Biochemical and Biophysical Research Communications 312 (2003) 179–184

BBRC
www.elsevier.com/locate/ybbrc

The term phytase comprises several different classes of enzymes

Edward J. Mullaney and Abul H.J. Ullah*
Southern Regional Research Center, Agricultural Research Center, United States Department of Agriculture, New Orleans, LA 70124, USA
Received 24 September 2003

Dedication reflect different methods by which organisms dephos-

phorylate phytate to produce myo-inositol phosphates.
This minireview on phytases’ classification is dedi- Research on the active site geometry and the mechanism
cated to honor Professor I.C. Gunsalus, a great mentor is emerging. Consequently, this knowledge allows us to
to one of us.1 Several of the research findings from our assign these enzymes to specific classes of enzymes.
laboratory had reached his office in the form of manu- In order for an enzyme to be a phytase it must display
scripts as editor of Biochemical Biophysical Research phosphatase activity. Depending on the pH versus ac-
Communications. Professor Gunsalus supported our tivity profile and the optimum pH for catalysis, these
work by encouraging us to publish in a timely fashion. enzymes are further broadly classified as acid, neutral,
He leaves a legacy of outstanding research and a will- or alkaline phosphatases. Since most of the recent in-
ingness to encourage others to conduct innovative work. terest generated in phytase research had centered on
We thank him for encouraging us to push the boundary identifying an enzyme that would function effectively in
of phytase research over the last decade. the digestive tract of monogastric animals, most of these
studies have focused on acid phosphatases with high
specific activity for the preferred substrate, phytic acid.
Introduction Within this subdivision, three structurally distinct clas-
ses of enzymes have been described to date as phytases.
Phytase is a generic term used to describe an enzyme These three classes include representatives of histidine
that hydrolyzes phosphomonoester bonds from phytic acid phosphatases (HAP), b propeller phytase (BPP),
acid (myo-inositol 1,2,3,4,5,6-hexakis dihydrogen phos- and purple acid phosphatases (PAP) (Table 1). In this
phate), thereby liberating inorganic orthophosphates [1]. minireview, examples and unique properties of each
The number of enzymes described as phytase or phytate- group will be presented.
degrading enzymes has increased over the last decade
[2]. However, not all of these enzymes are structurally
similar nor do they all cleave phosphate groups from Histidine acid phosphatase phytases
phytate with the same mechanism. The varied catalytic
properties and requirements observed in these enzymes Most known phytases are HAPs (EC 3.1.3.8). All
members of this class share both a common active site
motif, RHGXRXP, and a two-step mechanism that
*
Corresponding author. Fax: 1-504-286-4367. hydrolyzes phosphomonoesters [3,4]. However, not all
E-mail address: [email protected] (A.H.J. Ullah). HAPs are catalytically active as phytases. Recent re-
1
A.H.J.U. is grateful to Professor Gunsalus for the opportunity to
work in his laboratory as a postdoctoral fellow for over three years in search [5] has established a vital role for the enzyme’s
the early 1980s on linaloool 8-methyl hydroxylation system of substrate specificity site (SSS). Even among known HAP
Pseudomonas incognita. The training that I received during my tenure phytases, specific activity for phytic acid varies a great
at Urbana, Illinois, shaped my future research activities on phytase at deal. For example, Wyss and co-workers [6] compared
USDA’s Southern Research Research Center at New Orleans.
the catalytic properties of several fungal phytases and
Research on phytase is expanding because of its linkage to animal
nutrition, eutrophication, and phosphate depletion from soil. Research
proposed two classes of HAP phytases. One class has a
ideas obtained in Professor Gunsalus’s laboratory when applied to broad substrate specificity but a lower specific activity
phytase research in the mid-1980s made a significant impact. for phytic acid and the second class has a narrow sub-

0006-291X/$ - see front matter Ó 2003 Elsevier Inc. All rights reserved.
doi:10.1016/j.bbrc.2003.09.176
180 E.J. Mullaney, A.H.J. Ullah / Biochemical and Biophysical Research Communications 312 (2003) 179–184

Table 1
Classes of phytases
Enzyme Examples Molecular mass of NCBI
class monomer (kDa) structure No.
HAP PhyA 85a 1IHP
PhyB 68a 1QFX
AppA 45b 1DKP
BPP TS-Phy 44b 1H6L
PhyC 43b —

PAP GmPhy 72a —

a
Glycosylated.
b
Nonglycosylated.

strate specificity but a high specific activity for phytate.

No significant variation in the catalytic centers of these
fungal phytases has thus far been discovered. However,
when the amino acid residues comprising the SSS are
examined [5,7], a pattern of low specific activity for
phytase and a neutral amino acid occupying the residue
analogous to K300 in Aspergillus niger NRRL 3135
phytase (phyA) is evident [8]. When a three-dimensional
model of the phyA molecule was examined, the six
amino acids in its substrate specificity site, K91, K94,
E228, D262, K300, and K301, were found to encircle the
cavity containing the HAP active center [7]. They thus
appear to have a role as a gate-keeper for any substrate’s
access to the HAP active site. A. niger NRRL 3135
phytase (phyA) is widely known for its high specific
activity for phytic acid [9] and is commercially marketed Fig. 1. Computer-generated molecular models from National Center
as Natuphos as an animal feed additive to lower phos- for Biotechnology Information (NCBI)’s Web site (www.ncbi.nlm.-
phate levels in manure from poultry and swine. Ko- nih.gov) of representatives from three classes of phosphatases: (A)
1IHP, PhyA, a histidine acid phosphatase; (B) 1H6L, Ts-Phy, a b
strewa and co-workers [5] explain this high specific
propeller phytase; (C) 1KBP, KSPAP, a purple acid phosphatase.
activity for phytate by revealing its SSS that is optimized
for binding the negatively charged phytic acid. This
model also offers a reason that other less charged sub- is more electrostatically neutral and therefore can
strates do not bind as effectively to the SSS of A. niger accommodate a broader substrate spectrum [5,12] than
NRRL 3135 phyA. A three-dimensional model of A. A. niger phyA. It also means that at pH 2.5 the SSS of
niger NRRL 3135 phyA (1IHP) from the National A. niger T213 phyB is uncharged and will accept nega-
Center for Biotechnology Information’s (NCBI) Web tively charged phytate as a substrate; however, at pH
site is shown in Fig. 1A. 5.0, it is negatively charged and would thus repel a
A second phytase (phyB) has been isolated from A. negatively charged substrate [5].
niger NRRL 3135 [10]. Its catalytic properties are dis- Another difference between A. niger phyA and phyB
tinct from those of A. niger NRRL 3135 phyA. In fact, it is that the active form of phyA is a monomer, whereas
first appeared in the literature not as a phytase but as an the active form of phyB is a tetramer [12]. This tetra-
A. niger acid phosphatase optimum pH 2.5 [11]. When it meric structure initially provides phyB with thermosta-
was first isolated and tested for phytase activity, the bility, but it also explains why, unlike phyA, it is
activity assay was conducted at pH 5.0, the optimum pH incapable of proper refolding after it has been denatured
for A. niger NRRL 3135 phyA. However, the pH opti- by heating. The individual protomers are apparently
mum for phyB is lower, pH 2.5, and no or only minimal unable to properly reassociate into an active form after
activity is observed at pH 5.0. This lack of activity at pH the phyB tetramer is denatured.
5.0 for phyB can now be understood by examining its Thus far, phyA (HAP) has been isolated from fila-
SSS. Kostrewa and co-workers [5] identified the SSS of mentous fungi, bacteria, yeast, and plants [7]. Because of
phyB, which is different from the SSS of phyA. It is the proven efficacy of phyA as an animal feed additive
composed of just two amino acids; in A. niger T213 several of these phyA (HAP) phytases (Natuphos,
phyB they are D75 and E272. This means the phyB SSS RoNozyme, etc.) are now being marketed. These com-
 E.J. Mullaney, A.H.J. Ullah / Biochemical and Biophysical Research Communications 312 (2003) 179–184 181

Fig. 1. (continued)

mercial phytases are produced by large-scale fermenta- hydrolyzing its substrate. BPPs have been isolated and
tion coupled with overexpression. PhyA (HAP) has also their genes cloned from Bacillus subtilis (phyC) [19] and
been overexpressed in several transgenic plants [13]. Such Bacillus amyloliquefaciens (TS-Phy) [20]. A three-di-
transgenic plants have been suggested as an potential mensional model of its molecule displays a basic form
alternative method of phytase production for the animal similar to a propeller with six blades [21]. A search of the
feed industry [14] or a means of developing plant culti- Protein Data Base revealed no other known phospha-
vars that require less phosphorus fertilizer [15]. Recently, tase with this type of structure. The dependence on
the HAP phytase gene from E. coli, appA, has also been binding Ca2þ for thermostability and catalytic activity
successfully expressed in a transgenic pig [16]. further distinguishes phyC from other subclasses of
The phyB (HAP) has only been reported from As- phytases [22]. BPP has two phosphate binding sites [23].
pergillus niger [17,12] and Aspergillus awamori [18]. No The hydrolysis of its substrate occurs at the “cleavage
commercial form of this enzyme is currently marketed. site” and the adjacent “affinity site” which increases the
binding affinity for substrates like phytic acid that fea-
ture neighboring phosphate groups. The calcium ions
b Propeller phytase facilitate the binding by creating a favorable electro-
static environment [22]. Fig. 1B is a three-dimensional
Unlike HAP, BPP (EC 3.1.3.8) is a recently discov- model of b propeller phytase (1H6L) from the NCBI
ered class of enzyme with a novel mechanism for Web site.
182 E.J. Mullaney, A.H.J. Ullah / Biochemical and Biophysical Research Communications 312 (2003) 179–184

Fig. 1. (continued)

While BPP has no known homologous phosphatase, other bacterial species or in eukaryotes remains to be
a recently developed method analyzing side-chain determined.
patterns of proteins by a “multidimensional index tree”
has identified a class of enzymes, pyrophosphatase
(PPase), that share some structural features with b Purple acid phosphatase GmPhy—soybean (Glycine max
propeller phytase [24]. An examination of these two L. Merr.) phytase
classes of enzymes lends support to this sophisticated
analysis. First, while phytases hydrolyze phytate, pyro- Another phytase, GmPhy (EC 3.1.3.2), has recently
phosphatases hydrolyze inorganic pyrophosphate. Sec- been isolated from the cotyledons of germinating soy-
ond, the proposed catalytic mechanisms of these two beans [25]. GmPhy has the active site motif of a purple
enzymes are very similar. As in BBP, a water nucleophile acid phosphatase (PAP). This class of metalloenzyme
that is coordinated to two metal ions in PPase attacks has been widely studied [26]. Both its three-dimensional
the phosphorus atom of its substrate. Moreover, both structure [27] and a proposed mechanism of catalysis are
enzymes contain a “cleavage site” and an “affinity site.” known [28]. Searches of genomic databases have re-
The nucleophilic attack by the water molecule occurs in vealed PAP-like sequences in plants, mammals, fungi,
the former and the binding of a second phosphate group and bacteria [29]. The estimated size of purified GmPhy,
in the latter [24]. Whether BPP phytases are found in 70–72 kDa, suggests a molecular mass similar to other
 E.J. Mullaney, A.H.J. Ullah / Biochemical and Biophysical Research Communications 312 (2003) 179–184 183

plant PAPs [25]. However, GmPhy is the only known [8] E.J. Mullaney, C.B. Daly, T. Kim, J.M. Porres, X.G. Lei, K.
PAP reported to have significant phytase activity. An A. Sethumadhavan, A.H.J. Ullah, Site-directed mutagenesis of
Aspergillus niger NRRL 3135 phytase at residue 300 to enhance
niger NRRL 3135 PAP (Apase6) has previously been
catalysis at pH 4.0, Biochem. Biophys. Res. Commun. 297 (2002)
reported and displays only a minimum ability to utilize 1016–1020.
phytate as a substrate [30]. The lower catalytic activity [9] R.J. Wodzinski, A.H.J. Ullah, Phytase, Adv. Appl. Microbiol. 42
of GmPhy than of A. niger NRRL 3135 phyA may be (1996) 263–302.
necessary during germination in soybeans because the [10] A.H.J. Ullah, B.J. Cummins, Purification, N-terminal amino acid
sequence and characterization of the pH 2.5 optimum acid
germination process requires a steady breakdown of
phosphatase (E.C. 3.1.3.2) from Aspergillus ficuum, Prep. Bio-
phytate over a period of several days [31]. While a three- chem. 17 (1987) 397–422.
dimensional model of GmPhy is not available, a model [11] A.H.J. Ullah, B.Q. Phillippy, Substrate selectivity in Aspergillus
of another PAP, kidney bean purple acid phosphatase ficuum phytase and acid phosphatases using myo-inositol phos-
(KBPAP), is available. Its NCBI structural number is phates, J. Agric. Food Chem. 42 (1994) 423–425.
[12] M. Wyss, L. Pasamontes, R. Remy, J. Kohler, E. Kusznir, M.
1KBP. Both GmPhy and kidney bean PAP share the
Gadient, F. Muller, A.P.G.M. van Loon, Comparison of the
purple acid phosphatase active site motif and a common thermostability properties of three acid phosphatases from molds:
catalytic mechanism. Amino acid sequence comparison Aspergillus fumigatus phytase, A. niger phytase, and A. niger pH 2.5
revealed 33% homology between the two enzymes. acid phosphatase, Appl. Environ. Microbiol. 64 (1998) 4446–4451.
Fig. 1C features (1KBP) to underscore the structural [13] P.R. Day, Genetic modification of plants: significant issues and
hurdles to success, Am. J. Clin. Nutr. 63 (1996) 651S–656S.
differences in these three classes of enzymes.
[14] K. Gutknecht, Green genes: alfalfa biofarming is about to take
root, Wisconsin Agriculturist, Mid-March (1997) 8–10.
[15] A.E. Richardson, P.A. Hadobas, J.E. Hayes, Extracellular secre-
Conclusion tion of Aspergillus phytase from Arabidopsis roots enables plants
to obtain phosphorus from phytate, Plant J. 25 (2001) 641–649.
[16] S.P. Golovan, R.G. Meidinger, A. Ajakaiye, M. Cottrill, M.Z.
The term “phytase” has become a broad rubric that Wiederkehr, D.J. Barney, C. Plante, J.W. Pollard, M.Z. Fan,
includes several structurally different enzymes. A char- M.A. Hayes, J. Laursen, J.P. Hjorth, R.R. Hacker, J.P. Phillips,
acteristic of one phytase will not automatically be appli- C.W. Forsberg, Pigs expressing salivary phytase produce low-
cable to all other enzymes collectively grouped under this phosphorus manure, Nat. Biotechnol. 19 (2001) 741–745.
[17] K.C. Ehrlich, B.G. Montalbano, E.J. Mullaney, H.C. Dischinger
name. Individual researchers have a responsibility to en-
Jr., A.H.J. Ullah, Identification and cloning of a phytase gene
sure that they clearly identify the specific enzyme and its (phyB) from Aspergillus niger (ficuum), Biochem. Biophys. Res.
class in all their communications. It is likely that addi- Commun. 195 (1993) 53–57.
tional classes of enzymes will be found to have phytate- [18] C.S. Piddington, C.S. Houston, M. Paloheimo, M. Cantrell, A.
degrading capabilities and also be labeled as a “phytase.” Meittinen-Oinonen, H. Nevalainen, J. Rambosek, The cloning
and sequencing of the genes encoding phytase (phy) and pH 2.5-
optimum acid phosphatase (aph) from Aspergillus niger var.
awamori, Gene 133 (1993) 55–62.
References [19] J. Kerovuo, M. Lauraeus, P. Nurminen, N. Kalkkinen, J.
Apajalahti, Isolation, characterization, molecular gene cloning,
[1] G.C. Irving, D.J. Cosgrove, Inositol phosphate phosphatases of and sequencing of a novel phytase from Bacillus subtilis, Appl.
microbiological origin: some properties of the partially purified Environ. Microbiol. 64 (1998) 2079–2085.
phosphatase of Aspergillus ficuum NRRL 3135, Aust. J. Biol. Sci. [20] Y.O. Kim, H.K. Kim, K.S. Bae, J.H. Yu, T. Oh, Purification and
27 (1974) 361–368. properties of a thermostable phytase from Bacillus sp. DS11,
[2] U. Konietzny, R. Greiner, Molecular and catalytic properties of Enzyme Microb. Technol. 22 (1998) 2–7.
phytate-degrading enzymes (phytases), Int. J. Food Sci. Technol. [21] N.C. Ha, B.C. Oh, S. Shin, H.J. Kim, T.K. Oh, Y.O. Kim, K.Y.
37 (2002) 791–812. Choi, B.H. Oh, Crystal structures of a novel, thermostable
[3] A.H.J. Ullah, B.J. Cummins, H.C. Dischinger Jr., Cyclohexan- phytase in partially and fully calcium-loaded states, Nat. Struct.
edione modification of arginine at the active site of Aspergillus Biol. 7 (2000) 147–153.
ficuum phytase, Biochem. Biophys. Res. Commun. 178 (1991) 45– [22] B.C. Oh, B.S. Chang, K.H. Park, N.C. Ha, H.K. Kim, B.H. Oh,
53. T.K. Oh, Calcium-dependent catalytic activity of a novel phytase
[4] R.L. Van Etten, R. Davidson, P.E. Stevis, H. MacArthur, D.L. from Bacillus amyloliquefaciens DS11, Biochemistry 40 (2001)
Moore, Covalent structure, disulfide bonding, and identification 9669–9676.
of reactive surface and active site residues of human prostatic acid [23] S. Shin, N.C. Ha, B.C. Oh, T.K. Oh, B.H. Oh, Enzyme
phosphatase, J. Biol. Chem. 266 (1991) 2313–2319. mechanism and catalytic property of b propeller phytase, Struc-
[5] D. Kostrewa, M. Wyss, A. D’Arcy, A.P.G.M. van Loon, Crystal ture 9 (2001) 851–858.
structure of Aspergillus niger pH 2.5 acid phosphatase at 2.4 A  [24] T. Hamelrych, Efficient identification of side-chain patterns using
resolution, J. Mol. Biol. 288 (1999) 965–974. a multidimensional index tree, Proteins 51 (2003) 96–108.
[6] M. Wyss, R. Brugger, A. Kronenberger, R. Remy, R. Fimbel, G. [25] C.E. Hegeman, E.A. Grabau, A novel phytase with sequence
Oesterhelt, M. Lehmann, A.P.G.M. van Loon, Biochemical similarity to purple acid phosphatases is expressed in cotyledons
characterization of fungal phytases (myo-inositol hexakisphos- of germinating soybean seedlings, Plant Physiol. (2001) 1598–
phate phosphohydrolases): catalytic properties, Appl. Environ. 1608.
Microbiol. 65 (1999) 367–373. [26] T. Klabunde, B. Stahl, H. Suerbaum, S. Hahner, M. Karas, F.
[7] E.J. Mullaney, C.B. Daly, A.H.J. Ullah, Advances in phytase Hillenkamp, B. Krebs, H. Witzel, The amino acid sequence of the
research, Adv. Appl. Microbiol. 47 (2000) 157–199. red kidney bean Fe(III)–Zn(II) purple acid phosphatase: determi-
184 E.J. Mullaney, A.H.J. Ullah / Biochemical and Biophysical Research Communications 312 (2003) 179–184

nation of the amino acid sequence by a combination of matrix- [29] G. Schenk, L.W. Guddat, Y. Ge, L.E. Carrington, D.A. Hume, S.
assisted laser desorption/ionization mass spectrometry and Hamilton, J. de Jersey, Identification of mammalian-like purple
automated Edman sequencing, Eur. J. Biochem. 226 (1994) 369– acid phosphatase in a wide range of plants, Gene 250 (2000) 117–
375. 125.
[27] N. Strater, T. Klabunde, P. Tucker, H. Witzel, B. Krebs, [30] A.H.J. Ullah, B.J. Cummins, Aspergillus ficuum extracellular pH
Crystal structure of a purple acid phosphatase containing a 6.0 optimum acid phosphatase: purification, N-terminal amino
dinuclear Fe(III)–Zn(II) active site, Science 268 (1995) 1489– acid sequence and biochemical characterization, Prep. Biochem.
1492. 18 (1988) 37–65.
[28] T. Klabunde, N. Strater, R. Frohlich, H. Witzel, B. Krebs, [31] D.M. Gibson, A.H.J. Ullah, Purification and characterization of
Mechanism of Fe (III)–Zn(II) purple acid phosphatase based on phytase from cotyledons of germinating soybean seeds, Arch.
crystal structures, J. Mol. Biol. 259 (1996) 737–748. Biochem. Biophys. 260 (1988) 503–513.