Stable protease variants

Field of the invention

The present invention relates to the field of proteases. Background

Proteases are today used in large array of industrial applications, including animal feed, detergents, fruit and beverage processing, leather processing, production of protein hydrolysates, hard surface cleaning or biofilm cleaning, treatment of necrotic or burned tissue to promote wound healing and/or food preparation including baking dough preparation.

In many of these applications, improved stability of the enzyme is a significant advantage. Improved thermostability helps to increase the processability of the respective protease, because the latter oftentimes undergoes thermal treatment during the manufacturing process.

This applies, inter alia, for the use of proteases in animal feed where they help to improve the digestibility and nutrient exploitation of the feed.

During feed processing, the feed is often subjected to heat, e.g., by application of steam, to reduce or eliminate pathogens, increase storage life of the feed and optimized utilization of the ingredients leading to improved feed conversion. The conditioning time can vary from a few seconds up to several minutes depending on the type and formulation of the feed. The temperature during conditioning typically ranges from 70°C to 100°C. After conditioning, the feed is sometimes extruded through a pelleting die, which for a short time raises the temperature of the feed incrementally due to heat dissipation caused by friction.

Yet in other applications, protease enzymes are exposed to heat as well. This includes the use in detergents (e.g. exposure to hot water during laundry washing), fruit and beverage processing (heat exposure during the squeezing process or due to pasteurization or sterilization), leather processing, production of protein hydrolysates, hard surface cleaning or biofilm cleaning, treatment of necrotic or burned tissue to promote wound healing, processing aid in tissue engineering (sterilization, and denaturation of prion proteins) and/or food preparation including baking dough preparation.

Because proteases are proteins, they are susceptible to denaturation by heat and pressure. Denaturing essentially alters the structure of the enzyme, resulting in decreased activity levels and decreased efficacy of the enzyme.

There are different ways to improve protease stability or protect proteases from thermal impact. In animal feed applications, one option is Post-pellet liquid application, which is relatively complex and expensive because it requires the purchase and installation of specialized equipment, space in which to store the liquid enzyme and careful calculation of the amount of enzyme to apply.

Another option is the application of a protective coating before pelleting of the protease with other ingredients (e.g., in feed or detergents). This approach may reduce the efficacy of the enzyme because the coating may not fully dissolve, e.g., in the washing medium, or in the digestive tract of the animal. It is furthermore difficult to achieve a coating design that can withstand the high heat and moisture content of the pelleting process, but subsequently dissolve in the lower temperature and higher moisture conditions, e.g., in the animal's gut or the washing machine.

Another option is to use intrinsically thermostable proteases. These proteases are derived from thermophilic and hyper-thermophilic organisms and have unique structure and function properties of high thermostability. However, these proteases may suffer from other limitations, like suboptimal activity, specificity, bioavailability, pH-range or processability. It is hence one object of the present invention to provide stable protease variants which do not suffer from the above discussed limitations.

Summary of the invention

These and further objects are met with methods and means according to the independent claims of the present invention. The dependent claims are related to specific embodiments.

Embodiments of the invention

Before the invention is described in detail, it is to be understood that this invention is not limited to the particular component parts or structural features of the devices or compositions described or process steps of the methods described as such devices and methods may vary. It is also to be understood that the terminology used herein is for purposes of describing particular embodiments only, and is not intended to be limiting. The mere fact that certain measures are recited in mutually different dependent claims does not indicate that a combination of these measures cannot be used to advantage. Any reference signs in the claims should not be construed as limiting the scope. It must be noted that, as used in the specification and the appended claims, the singular forms "a," "an" and "the" include singular and/or plural referents unless the context clearly dictates otherwise. Further, in the claims, the word "comprising" does not exclude other elements or steps.

It is moreover to be understood that, in case parameter ranges are given which are delimited by numeric values, the ranges are deemed to include these limitation values.

It is further to be understood that embodiments disclosed herein are not meant to be understood as individual embodiments which would not relate to one another. Features discussed with one embodiment are meant to be disclosed also in connection with other embodiments shown herein. If, in one case, a specific feature is not disclosed with one embodiment, but with another, the skilled person would understand that does not necessarily mean that said feature is not meant to be disclosed with said other embodiment. The skilled person would understand that it is the gist of this application to disclose said feature also for the other embodiment, but that just for purposes of clarity and to keep the specification in a manageable volume this has not been done. According to one embodiment of the invention, a protease variant is provided which is at least 90% identical to the full length amino acid sequence of a Kumamolisin AS backbone as set forth in any of SEQ ID NOs 1 - 3, or a fragment, fraction or shuffled variant thereof maintaining proteolytic activity. The protease variant demonstrates altered or improved stability compared to

(i) the Kumamolisin AS wildtype as set forth in SEQ ID NO 4, or

(ii) the Kumamolisin AS backbone as set forth in any of SEQ ID NOs 1 - 3.

The term "shuffled variant" relates to a combination of such fragment or fraction with one or more fragments from other homologous enzymes, as long as such combination maintains proteolytic activity.

The term "homologous enzyme" describes enzymes belonging to the same structural fold as Kumamolisin and at least 40% sequence identity. This category encompasses Sedolisins as discussed below herein.

Some mutants of Kumamolisin AS have been described. The discovery of N291D mutant strain of Kumamolisin AS has been discussed to providing a useful treatment against celiac disease. There are many proposals that suggest creating a genetically modified organism that could produce N291D Kumamolisin AS protein in human's gastrointestinal tracts. See US application US 20140178355 Al .

Preferably, the Kumamolisin AS variant according to the invention has 93 % identity, more preferably 95 % identity, more preferably 98 % identity, most preferably 99 % identity.

The term "Kumamolisin" refers to acid proteases from the Sedolisin family of peptidases, also called S53 (MEROPS Accession MER000995, see also Wlodawer et al, 2003), comprising acid-acting endopeptidases and a tripeptidyl-peptidase. Sedolisins are endopeptidases with acidic pH optima that differ from the majority of endopetidases in being resistant to inhibition by pepstatin (Terashita et al, 1981; Oda et al, 1998). The activation of sedolisins involves autocatalytic cleavage at pH below pH 6.5, better below pH 3.5 (see also patent application EP16176044 and Okubo et al, 2016), which releases one or more peptides to deliver the maturated and active form. Said autocatalytic cleavage is inhibited under alkaline, neutral and lightly acidic conditions.

Sedolisins comprise a catalytic triad with Glu, Asp and Ser, which in Kumamolisin AS according to SEQ ID NO 1 reside in positions Glu267, Asp271 and Ser278. The Ser residue is the nucleophile equivalent to Ser in the catalytic triad Asp, His, Ser triad of subtilisin proteases (MEROPS family S8), and the Glu of the triad is a functional substitution for the His general base in subtilisin though not in structural equivalent positions.

The protein folds of sedolisins are clearly related to that of subtilisin, and both groups are sometimes called serine proteases. However, sedolisins have additional loops. The amino acid sequences are not closely similar to subtilisins, and this, taken together with the quite different active site residues and the resulting lower pH for maximal activity, justifies the separate families.

In one embodiment, a protease variant is provided which comprises an amino acid sequence derived from a Kumamolisin AS as set forth in SEQ ID NO. 1, or a fragment, fraction or shuffled variant thereof maintaining proteolytic activity, which protease variant has one or more amino acid substitutions at one or more residue positions in SEQ ID NO. 1 selected from the group consisting of D447, A449, A517, N510, V502, E453, E360, A514, A460, A392, A386, T301, D199, Q518, G266, P553, E269, R412, S435, G320, T326, T461, Q244, D293, A487, V274, A372, K283, T308, A418, 1391, A423, A331, S327, 1219, M333, A329, N515, A378, S434, E421, A433, S230, Q393, D399, Y490, G281, Y287, R516, A475, S354, S315P, W325, L442, A470, S324, Q361, A190, T196, Q202, E228, A229, A242, D251, S262, N291, L297, H305, D306, V314, A328, 1330, L338, A342, A351, D358, G388, D402, V455, E459, A478, K483, Q497, T507, L540, Q542, A548, P551, R166 and/or D265.

Note that, while the numbering set forth above refers to SEQ ID NO 1 or 4 (which are almost identical, with 4 being the wildtype and 1 being the actual backbone used for mutagenesis, the difference between the two being the N terminal AA residue), the claimed protease can be a fragment, fraction or shuffled variant thereof maintaining proteolytic activity. In such case, the resulting amino acid sequence is shorter than that of SEQ ID NO 1 or 4, while the numbering of the mutant residues still refers to the full length SEQ ID NO 1 or 4, and has to be translated respectively to the numbering of the shorter form.

In one embodiment, the protease variant demonstrates altered or improved stability compared to

(i) the Kumamolisin AS wildtype as set forth in SEQ ID NO 4, or

(ii) the Kumamolisin AS backbone as set forth in any of SEQ ID NOs 1 - 3,

In one embodiment, the protease variant has at least one amino acid substitution selected from the group consisting of D447S, A449Y, A517T, N510H, E360L, E360V, E360C, V502C, E453W, A514T, A514Y, A514D, A514S, A460W, A386I, A392V, A392L, A392I, A392M, T301S, D199E, Q518G, P553K, E269M, E269T, E269C, E269H, E269Q, G266A, D293Y, G320A, R412Q, E421R, A487Q, T461V, T461C, A331F, A331Y, A329Q, A329H, A329T, S435I, S435R, S435T, S435V, V274I, A372S, K283L, Q244C, Q244G, T308C, A418W, I391W, A423V, T326R, T326W, T326L, T326K, I219L, S327F, S327L, S327W, M333I, N515G, A378G, S434G, A433G, S230D, Q393S, D399S, Y490W, A190D, T196S, Q202D, E228Q, A229W, A242S, D251S, S262C, G281R, Y287K, N291T, N291S, D293F, L297T, T301C, T301M, H305F, H305W, D306S, V314M, V314L, S315P, G320Q, G320S, S324L, S324R, W325K, A328W, A328D, A328R, A328Y, I330L, M333Y, M333L, L338R, A342R, A351S, S354E, S354Q, D358G, Q361C, Q361L, A386L, A386V, A386M, G388C, D402E, R412M, R412E, R412D, L442W, L442W, D447C, D447A, A449L, A449M, A449E, A449N, E453Y, E453F, V455I, V455L, E459W, A460R, A470V, A475V, A478L, K483A, Q497Y, Q497M, Q497D, Q497R, V502T, T507L, R516L, R516E, R516I, A517S, L540V, Q542H, Q542D, Q542S, A548S, P551N, P551R, P553L, R166I, D265T, compared to the Kumamolisin as as set forth in SEQ ID NO 1 or 4.

These individual amino acid substitutions are shown in Table 1. Note that, while the numbering set forth above refers to SEQ ID NO 1 or 4, the claimed protease can be a fragment, fraction or shuffled variant thereof maintaining proteolytic activity. In such case, the resulting amino acid sequence is shorter, or longer, than that of SEQ ID NO 1 or 4, while the numbering of the mutant residues still refers to the full length SEQ ID NO 1 or 4. In one embodiment of the invention, the protease variant has at least one amino acid substitution compared to the Kumamolisin AS as set forth in SEQ ID NO 1 or 4, which substitution is selected from the group consisting of:

• A517T or A517S

• A514S, A514T or A514D

• N510H

• V502C

• A449Y, A449N or less preferred A449E

• D447S or D447C

• A392I, A392L, A392V or A392M

• E360L, E360V or E360C

• E269H, E269T , E269M, E269C or E269Q

• Q518G

• G320Q, G320A or less preferred G320S

• A3861, A386L, A386V or A386M

• G266A

• A372S

• E453Y, E453W or less preferred E453F

• A460W

• A329Q, A329H or A329T

• D293Y

• R412E, R412D, R412Q or R412M

• T301S

• D199E

• A331F or A331Y

• S435T, S435R or S435I

• V274I

• D399S

• S230D

• S434G

• M333I or M333L

• N515G • A418W

• I391W

• E421R

• A487Q

• A378G

• A423V

• T326K, T326L, T326 R or T326W

• A433G

• D399S

• Y490W

• R516E or R516I

• P553K

• V314L

• S327W, S327L or S327FA475V

• A342R

• S354E or S354Q

• S315P

Some of these substitutions cause a high ΔΙΤ50 when introduced individually into the Kumamolisin AS as set forth in SEQ ID NO 1 or 4, and are therefore preferred, while others have a high occurrence in the combinatorial and distinct clones of Tables 2a, 2b and 4 and some combinations, which have a combination of individual substitutions with a high overall ΔΙΤ50.

Some can interchangeably be used to stabilize the enzyme and some combinations results in other traits that are relevant for the production or performance in feed, like fermentation titers, the hydrolysis of anti-nutritive factors as protease inhibitors (soy bean Bowman-Birk and Kunitz-type trypsin and/or chymotrypsin inhibitors), pH profile, pH and pepsin stability, or stability against and performance under higher ionic strength.

Note that, while the numbering set forth above refers to SEQ ID NO 1 or 4, the claimed protease can be a fragment, fraction or shuffled variant thereof maintaining proteolytic activity. In such case, the resulting amino acid sequence is shorter than that of SEQ ID NO 1 or 4, while the numbering of the mutant residues still refers to the full length SEQ ID NO 1 or 4. In one embodiment of the invention, the protease variant has at least two amino acid substitutions compared to the Kumamolisin AS backbone as set forth in SEQ ID NO 1 or 4. Preferably the protease variant has at least three, more preferably at least four, more preferably at least five and most preferably at least six amino acid substitutions selected from said group. Preferably, these amino acid substitutions are combinations of the individual substitutions discussed above.

In one embodiment of the invention, the protease variant has at least two amino acid substitutions compared to the Kumamolisin AS backbone as set forth in SEQ ID NO 1 or 4, the at least 2 amino acid substitutions being at two or more residue positions in SEQ ID NO 1 or 4 selected from the group consisting of 447 and 449, 453, 502, 510, 517, 360, 460, 199, 266, 301, 386 and 514. Preferably the protease variant has at least three, more preferably at least four, more preferably at least five and most preferably at least six amino acid substitutions selected from said group.

In one preferred embodiment, the protease variant has at least one, preferably at least two, more preferably at least three, more preferably at least four, more preferably at least five, most preferably at least six amino acid substitutions selected from the group consisting of D447S, A449Y, A517T, N510H, E360L, E360V, E360C, V502C, E453W, A514T, A514Y, A460W, A386I, D199E, G266A, T301S.

Tables 2a, 2b and 4 show sets of such so-called "distinct clones" or "combinatorial clones" which have combinations of the individual mutations set forth above.

As used herein, the term "combinatorial clone or variant" means a clone or variant screened from a recombination library. Such a recombination library contains a population carrying different amounts and mutations selected from the group of table 1.

As used herein, the term "distinct clone or variant" means A clone constructed containing a defined set of mutations selected from the group of table 1 in a rational approach.

Preferably, said improved stability which the protease variant according to the invention has is improved thermostability (IT50). The thermostability of an enzyme is usually determined by measuring the inactivation temperature (IT 50). The "inactivation temperature" is defined as the temperature at which the residual activity of the enzyme after incubation for a certain duration and subsequent cooling to room temperature is 50% of the residual activity of the same enzyme incubated for the same duration under the same conditions at room temperature.

According to one embodiment, the protease variant has a set of substitutions at selected residues in the Kumamolisin AS backbone as set forth in SEQ ID NO 1 or 4, which set is at least one of the following a) 360, 447, 449 and 510

b) 447, 449 and 514, and/or

c) 447, 449, 453, and 517.

These three sets of simultaneously substituted residues occur in three sets of specific distinct or combinatorial clones which are particularly preferred (consensus mutations). See Table 2a/Fig 3, Table 2b/Fig. 4 and Table 4/Fig. 5. For these reasons, these sets of simultaneously substituted residues seem to be particularly synergistic when it comes to improvement of stability.

According to one embodiment said improved stability is improved thermostability (IT50) of either the activated enzyme or the zymogen. In one embodiment of, the protease variant has an IT50 of between > 75 and < 105 °C.

In some embodiments, for the activated enzyme an IT50 of between > 70 and < 90°C is provided, while a for the zymogen an IT50 of between > 80 and < 105°C is provided.

The Kumamolisin AS wildtype enzyme has an IT50 of 79,6 °C +/- 0,4°C (n = 46) as the zymogen, i.e., the inactive zymogen, and an IT50 of 59°C +/- 1°C (n= 10) as the activated enzyme. In the course of this specification, the different variants are either characterized by their IT50, or by ΔΙΤ50 (i.e., the difference compared to the wildtype IT50).

According to another embodiment of the invention, a nucleic acid molecule encoding a protease variant according the above description is provided. Furthermore, a plasmid or vector system comprising said nucleic acid molecule is provided, as well as a host cell being transformed with said plasmid or vector and/or comprising said nucleic acid molecule is provided. Further, a method for producing a protease or protease variant is provided, said method encompassing: a) cultivating said host cell, and

b) isolating the protease or protease variant from said host cell, or harvesting the protease or protease variant from the medium.

According to another embodiment of the invention, a composition comprising a protease variant according to the above description is provided, which composition has a pH of > 5.

Such composition is generally discussed - yet not with the specific protease variants disclosed herein - in EP application No 16176044.2-1375 and later applications claiming the priority thereof, the content of which is incorporated by reference herein.

According to another embodiment of the invention, a feed additive, feed ingredient, feed supplement, and/or feedstuff comprising a protease variant or a composition according to the above description is provided.

Further, the use of a protease variant according to the above description for the manufacture of a feedstuff is provided.

Such feed additive, feed ingredient, feed supplement, and/or feedstuff is preferably meant for monogastric poultry, pig, fish and aquaculture, where it helps to increase protein digestion and absorbance from the feedstuff, plus degrade proteinogenic compounds which are detrimental for animal health or digestion.

Furthermore, the use of a protease according to the above description is provided for at least one purpose or agent selected from the group consisting of:

• detergent

• fruit and beverage processing

• leather processing

• production of protein hydrolysates • hard surface cleaning or biofilm cleaning

• treatment of necrotic or burned tissue to promote wound healing,

• processing aid in tissue engineering and/or

• food preparation including baking dough preparation.

Likewise, an additive, ingredient or agent for one purpose or agent selected from the group consisting of:

• detergent

• fruit and beverage processing

• leather processing

• production of protein hydrolysates

• hard surface cleaning or biofilm cleaning

• treatment of necrotic or burned tissue to promote wound healing

• processing aid in tissue engineering and/or

• food preparation including baking dough preparation. is provided which additive, ingredient or agent comprises a composition according to the above description.

Furthermore, a process of generating a protease variant according to the above description is provided, which process comprises: i) mutagenizing a DNA, cDNA or mR A encoding a Kumamolisin AS amino acid sequence as set forth in any of SEQ ID NOs 1 - 4

ii) expressing one or more mutants of Kumamolisin AS thus obtained , and

iii) testing the one or more mutants of Kumamolisin AS for at least stability, preferably thermostability.

Preferably, in said method, the encoding nucleic acid sequence and/or the amino acid sequence of one or variants of Kumamolisin AS is determined. For this purpose, routine methods from the prior art can be used. Experiments and Figures

While the invention has been illustrated and described in detail in the drawings and foregoing description, such illustration and description are to be considered illustrative or exemplary and not restrictive; the invention is not limited to the disclosed embodiments. Other variations to the disclosed embodiments can be understood and effected by those skilled in the art in practicing the claimed invention, from a study of the drawings, the disclosure, and the appended claims. Any reference signs should not be construed as limiting the scope.

All amino acid sequences disclosed herein are shown from N-terminus to C-terminus; all nucleic acid sequences disclosed herein are shown 5'->3'.

1. Amino acid Sequences of the Kumamolisin AS backbone

SEQ ID NO 1 shows the proenzyme (propeptide plus enzyme, also called zymogen herein) sequence of the Kumamolisin AS backbone used herein. It is important to understand that, while the wildtype sequence of Kumamolisin AS has an N-terminal M residue, the Kumamolisin AS backbone used herein lacks said M, because the latter was replaced by a signal sequence that was later cleaved off. Such signal sequence is for example. The sacB signal peptide MNIKKFAKQATVLTFTTA LLAGGATQAFA.

In SEQ ID NO 1, the propeptide hence comprises AAs 2 - 189 (former N-terminal M which is lacking is yet considered as AA NO 1 in the numbering of SEQ ID NO), and the enzyme comprises AAs 190 - 553:

SDMEKPWKE g i¾¾ ^5QiS: HARAQAPQAV DKGPVAGDER MAVTVVLRRQ RAGELAAHVE 60

RQAAIAPHAR EHLKREAFAA SHGASLDDFA ELRRFADAHG LALDRANVAA GTAVLSGPVD 120

AINRAFGVEL RHFDHPDGSY RSYLGEVTVP AS IAPMIEAV LGLDTRPVAR PHFRMQRRAE 180

GGFEARSQAA APTAYTPLDV AQAYQFPEGL DGQGQC IAI I ELGGGYDEAS LAQYFASLGV 240

PAPQVVSVSV DGASNQPTGD PSGPDG A/EL DIEVAGALAP GAKFAVYFAP NTDAGFLDAI 300

TTAIHDPTLK PSVVS I SWGG PEDSWTSAAI AAMNRAFLDA AALGVTVLAA AGDSGSTDGE 360

QDGLYHVDFP AAS PYVLACG GTRLVASGGR IAQETVWNDG PDGGATGGGV SRI FPLPAWQ 420

EHANVPPSAN PGAS SGRGVP DLAGNADPAT GYEVVI DGEA TVIGGTSAVA PLFAALVARI 4 80

NQKLGKAVGY LNPTLYQLPA DVFHDI TEGN NDIANRAQIY QAGPGWDPCT GLGS PIGVRL 540

LQALLPSASQ PQP 553 The propeptide is shaded in grey. The catalytic triad SED (=Ser/Glu/Asp) consists of E267, D271 and S467, shown in italics. The positions where the inventors have found mutations that result in altered/improved properties are underlined.

2. Amino acid Sequences of the Kumamolisin AS backbone plus leader sequence and HisTag

In SEQ ID NO 2, the sacB leader sequence comprises AAs 1 - 29 (wavy underline) and replaces the original N-terminal M of the propeptide. The propeptide (shaded in grey) comprises AA 30 - 217, the activated enzyme comprises AA 218 - 581 and the His-tag comprises AAs 582 - 587 (double underline).

MJJIKKFAj^^^ DMEKPWKEGE EARAVLQGHA RAQAPQAVDK 60

GPVAGDERMA VTVVLRRQRA GELAAHVERQ AAIAPHAREH LKREAFAASH GASLDDFAEL 120

RRFADAHGLA LDRANVAAGT AVLSGPVDAI NRAFGVELRH FDHPDGSYRS YLGEVTVPAS 180

IAPMIEAVLG LDTRPVARPH FRMQRRAEGG FEARSQAAAP TAYTPLDVAQ AYQFPEGLDG 240

QGQCIAI IEL GGGYDEASLA QYFASLGVPA PQVVSVSVDG ASNQPTGDPS GPDGEVELDI 300

EVAGALAPGA KFAVYFAPNT DAGFLDAITT AIHDPTLKPS VVSISWGGPE DSWTSAAIAA 360

MNRAFLDAAA LGVTVLAAAG DSGSTDGEQD GLYHVDFPAA SPYVLACGGT RLVASGGRIA 420

QETVWNDGPD GGATGGGVSR IFPLPAWQEH ANVPPSANPG ASSGRGVPDL AGNADPATGY 480

EVVIDGEATV IGGTSAVAPL FAALVARINQ KLGKAVGYLN PTLYQLPADV FHDITEGNND 540

IANRAQIYQA GPGWDPCTGL GSPIGVRLLQ ALLPSASQPQ PHHHHHH 587

3. Amino acid Sequences of the activated Kumamolisin AS backbone devoid of propeptide

In SEQ ID NO 3, the activated Kumamolisin AS backbone enzyme is shown with AAs 1 - 364:

AAPTAYTPLD VAQAYQFPEG LDGQGQCIAI IELGGGYDEA SLAQYFASLG VPAPQVVSVS 60

VDGASNQPTG DPSGPDGEVE LDIEVAGALA PGAKFAVYFA PNTDAGFLDA ITTAIHDPTL 120

KPSVVSISWG GPEDSWTSAA IAAMNRAFLD AAALGVTVLA AAGDSGSTDG EQDGLYHVDF 180

PAASPYVLAC GGTRLVASGG RIAQETVWND GPDGGATGGG VSRIFPLPAW QEHANVPPSA 240

NPGASSGRGV PDLAGNADPA TGYEVVIDGE ATVIGGTSAV APLFAALVAR INQKLGKAVG 300

YLNPTLYQLP ADVFHDITEG NNDIANRAQI YQAGPGWDPC TGLGSPIGVRL LQALLPSAS 360 QPQP 364 4. Amino acid Sequence of the Kumamolisin AS Wildtype

SEQ ID NO 4 shows the proenzyme (propeptide plus enzyme) sequence of the Kumamolisin AS wildtype, as obtained from Alicyclobacillus sendaiensis (GenBank: AB085855.1). SEQ ID NO 4 differs from SEQ ID NO 1 , which shows the sequence of the Kumamolisin AS backbone used herein in that the latter lacks the N-terminal M still present in the Wildtype SEQ ID No 4. This is because the N-terminal M was replaced, in SEQ ID No 1 , by the sacB signal sequence, which was later cleaved off. In SEQ ID NO 4, the propeptide comprises AAs 1 - 189, and the enzyme comprises AAs 190 - 553:

GEEARAVLQG HARAQAPQAV DKGPVAGDER RAGELAAHVE 60

RQAAIAPHAR EHLKREAFAA SHGASLDDFA ELRRFADAHG LALDRANVAA GTAVLSGPVD 120

AINRAFGVEL RHFDHPDGSY RSYLGEVTVP ASIAPMIEAV LGLDTRPVAR PHFRMQRRAE 180

GGFEARSQAA APTAYTPLDV AQAYQFPEGL DGQGQCIAI I ELGGGYDEAS LAQYFASLGV 240

PAPQVVSVSV DGASNQPTGD PSGPDGEVEL DIEVAGALAP GAKFAVYFAP NTDAGFLDAI 300

TTAIHDPTLK PSVVSISWGG PEDSWTSAAI AAMNRAFLDA AALGVTVLAA AGDSGSTDGE 360

QDGLYHVDFP AASPYVLACG GTRLVASGGR IAQETVWNDG PDGGATGGGV SRIFPLPAWQ 420

EHANVPPSAN PGASSGRGVP DLAGNADPAT GYEVVIDGEA TVIGGTSAVA PLFAALVARI 480

NQKLGKAVGY LNPTLYQLPA DVFHDITEGN DIANRAQIY QAGPGWDPCT GLGSPIGVRL 540

LQALLPSASQ PQP 553

Again, the propeptide is shaded in grey. The catalytic triad SED (=Ser/Glu/Asp) consists of E267, D271 and S467, shown in italics.

Short description of the Figures

Fig. 1 shows the distribution of mutations in variants optimized for thermal stability of the zymogen and the activated enzyme.

Fig. 2 shows the effects of the ionic strength on stability and performance for the WT and top variants #1 to #7 from table 4.

Fig. 3 - 5 show the occurrence of substitutions at AA position in different sets of distinct clones and combinatorial clones. Example 1: Protease activity assay

Protease activity assays were carried out in microtiter plates a) AAPF assay 96 well formate

Assay buffer: 200 niM Sodium Acetate, 1 mM CaCl ₂ , 0,01 % Triton X- 100 at pH 3

depending on the experiment

Substrate stock solution: 100 mM in water free DMSO

Substrate working solution: Substrate Stock solution diluted 1 :50 in assay buffer,

Execution: Load 50 of the diluted sample into the wells of a Nunc 96 clear flat bottom plate. Dilution is made in water containing 0.01% Triton-XlOO corresponding to the volumetric activity of the sample. Start reaction by adding 50 of substrate working solution. Measure kinetics at 37°C by monitoring the increase in adsorption at 410 nm as a measure for enzymatic activity. The activity was calculated by building a calibration curve with a reference enzyme preparation of the backbone with known proteolytic activity measured by a reference method.

For assaying the protease activity at different pH values the following buffers were used, each 200 mM: glycine/HCL between pH 2.0 - 3.0, trisodium citrate/citric acid between 3.0 and 6.0 and Tris/maleic acid between 6.0 and 7.5. b) ITso

IT50 defines the temperature where 50% of the activity is inactivated under the conditions described above. Although not equivalent to, it is a measure for the thermal stability in the application, e.g. pelleting conditions or conditions in a detergent application, either dish washing or the cleaning of a fabric or hard surface and other technical applications.

The screening of enzyme variants under predictive conditions is essential. For proteases like those described herein, screening for thermally more stable variants by methods as also described herein can be affected by the self-hydrolysis of the protease. As already described in patent application EP16176044 Example 9, screening for variants with higher thermal stability under conditions where the protease is active results in a large number of false positives, as the result of a mixed effect of thermal inactivation and self-hydrolysis. The same applications teaches to circumvent this problem in the absence of small molecule reversible enzyme inhibitors, as is the case for the class of acid protease described herein, by executing the test for thermal stability of the enzyme and enzyme variants in the form of the inactive enzyme zymogen in the way described below.

Assay buffers: 50 mM sodium phosphate, 0,25mM CaCl ₂ pH6.5

800 mM glycine/HCl pH2.8

Thermal inactivation execution: Samples were diluted corresponding to the volumetric activity in potassium phosphate buffer. The pH of the final solution was checked to be above pH 6.3. The samples were transferred in replicates, 20 per well, into a 384 well PCR plate according to the direction of the temperature gradient of the PCR machine. The plates were sealed with an adhesive or hot melting cover foil and incubated on a thermal gradient cycler with a temperature gradient of +/- 12 °C around the expected IT50 value for 10 minutes. The samples were cooled to 8°C before measuring the residual activity of the samples with AAPF-pNA as followed. Samples, 15 each from the temperature incubation plate were transferred into a 384 well greiner clear flat bottom PS-microplate and 9 of glycine buffer was added to activate the protease during an incubation of 1 hour at 37°C. After the activation of the protease the assay was started by adding 24 μΕ of an AAPF-pNA solution (2 mM AAPF-pNA in water with 0.01% Triton-X100) and activity was measured by following the kinetics at 37°C. The normalized experimental data for residual activity at the inactivation temperatures were fitted to a four parameter logistics function to evaluate the IT50. c) IT50 without propeptide - activated enzyme protein:

Enzyme activation prior to thermal inactivation execution. Samples were diluted corresponding to the volumetric activity in glycine buffer pH 2.8 as described in 2b) and pH was checked to be equal or lower than pH 4.0. Samples were activated by an incubation for 1 hour at 37°C. After the incubation pH was set to above 7.0 by diluting the samples 1 :3 in 50 mM sodium phosphate buffer pH 8.0. Thermal inactivation of activated enzyme protein execution. Aliquots of the activated enzyme protein were transferred in replicates, 20 μΐ _^ per well, into a 384 well PCR plate according to the direction of the temperature gradient of the PCR machine. The plates were sealed with an adhesive or hot melting cover foil and incubated on a thermal gradient cycler with a temperature gradient of +/- 12 °C around the expected IT50 value for 10 minutes. The samples were cooled to 8°C before measuring the residual activity of the samples with AAPF-pNA as followed. Samples, 15 each from the temperature incubation plate were transferred into a 384 well greiner clear flat bottom PS-microplate and 9 of glycine/HCl buffer was added to adjust the pH to 3.0. The assay was started by adding 24 μΐ, of an AAPF- pNA solution (2 mM AAPF-pNA in water with 0.01% Triton-XlOO) and activity was measured by following the kinetics at 37°C. The normalized experimental data for residual activity at the inactivation temperatures were fitted to a four parameter logistics function to evaluate the IT50. d) pH-profile - activated enzyme protein

Undiluted bacterial supernatant containing enzyme protein was titrated with 1 M HC1 to pH 4 and enzyme was activated at 37°C for 60 min. 20 of sample were added to 200 μί Britton Robinson buffer with pH 1.8-7.0 (adjusted to conductivity of 15 mS/cm with NaCl). 20 μΐ, were then transferred into a 384-well Greiner flat bottom PS-microplate plus 20 μΐ substrate solution (2 mM AAPF-pNA in water with 0.01% Triton-XlOO) and activity was measured by monitoring the kinetics at 410 nm and 37°C as described in example la). Each kinetic experiment was run in quadruplet. e) pH/Pepsin-Resistance

Undiluted bacterial supernatant containing enzyme protein was titrated with 1 M HC1 to pH 2.5. 90 μΐ were then transferred to a Nunc 96-well clear flat bottom microtiter plate. 10 μΐ of a 250 μg/mL Pepsin stock solution in pH 2.5 buffer (final concentration in assay 25 μg/mL) or pH 2.5 buffer were added to each well and then incubated at 37°C for 30 min. Finally, 5 μΐ of a 100 μΜ Pepstatin A solution (final concentration 5 μΜ) was added to each well to stop the pepsin reaction. 25 μΐ of the sample were transferred in 175 μΐ glycine/HCl buffer pH 3.0 in a new Nunc 96-well clear flat bottom microtiter plate. 20 μΐ were then transferred into a 384-well Greiner flat bottom PS-microplate plus 20 μΐ substrate solution (2 mM AAPF-pNA in water with 0.01% Triton-XlOO) and activity was measured by monitoring the kinetics at 410 nm and 37°C as described in example 1 a). Each kinetic experiment was run in quadruplet. f) Conductivity dependency

20 μΐ undiluted bacterial supernatant was diluted in 180 μΐ, glycine/HCl buffer pH 3.0 adjusted with NaCl to conductivity of 2, 4, 6, 10, 20, 30, 40, 50 mS/cm in a Nunc 96-well clear fiat bottom microtiter plate. The samples were incubated at 37°C for 20 min and then 20 μΐ _^ sample were then transferred into a 384-well Greiner flat bottom PS-microplate plus 20 μΐ substrate solution (2 mM AAPF-pNA in water with 0.01% Triton-XlOO) and activity was measured by monitoring the kinetics at 410 nm and 37°C as described in example 1 a). Each kinetic experiment was run in quadruplet. g) BBI/KTI Hydrolysis - Functional Trypsin Assay

Bowman-Birk and Kunitz-type inhibitors (BBI/KTI) are strong inhibitors of serine proteases which are widely spread in seed of legumes and cereal grains. The assay principle is that a proteolytic degradation of the BBI/KTI by protease activity recovers the natural trypsin activity on Benzyl-Arginine-pNA (Bz-R-pNA) substrate without inhibitors. 90 μΐ _^ of bacterial supernatant containing enzyme protein was diluted in glycine/HCl buffer to pH 3.0 and then incubated at 37°C for 30 min. 20 μΐ of the sample was then mixed with 20 μΐ inhibitor solution (KTI: 8 μg/mL; BBI: 16 μg/mL; KTI/BBI: 4/8 μg/mL diluted in glycine buffer pH 3.0) and further incubated at 37°C for 60 min. 15 μΐ of the sample were transferred into a 384-well Greiner flat bottom PS-microplate and then 15 μΐ trypsin solution in pH 8.0 (final trypsin concentration 1 μg/mL; final pH 7.0 or pH 7.5) was added to each well and the plate was incubated at 37°C for 10 min. Finally, 30 μΐ _^ substrate solution (2 mM Bz-R-pNA in water with 0.01% Triton-XlOO) was added to each well and activity was measured by monitoring the kinetics at 410 nm and 37°C as described in example 1 a). Each kinetic experiment was run in quadruplet.

Example 2: Generation of genetic diversity

Initial genetic diversity was introduced by randomizing each position of the active enzyme core sequence of SEQ ID NO 1. Mutant enzyme single site saturation libraries were introduced in the gene carried on an E.coli / Bacillus shuttle vector using mutagenesis methods as described in Green & Sambrook (eds), Molecular Cloning, 4 ^th edition, CSHL and suitable mutagenic PCR methods as disclosed in Cadwell and Joyce (PCR Methods Appl. 3 [194], 136-140. Protease enzyme variants were characterized after heterologous expression in Bacillus subtilis and phenotypically optimized variants selected by the screening procedure outlined in Example 3.

In general, methods to mutagenize a protein, like an enzyme, to obtain a library of mutated proteins members of which may have altered characteristics, are well established. Methods to mutagenize a protein encompass site directed mutagenesis and others, as described e.g. in Hsieh & Vaisvila (2013), content of which incorporated herein by reference for enablement purposes.

Such methods are sometimes called "directed evolution", namely when the established library is then screened for particular features. Packer & Liu (2015) provide an overview of the respective methodology, content of which incorporated herein by reference for enablement purposes.

Example 3: Phenotypically screening for enzyme variants with increased thermal stability

The generated genetic diversity either in the initial stage in form of single site saturation libraries or in the subsequent stage in the form of recombination libraries or distinct clones was screened for variants with an optimized phenotype, i.e. increased thermal stability using the method as described in example lb) with adaptations required to run them in a fully automated robotic workstation at high throughput. These were mainly adaptation in incubation times, volumes, substrate and the main adaptation was to select optimized variants not by the thermal inactivation profile on a temperature gradient but by the residual activity after incubation at a single temperature, the temperature which was set to discriminate optimized variants from the average of the genetic diversity. Protease variants were derived which differed in one or more amino acid positions from SEQ ID NO 2, including two positions, three positions, n positions. Appropriate iterative rounds of the procedures described herein were performed to satisfy the demands of the application

Example 4:

The following individual mutations which increase the IT50 compared to the used backbone were identified. The IT50 was analyzed as described above and compared to the IT50 of the used backbone (=wildtype with missing N-terminal methionine) characterizing the variant by the corresponding ΔΙΤ50. The backbone has an IT50 of 79,6 °C +/- 0,4°C (n zymogen and an IT50 of 59°C +/- 1°C (n= 10) as activated enzyme.

Position Mutation ΔΙΤ50 ΔΙΤ50 activated

Zymogen Enzyme

A190 D 1,5 0,8

T196 S 0,7 0,3

D199 E 0,5 1,0

Q202 D 0,4 -0,3

1219 L 1,1 0,8

E228 Q 0,7 0,1

A229 W 0,2 n.d.

S230 D 2,8 -0,8

A242 S 0,3 -0,4

Q244 C 0,5 -3,6

Q244 G 0,7 1,5

D251 S 0,8 -0,3

S262 C 0,9 -0,3

G266 A 1,7 0,0

E269 M 2,4 -o,i

E269 T 2,6 -o,i

E269 c 2,1 -1,1

E269 H 4,0 -0,5

E269 Q 2,0 -1,4

V274 1 1,8 1,3

G281 2,0 5,4

K283 L 0,6 -0,2

Y287 K 0,2 5,2

N291 T 0,7 0,5

N291 s -0,2 1,0

D293 Y 0,8 1,0

D293 F 1,1 1,3

L297 T 1,2 0,2

T301 s 0,6 7,6

T301 c 0,8 1,0

T301 M 0,7 0,5

H305 F 0,4 -0,4

H305 W 0,1 -2,7

D306 S 0,3 -0,5

T308 C 0,5 -0,8

V314 M 0,6 0,3

V314 L 2,5 0,7

S315P P 0,8 3,0 G320 A 3,0 -0,2

G320 Q 3,6 1,5

G320 S 1,0 0,6

S324 L 0, 1 1,3

S324 0,7 2,0

W325 K -0,3 2,7

T326 R 1,7 1,2

T326 W 0,9 0,2

T326 L 1,7 1,6

T326 K 1,9 1,2

S327 F 1,2 0,6

S327 L 1,5 1, 1

S327 W 2,0 1,0

A328 W 0,6 0,5

A328 D 1,3 1, 1

A328 R 1, 1 0, 1

A328 Y 1,5 0,8

A329 Q 2,8 0,2

A329 H 2, 1 0,3

A329 T 1,0 0,9

1330 L 1, 1 0,8

A331 F 2,0 0,6

A331 Y 1,3 0,6

M333 1 2,5 -0,7

M333 Y 0,3 1,0

M333 L 2,4 -i,o

L338 R -0,5 1,5

A342 R -0,6 3,9

A351 S 1,3 -0,9

S354 E 1,6 3,3

S354 Q 2,0 0,3

D358 G -2,0 0,7

E360 L 1,4 3, 1

E360 V 2,4 2,9

E360 C 2,3 2,3

Q361 C 0,9 1,5

Q361 L 0,2 0, 1

A372 S 2,4 -0,7

A378 G 1,5 1,5

A386 1 3,6 0,5

A386 L 2,7 1,3

A386 V 2, 1 1,2

A386 M 1,7 0,0 G388 C 0,6 -3,5

1391 W 1,7 0,6

A392 V 2,8 0,7

A392 L 3,0 0,9

A392 1 3,7 2,4

A392 M 2,3 2,0

Q393 S 0,9 0,2

D399 S 2,3 2,1

D402 E 0,6 1,7 412 Q 0,5 2,4

R412 M 1,5 2,9

R412 E 1,8 4,4

R412 D 0,4 3,5

A418 W 2,8 0,2

E421 R 1,0 0,5

A423 V 1,1 0,8

A433 G 1,4 1,9

S434 G 1,9 0,7

S435 1 1,7 1,6

S435 R 1,8 0,5

S435 T 2,5 4,7

S435 V 1,6 2,1

L442 w 1,4 0,3

L442 w -0,7 2,4

D447 s 4,0 3,2

D447 c 3,0 1,4

D447 A 1,6 1,3

A449 Y 1,7 0,7

A449 L 0,8 0,3

A449 M 1,9 -0,9

A449 E 1,6 0,4

A449 N 1,6 3,3

E453 W 2,4 0,0

E453 Y 2,6 0,7

E453 F 1,1 -0,5

V455 1 1,2 0,3

V455 L 1,8 0,7

E459 W 0,9 -0,3

A460 W 2,6 0,5

A460 R 2,0 -0,6

T461 V 1,2 0,0

T461 C 1,2 0,6

A470 V 0,6 2,3

A475 V -0,3 3,7 A478 L 1,2 0,2

K483 A 1,5 0,7

A487 Q 0,0 1,6

Y490 W 1,5 0,3

Q497 Y 1,8 1,2

Q497 M 0,8 0,8

Q497 D 0,3 1,0

Q497 R 0,6 0,2

V502 C 2,3 1,9

V502 T 1,5 1,6

T507 L 0,2 1,0

N510 H 2,4 7,9

A514 T 2,2 1,3

A514 Y 1,3 -1,2

A514 D 1,5 1,2

A514 S 2,4 0,5

N515 G 2,0 -0,2 516 L 0,5 1,2

R516 E 1,1 3,5

R516 1 1,2 4,3

A517 T 1,3 3,9

A517 s 0,3 7,7

Q518 G 1,6 4,1

L540 V 0,7 0,5

Q542 H 0,9 -0,2

Q542 D 1,1 0,4

Q542 S 0,4 0,5

A548 S 0,2 n.d.

P551 N 0,9 -0,4

P551 R 0,6 0,3

P553 K 0,5 0,3

P553 L 0,8 0,2

R166 1 1,0 0,7

D265 T 1,7 n.d.

Table 1: Kumamolisin AS single amino acid substitutions relative to SEQ ID NO 1, and their ΔΙΤ50 compared to the backbone for the zymogen and the activated enzyme

Quite a few distinct clones and combinatorial clones as shown in Table 3 have substitutions in these positions, leading to synergistic effects in thermal stabilization, when two or more residues thereof are mutated simultaneously. Example 5

Distinct variants were generated by introducing selected distinct mutations into the Kumamolisin AS wild-type sequence via site-directed mutagenesis. Suitable mutagenic PCR methods known in the art and standard cloning techniques as described in Green & Sambrook (eds), Molecular Cloning, 4 ^th edition, CSHL were used. Protease enzyme variants were characterized after heterologous expression in Bacillus subtilis and phenotypically analysis using the methods described above.

Combinatorial libraries, combining mutations identified in the examples provided above and outlined in Table 1 were generated by well-known PCR methods as described in Yolov and Shabarova (1990) and standard cloning techniques as described in Green & Sambrook (eds), Molecular Cloning, 4th edition, CSHL were used. Combinatorial libraries were screened for optimized variants as described in example 3.

Example 6

Distinct clones and combinatorial clones comprising two or more mutations from Table 1 were identified, the IT50 analyzed as described above and compared to the IT50 of the used backbone (=wildtype with missing N-terminal methionine) characterizing the variant by the corresponding ΔΙΤ50. As the IT50 of the backbone was determined in the same experiment as the variant the measured IT50 of the backbone can be slightly different from the average value. Results are shown in the following Table 2a (Fig. 3 shows results in graphic form):

Mutations in distinct clones and selected combinatorial clones

Table 2a: Distinct clones comprising selected combinations of mutations from table 1, and their ΔΙΤ50 compared to the wildtype # Mutations i n distinct clones and selected combinatorial clones

1 D447S A449Y A460W V502C N510H

2 D447S A449Y E453W A460W V502C N510H

3 D447S A449Y E453W A460W V502C N510H

4 D447S A449Y A460W V502C N510H

5 D447S A449Y E453W A460W V502C N510H

6 D447S A449Y E453W A460W V502C N510H

7 D447S A449Y E453W V502C N510H

8 D447S A449Y E453W A460W V502C N510H

9 D447S A449Y E453W A460W N510H

10 D447S A449Y E453W A460W V502C N510H

11 D447S A449Y E453W A460W V502C N510H

12 D447S A449Y E453W A460W V502C N510H

13 D447S A449Y E453W A460W V502C N510H

14 D447S A449Y A460W V502C N510H

15 D447S A449Y E453W A460W V502C N510H

16 D447S A449Y A460W V502C N510H

17 D447S A449Y E453W A460W V502C N510H

18 D447S A449Y E453W A460W V502C N510H

19 D447S A449Y E453W A460W V502C N510H

20 D447S A449Y E453W A460W V502C N510H

21 D447S A449Y A460W V502C N510H

22 D447S A449Y A460W V502C N510H

23 D447S A449Y E453W A460W V502C N510H

24 D447S A449Y E453W A460W V502C N510H

25 D447S A449Y V502C N510H

26 D447S A449Y E453W A460W V502C N510H

27 D447S A449Y E453W A460W V502C N510H

28 D447S A449Y A460W V502C N510H

29 D447S A449Y E453W A460W V502C N510H

30 D447S A449Y E453W A460W V502C N510H

31 D447S A449Y E453W A460W V502C N510H

32 D447S A449Y E453W A460W V502C N510H

33 D447S A449Y E453W A460W V502C N510H

34 D447S A449Y E453W A460W V502C N510H

35 D447S A449Y E453W V502C N510H

36 D447S A449Y E453W A460W V502C N510H

37 D447S A449Y E453W A460W V502C N510H

38 D447S A449Y E453W A460W V502C N510H

39 D447S A449Y E453W A460W V502C N510H

40 D447S A449Y E453W A460W V502C N510H

Table 2a ctd': Distinct clones comprising selected combinations of mutations from table

1, and their ΔΙΤ50 compared to the wildtype # Mutations in distinct clones and selected IT50 ΔΙΤ50 IT50 active ΔΙΤ50 active combinatorial clones Zymogen Zymogen enzyme enzyme

1 A517T 95,5 17,0 90,1 30,6

2 A517T >95 >17 90,1 30,6

3 A517T 99,5 21,0 89,2 29,7

4 A517T 97,3 18,8 89,1 29,6

5 A517T 99,4 20,9 88,8 29,3

6 A517T 96,4 17,9 88,6 29,1

7 A517T 96,4 17,9 88,5 29,0

8 A517T 99,1 20,6 88,5 29,0

9 A517T Q518G 97,8 19,3 88,5 29,0

10 A517T 98,4 19,9 88,4 28,9

11 A517T 97,7 19,2 88,4 28,9

12 A517T 98,6 20,1 88,3 28,8

13 A517T 99,5 21,0 88,2 28,7

14 A517T >95 >17 88,2 28,7

15 A517T 98,3 19,8 88,1 28,6

16 A517T 95,8 17,3 88,0 28,5

17 A517T 97,2 18,7 88,0 28,5

18 A517T 97,6 19,1 87,8 28,3

19 A517T 98,5 20,0 87,8 28,3

20 A517T Q518G 97,0 18,5 87,8 28,3

21 A517T >95 >17 87,8 28,3

22 A517T >95 >17 87,8 28,3

23 A517T 97,1 18,6 87,8 28,3

24 A517T 99,0 20,5 87,8 28,3

25 A517T 94,0 16,0 87,7 27,0

26 A517T 97,4 18,9 87,7 28,2

27 A517T Q518G 98,0 19,5 87,7 28,2

28 A517T >95 >17 87,6 28,1

29 A517T 96,5 18,0 87,6 28,1

30 A517T 99,0 20,2 87,5 28,0

31 A517T 98,1 19,6 87,5 28,0

32 A517T Q518G 97,9 19,4 87,4 27,9

33 A517T Q518G 97,1 18,6 87,4 27,9

34 A517T 95,6 17,1 87,4 27,9

35 A517T 98,2 19,7 87,4 27,9

36 A517T 98,5 20,0 87,4 27,9

37 A517T Q518G 97,9 19,4 87,4 27,9

38 A517T >95 >17 87,3 27,8

39 A517T 96,2 17,7 87,2 27,7

40 A517T Q518G >95 >17 87,1 27,6

Table 2a ctd': Distinct clones comprising selected combinations of mutations from table

1, and their ΔΙΤ50 compared to the wildtype # Mutations in distinct clones and selected combinatorial clones

41 T301S E360L A386I

42 E360V A392I

43 E360V A386I A392I

44 E360L A392V

45 D199E E360V

46 T301S E360L A386I A392I

47 D199E G266A E360V A392V

48 G266A E360V A392V

49 E360L A392I

50 T301S E360V A386I

51 E360L A386I

52 E360L A386I

53 D199E G266A E360V A392V

54 E360V A386I A392I

55 D199E G266A E269H E360V A392L

56 E360V A386I

57 E360V A392V

58 T301S E360L A386I A392I

59 D199E E360V A386I

60 E360V A386I A392I

61 D199E E360V A386I

62 E360V A386I

63 D199E G266A T301S E360L

64 D199E G266A E269T G320A E360V A392L

65 E360L A386I

66 G266A E360V A392V

67 E360L A386I

68 D199E E360V

69 D199E E360V A386I

70 D199E E360L

71 D199E G266A E269H T301S E360L

72 D199E E360L

73 D199E G266A E360V

74 D199E G266A E269H E360V A392L

75 D199E E360V

76 T301S E360L A392I

77 D199E E360V

78 D199E E360L

79 E360V A386I

Table 2a ctd': Distinct clones comprising selected combinations of mutations from table

1, and their ΔΙΤ50 compared to the wildtype # Mutations i n distinct clones and selected combinatorial clones

41 D447S A449Y E453W A460W V502C N510H

42 D447S A449Y A460W N510H

43 D447S A449Y E453W V502C N510H

44 D447S A449Y E453W A460W N510H

45 D447S A449Y E453W A460W V502C N510H

46 D447S A449Y E453W A460W V502C N510H

47 D447S A449Y E453W A460W V502C N510H

48 R412E D447S A449Y E453W A460W

49 D447S A449Y E453W V502C N510H

50 D447S A449Y A460W V502C N510H

51 D447S A449Y E453W V502C N510H

52 D447S A449Y A460W V502C N510H

53 D447S A449Y E453W A460W V502C N510H

54 D447S A449Y E453W A460W V502C N510H

55 D447S A449Y E453W A460W V502C N510H

56 D447S A449Y E453W V502C N510H

57 D447S A449Y E453W A460W V502C N510H

58 D447S A449Y A460W V502C N510H

59 D447S A449Y V502C N510H

60 D447S A449Y E453W A460W V502C N510H

61 D447S A449Y E453W V502C N510H

62 D447S A449Y A460W V502C N510H

63 D447S A449Y E453W V502C N510H

64 D447S A449Y E453W A460W V502C N510H

65 D447S A449Y V502C N510H

66 D447S A449Y E453W A460W

67 D447S A449Y E453W V502C N510H

68 D447S A449Y E453W V502C N510H

69 D447S A449Y E453W A460W V502C N510H

70 D447S A449Y A460W V502C N510H

71 D447S A449Y E453W V502C N510H

72 D447S A449Y E453W V502C N510H

73 D447S A449Y E453W V502C N510H

74 D447S A449Y E453W A460W V502C N510H

75 D447S A449Y A460W V502C N510H

76 D447S A449Y E453W A460W N510H

77 D447S A449Y E453W V502C N510H

78 D447S A449Y E453W V502C N510H

79 D447S A449Y V502C N510H

Table 2a ctd': Distinct clones comprising selected combinations of mutations from table

1, and their ΔΙΤ50 compared to the wildtype # Mutations in distinct clones and selected IT50 ΔΙΤ50 IT50 active ΔΙΤ50 active combinatorial clones Zymogen Zymogen enzyme enzyme

41 A517T Q518G 98,4 19,9 87,0 27,5

42 A517T 92,7 14,2 86,9 27,4

43 A517T >95 >17 86,9 27,4

44 A517T 96,9 18,4 86,8 27,3

45 A517T 93,5 15,1 86,7 26,0

46 A517T Q518G 97,4 18,9 86,7 27,2

47 A517T Q518G 101,5 23,0 86,6 27,1

48 A517T Q518G 100,3 21,8 86,6 27,1

49 A517T Q518G >95 >17 86,6 27,1

50 A517T Q518G 94,7 16,2 86,5 27,0

51 A517T >95 >17 86,5 27,0

52 A517T >95 >17 86,4 26,9

53 A517T Q518G P553K 102,2 23,7 86,4 26,9

54 A517T Q518G >95 >17 86,4 26,9

55 A517T Q518G P553K 101,7 23,2 86,3 26,8

56 A517T 93,1 14,6 86,2 27,0

57 A517T Q518G >95 >17 86,2 26,7

58 A517T Q518G >95 >17 86,2 26,7

59 A517T 92,2 14,1 86,1 25,4

60 A517T Q518G 97,2 18,7 86,0 26,5

61 A517T 93,2 14,9 85,9 26,2

62 A517T 92,4 13,9 85,9 26,6

63 A517T 95,7 17,2 85,8 26,3

64 A517T Q518G P553K 100,1 21,6 85,8 26,8

65 A517T 94,0 15,9 85,8 25,1

66 R516I A517T Q518G 100,1 21,6 85,7 26,2

67 A517T >95 >17 85,7 26,2

68 A517T Q518G >95 >17 85,7 26,2

69 A517T 94,9 16,6 85,4 25,7

70 A517T -10,0 -10,0 85,4 25,9

71 A517T 95,8 17,3 85,4 25,9

72 A517T Q518G >95 >17 85,4 25,9

73 A517T Q518G >95 >17 85,4 25,9

74 A517T Q518G 100,4 21,9 85,3 25,8

75 A517T >95 >17 85,3 25,8

76 A517T Q518G 95,1 16,6 85,2 25,7

77 A517T 94,9 16,4 85,1 25,5

78 A517T >95 >17 85,1 25,6

79 A517T 93,0 15,0 85,0 25,7

Table 2a ctd': Distinct clones comprising selected combinations of mutations from table

1, and their ΔΙΤ50 compared to the wildtype G320 A 3,0 -0,2 2 186

Q 3,6 1,5 46

S 1,0 0,6 35

T326 1,7 1,2 11

W 0,9 0,2

L 1,7 1,6 1 6

K 1,9 1,2 1 1

T461 V 1,2 0,0 1 26

C 1,2 0,6 48

Q244 C 0,5 -3,6 46

G 0,7 1,5 1 1

D293 Y 0,8 1,0 1 24

F 1,1 1,3

A487 Q 0,0 1,6 1 24

V274 1 1,8 1,3 104

A372 S 2,4 -0,7 82

K283 L 0,6 -0,2 68

T308 C 0,5 -0,8 30

A418 W 2,8 0,2 12

H 1,1 1,3 16

1391 W 1,7 0,6 21

A423 V 1,1 0,8 18

A331 F 2,0 0,6 7

Y 1,3 0,6 9

S327 F 1,2 0,6

L 1,5 1,1 16

W 2,0 1,0

1219 L 1,1 0,8 16

M333 1 2,5 -0,7 16

A329 Q 2,8 0,2 5

H 2,1 0,3 3

T 1,0 0,9 7

N515 G 2,0 -0,2 13

A378 G 1,5 1,5 12

S434 G 1,9 0,7 12

E421 R 1,0 0,5 1 11

A433 G 1,4 1,9 11

S230 D 2,8 -0,8 9

Q393 S 0,9 0,2 3

D399 S 2,3 2,1 4

Y490 W 1,5 0,3 2

G281 R 2,0 5,4 Y287 K 0,2 5,2 516 1 1,2 4,3

E 1,1 3,5 L 0,5 1,2

A475 V -0,3 3,7

S354 E 1,6 3,3

S315P P 0,8 3,0

W325 K -0,3 2,7

L442 W -0,7 2,4

W 1,4 0,3

A470 V 0,6 2,3

S324 R 0,7 2,0 S324 L 0,1 1,3

Q361 C 0,9 1,5 Q361 L 0,2 0,1

A190 D 1,5 0,8

T196 S 0,7 0,3

Q202 D 0,4 -0,3

E228 Q 0,7 0,1

A229 W 0,2 n.d.

A242 S 0,3 -0,4

D251 S 0,8 -0,4

S262 C 0,9 -0,3

N291 T 0,7 0,5

N291 S -0,2 1,0

L297 T 1,2 0,2

H305 F 0,4 -0,4 H305 W 0,1 -2,7

D306 S 0,3 -0,5

V314 M 0,6 0,3 V314 L 2,5 0,7

A328 W 0,6 0,5 A328 D 1,3 1,1 A328 R 1,1 0,1 A328 Y 1,5 0,8

1330 L 1,1 0,8

M333 Y 0,3 1,0 M333 L 2,4 -i,o

L338 R -0,5 1,5

A342 R -0,6 3,9

A351 S 1,3 -0,9

S354 Q 2,0 0,3

D358 G -2,0 0,7 G388 C 0,6 -3,5

D402 E 0,6 1,7

V455 1 1,2 0,3

V455 L 1,8 0,7

E459 W 0,9 -0,3

A478 L 1,2 0,2

K483 A 1,5 0,7

Q497 Y 1,8 1,2

Q497 M 0,8 0,8

Q497 D 0,3 1,0

Q497 R 0,6 0,2

V502 T 1,5 1,6

T507 L 0,2 1,0

L540 V 0,7 0,5

Q542 H 0,9 -0,2

Q542 D 1,1 0,4

Q542 S 0,4 0,5

A548 S 0,2 n.d.

P551 N 0,9 -0,4

P551 R 0,6 0,3

P553 L 0,8 0,2

166 1 1,0 0,7

D265 T 1,7 n.d.

Table 3: Some preferred substitutions and their key characteristics

It is further to be understood that the mutations can have positive or negative effects on other enzyme parameters, as the producibility in fermentative microbial production systems or the stability against pH- conditions or endogenous proteases of the animal, like pepsin. Testing the stability of feed enzymes at low pH and in the presence of pepsin is a standard for feed enzymes and was performed in this study as outlined in example le. The stability against higher ionic strength is not a standard test for feed enzymes though high ion concentrations can interfere with the enzyme stability and with the enzyme performance under such conditions and can be found for example in the gut. The secretion of acid in the gut and the feed ingredients translate to an increased ionic strength.

Fig. 2 shows that the wildtype suffers from combined effects of stability and performance reduction in the presence of higher ionic strength. Fig. 2 also shows the effect of ionic strength on the top variants also shown in table 4, variants #1 to #7. The performance and stability in high ionic strength was tested as described in example Id. The pH profile was a control parameter and tested as described in example If. The digestion of proteinaceous antinutritive factors like the Trypsin/chymotrypsin inhibitors BBI and KTI (Bowman-Birk inhibitors and Kunitz-type inhibitors) is a potential beneficial performance characteristic of a protease which was tested as described in example lg.

From the 651 individual combinatorial and distinct variants tested in detail, Table 4 describes the variants consolidating a multitude of performance and stability parameters (Fig. 5 shows results in graphic form).

All variants shown in table 4 are better or equally well produced in a microbial production system than the wildtype and have no relevant changes in their pH activity profile tested as described in example Id. Table 4 ranks these variants based on the thermal stability of the activated enzyme, the pH/pepsin stability and the stability against and the performance under higher ionic strength.

It was further found that the best variants can hydrolyze BBI and KTI (Bowman-Birk inhibitors and Kunitz-type inhibitors) as tested in a functional trypsin inhibition assay, which differentiates these variants from the parent enzyme, beside the high thermal stability engineered into these variants.

Table 4: Some distinct and combinatorial clones with particularly good performance The following Table 5 shows the frequency of occurrence of given mutations preferred combinatorial and distinct variants. The frequency of occurrence is a measure for the role and importance of a given mutation.

Table 5: Frequency of occurrence of given mutations in preferred combinatorial and distinct variants. Frequency of occurrence is a measure for the role and importance of a given mutation. The following Table 6 shows the impact of single mutations on ΔΙΤ50 of the zymogen or the activated form. Again, the amount of impact of a single mutation on ΔΙΤ50 is a measure for the role and importance of a said mutation.

Table 6: Impact of single mutations on ΔΙΤ50 of the zymogen (left) or the activated form (right). The amount of impact of a single mutation on ΔΙΤ50 is a measure for the role and importance of a said mutation. Mutation ΔΙΤ50 Mutation ΔΙΤ50 Mutation ΔΙΤ50 Mutation ΔΙΤ50 Zymogen Zymogen activated activated

Enzyme Enzyme

A514T 2,2 S315P 0,8 V502C 1,9 L297T 0,2

E269C 2,1 A449L 0,8 A433G 1,9 A478L 0,2

A329H 2,1 Q497M 0,8 S324R 1,9 T326W 0,2

A331F 2,1 P553L 0,8 D402E 1,7 Q393S 0,2

A386V 2,1 T196S 0,7 T326L 1,6 P553L 0,2

E269Q 2,0 E228Q 0,7 S435I 1,6 Q497R 0,2

G281 2,0 Q244G 0,7 V502T 1,6 A331F 0,1

S327W 2,0 N291T 0,7 A487Q 1,6 A328R 0,1

S354Q 2,0 T301M 0,7 G320Q 1,5 E228Q 0,1

A460R 2,0 L540V 0,7 A378G 1,5 Q361L 0,1

N515G 2,0 K283L 0,6 Q361C 1,5 E453W 0

T326K 1,9 T301S 0,6 Q244G 1,5 G266A 0

S434G 1,9 V314M 0,6 L338R 1,5 A386M 0

A449M 1,9 S324R 0,6 D447C 1,4 I391W 0

V274I 1,8 A328W 0,6 A386L 1,3 T461V 0

R412E 1,8 G388C 0,6 A514T 1,3 E269T -o,i

S435R 1,8 D402E 0,6 V274I 1,3 E269M -o,i

V455L 1,8 A470V 0,6 D447A 1,3 Q542H -ο,ι

Q497Y 1,8 Q497R 0,6 D293F 1,3 G320A -0,2

G266A 1,7 P551R 0,6 S324L 1,3 N515G -0,2

T326R 1,7 D199E 0,5 A386V 1,2 K283L -0,2

T326L 1,7 Q244C 0,5 T326K 1,2 S262C -0,3

A386M 1,7 T308C 0,5 Q497Y 1,2 E459W -0,3

I391W 1,7 R412Q 0,5 T326R 1,2 D251S -0,3

S435I 1,7 R516L 0,5 A514D 1,2 Q202D -0,3

A449Y 1,7 P553K 0,5 R516L 1,2 P551N -0,4

D265T 1,7 Q202D 0,4 S327L 1,1 H305F -0,4

S354E 1,6 H305F 0,4 A328D 1,1 A242S -0,4

S435V 1,6 R412D 0,4 S327W 1,0 E269H -0,5

D447A 1,6 Q542S 0,4 D293Y 1,0 E453F -0,5

A449E 1,6 A242S 0,3 T301C 1,0 D306S -0,5

A449N 1,6 D306S 0,3 D199E 1,0 A460R -0,6

Q518G 1,6 M333Y 0,3 M333Y 1,0 M333I -0,7

A190D 1,5 Q497D 0,3 Q497D 1,0 A372S -0,7

S327L 1,5 A517S 0,3 T507L 1,0 S230D -0,8

A328Y 1,5 A229W 0,2 N291S 1,0 T308C -0,8

A378G 1,5 Y287K 0,2 A392L 0,9 A449M -0,9

Table 6 ctd': Impact of single mutations on ΔΙΤ50 of the zymogen (left) or the activated form (right). The amount of impact of a single mutation on ΔΙΤ50 is a measure for the role and importance of a said mutation. Mutation ΔΙΤ50 Mutation ΔΙΤ50 Mutation ΔΙΤ50 Mutation ΔΙΤ50 Zymogen Zymogen activated activated

Enzyme Enzyme 412M 1,5 Q361L 0,2 A329T 0,9 A351S -0,9

K483A 1,5 T507L 0,2 V455L 0,8 M333L -1,0

Y490W 1,5 A548S 0,2 A328Y 0,8 E269C -1,1

A514D 1,5 S324L 0,1 I330L 0,8 E269Q -1,4

E360L 1,4 A487Q 0 A423V 0,8 H305W -2,7

A433G 1,4 N291S -0,2 Q497M 0,8 G388C -3,5

L442W 1,4 W325K -0,3 A392V 0,7 Q244C -3,6

A328D 1,3 A475V -0,3 E453Y 0,7 D265T n.d.

A331Y 1,3 L338R -0,5 V314L 0,7 R166I n.d.

A351S 1,3 A342R -0,6 S434G 0,7 V410I n.d.

A514Y 1,3 L442W -0,7 A449Y 0,7 A229W n.d.

A517T 1,3 D358G -2 K483A 0,7 A548S n.d.

Table 6 ctd': Impact of single mutations on ΔΙΤ50 of the zymogen (left) or the activated form (right). The amount of impact of a single mutation on ΔΙΤ50 is a measure for the role and importance of a said mutation.

It is further to be understood that some mutations of Table 1 and Table 6 can interchangeably be used to engineer thermostability in Kumamolisin As. Table 7 shows a set of variants based on variant #1 of Table 7. In the course of engineering the mutations at position 502 and 510 seemed to change the activity at extrem acidic pH, below pH 2.

Excluding mutations at 502 and 510 reduced the thermostability significantly below the targeted temperature stability for the activated enzyme, as for example in Table 7, clone #2 which has a 7,8°C reduction in thermal stability compared to clone #1. A set of distinct variants were constructed by a rational approach taking advantage of the mutations identified and shown in Tables 1 and 6 to compensate for the effect of 502 and 510. With the exception of D399S substitutions can gradually or fully compensate the effect of mutations at 502 and 510.

References:

Wlodawer Al, Li M, Gustchina A, Oyama H, Dunn BM, Oda K., Acta Biochim Pol.

2003;50(1):81-102

TerashitaJ., Oda,K., Κοηο,Μ. & Murao,S., Agric Biol Chem (1981) 45, 1937-1943

Oda,K., Takahashi,S., Ito,M. & Dunn,B.M., Adv Exp Med Biol (1998) 436, 349-353 Packer & Liu, Methods for the directed evolution of proteins. Nature Reviews Genetics 16, 379-394 (2015)

Hsieh & Vaisvila, Protein engineering: single or multiple site-directed mutagenesis. Methods Mol Biol. 2013;978: 173-86

Cadwell and Joyce, Mutagenic PCR. PCR Methods Appl. 3, 1994, 136-140

Okubo et al, 2006 Jun;273(l l):2563-76.