PENICILLIN-G ACYLASES

Abstract

The present invention provides engineered penicillin G acylase (PGA) enzymes, polynucleotides encoding the enzymes, compositions comprising the enzymes, and methods of using the engineered PGA enzymes.

Claims

1. An engineered polynucleotide encoding an engineered penicillin G acylase variant having at least 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99% or more sequence identity to SEQ ID NO:6, and at least one substitution at a position selected from positions 54, 62, 115, 125, 127, 127, 185, 253, 254, 254/255, 254/255/370, 255, 256, 257, 257, 260, 268, 322, 325, 348, 369, 370, 372, 373, 377, 378, 384, 384/513/536, 388, 389, 391, 435, 461, 517, 530, 554, 556, 557, 559, 560, 600/623, 623, 624, 626, 627, 705, 706, 707, 723, 740, 748, and 752, wherein said positions are numbered with reference to SEQ ID NO:6.

2. The engineered polynucleotide encoding an engineered penicillin G acylase variant of claim 1, wherein said said engineered penicillin acylase variant has 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99% or more sequence identity to SEQ ID NO: 6, and at least one substitution selected from 54C, 62G, 115A/P, 125L/T, 127S/V, 185V, 253K/V, 254T, 254W/255G, 254W/255G/370I, 255L, 255M/Q/T/Y, 256Q, 257I, 257V, 260A/P, 2685/V, 322P, 325G, 348C, 348Q, 369L, 369P, 369V, 369W, 370F/G/S, 372A/H/L, 373F/M, 377P, 378H, 384A, 384F/513Q/536M, 384G/L, 388T, 389L, 391P/S, 435R, 461A, 517L/P, 530C/Y, 554A/E/P/V, 556G, 557G/S, 559P/S, 560I, 600T/623V, 623A/G/R/W, 624A, 626G, 627G/H, 705G/P, 706G, 707S, 723A/G, 740L, 748G, and 752E.

3. An engineered polynucleotide encoding an engineered penicillin G acylase variant having at least 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99% or more sequence identity to SEQ ID NO:8, and at least one substitution set selected from positions103/370/444/706/766, 103/369/370/442/444/536/556/766, 103/369/370/444, 103/369/370/444/556/706/766, 103/369/370/444/765/766, 103/369/370/765/766, 257/362/384/451, 257/362/384/451/723, 362/451/705, 369/370, 369/370/444/706/766, 369/370/556/766, 369/370/388/444/556/766, 369/370/444, 369/370/444/556/766, 369/370/556, 369/370/556/765, 369/370/556/766, 369/370/766, 369/370/444/556, 369/370/444/556/612/766, 369/370/444/556/706/765, 369/370/444/706/765/766, 372/373/384/513/560, 372/384/451/705, 372/384/560/705, 384/451/560/705/723, 384/451/705/723, 451/560/705/723, and 451/705/723, wherein said positions are numbered with reference to SEQ ID NO:8.

4. The engineered polynucleotide encoding an engineered penicillin G acylase variant of claim 3, wherein said substitution set is selected from 103V/370F/444S/706G/766G, 103V/369W/370F/442I/444S/536M/556G/766G, 103V/369W/370F/444S, 103V/369W/370F/444S/556G/706G/766G, 103V/369W/370F/444S/765P/766G, 103V/369W/370F/765P/766G, 257V/362V/384A/451R, 257V/362V/384L/451R/723L, 362V/451R/705D, 369P/370F, 369P/370F/444S/706G/766G, 369P/370F/556G/766G, 369V/370F/388T/444S/556G/766G, 369V/370F/444S, 369V/370F/444S/556G/766G, 369V/370F/556G, 369V/370F/556G/765P, 369V/370F/556G/766G, 369V/370F/766G, 369W/370F/444S/556G, 369W/370F/444S/556G/612A/766G, 369W/370F/444S/556G/706G/765P, 369W/370F/444S/706G/765P/766G, 372A/373M/384L/513Q/560G, 372A/384L/451R/705D, 372A/384L/560G/705D, 384A/451R/560G/705D/723L, 384L/451R/705D/723L, 451R/560G/705D/723L, and 451R/705D/723L.

5. An engineered polynucleotide encoding an engineered penicillin G acylase variant having at least 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99% or more sequence identity to SEQ ID NO:34, and at least one substitution set selected from 103/372/373/557, 253/322/369/623, 253/254/322/369/623, 253/254/369/391/623/723, 253/254/369/619/623/723, 253/254/369/623/723, 253/254/373/623/723, 253/254/255/369/623/723, 253/254/369, 253/322/369/373/723, 253/369/623/723, 253/373/623, 253/254/255/322/369/619/723, 260/372/373/556, 260/372/373/556/557/559, 322/369, 322/369/373/723, 322/369/623/723, and 369/373/556, wherein said positions are numbered with reference to SEQ ID NO:34.

6. The engineered polynucleotide encoding an engineered penicillin G acylase variant of claim 5, wherein said substitution set is selected from 103V/372S/373F/557G, 253H/322T/369W/623G, 253H/254Q/322T/369W/623G, 253H/254Q/369W/391A/623G/723A, 253H/254Q/369W/619R/623G/723A, 253H/254Q/369W/623G/723A, 253H/254Q/373L/623G/723A, 253H/254S/255V/369W/623S/723A, 253H/254S/369W, 253H/322T/369W/373W/723A, 253H/369W/623G/723A, 253H/373L/623S, 253S/254S/255V/322T/369W/619R/723A, 260S/372S/373F/556G, 260S/372S/373F/556G/557V/559S, 322T/369W, 322T/369W/373W/723A, 322T/369W/623G/723A, and 369W/373F/556G.

7. The engineered polynucleotide encoding an engineered penicillin G acylase variant of claim 1, wherein said engineered penicillin G acylase variant comprises a histidine tag.

8. The engineered polynucleotide encoding an engineered penicillin G acylase variant of claim 7, wherein said histidine tag is present at the C-terminus of said engineered penicillin G acylase variant.

9. The engineered polynucleotide encoding an engineered penicillin G acylase variant of claim 1, wherein said engineered penicillin G acylase variant comprises a polypeptide sequence set forth in variant numbers 1-308.

10. The engineered polynucleotide encoding an engineered penicillin G acylase variant of claim 1, wherein said engineered penicillin G acylase variant comprises a polypeptide sequence selected from the even-numbered sequences between SEQ ID NO:4 and SEQ ID NO:90.

11. The engineered polynucleotide sequence of claim 1, said wherein said sequence comprises a polynucleotide sequence that is at least 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99% or more identical to a sequence selected from the odd-numbered sequences between SEQ ID NO:5 and SEQ ID NO:89.

12. A vector comprising the engineered polynucleotide sequence of claim 1.

13. The vector of claim 12, further comprising at least one control sequence.

14. A host cell comprising the vector of claim 13.

15. A method for producing an engineered penicillin G acylase variant, comprising culturing said host cell of claim 14 under conditions that said engineered penicillin G acylase variant is produced by said host cell.

16. The method of claim 15, further comprising the step of recovering said engineered penicillin G acylase variant produced by said host cell.

Description

EXAMPLE 1

E. coli Expression Hosts Containing Recombinant PGA Genes

[0179] The initial PGA enzymes used to produce the variant enzymes of the present invention were obtained from variants disclosed in co-owned US Pat. Appin. Publ. No. 2016/0326508, incorporated herein by reference in its entirety and for all purposes. The PGA panel plate comprises a collection of engineered PGA polypeptides that have improved properties, as compared to the wild-type Kluyvera citrophila PGA. The wild type PGA gene is a heterodimer consisting of an alpha subunit (23.8 KDa) and a beta subunit (62.2KDa) that are linked by 54aa spacer region. Due to the presence of the spacer region, an autoprocessing step is required to form the active protein. During the development of the present invention, the wild-type gene was modified to eliminate the spacer region, thus eliminating the auto processing step. The PGA panel plate (Codexis) contains PGA variants that lack the spacer region (See e.g., US Pat. Appin. Publn. 2010/0143968, which is incorporated herein by reference in its entirety and for all purposes). A C-terminal histidine was added to the genes and the PGA-encoding genes were cloned into the expression vector pCK110900 (See US Pat. Appin. Publn. No. 2006/0195947 and 2016/0244787, both of which are incorporated herein by reference in their entireties and for all purposes), operatively linked to the lac promoter under control of the lacl repressor. The expression vector also contains the P15a origin of replication and a chloramphenicol resistance gene. The resulting plasmids were transformed into E. coli W3110, using standard methods known in the art. The transformants were isolated by subjecting the cells to chloramphenicol selection, as known in the art (See e.g., U.S. Pat. No. 8,383,346 and WO2010/144103, each of which is incorporated herein by reference in its entirety and for all purposes).

EXAMPLE 2

Preparation of HTP PGA-Containing Wet Cell Pellets

[0180] E. coli cells containing recombinant PGA-encoding genes from monoclonal colonies were inoculated into 180 μl LB containing 1% glucose and 30 μg/mL chloramphenicol into the wells of 96 well shallow-well microtiter plates. The plates were sealed with O.sub.2-permeable seals and cultures were grown overnight at 30° C., 200 rpm and 85% humidity. Then, 10 μl of each of the cell cultures were transferred into the wells of 96 well deep-well plates containing 390 mL TB and 30 μg/mL CAM. The deep-well plates were sealed with O.sub.2-permeable seals and incubated at 30° C., 250 rpm and 85% humidity until OD.sub.600 0.6-0.8 was reached. The cell cultures were then induced by IPTG to a final concentration of 1 mM and incubated overnight under the same conditions as originally used. The cells were then pelleted using centrifugation at 4000 rpm for 10 min. The supernatants were discarded and the pellets frozen at −80° C. prior to lysis.

EXAMPLE 3

Preparation of HTP PGA-Containing Cell Lysates

[0181] First, 2000 lysis buffer containing 50 mM Tris-HCl buffer, pH 7.5, 1 mg/mL lysozyme, and 0.5 mg/mL PMBS was added to the cell paste in each well produced as described in Example 2. The cells were lysed at room temperature for 2 hours with shaking on a bench top shaker. The plate was then centrifuged for 15 min at 4000 rpm and 4° C. The clear supernatants were then used in biocatalytic reactions to determine their activity levels.

EXAMPLE 4

Preparation of Lyophilized Lysates from Shake Flask (SF) Cultures

[0182] Selected HTP cultures grown as described above were plated onto LB agar plates with 1% glucose and 30 kg/ml CAM, and grown overnight at 37° C. A single colony from each culture was transferred to 6 ml of LB with 1% glucose and 30 μg/ml CAM. The cultures were grown for 18 h at 30° C., 250 rpm, and subcultured approximately 1:50 into 250 ml of TB containing 30 μg/ml CAM, to a final OD.sub.600 of 0.05. The cultures were grown for approximately 195 minutes at 30° C., 250 rpm, to an OD.sub.600 between 0.6-0.8, and induced with 1 mM IPTG. The cultures were then grown for 20 h at 30° C., 250 rpm. The cultures were centrifuged 4000 rpm×20 min. The supernatant was discarded, and the pellets were resuspended in 30 ml of 20 mM TRIS-HCl, pH 7.5. The cells were pelleted (4000 rpm x 20 min) and frozen at −80° C. for 120 minutes. Frozen pellets were resuspended in 30 ml of 20 mM TRIS-HCl pH 7.5, and lysed using a Microfluidizer® processor system (Microfluidics) at 18,000 psi. The lysates were pelleted (10,000 rpm×60 min) and the supernatants were frozen and lyophilized to generate shake flake (SF) enzymes.

[0183] The activity of selected shake flask PGA variants was evaluated based on the efficiency of the variants in removing the four/two phenyl acetate groups chemically attached to the A1/A1′ (glycine), and B1/B1′ (phenylalanine), residues of an insulin-dimer. Reactions using shake flask powders were carried out in 2 mL 96-well plates. In this assay, 200 μL reactions solutions consisting of 10-30 g/L tetra-protected insulin dimer (A1,A1′,B1,B1′-tetraphenylacetimido-insulin-dimer) or di-protected insulin dimer (A1,A1′-diphenylacetimido- insulin dimer), 0.15-3 g/L shake flask powder, 0.2 M triethanolamine (TEoA) buffer, pH 8.5, and 20-30% (v/v) DMSO were prepared. The reaction plates were sealed with a heat seal and incubated at 30° C. and agitated at 300 RPM in a Thermotron® shaker (2 mm throw, model # AJ185, Infors) for 25 h. Three 20 μL aliquots of each reaction were taken at 45 min, 4.75 and 25 h, and quenched 1:1 with MeCN, then diluted 1:5 with deionized water. The samples were analyzed by UPLC using parameters in Tables 12.1, 12.2 and 12.3.

EXAMPLE 5

Evaluation of Shake Flask Powders of SEQ ID NO: 4 and SEQ ID NO: 6 on Tetra-Protected Insulin Tethered-Dimer

[0184] In order to assess activity differences between SEQ ID NO: 4 and the engineered variant SEQ ID NO: 6, which contains a C-terminal polyhistidine affinity tag (His-tag, HT), reactions using shake flask powders (See, Example 4) were carried out in 96-well plates with 2 mL wells. In these assays, 200 μL reaction solutions consisting of 14.5 g/L tetra-protected insulin dimer substrate (A1, A1′, B1,B1′-tetraphenylacetimido-insulin tethered-dimer), 2.5 g/L shake flask powder, 0.2 M triethanolamine (TEoA) buffer, pH 8.5, and 20% (v/v) DMSO were prepared. Reaction solution-containing plates were sealed with a heat seal and incubated at 30° C. and agitated at 300 RPM in a Thermotron® shaker (2 mm throw, model # AJ185, Infors) for 25 h. Three 20 μL aliquots of each reaction were taken at 45 min, 4.75 and 25 h, and quenched 1:1 with MeCN, then diluted 1:5 with deionized water. The samples were analyzed by UPLC using a Waters Cortecs® C18 column and the method described in Table 12.1. The activity was determined by comparing percent conversion (n=3) to the insulin dimer (product). The results are presented in Table 5.1.

TABLE-US-00002 TABLE 5.1 Activity Assessment of SEQ ID NO: 4 Compared to SEQ ID NO: 6 % Conversion Enzyme 45 min 4.75 h 25 h SEQ ID NO: 4 58% 71% 83% SEQ ID NO: 6  3% 35% 78%

EXAMPLE 6

Improvements in the Deacylation of Insulin Compared to SEQ ID NO: 6 in High Throughput Screening

[0185] SEQ ID NO: 6 was selected as the next parent enzyme, based on the results described in Example 5. Libraries of engineered genes were produced using well-established techniques (e.g., saturation mutagenesis and recombination of previously identified beneficial mutations). The polypeptides encoded by each gene were produced in HTP as described in Example 2, and the soluble lysate was generated as described in Example 3.

[0186] HTP reactions were carried out in 96 well deep-well plates containing 200 μL of 10 g/L tetraprotected insulin dimer substrate, 200 mM TEoA buffer, pH 8.5, 20% DMSO and 10 μL HTP lysate. The HTP plates were heat sealed and incubated in Thermotron® shakers at 30° C., 300 rpm, for 18 hours. The reactions were quenched with 200 μL MeCN and mixed for 5 minutes using a bench top shaker. The plates were then centrifuged at 4000 rpm for 5 minutes, diluted 24× into water, and injected onto an UPLC for analysis using the parameters in Table 12.1.

[0187] The percent conversion relative to SEQ ID NO:6 (Percent Conversion FIOP) was calculated as the percent conversion of the product formed by the variant over the percent conversion produced by SEQ ID NO: 6 The results are shown in Table 6.1. The percent conversion was quantified by dividing the area of the product peak by the sum of the areas of the substrate, and product as determined by UPLC analysis.

TABLE-US-00003 TABLE 6.1 Activity of Deacylating Variants Relative to SEQ ID NO: 6 SEQ Amino Acid Differences Deacylation Percent Variant ID NO: (Relative to Conversion (FIOP).sup.1 NO: (nt/aa) SEQ ID NO: 6) Relative to SEQ ID NO: 6 4 13/14 A373M ++ 5 K369W ++ 6 L253V ++ 7 K369V ++ 8 L257V ++ 9 7/8 F254W/A255G/W370I ++ 10 T115A ++ 11 K369L ++ 12 Q626G ++ 13 F254T ++ 14 D623W ++ 15 D268S ++ 16 V391S ++ 17 T560I ++ 18 D623A ++ 19 N348Q ++ 20 N627G ++ 21 Q554P ++ 22 M600T/D623V ++ 23 S706G ++ 24 V391P ++ 25 15/16 K369P ++ 26 A255M ++ 27  9/10 F254W/A255G ++ 28 Q554V ++ 29 S740L ++ 30 N185V ++ 31 S530C ++ 32 Y752E + 33 A255Y + 34 T115P + 35 N348C + 36 G260P + 37 W370S + 38 L253K + 39 Q556G + 40 11/12 W370F + 41 N388T + 42 I624A + 43 Q554A + 44 T384L + 45 I127S + 46 Q559S + 47 W370G + 48 N125L + 49 N125T + 50 T705P + 51 S372A + 52 E377P + 53 I389L + 54 L557G + 55 A373F + 56 E707S + 57 T384F/P513Q/L536M + 58 R748G + 59 F256Q + 60 A517P + 61 17/18 T384A + 62 L557S + 63 D623R + 64 Q554E + 65 T384G + 66 K723G + 67 19/20 A255Q + 68 D268V + 69 Q559P + 70 S435R + 71 A255T + 72 K723A + 73 G260A + 74 T705G + 75 N627H + 76 L257I + 77 S530Y + 78 K322P + 79 A517L + 80 G54C + 81 I127V + 82 T62G + 83 G461A + 84 S325G + 85 S372H + 86 D623G + 87 A255L + 88 T378H + 89 S372L + .sup.1Levels of increased activity or selectivity were determined relative to the reference polypeptide of SEQ ID NO: 6 and defined as follows: “+” > than 1.2-fold but less than 2.5-fold increase; “++” > than 2.5-fold but less than 5-fold.

EXAMPLE 7

Improvements in the Deacylation of Insulin Compared to SEQ ID NO: 8 in High Throughput Screening

[0188] SEQ ID NO: 8 was selected as the next parent enzyme, based on the results described in Example 6. Libraries of engineered genes were produced using well-established techniques (e.g., saturation mutagenesis and recombination of previously identified beneficial mutations). The polypeptides encoded by each gene were produced in HTP as described in Example 2, and the soluble lysate was generated as described in Example 3.

[0189] HTP reactions were carried out in 96 well deep-well plates containing 200 μL of 10 g/L tetraprotected insulin dimer substrate, 200 mM TEoA buffer, pH 8.5, 30% DMSO and 10 μL HTP lysate. The HTP plates were heat sealed and incubated in Thermotron® at 30° C., 300 rpm, for 18 hours. The reactions were quenched with 200 μL MeCN and mixed for 5 minutes using a bench top shaker. The plates were then centrifuged at 4000 rpm for 5 minutes, diluted 24× into water, and injected onto an UPLC for analysis using the parameters in Table 12.1.

[0190] The percent conversion relative to SEQ ID NO:8 (Percent Conversion FIOP) was calculated as the percent conversion of the product formed by the variant over the percent conversion produced by SEQ ID NO: 8 The results are shown in Table 7.1. The percent conversion was quantified by dividing the area of the product peak by the sum of the areas of the substrate, and product peaks as determined by HPLC analysis.

TABLE-US-00004 TABLE 7.1 Activity of Deacylating Variants Relative to SEQ ID NO: 8 Deacylation Percent Conversion SEQ ID (FIOP).sup.1 Variant NO: Amino Acid Differences Relative to SEQ NO: (nt/aa) (Relative to SEQ ID NO: 8) ID NO: 8  90 29/30 K103V/K369W/I370F/G444S/Q556G/S706G/H766G ++  91 25/26 K369W/I370F/G444S/Q556G/V612A/H766G ++  92 K369W/I370F/G444S/S706G/H765P/H766G ++  93 31/32 K369V/I370F/N388T/G444S/Q556G/H766G ++  94 27/28 K369W/I370F/G444S/Q556G/S706G/H765P ++  95 K369V/I370F/Q556G ++  96 K369W/I370F/G444S/Q556G ++  97 K369V/I370F/G444S/Q556G/H766G ++  98 K369P/I370F/Q556G/H766G ++  99 K103V/K369W/I370F/G444S/H765P/H766G ++ 100 K369V/I370F/Q556G/H766G ++ 101 K103V/K369W/I370F/V442I/G444S/L536M/Q556G/H766G ++ 102 T384A/A451R/T560G/T705D/K723L ++ 103 K103V/I370F/G444S/S706G/H766G ++ 104 A451R/T560G/T705D/K723L ++ 105 K369V/I370F/G444S ++ 106 A451R/T705D/K723L ++ 107 K369V/I370F/H766G ++ 108 L257V/A362V/T384A/A451R ++ 109 T384L/A451R/T705D/K723L ++ 110 S372A/T384L/A451R/T705D ++ 111 K369P/I370F/G444S/S706G/H766G ++ 112 K103V/K369W/I370F/G444S ++ 113 K369V/I370F/Q556G/H765P ++ 114 L257V/A362V/T384L/A451R/K723L ++ 115 K369P/I370F ++ 116 K103V/K369W/I370F/H765P/H766G ++ 117 33/34 S372A/A373M/T384L/P513Q/T560G ++ 118 21/22 A362V/A451R/T705D ++ 119 23/24 S372A/T384L/T560G/T705D + .sup.1Levels of increased activity or selectivity were determined relative to the reference polypeptide of SEQ ID NO: 8 and defined as follows: “+” > than 1.2-fold but less than 2.5-fold increase; “++” > than 2.5-fold but less than 5-fold; “+++” > than 5-fold.

EXAMPLE 8

Improvements in the Deacylation of Insulin Compared to SEQ ID NO: 34 in High Throughput Screening in 30% DMSO

[0191] SEQ ID NO: 34 was selected as the next parent enzyme, based on the results described in Example 7. Libraries of engineered genes were produced using well-established techniques (e.g., saturation mutagenesis and recombination of previously identified beneficial mutations). The polypeptides encoded by each gene were produced in HTP as described in Example 2, and the soluble lysate was generated as described in Example 3.

[0192] For Table 8.1, HTP reactions were carried out in 96 well deep-well plates containing 2004 of 20 g/L diprotected insulin dimer substrate (A1, A1′-diphenylacetimido-insulin tethered-dimer), 200 mM TEoA buffer, pH 8.5, 30% DMSO and 2.54 HTP lysate. The HTP plates were heat sealed and incubated in Thermotron® shakers at 30° C., 300 rpm, for 3.5 hours. The reactions were quenched with 1:5 DMAc and mixed for 5 min using a bench top shaker. The plates were then centrifuged at 4000 rpm for 5 min, and injected onto an UPLC for analysis using the parameters in Table 12.2.

[0193] For Table 8.2, HTP reactions were carried out in 96 well deep-well plates containing 200 of 20 g/L Diprotected Insulin Dimer substrate, 200 mM TEoA buffer, pH 8.5, 20% DMSO and 2.54 HTP lysate. The HTP plates were heat sealed and incubated in Thermotron® shakers at 30° C., 300 rpm, for 5 hours. The reactions were quenched with 1:5 DMAc and mixed for 5 min using a bench top shaker. The plates were then centrifuged at 4000 rpm for 5 min, and injected onto an UPLC for analysis using the parameters in Table 12.2.

[0194] The percent conversion relative to SEQ ID NO:34 (Percent Conversion FIOP) was calculated as the percent conversion of the product formed by the variant over the percent conversion produced by SEQ ID NO: 34 The results are shown in Tables 8.1 and 8.2. The percent conversion was quantified by dividing the area of the product peak by the sum of the areas of the substrate, and product peaks as determined by UPLC analysis.

TABLE-US-00005 TABLE 8.1 Activity of Deacylating Variants Relative to SEQ ID NO: 34 Amino Acid Deacylation SEQ ID Differences Percent Conversion Variant NO: (Relative to SEQ (FIOP).sup.1 Relative to NO: (nt/aa) ID NO: 34) SEQ ID NO: 34 120 D403T + 121 P275E + 122 A664G + 123 A747G + 124 K622R + 125 Q541A + 126 Q759N + 127 L55V + 128 E482A + 129 P496K + 130 A616G + 131 E482S + 132 S639G + 133 K619N/A664G + .sup.1Levels of increased activity or selectivity were determined relative to the reference polypeptide of SEQ ID NO: 34 and defined as follows: “+” > than 1.2-fold but less than 2.5-fold increase; “++” > than 2.5-fold but less than 5-fold; “+++” > than 5-fold.

TABLE-US-00006 TABLE 8.2 Activity of Deacylating Variants Relative to SEQ ID NO: 34 Deacylation Percent Conversion SEQ (FIOP).sup.1 Variant ID NO: Amino Acid Differences Relative to SEQ NO: (nt/aa) (Relative to SEQ ID NO: 34) ID NO: 34 134 L253H/K369W/D623G/K723A ++ 135 L253H/W254Q/K322T/K369W/D623G ++ 136 41/42 K103V/A372S/M373F/L557G ++ 137 37/38 G260S/A372S/M373F/Q556G/L557V/Q559S + 138 L253H/K322T/K369W/M373W/K723A + 139 L253H/W254Q/K369W/K619R/D623G/K723A + 140 L253H/K322T/K369W/D623G + 141 39/40 G260S/A372S/M373F/Q556G + 142 47/48 L253H/W254S/K369W + 143 43/44 L253H/W254Q/K369W/V391A/D623G/K723A + 144 35/36 K369W/M373F/Q556G + 145 45/46 L253H/W254Q/K369W/D623G/K723A + 146 L253S/W254S/G255V/K322T/K369W/K619R/K723A + 147 L253H/M373L/D623S + 148 K322T/K369W/D623G/K723A + 149 L253H/W254Q/M373L/D623G/K723A + 150 K322T/K369W/M373W/K723A + 151 L253H/W254S/G255V/K369W/D623S/K723A + 152 K322T/K369W + .sup.1Levels of increased activity or selectivity were determined relative to the reference polypeptide of SEQ ID NO: 34 and defined as follows: “+” > than 1.2-fold but less than 2.5-fold increase; “++” > than 2.5-fold but less than 5-fold; “+++” > than 5-fold.

EXAMPLE 9

Improvements in the Deacylation of Insulin Compared to SEQ ID NO: 46 in High Throughput Screening

[0195] SEQ ID NO: 46 was selected as the next parent enzyme, based on the results described in Example 8. Libraries of engineered genes were produced using well-established techniques (e.g., saturation mutagenesis and recombination of previously identified beneficial mutations). The polypeptides encoded by each gene were produced in HTP as described in Example 2, and the soluble lysate was generated as described in Example 3.

[0196] HTP reactions were carried out in 96 well deep-well plates containing 200 μL of 30 g/L diprotected insulin dimer substrate, 200 mM TEoA buffer, pH 8.5, 30% DMSO and 10 μL HTP lysate. The HTP plates were heat sealed and incubated in Thermotron® shakers at 30° C., 300 rpm, for 5 hours. The reactions were quenched with 1:5 DMAc and mixed for 5 min using a bench top shaker. The plates were then centrifuged at 4000 rpm for 5 min, and injected onto an UPLC for analysis using the parameters in Table 12.2 or 12.3.

[0197] The percent conversion relative to SEQ ID NO:46 (Percent Conversion FIOP) was calculated as the percent conversion of the product formed by the variant over the percent conversion produced by SEQ ID NO: 46 The results are shown in Table 9.1. The percent conversion was quantified by dividing the area of the product peak by the sum of the areas of the substrate, and product peaks as determined by UPLC analysis.

TABLE-US-00007 TABLE 9.1 Activity of Deacylating Variants Relative to SEQ ID NO: 46 SEQ Deacylation Percent ID Conversion (FIOP).sup.1 Variant NO: Amino Acid Differences Relative to SEQ ID NO: (nt/aa) (Relative to SEQ ID NO: 46) NO: 46 153 A71F ++ 154 M373A/E482C/Y569W/K619N/A764S ++ 155 T176S/M373F/E482A/K622V ++ 156 Q233E/M373F/E482A/K622V/A664G ++ 157 T176S/M373F/E482A/K622F/A664G ++ 158 Q233E/P275E/E482C/K619N ++ 159 T176S/M373F/E482A/Y569W ++ 160 K103V/G260S/K322T/N348A/G444S/Q556G/L557G/G623D ++ 161 T176S/E482A ++ 162 K146M/N309D/Q556N/K619S/R748A ++ 163 T176S/Q233E/M373A/K619N/A664R ++ 164 Q233E/P275E/E482A/Y569W/A664G ++ 165 51/52 K103V/L257V/G260S/K322T/N348A/L384T/G444S/ ++ Q556G/G623D 166 55/56 N9K/K103V/H253S/K322T/N348A/G444S/Q556G/L557G/ ++ G623D 167 N9K/G25V/K103V/H253S/N348A/G444S/L557G/G623D ++ 168 E482S/G623D ++ 169 K304I/P496K/A616S/K619N/A664E/A747P/F756P/Q759E ++ 170 57/58 K103V/G260S/K322T/N348A/M373A/V391A/G444S/ ++ Q556G/L557G/G623D 171 N9K/K103V/K322T/V391A/G444S/L557G/G623D ++ 172 L225T/K304I/N494E/A616G/K619N/A664G/A747P/ ++ Q759E 173 N494E/P496K/A616S/K619N/A664E ++ 174 L225K/K304C/N309V/Q556N/L557R/K619S/R748A ++ 175 T176S/M373F/E482C/Y569W/K622C/G623D/A764S ++ 176 G25V/K103V/N241K/H253S/K322T/N348A/G444S/ ++ Q556G/L557G/G623D 177 L225T/K304I/K322T/N494E/P496N/A616G/K619N/ ++ A664G/A747S/F756P 178 K103V/L257V/G260S/K322T/N348A/G444S/L557G ++ 179 53/54 K103V/G260S/K322T/N348A/G444S/G623D ++ 180 59/60 K322T/N348A/M373A/V391A/G444S/Q556G/G623D ++ 181 61/62 K103V/K322T/N348A/M373A/G444S/Q556G/L557G + 182 49/50 K322T/N348A/G444S/L557G + 183 A71F ++ 184 K619P + 185 K619V + 186 A71C + 187 K619A + 188 A71L + 189 K619H + 190 K619S + 191 T705N + 192 K128H + 193 A71G + 194 Q626E + 195 F617W + 196 A616D + 197 W369A + 198 M373G + 199 K619L + 200 N28A + 201 T129E + 202 A616E + 203 I370M + 204 W369L/A764G + 205 A616N + 206 I370Q + 207 A451H + 208 W369L + 209 I389V + 210 K622V + 211 T379S + 212 I77V + 213 K622I + 214 Q380D + 215 G111S + 216 N28S + 217 A616G + 218 N28Q + 219 W369E + 220 T131D + 221 W369V + 222 I77T + 223 R471V + 224 Q626D + 225 A616T + 226 T379D + 227 A616Q + 228 N28C + .sup.1Levels of increased activity or selectivity were determined relative to the reference polypeptide of SEQ ID NO: 46 and defined as follows: “+” > than 1.2-fold but less than 2.5-fold increase; “++” > than 2.5-fold but less than 5-fold; “+++” > than 5-fold.

EXAMPLE 10

Improvements in the Deacylation of Insulin Compared to SEQ ID NO: 54 in High Throughput Screening

[0198] SEQ ID NO: 54 was selected as the next parent enzyme, based on the results described in Example 9. Libraries of engineered genes were produced using well-established techniques (e.g., saturation mutagenesis and recombination of previously identified beneficial mutations). The polypeptides encoded by each gene were produced in HTP as described in Example 2, and the soluble lysate was generated as described in Example 3.

[0199] HTP reactions were carried out in 96 well deep-well plates containing 200 μL of 30 g/L diprotected insulin dimer substrate, 200 mM TEoA buffer, pH 8.5, 30% DMSO and 2.5 μL HTP lysate. The HTP plates were heat sealed and incubated in Thermotron® shakers at 30° C., 300 rpm, for 5 hours. The reactions were quenched with 1:5 DMAc and mixed for 5 min using a bench top shaker. The plates were then centrifuged at 4000 rpm for 5 min, and injected onto an UPLC for analysis using the parameters in Table 12.3.

[0200] The percent conversion relative to SEQ ID NO:54 (Percent Conversion FIOP) was calculated as the percent conversion of the product formed by the variant over the percent conversion produced by SEQ ID NO: 54. The results are shown in Table 10.1. The percent conversion was quantified by dividing the area of the product peak by the sum of the areas of the substrate, and product peaks as determined by HPLC analysis.

TABLE-US-00008 TABLE 10.1 Activity of Deacylating Variants Relative to SEQ ID NO: 54 SEQ Deacylation Percent ID Conversion (FIOP).sup.1 Variant NO: Amino Acid Differences Relative to SEQ ID NO: (nt/aa) (Relative to SEQ ID NO: 54) NO: 54 229 73/74 A71L/K128H/M373A/E482S/A664E/P753C +++ 230 75/76 A71L/K128H/T176S/E482S/P496K +++ 231 71/72 A71L/K128H/T176S/M373A/E482C/P496K/K619S +++ 232 A71F/T176S/P275C/E482S +++ 233 69/70 K128H/T176S/M373A/E482S/A664E +++ 234 A71L/T176S/E482A/K619P/A664D/Q759E +++ 235 A71L/T176S/A451H/E482A/K619V/Q759E +++ 236 A71L/K128H/T176S/M373A/E482S/P496K/Y569C +++ 237 A71L/T176S/E482A +++ 238 A71L/Q233E/S260G/E482A/L557G/Q759E +++ 239 K128H/T176S/Q233E/M373A/E482S/Q626E/P753C +++ 240 63/64 A71L/T176S/S260G/P275C/E482A/L557G/Q759E +++ 241 A71L/K128H/T176S/P496K/A664E +++ 242 K128H/T176S/Q233E/P496K/A664E/P753C +++ 243 A71F/T176S/S260G/A451H/K619V +++ 244 67/68 A71L/T176S/Q233E/S260G/A451H/E482S/A664C/Q759E ++ 245 65/66 A71F/T176S/Q233E/S260G/P275C/E482S/K619N/Q759D ++ 246 A71L/T176S/M373A/Q626E/A664E/P753C ++ 247 A71F/T176S/P275E/A664D ++ 248 A71L/T176S/Q233E/M373A/E482C/Y569C/P753C ++ 249 A71L/T176S/S260G/E482A ++ 250 N28A/A71L/K128H/T176S/Q626D/P753C ++ 251 K128H/T176S/M373A/P496K/P753C ++ 252 A71L/T176S/S260G/E482A/L557G/K619P/A664D ++ 253 N28A/A71L/K128H/T176S/K619N/A664E ++ 254 T176S/Q233E/A451H/E482S/K619N/A664C/Q759D ++ 255 A71F/T176S/Q233E/E482A ++ 256 A71L/M373A/F756C ++ 257 A71L/S260G/A451H/E482A/A664D/Q759E ++ 258 T176S/Q233E/S260G/P275E/E482C/A664E/Q759D ++ 259 A71M ++ 260 A71F ++ 261 A71G + 262 A71L + 263 Y180F + 264 A71I + 265 L122M + 266 L82V + 267 P739D + 268 A71V + 269 W658C + 270 F679L + 271 P496K + 272 V184F + 273 V184A + 274 P739S + 275 H472F + 276 H472V + 277 P686A + 278 V126L + .sup.1Levels of increased activity or selectivity were determined relative to the reference polypeptide of SEQ ID NO: 54 and defined as follows: “+” > than 1.2-fold but less than 2.5-fold increase; “++” > than 2.5-fold but less than 5-fold; “+++” > than 5-fold.

EXAMPLE 11

Improvements in the Deacylation of Insulin Compared to SEQ ID NO: 74 in High Throughput Screening

[0201] SEQ ID NO: 74 was selected as the next parent enzyme, based on the results described in Example 10. Libraries of engineered genes were produced using well-established techniques (e.g., saturation mutagenesis and recombination of previously identified beneficial mutations). The polypeptides encoded by each gene were produced in HTP as described in Example 2, and the soluble lysate was generated as described in Example 3.

[0202] HTP reactions were carried out in 96 well deep-well plates containing 200 μL of 30 g/L diprotected insulin dimer substrate, 200 mM TEoA buffer, pH 9, 30% DMSO and 1.25 μL HTP lysate.

[0203] The HTP plates were heat sealed and incubated in Thermotron® shakers at 35 ° C., 300 rpm, for 5 hours. The reactions were quenched with 1:5 DMAc and mixed for 5 min using a bench top shaker. The plates were then centrifuged at 4000 rpm for 5 min, and injected onto an UPLC for analysis using the parameters in Table 12.3.

[0204] The percent conversion relative to SEQ ID NO:74 (Percent Conversion FIOP) was calculated as the percent conversion of the product formed by the variant over the percent conversion produced by SEQ ID NO: 74 The results are shown in Table 11.1. The percent conversion was quantified by dividing the area of the product peak by the sum of the areas of the substrate, product and impurities/side product peaks as observed by UPLC analysis.

TABLE-US-00009 TABLE 11.1 Activity of Deacylating Variants Relative to SEQ ID NO: 74 Deacylation Percent Conversion (FIOP).sup.1 Variant SEQ ID NO: Amino Acid Differences Relative to SEQ ID NO: (nt/aa) (Relative to SEQ ID NO: 74) NO: 74 279 77/78 T176S/P275C/Y569W/Q759D +++ 280 81/82 T176S/L557G/Y569W/A616T +++ 281 83/84 T176S/Q759D ++ 282 T176S/A616S ++ 283 T176S/A348M/L557G/Y569W/A616G ++ 284 T176S/L557G/Y569W/A616G/I708L ++ 285 T176S ++ 286 T176S/P275C/A348M/L557G/Q759D ++ 287 T176S/Q233E ++ 288 T176S/Q233E/Q759E ++ 289 85/86 T176S/T352S ++ 290 T176S/P275E ++ 291 89/90 L71C/A451H/R748A ++ 292 T176S/L557G/K619G ++ 293 L71F/S353A/R357A/A451H/T705N/R748A ++ 294 87/88 G111S/T176S/T352S ++ 295 T176S/A616G/K619R ++ 296 T176S/L557G/I708L ++ 297 T176S/A616T ++ 298 T176S/Q233E/T352S ++ 299 79/80 L71F/A451H/Q556N/T705N/R748A ++ 300 T176S/P275E/L557G/Q759E ++ 301 T176S/Q233E/L557G/K619G/Q759D ++ 302 T176S/A616G ++ 303 T176S/L557G/A616N ++ 304 T176S/A361T ++ 305 T176S/Y569W/A616G/K619S/Q759D ++ 306 L71C/T352S ++ 307 T176S/S482C/A616G/Q759E ++ 308 I77T/T176S/A712V ++ .sup.1Levels of increased activity or selectivity were determined relative to the reference polypeptide of SEQ ID NO: 74 and defined as follows: “+” > than 1.2-fold but less than 2.5-fold increase; “+” > than 2.5-fold but less than 5-fold; “+++” > than 5-fold.

EXAMPLE 12

Analytical Detection of Insulin Dimer and its Deacylated Products

[0205] Data described in Examples 5-11 were collected using analytical methods in Tables 12.1, 12.2, 12.3. The methods provided herein all find use in analyzing the variants produced using the present invention. However, it is not intended that the methods described herein are the only methods applicable to the analysis of the variants provided herein and/or produced using the methods provided herein.

TABLE-US-00010 TABLE 12.1 Analytical Method Instrument Thermo Scientific Vanquish ™ UPLC Column Waters Cortecs ® C18, 2.7 × 50 mm, 1.6 μM Mobile Phase Gradient I (A: 0.1% TFA in water; B: 0.1% TFA in MeCN) Time(min) % A 0.000 69 1.500 50 1.550 5 1.950 5 2.000 69 Flow Rate 1.000 mL/min Run Time 2.400 min Product Elution Insulin dimer (0.77 min); A1-acylated insulin dimer (0.83 min); B1-acylated order insulin dimer (0.91 min); A1,A1′-diacylated insulin dimer (0.99 min); B1,B1′-diacylated insulin dimer (1.1 min); A1, A1′, B1-triacylated insulin dimer & A1, B1, B1′-triacylated insulin dimer (1.15 min); A1,A1′,B1,B1′- tetraacylated insulin dimer (1.25 min). Column 40° C. Temperature Injection Volume 1.0 μL Detection UV 218 nm; Detector: MWD-Data Collection Rate: 20 Hz

TABLE-US-00011 TABLE 12.2 Analytical Method Instrument Thermo Scientific Vanquish ™ UPLC Column Waters Cortecs ® C18, 2.7 × 50 mm, 1.6 μM Mobile Phase Gradient I (A: 0.1% TFA in water; B: 0.1% TFA in MeCN) Time(min) % A 0.000 69 0.900 50 0.950 5 1.300 5 1.350 100 1.650 69 Flow Rate 0.800 mL/min Run Time 2.000 min Product Insulin dimer (0.75 min); A1-acylated insulin dimer Elution order (0.82 min); A1,A1′-diacylated insulin dimer (0.89 min). Column 40° C. Temperature Injection 1.0 μL Volume Detection UV 218 nm; Detector: MWD-Data Collection Rate: 20 Hz

TABLE-US-00012 TABLE 12.3 Analytical Method Instrument Thermo Scientific Vanquish ™ UPLC Column Thermo Hypersil ™ Gold C18, 2.1 × 50 mm, 1.9 μM Mobile Phase Gradient I (A: 0.1% TFA in water; B: 0.1% TFA in MeCN) Time(min) % A 0.000 70 0.900 53 0.950 5 1.300 5 1.350 70 Flow Rate 0.950 mL/min Run Time 1.500 min Product Insulin dimer (0.61 min); A1-acylated insulin dimer (0.65 Elution order min); A1,A1′-diacylated insulin dimer (0.69 min). Column 40° C. Temperature Injection 1.0 μL Volume Detection UV 218 nm; Detector: MWD-Data Collection Rate: 20 Hz

[0206] All publications, patents, patent applications and other documents cited in this application are hereby incorporated by reference in their entireties for all purposes to the same extent as if each individual publication, patent, patent application or other document were individually indicated to be incorporated by reference for all purposes.

[0207] While various specific embodiments have been illustrated and described, it will be appreciated that various changes can be made without departing from the spirit and scope of the invention(s).