population of mutated polynucleotide sequences. The selected, isolated polynucleotide sequence is then mutated to yield a second population of mutated polynucleotide sequences. Preferably, a mixture of selected isolated polynucleotide sequences are mutated to yield a second population of mutated polynucleotide sequences. The process may be repeated until a further polynucleotide sequence is obtained, e.g., selected and/or isolated, which further polynucleotide sequence encodes an enzyme which has at least one of the desired properties. As used herein, the terms “isolated and/or “purified” refer to in vitro isolation of a RNA, DNA or polypeptide molecule from its natural cellular environment, and from association with other components of the cell, such as nucleic acid or polypeptide, e.g., so that it can be sequenced, replicated, and/or expressed.
The invention relates to enzymes, e.g., beetle luciferases, that are created by mutations made in the encoding genes, generally by recursive mutagenesis, which mutated enzymes have one or more desired properties, for example, increases thermostability, increased resistance to inhibitors, and/or enhanced enzymological properties, relative to a reference enzyme, e.g., the wild-type enzyme. The polynucleotide sequence which encodes an enzyme of the invention comprises mutations that encode a plurality of amino acid substitutions relative to the polynucleotide sequence encoding the enzyme from which the enzyme of the invention was derived. For example, the invention relates to enzymes, e.g., luciferases, that are thermostable. The increased thermostability allows storage of enzymes such as luciferases without altering its activity, and improves reproducibility and accuracy of assays using the mutated luciferases. Thus, one embodiment of the invention comprises isolated polynucleotide sequences (cDNAs) which encode mutant luciferases with increased thermostability, vectors containing the polynucleotide sequences, and hosts transformed to express the polynucleotide sequences. Table 1 shows results of about 250 clones and characteristics of the luciferases from the clones including thermostability. The invention also encompasses the use of the mutant luciferases in any application where luciferases are conventionally utilized, and kits useful for some of the applications.
Unexpectedly, beetle luciferases with the sought after improved thermostability were achieved in the present invention through a process of recursive mutagenesis and selection (sometimes referred to as “directed evolution”). A strategy of recursive mutagenesis and selection is an aspect of the present invention, in particular the use of multi-parameter automated screens. Thus, instead of screening for only a single attribute such as thermostability, simultaneous screening was done for additional characteristics of enzyme activity and efficiency. By this method, one property is less likely to “evolve” at the expense of another, resulting in increased thermostability, but decreased activity, for example.
Table 1 presents examples of parameter values (Li, Tau, Km and S, see below) derived from experiments using different luciferases as starting (parent) sequences. The subtitles refer to designations of the temperature at which the enzyme stability was measured and the starting luciferase, e.g., Luc39-5B10 at 51° C. and so forth. All parameters in each experiment are recorded as relative values to the respective starting sequence, e.g., the parameter values for the starting sequence in any experiment equals “1.” (See Example 2 herein for definitions.)
Thermostability has evolved in nature for various enzymes, as evidenced by thermostable isozymes found in thermophilic bacteria. Natural evolution works by a process of random mutagenesis (base substitutions, gene deletions, gene insertions), followed by selection of those mutants with improved characteristics. The process is recursive over time. Although the existence of thermostable enzymes in nature suggests that thermostability can be achieved through mutagenesis on an evolutionary scale, the feasibility of achieving a given level of thermostability for a particular class of enzymes by using short term laboratory methods was unpredictable. The natural process of evolution, which generally involves extremely large populations and many millions of generations and genes, by mutation and selection cannot be used to predict the capabilities of a modern laboratory to produce improved genes by directed evolution until such mutants are produced.
After such success, because the overall three-dimensional structure of all beetle luciferases are quite similar, having shown it possible for one member of this class makes it predictable that high thermostability can be achieved for other beetle luciferases by similar methods.
Enzymes belong to different structural classes based on the three-dimensional arrangement of secondary elements such as helices, sheets, and loops. Thermostability is determined by how efficiently the secondary elements are packed together into a three-dimensional structure. For each structural class, there also exists a theoretical limit for thermostability. All beetle luciferases belong to a common structural class as evident by their common ancestry (
The application of a limited number of amino acid substitutions by mutagenesis is unlikely to significantly affect the overall three-dimensional architecture (i.e., the structural class for mutant luciferases is not expected to change.) Because the theoretical limit for thermostability for any structural class is not known, the potential thermostability of beetle luciferases was not known until demonstrations of the present invention.
A priori difficulties in achieving the goals of the present invention included:
1. The types of mutations which can be made by laboratory methods are limited.
2. The number of possible luciferase mutants that can be screened is limited.
3. The probability of finding functional improvement requiring cooperative mutations is rare (the probability of finding a specific cooperative pair is 1 out of 108 clones).
Thus, even if the theoretical limits of thermostability were known, because only a very small number of the possible luciferase mutants can be screened, the a priori probability of finding such a thermostable enzyme was low.
However, the present invention now shows that it is possible and feasible to create novel beetle luciferases having high thermostability.
To compensate for a statistical effect caused by the large number of deleterious random mutations expected relative to the beneficial mutations, methods were employed to maximize assay precision and to re-screen previously selected mutations in new permutations. Among the methods for maximizing assay precision were closely controlling culture conditions by using specialized media, reducing growth rates, controlling heat transfer, and analyzing parameters from mid-logarithmic phase growth of the culture. The robotic processes maximized for precision include controlling mixing, heat transfers, and evaporation of samples in the robotic screening process; and normalizing data to spatially distributed control samples. New permutations of the selected mutations were created by a method of DNA shuffling using proof-reading polymerases.
The difficulty in predicting the outcome of the recursive process is exemplified by the variable success with the other characteristics of luciferase that were also selected for. Although the primary focus was on the enzyme thermostability, selection for mutants producing brighter luminescence, with more efficient substrate utilization, and an extended luminescence signal was also attempted. The definitions are given by equations herewith. The selection process was determined by changes relative to the parent clones for each iteration of the recursive process. The amount of the change was whatever was observed during the screening process. The expression of luciferase in E. coli was relatively inefficient, for LucPpe2, compared to Luc+. Other luciferases varied (see
To improve the overall efficiency of substrate utilization, reduction in the composite apparent utilization constant (i.e., Km[ATP+luciferin]) for both luciferin and ATP was sought. Although there was an unexpected systematic change in each utilization constant (Km[ATP], Km[luciferin]), there was little overall change. Finally, the luminescence signal could only be moderately affected without substantially reducing enzyme efficiency. Thus, although the enzyme thermostability was greatly increased by methods of the present invention, other characteristics of the enzyme were much less affected.
Other aspects of the invention include methods that incorporate the thermostable luciferases, specifically beetle luciferases having high thermostability, as well as methods to prepare an enzyme, including a luciferase, having one or more desired properties, e.g., resistance to inhibition by a substrate inhibitor, or enhanced enzymological properties. Thus, the invention also provides a method to prepare an enzyme which has at least one enhanced enzymological property. From a population of polynucleotide sequences encoding the enzyme which is derived from a first polynucleotide sequence encoding the enzyme which is subject to mutation, at least one polynucleotide sequence encoding an enzyme which has the enhanced enzymological activity is selected and isolated. In one embodiment, oligonucleotide-mediated mutagenesis is then employed to introduce at least one codon which encodes a consensus amino acid to at least one of the selected, isolated polynucleotide sequences encoding the enzymes to yield a further polynucleotide sequence encoding the enzyme and having the codon which encodes the consensus amino acid, wherein the codon which is introduced is not present in the first polynucleotide sequence.
Production of Luciferases of the Present Invention
The method of making luciferases with increased thermostability is recursive mutagenesis followed by selection. Embodiments of the highly thermostable mutant luciferases of the invention were generated by a reiterative process of random point mutations beginning with a source nucleotide sequence, e.g., the lucPpe2 [T249M] cDNA. Recombination mutagenesis is a part of the mutagenesis process, along with point mutagenesis. Both recombination mutagenesis and point mutagenesis are performed recursively. Because the mutation process causes recombination of individual mutants in a fashion similar to the recombination of genetic elements during sexual reproduction, the process is sometimes referred to as the sexual polymerase chain reaction (sPCR). See, for instance, Stemmer, U.S. Pat. No. 5,605,793, issued Feb. 25, 1997.
Taking the lucPpe2 cDNA sequence as a starting point, the gene was mutated to yield mutant luciferases which are far more thermostable. A single point mutation to the lucPpe2 sequence yielded the luciferase whose sequence is depicted as T249M. This mutant is approximately 5 times brighter in vivo than that of lucPpe2, it was utilized as a template for further mutation. It was also used a baseline for measuring the thermonstability of the other mutant luciferases described herein.
Embodiments of Sequences of Luciferases of the Present Invention
Directed Evolution, A Recursive Process
Directed evolution is a recursive process of creating diversity through mutagenesis and screening for desired changes. For enzymological properties that result from the cumulative action of multiple amino acids, directed evolution provides a means to alter these properties. Each step of the process typically produces small changes in enzyme function, but the cumulative effect of many rounds of this process can lead to substantial overall change.
The characteristic, “thermostability” is a candidate for directed evolution because it is determined by the combined action of many of the amino acids making up the enzyme structure. Luminescence output and efficiency of substrate binding of the modified luciferase were also screened. This was to ensure that changes in thermostability did not also produce undesirable changes in other important enzymological properties.
Because the frequency of deleterious mutations is much greater than useful mutations, it is likely that undesirable clones are selected in each screen within the precision limits of the present invention. To compensate for this, the screening strategy incorporated multiple re-screens of the initially selected mutations. However, before re-screening, the selected mutations were “shuffled” to create a library of random intragenetic recombinations. This process allows beneficial mutations among different clones to be recombined together into fewer common coding sequences, and unlinks deleterious mutations to be segregated and omitted. Thus, although essentially the same set of selected mutations was screened again, they were screened under different permutations as a result of the recombination or shuffling.
Although results of each step of the evolutionary process were assayed by quantitative measurements, these measurements were mutually made in cell lysates rather than in purified enzymes. Furthermore, each step only measured changes in enzyme performance relative to the prior step, so global changes in enzyme function were difficult to judge.
Table 1 summarizes the characteristics of various clones obtained using the methods of the invention.
To evaluate the impact of directed evolution on enzyme function, clones from the beginning, middle and end of the process (Table 2) were purified and analyzed. The clones selected for this analysis were Luc[T249M], Luc49-7C6, and Luc78-0B10. Another clone, Luc90-1B5, created by a subsequent strategy of oligonucleotide-directed mutagenesis and screening was also purified for analysis.
The effect of directed evolution on thermostability was dramatic. At high temperatures, where the parent clone was inactivated almost instantaneously, the mutant enzymes from the related clones showed thermostability over several hours (see also Table 1, and
Kajiyama and Nakamo (1993) showed that a single amino acid substitution of A at position 217; to either I, L, or V, in the firefly luciferase from Luciola lateralis, resulted in a luciferase having increased thermostability. Substitution with leucine produced a luciferase that maintained 70% of its activity after incubation for 1 hour at 50° C. All of the enzymes of the present invention created through directed evolution, are much more stable than this L. lateralis mutant. One clone, Luc90-1B5, maintains 75% activity after 120 hours (5 days) incubation under similar conditions (50° C., 25 mol/L citrate pH 6.5, 150 mmol/L NaCl, 1 mg/mL BSA, 0.1 mmol/L EDTA, 5% glycerol). Interestingly, the LucPpe2 reported by Leach et al. already contains isoleucine at the homologous position described for the L. lateralis mutant.
Although thermostability was the characteristic of interest, clones were selected based on the other enzymological parameters in the screens. By selecting clones having greater luminescence expression, mutants were found that yielded greater luminescence intensity in colonies of E. coli. However, the process showed little ability to alter the kinetic profile of luminescence by the enzymes. This failure suggests that the ability to support steady-state luminescence is integral to the catalytic mechanism, and is not readily influenced by a cumulative effect of many amino acids.
Substrate binding was screened by measuring an apparent composite Km (see Example 2) for luciferin and ATP. Although the apparent composite Km remained relatively constant, later analysis showed that the individual Km's systematically changed. The Km for luciferin rose while the Km for ATP declined (Table 3). The reason for this change is unknown, although it can be speculated that more efficient release of oxyluciferin or luciferin inhibitors could lead to more rapid enzyme turnover.
Each point mutation, on its own, increases (to a greater or lesser extent) the thermostability of the mutant enzyme relative to the wild-type luciferase. The cumulative effect of combining individual point mutations yields mutant luciferases whose thermostability is greatly increased from the wild-type, often on the order of a magnitude or more.
The following examples illustrate the methods and compositions of the present invention and their embodiments.
An illustrative mutagenesis strategy is as follows: From the “best” wild-type luciferase clone, that is a clone with increased thermostability and not appreciably diminished values for other parameters, random mutagenesis was performed by three variations of error-prone PCR. From each cycle of random mutagenesis, 118 of the best clones were selected. DNA was prepared from these clones yielding a total of 54 clones. These clones represent new genetic diversity.
These 54 clones were combined and recombination mutagenesis was performed. The 18 best clones from this population were selected.
These 18 clones were combined with the 18 clones of the previous population and recombination mutagenesis was performed. From this screening, a new luciferase population of 18 clones was selected representing 6 groups of functional properties.
In this screening the new mutations of the selected 54 clones, either in their original sequence configurations or in recombinants thereof, were screened a second time. Each mutation was analyzed on the average about 10 times. Of the 90 clones used in the recombination mutagenesis, it was likely that at least 10 were functionally equivalent to the best clone. Thus, the best clone or recombinants thereof should be screened at least 100 times. Since this was greater than the number of clones used in the recombination, there was significant likelihood of finding productive recombination of the best clone with other clones.
Robotic Processing Methods.
Heat transfers were controlled in the robot process by using thick aluminum at many positions where the 96-well plates were placed by the robotic arm. For example, all shelves in the incubators or refrigerator were constructed from ¼ inch aluminum. One position in particular, located at room temperature, was constructed from a block of aluminum of dimensions 4.5×7×6.5 inches. When any 96-well plate was moved from a high temperature (e.g., incubators) or low temperature (e.g., refrigerator) to a device at room temperature, it was first placed on the large aluminum block for temperature equilibration. By this means, the entire plate would rapidly reach the new temperature, thus minimizing unequal evaporation for the various wells in the plate due to temperature differences. Heat transfers in a stack of 96-well plates placed in an incubator (e.g., for overnight growth of E. coli) were controlled by placing 1 mm thick sheets of aluminum between the plates. This allowed for more efficient heat transfer from the edges of the stack to the center. Mixing in the robotic process was controlled by having the plate placed on a shaker for several second after each reagent addition.
Please refer to
1. Culture Dilution Method.
A plate with lid (Falcon 3075) containing cells (E. coli JM109) is placed on a shaker and mixed for 3-5 minutes.
A plate (with lid) is obtained from a carousel and placed in the reagent dispenser. 180 μl of media (M9 minimal media) is added after removing the lid and placing on the locator near the pipetter. The plate is then placed in the pipetter.
The plate on the shaker is placed in the pipetter, and the lid removed and placed on the locator. Cells are transferred to the new plate using pipetting procedure (see “Dilution of Cells into New Cell Plate”).
The lids are replaced onto both plates. The new plate is placed in the refrigerator and the old plate is returned to the carousel.
2. Luminescence Assay Method.
A plate containing cells is retrieved from the carousel and placed on the shaker for 3-5 minutes to fully mix the cells. The cells tend to settle from solution upon standing.
To measure Optical Density (O.D.), the plate is moved from the shaker to the locator near the luminometer; the lid is removed and the plate placed into the luminometer. The O.D. is measured using a 620 nm filter.
When it is finished, the plate is then placed in the refrigerator for storage.
The above steps are completed for all plates before proceeding with subsequent processing.
To prepare a cell lysate, the plate of cells is first retrieved from the refrigerator and mixed on the shaker to resuspend the cells. A new plate from the carousel without a lid is placed in the reagent dispenser and 20 μl of Buffer A is added to each well. This is placed in the pipetting station.
The plate of cells in the shaker is placed in the pipetting station. A daughter plate is prepared using pipetting procedure (see “Pipetting Cells into the Lysis Plate”) to prepare a daughter plate of cells.
After pipetting, the new daughter plate is placed on the shaker for mixing.
After mixing, the Lysate Plate is placed into a solid CO2 freezing station to freeze the samples. The plate is then moved to the thaw block to thaw for 10 minutes.
The plate is then moved to the reagent dispenser to add 175 μl of Buffer B, and then mixed on the shaker for about 15 minutes or more. The combination of the freeze/thaw and Buffer B will cause the cells to lyse.
A new plate with a lid from the carousel is used to prepare the dilution plate from which all assays will be derived. The plate is placed in the reagent dispenser and the lid removed to the locator near the pipetter. 285 μl of Buffer C is added to each well with the reagent dispenser, then the plate is placed in the pipetting station.
The Lysate Plate in the shaker is moved to the pipetting station and pipetting procedure (see “Dilution from Lysis Plate to Incubation Plate”) is used. After pipetting, the new daughter plate is placed on the shaker for mixing. The Lysate Plate is discarded.
Two white assay plates (Labsystems #9502887) are obtained from the plate feeder and placed in the pipetter. The incubation plate from the shaker is placed in the pipetter, and the lid removed and placed on the nearby locator. Two daughter plates are made using the pipetting procedure (see “Create Pair of Daughter Plates from Incubation Plate”). Afterwards, the lid is replaced on the parent plate, and the plate is placed in a high temperature incubator. [ranging from 31° C. to about 65° C. depending on the clone.]
One daughter plate is placed in the luminometer and the 1× assay method is used. After the assay, the plate is placed in the ambient incubator, and the second daughter plate is placed in the luminometer. For the second plate, the 0.02× assay method is used. This plate is discarded, and the first plate is returned from the incubator to the luminometer. The repeat assay method is used (i.e., no reagent is injected). Afterwards, the plate is again returned to the ambient incubator.
The above steps are completed for all plates before proceeding with processing.
To begin the second set of measurements, the plate from the high temperature incubator is placed in the shaker to mix.
The plate in the ambient incubator is returned to the luminometer and the repeat assay method is again used. The plate is returned afterwards to the ambient incubator.
Two white assay plates again are obtained from the plate feeder and placed in the pipetter. The plate on the shaker is placed in the pipetter, and the lid removed and placed on the nearby locator. Two daughter plates are again made using the pipetting procedure (see “Create Pair of Daughter Plates from Incubation Plate”). Afterwards, the lid is replaced on the parent plate, and the plate is returned to the high temperature incubator.
One daughter plate is placed in the luminometer and the 1× assay method is again used. The plate is discarded after the assay. The second daughter plate is then placed in the luminometer and the 0.06× assay method is used. This plate is also discarded.
The above steps are completed for all plates before proceeding with processing.
In the final set of measurements, the plate from the high temperature incubator is again placed in the shaker to mix.
The plate in the ambient incubator is returned to the luminometer and the repeat assay method is again used. The plate is discarded afterwards.
One white assay plate is taken from the plate feeder and placed in the pipetter. The plate from the shaker is placed in the pipetter, and the lid removed and placed on the nearby locator. One daughter plate is made using the pipetting procedure (see “Create Single Daughter Plate from Incubation Plate”). The lid is replaced on the parent plate and the plate is discarded.
The daughter plate is placed in the luminometer and the 1× assay method is used. The plate is discarded after the assay.
Buffers and Assay Reagents
Non-aseptic procedure using fixed tips
On the Pipetter Deck:
On the Pipetter Deck:
This procedure is done twice
On the Pipetter Deck:
On the Pipetter Deck:
Aseptic procedure using fixed tips
On the Pipetter Deck:
This procedure is used to prepare the cell plates used in the main analysis procedure. 180 μl of M9 minimal growth medium is added by the reagent dispenser to each of the new cell plates just prior to initiating the pipetting procedure. The dispenser is flushed with 75% isopropanol before priming with medium. The medium also contains selective antibiotics to reduce potential contamination.
A. 1× Assay Method
Five to seven nitrocellulose disks having 200-500 colonies per disk (1000-3500 colonies total) are screened per 2 microplates (176 clones) (Wood and DeLuca, 1987). The clones are screened at high temperatures using standard screening conditions.
Eight positions in each microplate are reserved from a reference clone using the “best” luciferase (the parent clone for random mutagenesis and codon mutagenesis). The positions of the reserved wells is shown as “X” below.
The reference clones are made by placing colonies from DNA transformed from the parent clone into the reference wells. To identify these wells prior to inoculation of the microplate, the wells are marked with a black marking pen on the bottom of each well.
Screening Selection Criteria
The following criteria were used for screening purposes. The temperature chosen for the enzyme stability parameter was such that the parent enzyme would decay 100 to 1000 fold over 10 hours (see Table 1). Criteria 1 is achieved manually; data for criteria 2-6 is generated by robotic analysis. For all criteria, the maximum value as described is selected.
6. Composite fitness. The values of criteria 2 through 5 are combined into a single composite value of fitness (or commercial utility). This value is based on a judgment of the relative importance of the other criteria. This judgment is given below:
The composite, C=Sum (criteria 2-5 weighted by relative value, e.g., more weight is on stability because that was a major goal).
Organize Data into SOL Database
Each file created by a luminometer (96 well, Anthos, Austria) represents the data from one microplate. These files are stored in the computer controlling the luminometer, and connected to the database computer by a network link. From each microplate of samples, nine microplates are read by the luminometer (the original microplate for optical density and eight daughter microplates for luminescence).
Ninety files are created in total; each containing data sets for 96 samples. Each data set contains the sample number, time of each measurement relative to the first measurement of the plate, luminometer reading, and background corrected luminometer reading. Other file header information is also given. The time that each microplate is read is also needed for analysis. This can be obtained from the robot log or the file creation time. A naming convention for the files is used by the robot during file creation that can be recognized by SQL (e.g. YYMMDDPR.DAT where YY is the year, MM is the month, DD is the day, P is the initial plate [0-9], and R is the reading [0-8]).
Data Reduction and Organization
Normalize luminescence data: For each measurement of luminescence in the eight daughter plates, the normalized luminescence is calculated by dividing the relative light units by the optical density of the original plate. If any value of normalized luminescence is less than zero, assign the value of 0.1 sL where sL is the standard deviation for measurements of normalized luminescence.
Calculate relative measurement time: For each normalized luminescence measurement, the time of the measurement is calculated relative to the first measurement of the sample. For example, the times of all luminescence measurements of sample B6 in plate 7 (i.e., 7:B06) are calculated relative to the first reading of 7:B06. This time calculation involves both the time when the plate is read and the relative time of when the sample is read in the plate.
Calculate enzyme stability (τ): For each sample, use linear regression to fit ln(L1×)=ln(L0)−(t/τ) using the three luminescence measurements with 1× substrate concentrations (Plates 1, 5, 8). Also calculate the regression coefficient.
Calculate substrate binding (Km:app,total): Using microplates from the first set of readings (Plates 1 and 2), calculate the L0.2×,rel by dividing measurements made with substrate concentrations of 0.02× by those of 1×. Similarly, calculate the L0.06×,rel using microplates of the second set of readings (Plates 5 and 6), by dividing measurements made with substrate concentrations of 0.06× by those of 1×.
For each sample, use linear regression to fit 1/L=(Km:app,total/Lmax:app)(1/[S])+(1/Lmax:app) using
Km:app,total is calculated as the slope/intercept. Also calculate the regression coefficient.
Calculate signal stability (S): For each sample, use linear regression to fit ln(L)=ln(L0)−(t/τ) using the four luminescence measurements of the initial microplate with 1× substrate concentrations (Plates 1, 3, 4, and 7). Also calculate the regression coefficient. From the calculated values of τ and L0, calculate the integral of luminescence by int(L)=τL0(1−exp(−tf/τ)), where tf is the average time of the last measurement (e.g., 15 hours). The signal stability is calculated as S=(1−int(L)/Litf)2, where Li is the initial measurement of normalized luminescence with 1× substrate concentration (Plate 1).
[Note: To correct for evaporation, an equation S=(1+K−int(L)/Litf)2, may be used where 1/K=2(relative change of liquid volume at tf).]
Calculate the reference value surfaces: A three dimensional coordinate system can be defined by using the grid positions of the samples within a microplate as the horizontal coordinates, and the calculated values for the samples (Li, τ, Km:app,total, or S) as the vertical coordinates. This three dimensional system is referred to as a “plate map”. A smooth surface in the plate maps representing a reference level can be determined by least squares fit of the values determined for the 8 reference clones in each microplate. For each of the 10 initial microplates of samples, respective reference surfaces are determined for the criteria parameters Li, τ Km:app,total, and S (40 surfaces total).
In the least squares fit, the vertical coordinates (i.e., the criteria parameters) are the dependent variables, the horizontal coordinates are the independent variables. A first order surface (i.e., z=ax+by+c) is fitted to the values of the reference clones. After the surface is calculated, the residuals to each reference clone are calculated. If any of these residuals is outside of a given cutoff range, the reference surface is recalculated with omission of the aberrant reference clone.
If a first order surface does not sufficiently represent the values of the reference clones, a restricted second order surface is used (i.e., z=a(x2+ky2)+bx+cy+d, where k is a constant).
Calculate the reference-normalized values: For the criteria parameter of each sample, a reference-normalized value is determined by calculating the ratio or inverse ratio with the respective reference value. The reference-normalized values are Li/Lir, τ/τr, Kmr/Km:app,total, and Sr/S, where reference values are calculated from the equations of the appropriate reference surface.
Calculate the composite scores: For each sample, calculate C=5(τ/τr)+2(Sr/S)+2(Kmr/Km:app,total)+(Li/Lir).
Determine subgroupings: For the criteria parameters Li, τ, Km:app,total, S, and C, delimiting values (i.e., bin sizes) for subgroupings are defined as gL, gτ, gKm, gS, and gC. Starting with the highest values for Li, τ, or C, or the lowest values of Km:app,total or S, the samples are assigned to bins for each criteria parameter (the first bin being #1, and so on).
Display sorted table of reference-normalized values: Present a table of data for each sample showing in each row the following data:
The table is sorted by the composite score (C).
Present Sorted Table of Criteria Parameters.
Present a table of data for each sample showing in each row the following data:
The table is sorted by the composite score (C); the reference clones are excluded from the table. Same entry coding by standard deviation as described above.
Present Sorted Table of Reference-Normalized Values
This is the same procedure as the final step of the data reduction procedure. The table will show:
The table is sorted by the composite score (C); the reference clones are excluded from the table. Same entry coding by standard deviation as described above.
Present Sorted Table of Criteria Parameters for Reference Clones
This is the same procedure as described above for criteria parameters, except for only the reference clones. The table will show:
The table is sorted by the composite score (C). Same entry coding by standard deviation as described above.
Present Sorted Table of Reference-Normalized Values
This is the same procedure as described above for reference-normalized values, except for only the reference clones. The table will show:
The table is sorted by the composite score (C). Same entry coding by standard deviation as described above.
Any table may be sorted by any entries as primary and secondary key.
Display Histogram of Table
For any table, a histogram of criteria parameter vs. bin number may be displayed for any criteria parameter.
Display Plate Map
For any plate, a plate map may be displayed showing a choice of:
For Li, τ, Km:app,total, and S, any entry value in a table may be selected to display the luminescence and optical density reading underlying the value calculation, and a graphical representation of the curve fit where appropriate. Preferably the equations involved and the final result and correlation coefficient will also be displayed.
Li or Li/Lr. Display the optical density and luminescence value from the chosen sample in Plate 0 and Plate 1.
τ or τ/τr. Display the optical density and luminescence value from the chosen sample in Plate 0, Plate 1, Plate 5, and Plate 8. Display graph of ln(L1×) vs. t, showing data points and best line.
S or Sr/S. Display the optical density and luminescence value from the chosen sample in Plate 0, Plate 1, Plate 3, Plate 4, and Plate 7. Display graph of ln(L) vs. t, showing data points and best line.
Km:app,total or Kmr/Km:app,total. Display the optical density and luminescence value from the chosen sample in Plate 0, Plate 1, Plate 2, Plate 5, and Plate 6. Display graph of 1/L vs. 1/[S], showing data points and best line.
The gene shown in
To eliminate the peroxisome targeting signal (SKL), the L was mutated to a STOP codon and the 3 codons immediately upstream were randomized according to the oligonucleotide mutagenesis procedure described herein. The mutagenic oligonucleotide designed to accomplish this also introduces a unique SpeI site to allow mutant identification without sequencing. The mutants were screened in vivo and 13 colonies picked, 12 of which contained the SpeI site.
To test if expression could be improved, the 3 codons immediately downstream from the initiation Met were randomized as described herein. The mutagenic oligo designed to accomplish this also introduces a unique ApaI site to allow mutant identification without sequencing. Seven clones were selected, and six of the isolated plasmids were confirmed to be mutants.
Shuffling of C- and N-Terminus Mutants
The C- and N-terminus mutagenesis were performed side-by-side. To combine the N- and C-terminus mutations, selected clones from each mutagenesis experiment were combined with the use of recombination mutagenesis according to the recombination mutagenesis protocol described herein. The shuffled mutants were subcloned into ampS pRAM backbone and screened in DH5 F'IQ (BRL; Hanahan, 1985). A total of 24 clones were picked, only 4 contained both the N- and C-terminus mutations. These 4 clones were used as templates for randomization of the cysteine positions in the gene.
Mutagenesis to Randomize Cysteine Positions/Random Mutagenesis and Recombination Mutagenesis in the Luc Gene
There are 7 cysteine positions in LucPpe2. It is known that these positions are susceptible to oxidation which could cause destabilization of the protein. Seven oligonucleotides were ordered to randomize the cysteine positions.
The oligonucleotides were organized into two groups based upon the conservation of cysteine in other luciferase genes from different families. Group 1 randomizes the conserved cysteine positions C-60, C-80, and C-162. Group 2 randomizes cysteines that are not strictly conserved at positions C-38, C-127, C-221, and C-257.
The four selected templates from the N- and C-terminus mutagenesis were sub-cloned into an ampicillin-sensitive backbone and single-stranded DNA was prepared for each of the templates. These templates were combined in equal amounts and oligonucleotide mutagenesis was completed as described herein. It was determined by plating an aliquot of the mutS transformation prior to overnight incubation that each of the 2 groups contained 2×104 independent transformants. MutS-DNA was prepared for the 2 groups and was then transformed into JM109 cells for screening. Mutants from group 1 were screened in vivo and picks were made for a full robotic run. Five clones were selected that had improved characteristics. Mutants from group 2 were screened in vivo and picks were made for a full robotic run. The temperature incubator on the robot was set at 33° C. for this set of experiments. Ten clones were selected that had improved characteristics. The fifteen best picks from both groups of the cysteine mutagenesis experiments were shuffled together as described herein and 18 of the best clones were selected after robotic processing.
The “best” clone from the above experiment (Luc31-1G8) was selected as a template for subsequent rounds of mutagenesis. (The high temperature robot incubator temperature was set to 42° C.) Another complete round of mutagenesis was completed.
The 18 best clones from the above mutagenesis were picked and clone (Luc39-5B10) was selected as the best clone and was used as a template for another round of mutagenesis. (The high temperature robot incubator temperature was set at 49° C.).
After this cycle, 6 of the best clones were selected for sequencing (the nucleotide sequence and inferred amino acid sequence of five of the clones is shown in
The randomization oligonucleotides were divided into 4 groups, and transformants from these experiments were picked and two robotic runs were completed. Ten clones were selected from the two experiments. (The high temperature robot incubator temperature on robot was set at 56° C.).
The best 10 picks from the above two experiments, and the best 18 picks from the previous population of clones were shuffled together (recombination mutagenesis protocol).
The 18 best clones were selected and clone Luc58-0A5 was determined to be the best clone. This clone was then used as a template for another round of mutagenesis. The high temperature robot incubator temperature was set at 58° C. Clone Luc71-504 was selected as a new lead clone and another round of mutagenesis was completed. Incubator set at 60° C.
The best 18 picks were selected. The nucleotide sequence and inferred amino acid sequence of 4 clones from experiment 78 are shown in
Twenty-three oligonucleotides were prepared to change 28 positions to consensus. All of the oligonucleotides were tested individually using oligonucleotide directed mutagenesis with single stranded DNA from clone luc78-0B10 as a template to determine which oligonucleotides gave an improvement in thermostability. Table 4 lists the mutagenic oligonucleotides.
Three oligonucleotide-directed mutagenesis experiments with clone luc78-0B10 as a template were completed. The oligonucleotides for these experiments were divided in the following manner:
Selections from the three experiments listed above were screened with the robotic screening procedure (experiment 84, see Table 1) using luc78-0B10 as a control. Selections from experiment 84 were recombined using the recombination mutagenesis procedure and then screened with the robotic screening procedure (experiment 85).
Single stranded DNA was prepared from three clones, luc85-3E12, luc85-4F12, luc85-5A4. The nucleotide sequence and inferred amino acid sequence of luc85-4F12 are shown in
In the first experiment, the three templates listed above from experiment 85 were combined and used as a templates for oligonucleotide-directed mutagenesis. All of the oligonucleotides were combined in one experiment and clones resulting from oligonucleotide-directed mutagenesis were screened using the robotic screening procedure as experiment 88. There were a low percentage of luminescent colonies that resulted from this experiment, so another oligonucleotide-directed mutagenesis experiment was completed in which the oligonucleotides were combined in the following groups:
It was discovered that samples from group b had a low number of luminescent colonies, and it was hypothesized that one of the oligonucleotides in group b was causing problems. Selections were made from all of the experiments with the exception of experiment b. Samples were then run through the robotic screening procedure (experiment 89). Selections from experiments 88 and 89 were shuffled together with the recombination mutagenesis protocol and were then screened with the robotic screening procedure (experiment 90).
Materials and Methods
A. Mutagenesis Protocol
The mutant luciferases disclosed herein were produced via random mutagenesis with subsequent in vivo screening of the mutated genes for a plurality of characteristics including light output and thermostability of the encoded luciferase gene product. The mutagenesis was achieved by generally following a three-step method:
Both the N-terminus and the C-terminus of the starting sequence were modified by oligonucleotide-directed mutagenesis to optimize expression and remove the peroxisomal targeting sequence. At the N-terminus, nine bases downstream of the initiation codon were randomized. At the C-terminus, nine bases upstream of the termination codon were randomized. Mutants were analyzed using an in vivo screen, resulting in no significant change in expression.
Six clones from this screen were pooled, and used to mutate the codons for seven cysteines. These codons were randomized using oligonucleotide-directed mutagenesis, and the mutants were screened using the robotic screening procedure. From this screen, fifteen clones were selected for directed evolution.
C. Generating and Testing Clones
Several very powerful and widely known protocols are used to generate and test the clones of the present invention. Unless noted otherwise, these laboratory procedures are well known to one of skill in the art. Particularly noted as being well known to the skilled practitioner is the polymerase chain reaction (PCR) devised by Mullis and various modifications to the standard PCR protocol (error-prone PCR, sPCR, and the like), DNA sequencing by any method (Sanger or Maxxam & Gilbert's methodology), amino acid sequencing by any method (e.g., the Edman degradation), and electrophoretic separation of polynucleotides and polypeptides/proteins.
D. Vector Design
A preferred vector (PRAM) (see
The pRAM vector contains a filamentous phage origin, f1, which is necessary for the production of single-stranded DNA.
Two SfiI sites flank the gene. These sites were designed by so that the gene to be subcloned can only be inserted in the proper orientation.
The vector contains a tac promoter.
Templates to be used for oligonucleotide mutagenesis contain a 4 base-pair deletion in the bla gene which makes the vector ampicillin-sensitive. The oligonucleotide mutagenesis procedure uses a mutant oligonucleotide as well as an ampicillin repair oligonucleotide that restores function to the bla gene. This allows for the selection of a high percentage of mutants. (If selection is not used, it is difficult to obtain a high percentage of mutants.)
E. Uses of Luciferases
The mutant luciferases of the present invention are suitable for use in any application for which previously known luciferases were used, including the following:
ATP Assays. The greater enzyme stability means that reagents designed for detection of ATP have a greater shelf-life and operational-life at higher temperatures (e.g., room temperature). Therefore, a method of detecting ATP using luciferases with increased thermostability is novel and useful.
Luminescent labels for nucleic acids, proteins, or other molecules. Analogous to advantages of the luciferases of the present invention for ATP assays, their greater shelf-life and operational-life is a benefit to the reliability and reproducibility of luminescent labels. This is particularly advantageous for labeling nucleic acids in hybridization procedures where hybridization temperatures can be relatively high (e.g., greater than 40° C.). Therefore, a method of labeling nucleic acids, proteins, or other molecules using luciferases of the present invention is novel and useful.
Genetic reporter. In the widespread application of luciferase as a genetic reporter, where detection of the reporter is used to infer the presence of another gene or process of interest, the increased thermostability of the luciferases provides less temperature dependence of its expression in living cells and in cell-free translations and transcription/translation systems. Therefore, a method using the luciferases of the present invention as genetic reporters is novel and useful.
Enzyme immobilization. Enzymes in close proximity to physical surfaces can be denatured by their interaction with that surface. The high density immobilization of luciferases onto a surface to provide strong localized luminescence is improved by using thermostable luciferases. Therefore, a method of immobilizing luciferases onto a solid surface using luciferases of the present invention is novel and useful.
Hybrid proteins. Hybrid proteins made by genetic fusion genes encoding luciferases and of other genes, or through a chemical coupling process, benefit by having a greater shelf-life and operational-life. Therefore, a method of producing hybrid proteins through genetic means or chemical coupling using the luciferases of the present invention is novel and useful.
High temperature reactions. The light intensity of a luciferase reaction increases with temperature until the luciferase begins to denature. Because the use of thermostable luciferases allows for use at greater reaction temperatures, the luciferases of the present invention are novel and useful for performing high temperature reactions.
Luminescent solutions. Luminescence has many general uses, including educational, demonstrational, and entertainment purposes. These applications benefit from having enzymes with greater shelf-life and operational-life. Therefore, a method of making luminescent solutions using the luciferases of the present invention is novel and useful.
F. Firefly Luciferase
The firefly luciferase gene chosen for directed evolution was LucPpe2 isolated from Photuris pennsylvanica. The luciferase was cloned from fireflies collected in Maryland by Wood et al. and later was independently cloned by Dr. Leach using fireflies collected in Oklahoma (Ye et al., 1997). A mutant of this luciferase (T249M) was made by Wood et al. and used in the present invention because it produced approximately 5-fold more light when expressed in colonies of E. coli.
Overview of Evolution Process: Directed evolution was achieved through a recursive process, each step consisting of multiple cycles of 1) creating mutational libraries of firefly luciferase followed by 2) screening the libraries to identify new mutant clones having a plurality of desired enzymological characteristics.
To begin the process, three mutational libraries were created using error-prone PCR (Fromant et al., 1995). Each library was screened first by visual evaluation of luminescence in colonies of E. coli (Wood and De Luca, 1987), and then by quantitative measurements of enzymological properties in E. coli cell lysates. Approximately 10,000 colonies were examined in the visual screen, from which 704 were selected for quantitative analysis. From each quantitative screen 18 clones were selected. The three sets of 18 clones each were pooled together, and a new mutational library was created using DNA shuffling to generate intragenetic recombinations (sPCR; Stemmer, 1994). The results were screened to yield another set of 18 clones. The entire process was completed by combining this set of 18 clones with 18 clones from the previous round of evolution, creating another mutational library by DNA shuffling, and screening as before.
Screening method: In the qualitative visual screen, colonies were selected only for their ability to sustain relatively bright luminescence. The thermal stability of the luciferase within the colonies of E. coli was progressively challenged in successive rounds of evolution by increasing the temperature of the screen. The selected colonies were inoculated into wells of 96-well plates each containing 200 μl of growth medium.
In the quantitative screens, lysates of the E. coli cultures were measured for 1) luminescence activity, 2) enzyme stability, 3) sustained enzymatic turnover, and 4) substrate binding.
“Luminescence activity” was measured as the ratio of luminescence intensity to the optical density of the cell culture.
“Enzyme stability” was determined by the rate of activity loss from cell lysates over 10 hours. In successive rounds of evolution the incubation temperature of the lysates was increased.
“Sustained enzymatic turnover” was determined by the rate of luminescence loss of a signal enzymatic reaction over 10 hours at room temperature.
“Substrate binding” was determined by the relative activity of the lysate when assayed with diluted substrate mixtures. Of these four parameters, the highest priority for selection was placed on thermostability.
Robotic Automation: Robotic automation was used in the quantitative screens to accurately perform the large number of required quantitative assays on the cultured cells. Overnight cultures were first diluted into fresh medium and grown for 3 hours to produce cultures in mid-log phase growth. The optical densities of each culture was then measured, and aliquots of the cultures were lysed by freeze/thaw and lysozyme. The resulting lysates were further diluted before analysis and incubated at elevated temperatures. Luminescence was measured from aliquots of the diluted lysates, taken at various times, and measured under various conditions as prescribed by the analytical method (see Example 2). Computer analysis of this data yielded the quantitative selection criteria described herein.
Summary of evolutionary progression: After mutagenesis of the N- and C-termini, and randomization of the cysteine codons, a pool of 15 clones was subjected to two rounds of directed evolution as described herein. Five of the 18 clones resulting from this process were sequenced to identify mutations. One of these clones designated, Luc49-7C6, was chosen for more detailed analysis and further mutagenesis. This clone contained 14 new amino acid substitutions compared to the luciferase Luc[T249M].
To assess the potential for other amino acid replacements at the sites of these substitutions, oligonucleotide-directed mutagenesis was used to randomize these codons. The resulting clones were screened as described herein, and 18 selected clones were used to initiate two new rounds of directed evolution. Of the 18 clones resulting from this second set of rounds, the clone designated Luc78-0B10 was chosen for additional study and mutagenesis. This clone encoded a luciferase that contained 23 new amino acid substitutions compared to Luc[T249M].
Using oligonucleotide directed mutagenesis with Luc78-0B10 as the template, codons were selected for substitution to consensus amino acids previously known among beetle luciferases. Selections from this mutagenesis experiment were shuffled together and three clones, determined to be the most stable were then used as templates for oligonucleotide mutagenesis to improve codon usage in E. coli. A clone designated Luc90-1B5 selected from this experiment, contained 34 amino acid substitutions relative to Luc[T249M] (see
Protein purification: Four mutants that are described herein (Luc[T249M], Luc49-7C6, Luc78-0B10, and Luc90-1B5) were purified using a previously published procedure (Hastings et al., 1996).
Enzymological characterization: Purified proteins were diluted in 25 mmol/L HEPES pH 7.8, 150 mmol/L NaCl, 0.1 mmol/L EDTA, 1 mg/ml BSA. Enzyme stability was determined from diluted proteins incubated at different temperatures, and aliquots were removed at different time points. A linear regression of the natural log of the luminescence and time was calculated. Half-life was calculated as the ln(0.5)/slope of the regression.
G. PCR Mutagenesis Protocol (Random Mutagenesis)
PCR Mutagenesis Reactions
There are three groups of mutagenic conditions using different skewed nucleotide concentrations.
The conditions listed herein yield in the range of from 8-10% wild-type Luc colonies after subcloning phenotypic for each generated parent clone. The rate of mutagenesis is estimated by the number of luminescent colonies that are present after mutagenesis. Based upon results of clones mutated in the range of 8-10%, it was determined that this level of mutagenesis produces on average approximately 2-3 amino acid changes per gene. If the mutagenesis rate is selected so that on average there is one amino acid change per gene, then on average 50% of the clones will have no mutations. (Bowie et al., 1990).
For the master mix: add all components (see Table 6) except polymerase, vortex, spin briefly, add polymerase, and mix gently.
Cycle in thermal cycler: 94° C. for 1 minute (94° C. for 1 minute, 72° C. for 10 minutes) 10×
1. Set up five 50 μl reactions (see Table 7) per group.
DNase I Digestion of Plasmid DNA
3. Prepare 1 U/μl DNase dilution on ice according to the table below:
Set up 4 reactions (see Table 9) and pool when completed.
Cycling conditions: 94° C. for 30 seconds (94° C. for 20 seconds, 65° C. for 1 minute, 72° C. for 2 minutes) 25×
Amplification of Assembly
Usually 5 amplification reactions (see Table 10) will produce enough DNA for a full 8 plate robotic run.
Purify amplification products with Wizard PCR purification:
Prepare Ampicillin-sensitive single stranded DNA of the template to be mutated. Design a mutagenic primer that will randomly generate all possible amino acid codons.
Heat reaction at 60° C. for 15 minutes and then immediately place on ice.
JM109 clones (from a transformation reaction) are plated onto nitrocellulose filters placed on LB amp plates at a screening density of about 500 colonies per plate.
As listed in the Random Mutagenesis procedure, approximately 10% of the clones to be selected will have to be as stable as the same sequenced or better than source. Or stated another way, about 50 colonies per plate will be suitable for selection. There are 704 wells available for a full eight plate robotic run, so at least 15 LB amp plates will be needed for a full robotic run.
After overnight growth at 37° C. the plates containing the transformants are removed from the incubator and placed at room temperature.
The nitrocellulose filter is lifted on one side and 500 μl of 10 mM IPTG is added to each of the plates. The filter is then placed back onto the plate to allow diffusion of the IPTG into the colonies containing the different mutant luciferase genes. The plates are then incubated for about 4 hours at room temperature.
One (1) ml of a solution contains 1 mM luciferin and 100 mM sodium citrate is pipetted onto a slide warmer that is set at 50° C. A nitrocellulose filter that contains mutant luciferase colonies and has been treated with IPTG is then placed on top of the luciferin solution. After several minutes, the brightest colonies are picked with tooth picks which are used to inoculate wells in a microtiter plate that contain M9—minimal media with 1% gelatin.
After enough colonies are picked to 8 microtiter plates, the plates are placed in an incubator at 350 rpm at 30° C. incubation and are grown overnight.
In the morning the overnight plates are loaded onto the robot and the cell dilution procedure is run. (This procedure dilutes the cultures 1:10 into induction medium). The new plates are grown for 3 hours at 350 rpm at 30° C.
After growth, the plates are loaded to the robot for the main assay procedure.
Upon storage, luciferin degrades and the degradation products inhibit luciferase. The production of inhibitors causes an apparent instability in the reagent containing both luciferase and luciferin. There are two ways to reduce this problem: 1) Store the luciferin and luciferase at pH 5.5-6.0 to reduce the rate of luciferin degradation, and/or 2) Evolve an enzyme that is resistant to the luciferin degradation products.
LucPpe2 mutants that were evolved after clone Luc90-1B5 were evolved to be more stable at low pH and have resistance to luciferin degradation products. These mutant enzymes are useful, for example, in an ATP detection kit. One embodiment of such a kit comprises a mixture of luciferin and luciferase. A luminescent reaction occurs when a sample comprising ATP is added to the mixture.
Three populations of random mutants were produced using clone Luc90-1B5 as a template. These three populations were screened on the robot as experiments 114, 115, and 117. Robotic screens for experiments 114, 115, 116, 117, 118, 119, and 122 were completed as described previously except that buffer C was prepared with citrate buffer pH 4.5 instead of HEPES buffer pH 7.8, and the assay reagent was prepared with HEPES pH 7.1 with 10 μM ATP instead of Tricine pH 8.0 and 175 μM ATP. These screening conditions were biased to select clones that have increased retention of luminescence activity over time at pH 4.5 at 48° C. and increased luminescence activity when assayed at pH 7.1 with 10 μM ATP. Seventeen clones from experiment 114, seven clones from experiment 115, and ten clones from experiment 116 were shuffled together using sPCR and selected mutants from this screen were run on the robot as experiment 117. Eighteen clones were selected from experiment 117.
The clone that was determined to have the most improved characteristics (increased retention of luminescence activity over time at pH 4.5 and 48° C. and increased luminescence activity when assayed at pH 7.1 with 10 μM ATP) was clone Luc117-3C1 and it was selected as a template for random mutagenesis. Two populations of random mutants were screened and then run on the robot as experiments 118 and 119. Seven clones from experiment 118 and five clones from experiment 119 were saved.
Clones from experiments 114, 115, 116, 117, 118, and 119 were selected based upon the following characteristics: brighter luminescence than Luc90-1B5, and increased retention of luminescence activity over time at pH 4.5. These select clones were shuffled together and were run on the robot as experiment 122. Eleven clones from this experiment were saved.
Three populations of random mutants were prepared from clone Luc122-4D5 and run on the robot as experiments 125, 126, and 127. Thirteen clones from experiment 125, four clones form experiment 126, and three clones from experiment 127 were shuffled together and run on the robot as experiment 128. For experiments 125, 126, 127 and 128 the screen for Km was altered to select for clones that are more resistant to luciferin degradation products. The clones were also screened for retention of luminescence over time at pH 4.5.
Instead of screening for substrate utilization, a screen for resistance to inhibitor was conducted. In place of the 0.06× dilution of substrates, a 75:25 mix of D to L luciferin in 1× assay buffer was used and designated as “0.75×”. In place of the 0.02× dilution of substrates, a 50:50 mix of D to L luciferin in 1× assay buffer and was designated as “0.5×”. The 1× assay buffer in these experiments contained the following: 10 μM ATP, 50 mM HEPES pH 7.8, 8 mM MgSO4, and 0.1 mM EDTA. The 0.75× sample contained 75 μM D-luciferin and 25 μM L-luciferin. The 0.5× sample contained 50 μM D-luciferin and 50 μM L-luciferin. The 1× sample contained 250 μM D-luciferin. A Km regression was used as before and a Km value was calculated. Normalized values of greater than 1 indicate more resistance to inhibitor. Clones from these experiments that were shown to have greater resistance to L-luciferin were also more resistant to luciferin degradation products.
To more easily measure resistance to inhibitor on the robotic system, a new variable “Q” was designated. The “Q” variable replaces the Km variable used previously. The luminescence ratio is calculated the same as in the Km measurement, then the natural log (ln) of each luminescence ratio is calculated (Y-axis). The X-axis is an arbitrary time that is entered by the user. The first time point is zero and the samples are measured with 1× assay buffer that contains 250 μM D-luciferin. The next two time points have the same time value (i.e., 4 hours to simulate incubation of luciferin) and samples are measured with 1× assay buffer that contains a 50:50 mixture (as described above) of D-luciferin to L-luciferin. A linear regression correlating ln(lum ratio) to time is calculated. Q is calculated as the ln(0.5)/slope. Normalized values of “Q” greater than 1 indicate more resistance to inhibitor. Experiments 133 and higher were run using this program.
Sixteen clones from experiment 128 were shuffled with clones from experiment 122 and run on the robot as experiment 133. Two samples, Luc133-1B2 and Luc133-0D11, were selected as templates for random mutagenesis and run on the robot as experiments 145 and 146, respectively. The clone that showed an increased retention of luminescence over time at pH 4.5 and the most resistance to inhibitor was clone Luc146-1H2. Moreover, at pH 4.5 and 48° C., Luc133-1B2 and Luc146-1H2 had increased thermostability relative to Luc90-1B5, and increased resistance to inhibitor (
Materials and Methods
Assay to Detect Resistance to Luciferase Inhibitor
A 10 mM stock solution of luciferin is incubated at 50° C. in 50 mM HEPES, pH 7.8, to accelerate the production of luciferin breakdown products. At different time points an aliquot is removed and then placed at −20° C. After incubation is complete, assay reagent (100 μM Luciferin, 1 μM ATP, 50 mM HEPES, pH 7.8 and 8 mM MgSO4) is prepared with luciferin from each of the different time points and a diluted lysate is then assayed with each assay reagent.
The lysate is prepared as follows. Overnight cultures of clones to be tested are prepared in LB supplemented with 100 μg/ml AMP. The cultures are diluted 1:10 in M-9 minimal media supplemented with 1 mM IPTG, 100 μg/ml AMP and grown for 3 hours at 30° C. Forty-five μl of cells is mixed with 20 μl of Buffer A and frozen. The mixture is thawed, 175 μl of Buffer B added, and the resulting mixture diluted 1:10 in Buffer C. A regression of luminescence versus time of luciferin incubation is then calculated, and from this graph half-life is extrapolated. A longer half-life means that the mutant being tested is more resistant to luciferin breakdown products.
The luciferases from the luminous beetle, Pyrophorus plagiophthalamus, had been shown previously to generate different colors of luminescence (LucPpl). Analysis of these luciferases revealed that the different colors were caused by discreet amino acid substitutions to their protein sequences. This allowed the possibility to make a pair of genetic reporters capable of emitting a multiplexed luminescent signal, thus enabling quantitation of two biomolecular events simultaneously from within the same living system.
Amino acid substituted LucPpl were prepared which have the following properties:
Physical Stability of the Luciferases
Although the luminescence activity of LucPpl within colonies of E. coli appeared to be thermostable to above 60° C., in lysates these luciferases had relatively low stability. They were particularly unstable in the presence of Triton X-100 detergent. When lysates are prepared containing the commonly used firefly luciferase, the enzyme retains greater than 90% activity over 5 hours at room temperature. In contrast, the activity of the LucPpl luciferases would decrease several fold over the same period.
The thermostabilities of the LucPpl luciferases are also near the physiological temperature of mammalian cells. The green-emitting luciferase (LucPplGR) and red-emitting luciferase (LucPplRD) have different thermostabilities which may cause differences in the behaviors as genetic reporters within cells. The influence of temperature should be greatest near the point of denaturation for the enzymes, where small changes in temperature will have the greatest effect on protein structure. In contrast, temperature will have much less affect on protein structure when it is much below the denaturation point. Thus, the differential effect on two enzymes having slightly different denaturation temperatures will be less at relatively lower temperatures. It might therefore be preferable to have the denaturation temperature of the reporter enzyme significantly above the growth temperature of mammalian cells.
Spectral Overlap Between the Luciferases
Although a method was developed to quantify each luciferase in a mixture by using colored filters, the ability to discriminate between the luciferases is limited by their spectral overlap. This overlap reduces the ability to accurately measure both luciferases if their luminescence intensities differ by more than 10 fold. If the intensities differ by more than 50 fold, the luminescence signal of the dimmer luciferase is obscured by the other. Thus, it would be preferable to further separate the luminescence spectra of the two luciferases.
Many different mutations were identified which shifted the luminescence spectrum towards the red. But the limit for red luminescence appears to be about 620 nm. Further effort at shifting the spectrum of the red-emitting luciferase into still longer wavelengths might have some benefit. It was found that green-shifting mutations were rare, however, an extensive analysis was not conducted. Measurements from native luciferases show some examples of luminescence below 530 nm, about 15 nm less than the green-emitting prototype enzyme.
Differential Physical and Enzymological Characteristics
Ideally the two luciferase reporters would be identical in all characteristics except for the color of luminescence. However, as noted above, the physical stability of the luciferases was not identical. It was also found that mutations resulting in red-shifted luminescence also caused an increase in the KM for luciferin. Although some of these differences may be unavoidable, it is not clear whether the properties are fundamentally associated. For instance, luciferases from different beetle species sometimes have significantly differing KM even though their luminescence spectra are similar. It may be that much of the differences associated with development of the red-emitting luciferase are due to concomitant perturbations to the integrity of the enzyme structure, as the thermostability of a prototype of the red-emitting luciferase was increased without significantly altering the luminescence spectrum.
Stable Luminescence Signals
When firefly luciferase was first described as a genetic reporter, the luminescent signal was a relatively brief flash initiated upon injection of the reaction substrates. Subsequent development of the luminescent chemistry made the assay more convenient by enabling a stable signal for several minutes. Presently, such stabilized assays are standard for general laboratory applications. However, to allow high throughput screening in pharmaceutical research, the luminescence signal was further stabilized to extend for over an hour. This was necessary to allow sufficient time to assay several thousand samples in a batch. Although the luminescent signal of the new multiplexed luciferases was stable for minutes, they did not provide the extended signal stability needed for high throughput screening. It would be preferable if the signal stability could be further increased while optimizing other properties.
Methods to Optimize Luciferase Performance
To prepare luciferases having certain performance, a method for in vitro evolution of enzyme function, as described above, was employed. Briefly described, the method is a recursive process of generating random mutations and screening for desirable properties. It was originally developed primarily to increase the thermostability of luciferases, although other enzymological characteristics are also subject to optimization by the screening criteria. A slightly different strategy was used to achieve the properties described above since two related luciferases needed to be optimized concomitantly.
Initially, a single prototype enzyme is subjected to in vitro evolution to optimize physical stability and the luminescence signal. In the process, the mutant libraries are also screened for any new mutations causing changes in color. Particular emphasis is placed on isolating green shifted mutants. After initial optimization of a common prototype, a green- and red-emitting form of the enzymes is created, and these are further optimized separately to harmonize their physical and chemical properties. Particular attention is given to matching their physical stabilities and their substrate binding constants, especially for luciferin.
The choice for the initial prototype for optimization was the wild-type yellow-green-emitting luciferase isolated from the luminous beetles (LucPplYG). Of the luciferases originally cloned from P. plagiophthalamus, this one produced the brightest luminescence when expressed in E. coli. Furthermore, there was concern that the lack of green-shifting mutation resulted because the prior mutagenesis studies were done using a luciferase that already had the greenest luminescence. It was possible that if additional green-shifting mutations existed, they might be more evident when screened in a red-shifted background. The mutagenesis was performed as follows:
Remove Peroxisomal Targeting Sequence
The translocation signal at the C-terminus of the luciferases was removed. This was done using oligonucleotide-directed mutagenesis to convert the normal -KSKL to -XXX* (where X represents any amino acid, and * represents a termination codon). Several colonies yielding bright luminescence were selected and used as templates for the next stage of mutagenesis.
Removal of Sensitive Cysteines
The luciferases from P. plagiophthalamus have 13 cysteines, which are potentially sensitive to oxidation. This is in contrast to the commonly used firefly luciferase, which has only 4 cysteines. To remove any cysteines that may limit enzyme stability, oligonucleotide-directed mutagenesis was used to randomize the cysteine codons. Three sets of oligonucleotides were used: non-conserved cysteines in regions of low sequence homology (positions 69, 114, 160, 194, 335, and 460), non-conserved cysteines in regions of higher sequence homology (positions 127, 213, 310, and 311), and highly conserved cysteines (positions 60, 80, and 388). The best clones from each of these screens were isolated, and a new mutant library made by sPCR and screened again. At the screening temperature of 29° C., the activity of the wild-type yellow-green-emitting luciferase decreased about 500-fold over 10 hours. The activity of the most stable mutant (Luc20-4C10) was more stable, decreasing only about 2 fold.
First Cycle of Random Mutagenesis
Using the procedure developed previously, three mutant libraries were generated using error-prone PCR and screened. The best mutants from these were recombined into a new library by sPCR and screened again. Finally, the best clones of this screen were recombined with the best clones from the previous oligonucleotide-directed mutagenesis by sPCR, and screened again. At 41° C., the activity of the best mutant from this process (Luc30-4B02) decreased 63 fold over 10 hours, whereas the activity of the parent mutant (Luc20-4C10) decreased greater than 100,000 fold.
Six of the best mutants from the last screen were isolated and sequenced. This revealed that the amino acids at 16 positions had been changed among the six clones. Thirteen positions had been changed in the preferred mutant, Luc30-4B02. Four of the changes were at the C-terminus in all the isolated mutants, where oligonucleotide mutagenesis had changed the wild-type sequence of -KSKL to -AGG*. Only two of the cysteines had been changed by the previous oligonucleotide mutagenesis; one highly conserved cysteine at position 60 was changed to valine and one moderately conserved cysteine at position 127 was changed to threonine. The remaining amino acid changes were all due to point mutations in the DNA, consistent with error-prone PCR. Interestingly, three of these changed the amino acid into that found in the wild-type green-emitting luciferase (two in mutant Luc30-4B02). Four of the remaining changes brought the mutant sequences closer to the consensus amino acid among other cloned beetle luciferases (two in mutant Luc30-4B02) (
To further explore the potential of the mutations identified in the sequenced mutants, additional mutagenesis experiments were performed using oligonucleotides. Eight of the codons mutated by the error-prone PCR were randomized or partially randomized using oligonucleotide-directed mutagenesis. Four of the remaining cysteine codons were randomized; two highly conserved cysteines (positions 80 and 388) and two cysteines in a region of sequence homology (positions 310 and 311). One leucine was mutated to a leucine/proline; proline is the consensus amino acid among other beetle luciferases.
The mutagenesis was performed with four sets of oligonucleotides (Table 13), and the best clones from each set were selected. These were recombined by sPCR together with the selected clones from the previous random mutagenesis and screened again. The activity of the best clone from this process (Luc47-7A11) decreased 2.3 fold at 42° C.; the activity of the parent clone (Luc30-4B02) decreased greater than 2000 fold.
The random mutagenesis process using error-prone PCR was applied again to the best clone from the oligonucleotide-directed mutagenesis (Luc47-7A11). Three libraries were again created and screened, and selected mutants were recombined by sPCR and screened again. Following recombination, the activity of the best mutant (Luc53-0G01) decreased 1.2 fold at 43° C. The parent clone (Luc47-7A11) decreased 150 fold. After recombining the best of these new mutants with the best mutants from the previous oligonucleotide-directed mutagenesis, the activity of the new best mutant (Luc55-2E09) decreased 31 fold at 47° C., compared to 80 fold for the parent (Luc53-0G01).
Third Cycle of Random Mutagenesis
The random mutagenesis process was repeated using the best clone from the previous cycle of mutagenesis (Luc53-0G01). After recombining the selected mutants with the mutants from the second cycle of mutagenesis, the activity of the best clone (Luc81-6G01) decreased 100-fold at 47° C., compared with 750 fold for the parent (Luc53-0G01). The discrepancy in measured activity of Luc53-0G01 in this cycle of mutagenesis compared to the previous cycle may be due to changes in the assay procedure and recalibration of the incubator temperature. It should be noted that the recorded thermostabilities from each stage of mutagenesis are calculated from robotic data using abbreviated assay procedures. The data are intended to indicate the relative stabilities of enzyme mutants when assayed in the same screen, rather than providing an accurate quantitation of thermostability.
Before making a final selection of the best clone from which to create the green- and red-emitting luciferases, further analysis was done on the best clones from the final screen. Three clones in particular were strong candidates as the final choice: Luc81-0B11, Luc81-5F0, and Luc81-6G01. The luminescence properties of these three mutant enzymes were compared among one another. They were also compared to the wild-type yellow-green-emitting luciferase to gauge the effect of the in vitro evolution process.
From colonies of E. coli expressing the luciferases, Luc81-5F01 and Luc81-6G01 produced luminescence most rapidly at room temperature upon addition of luciferin. The luminescence was more rapid and brighter than colonies expressing the wild-type green- and yellow-green-emitting luciferases. The luminescence from all the selected colonies appeared green-shifted compared to the yellow-green parent clone. When the colonies are heated to 65° C., the yellow-green clone looses most luminescence and the green clone becomes dimmer. Some of the mutant clones loose their luminescence at 65° C., but the three preferred clones remain bright above 70° C. No spectral changes upon heating the colonies were evident until above 70° C., where those clones still retaining activity began to red-shift slightly (sometime, the initial phases of enzyme denaturation are accompanied by a red shift in the luminescence). The luminescence characteristics of the three preferred mutants are quite similar.
The thermostability of the mutant luciferases in cell lysates was compared at room temperature (
For luminescence assays requiring only a few second, the wild-type yellow-green-emitting luciferase produces a very stable signal (the initial rise in the signal evident in the first 2 seconds is due to the response time of the luminometer, not the kinetics of the luminescence reaction). However, the signal intensity was reduced about 30% by the presence of 1% Triton X-100 in the lysate (diluted 1:5 with the addition of assay reagent). In contrast, the luminescence intensity of the mutant luciferases was unaffected by the presence of Triton X-100. Under these conditions, the most stable signal was produced by Luc81-6G01, although the signal intensity was somewhat brighter for Luc81-5F01. However, the data are not corrected for the efficiency of enzyme expression in E. coli. Thus, differences in luminescence intensity may not correlate to changes in enzyme specific activity, nor is the expression efficiency in E. coli necessarily relevant to expression in mammalian cells.
For batches of assays requiring more than an hour to process, the signal stability of the yellow-green-emitting luciferase is inadequate under the conditions tested. The luminescence intensity decreases several fold per hour (
From these results, the mutant Luc81-6G01 was chosen as the best clone from which to subsequently create the green- and red-emitting luciferases. The sequence of Luc81-6G01 (
Interestingly, in either the sequences of Luc81-6G01 or Luc8]-0B11, there is no evidence that the prior oligonucleotide-directed mutagenesis had any beneficial effect. No novel nucleotide sequences appear at any of the targeted codons. The improved enzyme performance following the oligonucleotide-directed mutagenesis apparently was due to recombination of previously acquired mutations. All of the novel amino acid changes in Luc81-6G01 and Luc81-0B11 are at sites not targeted by the oligonucleotides and are due to single-base modifications of the codons, consistent with error-prone PCR. Even though the novel mutations in Luc81-6G01 were not found in the earlier sequence data, it is not certain when they were generated in the process. Most likely they were produced in the second and third cycles of random mutagenesis; however, they may have been present among other selected mutants prior to Luc30-4B02. Relative to the initial yellow-green-emitting luciferase, the Luc81-6G01 mutant has acquired 17 amino acid changes and 3 codon mutations not affecting the amino acid sequence.
The observation that the onset of luminescence within colonies of E. coli is faster for the new mutants, and that the luminescence is brighter at higher temperatures, is probably not due to differences in protein expression. Immunoblot analysis of cell expressing the different luciferases showed no significant differences in the amount of polypeptide present. As noted above, the greater light intensity at higher temperatures is due to the increased thermostability of the mutant luciferases. The apparent KM's for ATP and luciferin have also changed during the course of the in vitro evolution (Table 14). To estimate the KM values, the mutant luciferases were partially purified from lysates of E. coli by differential precipitation using ammonium sulfate (40-65% saturation fraction). The results show that the KM'S for both ATP and luciferase are more than 10-fold lower.
When luciferin is added to an E. coli colony expressing luciferase, the intracellular concentration of luciferin slowly increases as it diffuses across the cell membrane. Thus, the intracellular concentration of luciferin reaches saturation sooner for those luciferases having the lowest KM's. Hence, the mutant luciferases appear brighter sooner than the wild-type parent clone. This also explains why the luminescence of the red-emitting prototype clone appears in E. coli colonies much more slowly than the green-emitting luciferase. Analysis of KM shows that the mutations causing the red luminescence also substantially increase the KM for luciferin.
From the analysis of luminescence signal in vitro, the luminescence from the mutant luciferases might be expected to fade more quickly than the wild-type luciferase during the first 30 minutes. Following this, the luminescence should be most stable in the mutants. However, this has not been noticed in the colonies of E. coli, and it may be that the kinetics of luminescence are different within cells compared to diluted enzyme in buffer.
All publications, patents and patent applications are incorporated herein by reference. While in the foregoing specification, this invention has been described in relation to certain preferred embodiments thereof, and many details have been set forth for purposes of illustration, it will be apparent to those skilled in the art that the invention is susceptible to additional embodiments and that certain of the details herein may be varied considerably without departing from the basic principles of the invention.