OUP user menu

Sugar non-specific endonucleases

E. Srinivasan Rangarajan, Vepatu Shankar
DOI: http://dx.doi.org/10.1111/j.1574-6976.2001.tb00593.x 583-613 First published online: 1 December 2001


Sugar non-specific endonucleases are multifunctional enzymes and are widespread in distribution. Apart from nutrition, they have also been implicated in cellular functions like replication, recombination and repair. Their ability to recognize different DNA structures has also been exploited for the determination of nucleic acid structure. Although more than 30 non-specific endonucleases have been isolated to date, very little information is available regarding their structure–function correlations except that of staphylococcal and Serratia nucleases. However, during the past few years, the primary structure, nature of the active site based on sequence homology, and the probable mechanism of action have been postulated for some of the enzymes. This review describes the purification, characteristics, biological role and applications of sugar non-specific endonucleases.

  • Endonuclease
  • Physicochemical property
  • Biological role
  • Application

1 Introduction

Nucleic acids act as carriers of genetic information. For the effective transfer of genetic information from one generation to another, they have to undergo processes such as replication, recombination and repair. All living organisms contain a set of enzymes, namely nucleases, which hydrolyze phosphodiester linkages in nucleic acids. The involvement of enzymes in the break down of nucleic acids was first observed by Araki [1] and the term ‘nucleases’ was coined by Iwanoff [2]. Kunitz [3], in 1940, described two classes of nucleases based on their sugar specificity. Subsequently, different schemes of classification were proposed in an attempt to overcome the shortcomings of the earlier ones [46]. However, with the discovery of newer nucleases and multifunctional enzymes, such as micrococcal nuclease [7] and snake venom phosphodiesterase [8], the classification of Kunitz [3, 4] was found to be inadequate. In order to overcome these inadequacies Bernard [9] and Laskowski [5, 10] came up with consensus criteria for the classification of nucleases on the basis of:

  1. the nature of substrate hydrolyzed (DNA, RNA);

  2. the type of nucleolytic attack (exonuclease or endonuclease);

  3. the nature of the hydrolytic products formed (i.e. mono- or oligonucleotides terminating in a 3′- or a 5′-phosphate); and

  4. the nature of the bond hydrolyzed.

Though the above classification could accommodate enzymes with exceptional properties into a particular group, discrepancies still existed as the complex nature of the catalytic activities of different nucleases was unraveled. Properties related to strand specificity, site specificity and sequence specificity did not find a place in the above classification scheme. Additionally, nucleases from Neurospora crassa [11] and BAL 31 nuclease from Alteromonas espejiana [12] exhibited mixed activities, depending on the substrate, making it difficult for them to be classified under a particular scheme. Taking all these aspects into consideration, Linn [13] formulated the most acceptable system of classification of nucleases (Fig. 1).

Figure 1

Schematic representation of the classification of nucleases. Based on the tabulation of nucleases by Linn [13].

Nucleases are ubiquitous in distribution and among them sugar non-specific endo- and exonucleases have been implicated in cellular functions like replication, recombination and repair. Their ability to recognize different DNA structures has made them important analytical tools for the determination of nucleic acid structures [1416], mapping mutations [17] and studying the interactions of DNA with various intercalating agents [18, 19]. These enzymes are characterized by their ability to hydrolyze both DNA and RNA. Fraser and Low [20] further subdivided the major sugar non-specific nucleases into three distinct but distantly related groups viz.:

  1. secreted fungal single-strand-specific endonucleases;

  2. protease-sensitive multifunctional endo–exonucleases; and

  3. mitochondrial nucleases that are closely related to bacterial Serratia marcescens nuclease.

Although more than 30 sugar non-specific endonucleases have now been isolated (Table 1), only staphylococcal nuclease [5557] and S. marcescens nuclease [58] have been extensively characterized. However, the past few years have witnessed significant progress in the molecular enzymology of some of these enzymes. The present review gives a comprehensive account of sugar non-specific endonucleases with respect to their occurrence, purification, physicochemical properties, biological role and applications.

View this table:
Table 1

Sugar non-specific endonucleases

Staphylococcal nucleaseS. aureus[7]
Azotobacter nucleaseA. agilis[21]
Sm nucleaseS. marcescens[22, 23]
Neurospora nucleaseN. crassa mitochondria[11, 24]
Bacillus nucleasesB. subtilis[25]
BAL 31 nucleaseA. espejiana[12]
Marine bacterium nucleaseVibrio sp.[26]
Lysobacter nucleaseL. enzymogens[27]
Yeast nucleaseS. cerevisiae mitochondria[2831]
Aspergillus nucleaseA. nidulans[32]
Anabaena nucleaseAnabaena sp. PCC 7120[33]
Sr nucleaseS. racemosum[34]
Schizosaccharomyces nucleaseS. pombe mitochondria[35]
Streptococcus MFS. pyogenes[36]
Leishmania nucleaseLeishmania sp.[37]
C1 nucleaseC. echinulata var. echinulata[38]
Streptomyces nucleasesS. antibioticus[39]
Nuclease RsnR. stolonifer[40]
Potato nucleasePotato tubers[41]
Tea leaves nucleasesTea leaves[42]
Pollen nucleaseNicotiana tabacum pollen[43]
Barley nucleaseBarley aleurone layers[44]
Pollen nucleaseP. hybrida pollen[45]
Wheat nucleaseWheat chloroplasts[46]
Rye germ nuclease-IRye germ ribosomes[47]
Rye germ nuclease-IIRye germ ribosomes[48]
Barley microspores nucleaseHordeum vulgare L. uninucleate microspores[49]
Rat liver nucleaseRat liver nuclei[50]
Endonuclease GB. taurus mitochondria[51]
Shrimp nucleaseShrimp hepatopancreas[52]
Drosophila nucleaseD. melanogaster[53]
Drosophila nucleaseD. melanogaster embryos[54]

2 Occurrence and localization

It is well known that nucleases play an important role in four Rs, i.e. replication, recombination, restriction and repair. Hence, every living organism must produce at least one type of nuclease. Sugar non-specific endonucleases have been isolated from a wide variety of sources (Table 1). The majority of these enzymes are intracellular but enzymes from Staphylococcus aureus [7], A. espejiana [12], S. marcescens [22], Bacillus subtilis [25], Vibrio sp. [26], Lysobacter enzymogens [27], Saccharomyces cerevisiae [30], Anabaena sp. PCC 7120 [33], Syncephalastrum racemosum [34], Cunninghamella echinulata [38], Streptomyces antibioticus [39] and Rhizopus stolonifer [40] are extracellular in nature. Endonucleases of mitochondrial origin have been reported from N. crassa [11, 24], S. cerevisiae [28, 29, 31] and Aspergillus nidulans [32]. In the cases of N. crassa [24] and A. nidulans [32] their presence has been shown in vacuoles, conidia, mycelia and nuclei. Additionally, in N. crassa, the mitochondrial nuclease has been shown to occur as membrane-bound as well as soluble forms in approximately equal proportions [59].

In plants, sugar non-specific endonucleases have been isolated from tea leaves [42], pollen grains of tobacco [43] and Petunia hybrida [45], wheat chloroplasts and its organelles [46], rye germ ribosomes [47, 48] and barley microspores [49]. The presence of these enzymes has also been shown in the lumen of the endoplasmic reticulum, in the Golgi apparatus, in protein bodies and vacuoles of barley aleurone layers [60].

In the kinetoplastid parasite, Leishmania sp., the localization of the endonuclease has been tentatively assigned to nuclei and kinetoplasts [37]. In animals, they have been isolated from various organelles, for example, the hepatopancreas of shrimp [52], rat liver nuclei [50], Bos taurus mitochondria [61] and embryos of Drosophila melanogaster [54].

3 Assay

3.1 Viscometry

This method is based on the measurement of decrease in viscosity of the nucleic acid samples following the action of nucleases [62].

3.2 Spectrophotometric methods

The increase in the amount of the acid soluble nucleotides (deoxyribo- or ribonucleotides) produced following the hydrolysis of DNA/RNA is measured at 260 nm [4, 63]. A unit of the enzyme is defined on the basis of μmol of acid soluble nucleotides liberated [64] or μg of RNA or DNA digested [65].

Viscometric and hyperchromicity measurements can also provide an insight into the mode of action of the enzyme. A sudden drop in the viscosity of the DNA solution without a significant increase in the hyperchromicity suggests an endonucleolytic cleavage, whereas a gradual decrease indicates an exo mode of action. However, an endo–exonuclease can produce significant changes in the hyperchromicity as well as a drop in the viscosity.

3.3 Radioactive measurements

The sensitivity of the assay can be enhanced by the use of radiolabelled substrates, where the increase in the acid soluble radioactivity or decrease in the acid insoluble radioactivity following the action of nuclease is measured. A unit of the enzyme is defined as the amount of enzyme required to render 1 nmol of labelled substrate acid soluble, under the assay conditions [66].

Another rapid and sensitive assay that measures endonucleolytic activity on DNA utilizes the fact that nitrocellulose membrane can retain only large fragments of denatured DNA. In this procedure, following enzyme action, radiolabelled denatured DNA is passed through nitrocellulose filters and the decrease in the retention of radioactivity on the nitrocellulose membrane is measured [67].

3.4 Gel electrophoresis

Agarose gel electrophoresis is a convenient and rapid technique for studying the extent and nature of single- or double-strand breaks, the frequency of damage, and the pattern of distribution of breaks in the substrate [68]. Although developed initially as a qualitative technique, it can be augmented and used as a quantitative method by end-labelling the substrate [69] or by densitometric scanning following electrophoresis in agarose gels [70]. A unit of the enzyme is defined as the amount of enzyme required to produce 1 fmol of nicks in the plasmid DNA under the assay conditions. Separation of the cleavage products on polyacrylamide gel electrophoresis (PAGE) followed by autoradiography can provide information regarding the cleavage site [71].

3.5 Phosphomonoesterase activity

Phosphomonoesterase activity, associated with some of the enzymes, is assayed by measuring the inorganic phosphate liberated following the hydrolysis of either 3′-AMP or 5′-AMP. A unit of the enzyme is defined on the basis of μmol of inorganic phosphate liberated [72].

4 Detection

4.1 Agar plate method

For qualitative detection of nucleases, the nucleic acid (DNA/RNA) is incorporated into the growth medium, in agar plates, along with a dye such as methyl green. The culture to be tested for nuclease production is spotted onto the plate and/or the sample is added in wells on the plates. A clear zone around the growth or the well, after precipitation of the unhydrolyzed methyl green nucleic acid complex with HCl, indicates the presence of nuclease activity [73, 74].

4.2 Zymogram analysis

The feasibility of detecting nuclease activity in gels containing nucleic acids was first demonstrated by Boyd and Mitchell [75]. Following electrophoresis/isoelectric focussing, the gels are incubated in appropriate buffers for enzyme action and then stained with suitable dyes, such as toluidine blue, acridine orange, methyl green or pyronin Y. A clear band against a colored background indicates the presence of nuclease. Alternatively, nucleases that renature after sodium dodecyl sulfate (SDS) treatment can be separated on SDS–PAGE containing the nucleic acid. After electrophoresis, the digested regions in the gel are detected as clear bands against the fluorescent background of ethidium bromide (EtBr) bound to nucleic acids [76].

5 Purification

Since the majority of sugar non-specific endonucleases are intracellular, most of the purification procedures, irrespective of the source, involve steps like lysis of cells, isolation of organelles (by differential centrifugation), concentration of the crude extract by salt precipitation followed by conventional purification methods, such as ion-exchange chromatography and gel filtration. During the initial purification steps, one of the primary aims is to get rid of the colored impurities contributed by pigments of the organelles and this is achieved either by extraction with acetone–water mixture (4:1 v/v) [42] or by a simple wash with ammonium chloride followed by high speed centrifugation [47, 48]. In most cases apart from sonication [52], grinding with glass beads or sand [11, 21, 41, 42] has also been used for disrupting the cells. In general, ammonium sulfate precipitation is also used as a preliminary purification step. In the case of membrane-bound enzymes, solubilization is achieved using non-ionic detergents like Triton X-100 or Nonidet P-40 and/or a high salt wash [31, 35, 47, 77]. Moreover, in certain cases, inhibitors like phenylmethylsulfonyl fluoride (PMSF), pepstatin A and leupeptin are used to overcome proteolytic activity [30, 31, 77].

Although ion-exchangers like DEAE- and CM-cellulose have been widely used for the purification of these enzymes, phosphocellulose has also been used in certain cases. For example, potato tuber nuclease, despite its net negative charge at pH 7.5, binds to phosphocellulose due to its affinity towards phosphate groups in phosphocellulose [41]. In this manner, this support not only acts as a cation-exchanger but also as an affinity matrix. Gel filtration has been used as one of the steps for the purification of nucleases from Vibrio sp. [26], L. enzymogens [27], S. racemosum [34], S. antibioticus [39], tea leaves [42] and rye germ ribosomes [48].

Hydroxylapatite has been widely used for the purification of several enzymes including nucleases [78]. This adsorbent has been used successfully for the purification of endonucleases from S. cerevisiae [28, 30], shrimp hepatopancreas [52], S. racemosum [34], Leishmania sp. [37] and B. taurus [77]. Endonucleases from S. racemosum [34], C. echinulata [38], rye germ ribosomes [48] and shrimp hepatopancreas [52] have been purified on hydrophobic matrices like phenyl- and octyl-Sepharose.

Affinity chromatography has been used extensively for the purification of sugar non-specific endonucleases. Endonucleases from barley aleurone layers [44], S. cerevisiae mitochondria [31], C. echinulata [38] and R. stolonifer [40] were purified on heparin-agarose and/or reactive blue 2-agarose/Cibacron blue-agarose. Other affinity adsorbents include deoxythymidine 3′,5′-diphosphate (pdTp)-aminophenyl-Sepharose [79], NADP-agarose [80] and 5′-AMP-agarose [81]. The affinity of sugar non-specific endonucleases for single-stranded nucleic acids has been utilized for the purification of endonucleases of S. cerevisiae [30], A. nidulans [32], D. melanogaster [53] and Leishmania sp. [37] on single-stranded DNA (ssDNA) bound to cellulose. In this case, the chromatographic operations are generally carried out under conditions where the enzyme is either not active or shows very little activity.

Modern purification techniques like FPLC have been utilized successfully for the purification of sugar non-specific endonucleases from S. antibioticus [39], D. melanogaster embryos [54] and B. taurus [77].

6 Physical properties

6.1 Molecular mass and subunit structure

The molecular masses (Mr) of sugar non-specific endonucleases range from 14 to 140 kDa, but the majority of them are between 28 and 44 kDa (Table 2). The enzymes from the marine bacterium Vibrio sp. [26], yeast [29] and A. espejiana F (fast) form [82] are high molecular mass proteins with Mr values of 100, 140 and 109 kDa, respectively. Nucleases from S. aureus [83], yeast mitochondria [28], rye germ ribosomes [47] and S. antibioticus [39] are comparatively low Mr proteins in the range of 14–20 kDa.

View this table:
Table 2

Properties of sugar non-specific endonucleases

EnzymeMolecular mass (kDa)Optimum pHOptimum temperature (°C)Metal ion requirementReference
Azotobacter nuclease7.7Mg2+, Mn2+, Co2+[21]
Sm nuclease528.0–8.535–44Mg2+, Mn2+, Co2+[23, 58]
Neurospora nuclease6.0–7.537Mg2+, Mn2+, Co2+[11]
Neurospora nuclease668.0Mg2+, Mn2+[24]
Bacillus nucleases9.0Ca2+[25]
Streptomyces nucleases188.0–8.5Mg2+[39]
C1 nuclease307.0–8.550–55Mg2+, Mn2+[38]
Aspergillus nuclease286.8–8.0Mg2+, Mn2+, Zn2+[32]
Yeast nuclease767.0–7.535Mg2+, Mn2+, Co2+, Ca2+, Zn2+[31]
Yeast nuclease1406.5–7.0Mg2+, Mn2+[29]
Yeast nuclease147.030–40Mg2+, Mn2+[28]
Leishmania nuclease527.550Mg2+, Mn2+[37]
Staphylococcal nuclease16.89.0–10.0Ca2+[7]
BAL 31 nuclease[82]
F (fast) form1098.0–8.8Mg2+, Ca2+
S (slow) form858.0–8.8Mg2+, Ca2+
Anabaena nuclease297.535Mg2+, Mn2+, Co2+[33]
Yeast nuclease (influenced by RAD52)727.5Mg2+, Mn2+[30]
Lysobacter nuclease22–288.0Mg2+, Mn2+[27]
Vibrio nuclease1008.030–40None[26]
Schizosaccharomyces nuclease327.0–8.0Mg2+, Mn2+[35]
Nuclease Rsn677.040–45Mg2+, Mn2+, Co2+[40]
Sr nuclease567.0–7.2Mg2+, Mn2+[34]
Rye germ nuclease-I205.0, 6.5None[47]
Rye germ nuclease-II57/625.0, 7.8None[48]
Petunia pollen nuclease345.0None[45]
Barley nuclease366.055None[44]
Wheat nuclease296.8–7.850None[46]
Potato nuclease336.5–8.0None[41]
Tobacco pollen nuclease605.2None[43]
Barley microspore nuclease30Mg2+[49]
Tea leaves[42]
nuclease A355.5–6.570None
nuclease B335.5–6.560–70None
Drosophila nuclease327.0–8.5Mg2+, Mn2+[53]
Drosophila embryo nuclease446.5–7.4Mg2+, Mn2+[54]
Shrimp nuclease457.5Mg2+, Mn2+, Ca2+[52]
Rat liver nuclease506–7Mg2+, Mn2+, Co2+[50]
Endonuclease G507.5–8.0Mg2+, Mn2+[77]

Most of the non-specific nucleases consist of a single polypeptide chain, but the enzymes from N. crassa mitochondria [24], S. racemosum [34], B. taurus [77] and S. marcescens [8486] are dimers made up of two identical subunits of 33, 28, 26 and 26 kDa, respectively. Moreover, the dimeric forms of non-specific endonucleases from S. racemosum [34] and S. marcescens [87] are held together purely by non-covalent interactions and not by disulfide linkages. The endonuclease from R. stolonifer is a tetramer and each protomer is made up of two non-identical subunits of Mr 21 and 13 kDa. Since the amino acid composition revealed the absence of cysteine [88], it can be assumed that the subunits are held together by non-covalent interactions. A genetically engineered monomeric variant of S. marcescens nuclease, H184R, was shown to exist as a monomer even at high protein concentrations. This monomeric variant exhibited similar secondary structure, stability towards chemical denaturants, and activity to the wild-type enzyme [86]. Although the monomeric variants were functionally independent, in the presence of low enzyme concentration and high molecular mass DNA the native dimeric form of S. marcescens nuclease was relatively more active than the monomeric form or the heterodimer with one inactive subunit [89].

In S. cerevisiae and N. crassa multiple nucleases having similar catalytic properties, but with subtle differences in their physical properties, have been reported. Three endonucleases with Mr of 14, 38 and 57 kDa have been isolated from yeast mitochondria [28, 29, 31]. Among them, the 38- and 57-kDa enzymes exist as dimers under natural conditions. The majority of the nuclease activity in yeast mitochondria, attributed to the 38-kDa enzyme, is located in the mitochondrial inner membrane. Additionally, the presence of multiple forms of the enzyme was correlated to proteolytic cleavage [31]. S. antibioticus produces two extracellular nucleases of Mr 18 and 34 kDa, exhibiting similar properties. The 18-kDa protein was reported to be formed from a 74-kDa precursor, as evidenced by the cross-reactivity against the anti-18-kDa antibodies. It was also shown that the 34-kDa protein is not formed from the 74-kDa protein and the former is not a precursor of the 18-kDa enzyme [39]. In A. nidulans, the 28-kDa endo–exonuclease is formed due to the proteolytic digestion of the 38-kDa polypeptide via the formation of a 33-kDa intermediate. Moreover, crude extracts, obtained from cells grown in the presence of PMSF, showed the presence of an immunoreactive 90-kDa polypeptide which is a putative precursor of the 28-kDa endo–exonuclease [32]. In N. crassa 90% of the intracellular neutral ssDNase activity is associated with endo–exonucleases localized in mitochondria, vacuoles and nuclei. In addition, an unstable and trypsin-activatable form of endo–exonuclease found in cytosol, nuclei and mitochondria gave rise to two nucleases similar to endo–exonucleases reported by Linn and Lehman [90, 91] and Fraser et al. [92]. From mitochondria, a strand non-specific endonuclease [11] and a ssDNA binding endo–exonuclease not inhibited by ATP [24] have also been isolated. Fraser et al. [93] demonstrated by immunochemical techniques that all the above enzymes from N. crassa are derived from one or more related large inactive (precursor) polypeptides which undergo proteolysis to produce various active forms of nucleases. This inactive precursor was shown to have a Mr of 93–95 kDa [94]. Additionally, Fraser et al. [93] showed that endo–exonucleases from A. nidulans and S. cerevisiae are immunologically related to N. crassa endo–exonuclease but not to single-strand-specific nucleases like S1, P1 and mung bean nuclease. Similarly, isoforms of the nucleases were observed in S. aureus [95], B. subtilis [25], A. espejiana [12], S. marcescens kums 3958 [23], tobacco pollen [43], P. hybrida pollen [45] and S. marcescens [9698]. In case of staphylococcal nuclease, Taniuchi and Anfinsen [99] showed that limited proteolysis of the enzyme, in the presence of Ca2+ and pdTp, produces an active non-covalently bonded derivative, termed nuclease-T, which shows a similar level of activity against DNA and RNA as that of the native enzyme.

In contrast to protease-mediated formation of multiple forms, shrimp nuclease exhibited multiple forms in the presence of SDS and 2-mercaptoethanol. The enzyme is a monomer of Mr 45 kDa cross-linked with a large number of disulfide bridges. It exhibits four different forms on SDS–PAGE, depending on the treatment of the sample (Fig. 2). The native Form I of the protein (which is acidic in nature) is represented by a structure in which the acidic groups (carboxylate residues) are exposed and the disulfides are buried. When the protein was subjected to heat treatment with 2-mercaptoethanol, in the presence or absence of SDS, it changed into an inactive Form III (45 kDa, reaction a). Form III could not refold and re-oxidize to form active Form I even after the removal of SDS and 2-mercaptoethanol (reaction a′). However, when Form I was subjected to heat treatment with SDS in the absence of 2-mercaptoethanol, the resulting Form II (39 kDa, reaction b) became catalytically active only after the removal of SDS (reaction b′). The native protein on heat treatment in the absence of SDS became irreversibly denatured (Form II′, reaction c) which could not refold back to the active Form I (reaction c′). However, the Form II′ could not be converted to Form II, even after treatment with SDS (reaction d). Based on these observations the authors opined that the irreversible denaturation on heating in the absence of SDS is due to the formation of the incorrect inside-out structure, with an exposed hydrophobic domain which has a tendency to aggregate and cannot revert to a catalytically active form [100].

Figure 2

A scheme for folding and unfolding pathways of shrimp nuclease in the presence and absence of SDS and 2-mercaptoethanol (reprinted from Lin et al. [100], Copyright 1994, with permission from Elsevier Science).

6.2 Isoelectric point and glycoprotein nature

The isoelectric points of the majority of sugar non-specific endonucleases have not been reported, but in the case of some of the well characterized enzymes they are in the range 4.0–9.6. Nucleases from shrimp hepatopancreas [52], A. espejiana [82], R. stolonifer [40], rye germ ribosomes [47], D. melanogaster embryos [54] and S. racemosum [34] are acidic proteins with pI values of 4.06, 4.2, 4.2, 4.8, 4.9 and 5.0, respectively. S. marcescens nuclease and its isoforms are neutral proteins with pI values of 6.8, 7.3, 7.4 and 7.5 [23, 97]. However, the endonuclease SM1 from S. marcescens kums 3958 [23] and S. aureus nuclease [101] are basic proteins with pI values of 8.1 and 9.6, respectively.

Nucleases from barley aleurone layers [102], S. racemosum [52] and rye germ ribosomes [47] are glycoproteins. Rye germ ribosomal nuclease I contains 28% carbohydrate consisting of fucose, mannose and glucosamine [47].

7 Catalytic properties

7.1 Optimum pH

In general, the pH optima of sugar non-specific endonucleases are in the range 5.0–10.0 (Table 2) and most of the enzymes show the same optimum pH for the hydrolysis of both polymeric and monomeric substrates. However, nucleases from A. espejiana [81], N. crassa (mitochondria) [11] and tea leaves [42] show different pH optima for the hydrolysis of ssDNA (8.8, 6.5–7.5 and 6.0) and double-stranded DNA (dsDNA) (8.0, 5.5–6.5 and 5.5, respectively). On the other hand, 3′-nucleotidase-nuclease from potato tubers showed different pH optima for the nucleotidase (pH 8.0) and nuclease (pH 6.5–7.5) activities [41]. Similarly, nuclease I from rye germ ribosomes exhibited pH optima of 6.0 and 6.5 for the nuclease and 3′-nucleotidase activities, respectively. Nucleases from wheat chloroplasts [46] and rye germ ribosomes [48] showed optimum pH values of 7.8 and 5.0 for the hydrolysis of denatured DNA and 6.8 and 7.8 for the hydrolysis of RNA, respectively. However, rye germ ribosomal nuclease II degraded poly(I).poly(C) optimally at pH 8.5 [48]. Interestingly, nucleases (A and B) from tea leaves exhibited different pH optima for the hydrolysis of different 3′-mononucleotides (i.e. pH 6.0 for 3′-GMP and 3′-CMP and pH 6.5 for 3′-AMP and 3′-UMP), a property associated with the 3′-nucleotidase activity of single-strand-specific nucleases like Le1 and Le3 from Lentinus edodes [103, 104] and P1 from Penicillium citrinum [64].

In certain cases, the optimum pH of sugar non-specific endonucleases is influenced by the type of metal ions involved. For example, S. racemosum nuclease exhibited optimum pH values of 7.0 and 7.2 in the presence of 5 mM Mg2+ or Mn2+, respectively [34]. However, the DNase activity of staphylococcal nuclease showed an optimum pH of 10.0 or 9.5 in the presence of 1 or 10 mM Ca2+, respectively, but under similar conditions the RNase activity was optimal at pH 9.5 or 9.0 [105].

7.2 Optimum temperature and temperature stability

The temperature optima of the majority of sugar non-specific endonucleases have not been reported but in the cases of some of the well characterized enzymes, they are in the range of 30–70°C (Table 2). Most of them exhibit the same optimum temperature for all the substrates. However, C. echinulata nuclease showed an optimum temperature of 50°C and 55°C for DNA hydrolysis in the presence of Mg2+ and Mn2+, respectively [38]. Nuclease Rsn from R. stolonifer showed an optimum temperature of 40°C for ssDNA hydrolysis in the presence of Mg2+, Mn2+ and Co2+ and for dsDNA hydrolysis in the presence of Mg2+, but it showed a higher optimum temperature (45°C) for dsDNA hydrolysis in the presence of Mn2+ and Co2+[40]. On the contrary, the RNase activity of nuclease Rsn exhibited an optimum temperature of 35°C [106].

The majority of sugar non-specific endonucleases are thermolabile enzymes, but staphylococcal nuclease showed high stability and retained a significant amount of its activity in the presence of 1 N HCl for 6 h or at 100°C for 20 min. Similarly, tea leaf nucleases were stable at 60°C for 15 min [42], whereas nucleases SM1 and SM2 from S. marcescens kums 3958 retained their activity for 24 h at 25°C [23]. Interestingly, a mitogenic factor of Streptococcus pyogenes exhibiting nuclease activity was resistant to inactivation at 100°C for 10 min and showed a biphasic temperature stability. The enzyme was stable up to 50°C (for 10 min) but lost more than 90% of its activity at 60°C. However, it regained approximately 50% of its original activity at 80°C [36].

7.3 Metal ion requirement

Sugar non-specific endonucleases, with the exception of nucleases from Vibrio sp. [26], potato tubers [41], tea leaves [42], barley aleurone layers [44], P. hybrida [45], wheat chloroplasts [46] and rye germ ribosomes [47, 48], are metal requiring enzymes. Although wheat chloroplast nuclease [46] did not require metal ions for its activity, the ssDNase activity showed slight stimulation (20%) in the presence of Mg2+. Similarly, the activity of endonuclease from Vibrio sp. was stimulated by Mg2+ and Ca2+[26]. In contrast, nuclease I of rye germ ribosomes [47], pollen nucleases of tobacco [43] and P. hybrida [45] were stimulated by Zn2+, while nuclease II from rye germ ribosomes was stimulated by Mn2+[48]. Extracellular nucleases (Bs-IA, Bs-IB and Bs-II) from B. subtilis required Ca2+ for the hydrolysis of native DNA. However, low rates of hydrolysis of denatured DNA and rRNA were observed in the absence of added metal ion [25]. The action of N. crassa nuclease on dsDNA is dependent on Mg2+ concentration but its activity on ssDNA is independent of Mg2+ concentration, though it is stimulated to some extent [107]. Addition of 10 mM Mg2+, Ca2+ or Fe2+ resulted in a 2.5-fold stimulation of the ssDNase activity of N. crassa enzyme, but it also brought about approximately 40% inhibition of the RNase activity. The selective inhibition of the RNase activity in the presence of metal ions was attributed to the induction of secondary structures in RNA by these metal ions. On the contrary, nuclease Rsn from R. stolonifer required Mg2+, Mn2+ and Co2+ for its DNase and RNase activities. Although Mg2+ and Mn2+ did not affect the ssDNase to dsDNase ratio, the enzyme showed a higher preference for ssDNA in the presence of Co2+. Moreover, there was no synergism in the presence of either of the metal ions but a higher ssDNase:dsDNase ratio was observed when Co2+ was used in combination with Mg2+ or Mn2+[40].

Many of the metal requiring endonucleases show synergism when metal ions are used in combination. For example, nucleases from S. aureus [105], A. espejiana [12], A. nidulans [32], shrimp hepatopancreas [52], mitogenic factor of S. pyogenes [36] and S. antibioticus (18-kDa nuclease) [39] exhibited maximum activity in the presence of Ca2+ and Mg2+.

7.4 Effect of salt concentration

It has been reported that salt concentration in the reaction mixture can influence the activity of sugar non-specific endonucleases. For example, the activity of BAL 31 nuclease is maximum in the range of 0–2 M NaCl and the enzyme shows only 40% of its ssDNase activity in the presence of 4.4 M NaCl [12]. In the case of endonucleases from N. crassa [107] and yeast mitochondria [30], 200 mM NaCl completely inhibited the dsDNase activity but it had only a marginal effect on the ssDNase activity. Similarly, D. melanogaster endonuclease showed 50% inhibition of its dsDNase activity in the presence of 30 mM NaCl but it required 100 mM NaCl to bring about the same level of inhibition of the ssDNase activity [53]. The inhibition of dsDNase activity, in the presence of high salt concentrations, was attributed to the stabilization of the AT-rich regions in dsDNA [108, 109]. The ssDNase and RNase activities of yeast mitochondrial nuclease are maximal between 1 and 300 mM KCl while KCl (>150 mM) inhibited the dsDNase activity. However, NaCl up to 100 mM did not have any effect on the enzyme activities [31]. Similarly, staphylococcal nuclease [55] and endonuclease G from B. taurus [51] were inhibited at NaCl concentrations above 100 mM. On the contrary, Leishmania endonuclease showed optimal activity in the presence of 100–150 mM NaCl/KCl but subsequent increase in the salt concentration progressively inhibited the enzyme activity [37]. Nucleases from S. antibioticus [39] and yeast mitochondria [28], however, showed maximum activity in the presence of 20–30 mM NaCl and 50 mM KCl, respectively. The dsDNase activity of tobacco pollen nuclease was inhibited in the presence of 200 mM NaCl [43] while Serratia nuclease retained 25% of its activity in the presence of 1 M NaCl [110]. For the mitogenic factor of S. pyogenes concentrations of NaCl/KCl greater than 60 mM brought about 90% inhibition of the nuclease activity [36]. Inhibition in the presence of high concentrations of NaCl/KCl was also observed with endonucleases from rat liver nuclei [50], Anabaena sp. [33], Schizosaccharomyces pombe [35] and R. stolonifer [40, 106].

7.5 Effect of denaturants

In the presence of 7 M urea, S. marcescens endonuclease showed approximately 2-, 1.7- and 1.5-fold increases in its dsDNase, ssDNase and RNase activities, respectively [111]. However, Azotobacter agilis endonuclease showed a 5–10-fold stimulation of its activity towards poly(A) in the presence of 2 M urea [21]. Gray et al. [12] showed that the S (slow) form of BAL 31 nuclease is active in the presence of 5% (w/v) SDS and can be incubated with the detergent without loss of activity if Ca2+ and Mg2+ are present at concentrations of 12.5 mM before the addition of the detergent. Purified S form BAL 31 nuclease retained 60% of its optimal activity in the presence of 4 M urea. In the case of endonuclease from barley aleurone layers, complete inactivation of the DNase and RNase activities was observed in the presence of 1% (w/v) SDS [102]. Similarly, the DNase activity of nuclease Rsn showed high sensitivity towards denaturants and lost more than 80% of its activity in the presence of low concentrations of SDS (0.02% w/v), guanidine hydrochloride (0.2 M) or urea (3 M) [40].

7.6 Activators and inhibitors

Polyamines such as spermine and spermidine, which bind to double-stranded nucleic acids, also inhibit the ssDNase activity of nucleases. Spermine stimulated the exonuclease activity of BAL 31 nuclease but the cleavage specificity was considerably reduced in its presence [112]. The DNase and RNase activities of yeast mitochondrial endonuclease [31] and the RNase activity of rye germ ribosomal nuclease I [47] were stimulated in the presence of 2 and 2.5 mM spermidine, respectively. Moreover, a low concentration (0.1 mM) of polyamines such as putrescine and spermidine inhibited the RNase activity of rye germ ribosomal nuclease I [47]. On the contrary, the DNase and RNase activities of S. aureus nuclease were stimulated by low amounts of putrescine, spermine and spermidine, whereas at high concentrations they inhibited both the activities [113]. However, endonuclease from barley aleurone layers was not affected by putrescine, spermidine and spermine at 1 mM [102]. The ability of polyamines to either activate or inhibit the nuclease activity in a concentration-dependent manner was correlated to the regulation of nucleic acid levels within the cells by controlling the nuclease activity [114].

Low concentrations of EtBr (1–10 μg ml−1) brought about marginal stimulation of the dsDNase activity of yeast mitochondrial endonuclease but it had no effect on the RNase activity [31]. Although 10 μM EtBr did not affect the ssDNase activity of yeast mitochondrial endonuclease, slight inhibition (30%) of the dsDNase activity was observed with 2 μM EtBr [28].

Endonucleases from D. melanogaster [53] and S. antibioticus [39] showed optimal activity in the presence of either dithiothreitol (DTT) or 2-mercaptoethanol. Although N. crassa nuclease [11] and rye germ ribosomal nuclease I [47] did not show an obligate requirement for thiol reagents for their activity, considerable stimulation of the activity was observed in the presence of DTT and 2-mercaptoethanol. In contrast, DTT and 2-mercaptoethanol brought about significant inhibition of both the DNase and RNase activities of wheat chloroplast nuclease [46]. Ten mM DTT inhibited barley aleurone layer nuclease whereas 2-mercaptoethanol, at the same concentration, had no effect on the enzyme activity [102]. However, sulfhydryl reagents had no effect on endonucleases from S. marcescens kums 3958 [23], yeast mitochondria [31] and shrimp hepatopancreas [52]. Interestingly, 2-mercaptoethanol had no effect on the activity of S. marcescens endonuclease [110], though disulfide bonds are essential for the enzyme activity [87]. Subsequently, Filiminova et al. [111, 115] demonstrated that substantial inactivation of the enzyme activity occurred when a high concentration of 2-mercaptoethanol (640 mM) was used in the presence of 2 M urea. The inability of 2-mercaptoethanol to inactivate the native enzyme was correlated to the masking of the sulfhydryl groups in the native conformation.

The majority of the sugar non-specific endonucleases are metal requiring enzymes and hence they are strongly inhibited by metal chelators like EDTA and EGTA. In most cases, the inactivation can be readily reversed by the addition of divalent cations. The ssDNase activity of wheat chloroplast nuclease was strongly inhibited by EDTA but it had no significant effect on its RNase activity [46]. Endonucleases from rat liver nuclei [50] and S. marcescens kums 3958 [23] were inhibited by pyrophosphate, whereas the enzyme from tobacco pollen was inhibited by inorganic phosphate [43]. While inorganic phosphate significantly inhibited the DNase and RNase activities of B. subtilis nuclease [25], it only inhibited the RNase activity of rye germ ribosomal nuclease I [47]. On the other hand, inorganic phosphate had no effect on staphylococcal nuclease [105]. However, inorganic phosphate and pyrophosphate inhibited both the DNase and RNase activities of nuclease Rsn [40, 106].

Divalent cations like Mn2+, Co2+ and Zn2+ inhibited nucleases from potato tubers [41] and B. subtilis [25] while wheat chloroplast nuclease was inhibited by Cu2+, Co2+ and Zn2+[46]. Nucleases from S. marcescens kums 3958 [23], shrimp hepatopancreas [52] and D. melanogaster [53] were inhibited by Ca2+. Though Zn2+ preferentially inhibited the exonuclease activity on dsDNA of N. crassa mitochondrial nuclease [107] and the RNase activity of rye germ ribosomal nuclease I [47], it inhibited both the DNase and RNase activities of S. aureus [105] and S. antibioticus [39] nucleases. Additionally, Hg2+ and Cd2+ inhibited the DNase and RNase activities of staphylococcal nuclease, whereas Ba2+ and Sr2+ inhibited only its RNase activity [105]. The inhibition of the RNase activity, in the presence of Mg2+ and Ca2+, was also observed with rye germ ribosomal nuclease I [47]. While Mn2+ inhibited the endonuclease activity of barley microspores nuclease [49], tea leaf nucleases (A and B) were inhibited by Mg2+, Mn2+, Co2+, Cu2+ and Fe3+[42]. Interestingly, in case of tea leaf nucleases, Zn2+ did not inhibit the dsDNase activity but showed considerable inhibition of the ssDNase, RNase and 3′-nucleotidase activities. However, Mg2+ selectively inhibited the DNase and RNase activities with no effect on the 3′-nucleotidase activity [43]. In case of nuclease Rsn, Zn2+, Cu2+ and Hg2+ inhibited the DNase activity [40] but Cu2+ and Hg2+ had no effect on the RNase activity [106].

Nucleotides and their analogues are potent inhibitors of sugar non-specific endonucleases. For example, staphylococcal nuclease was inhibited by pdTp, 5′-dAMP and 5′-dTMP. While 5′-dCMP and 5′-dGMP inhibited the RNase activity, the DNase activity was not affected. Moreover, among ribonucleotides, only 5′-AMP could strongly inhibit both activities of the enzyme [105]. Thymidine 3′,5′-diphosphate and ATP also inhibited nucleases SM1 and SM2 from S. marcescens kums 3958, whereas adenosine had no effect [23]. In contrast, ATP did not have any effect on the activity of yeast mitochondrial endonucleases [30, 31] and endonuclease G from B. taurus [51]. In the case of potato tuber nuclease, the 3′-nucleotidase activity was inhibited competitively by 5′-mononucleotides, RNA and poly(A) [116]. S. antibioticus nucleases were inhibited by aurin tricarboxylic acid [39] while heparin, Cibacron blue and aurin tricarboxylic acid inhibited yeast mitochondrial endonuclease [31]. Rat liver endonuclease showed inhibition in the presence of polydextran sulfate [50].

The endo–exonuclease from N. crassa was inhibited by a heat-stable, trypsin-sensitive, cytosolic 24-kDa polypeptide. The protein inhibited the ssDNase activity non-competitively but the dsDNase activity was inhibited competitively. In addition, the inhibitor blocked the formation of site-specific double-strand breaks and nicking of linearized pBR322 DNA. It also inhibited the RNase activity of N. crassa nuclease as well as the immunochemically related nuclease from A. nidulans [117]. Moreover, the antibodies raised against N. crassa endo–exonuclease inhibited all the activities of yeast mitochondrial nucleases [30, 31]. The endonuclease NucA from Anabaena sp. was inhibited by its polypeptide inhibitor NuiA, while the related Serratia nuclease was not inhibited with a 10-fold excess of the inhibitor [118]. Cleavage of the monomeric substrate deoxythymidine 3′,5′-bis-(p-nitrophenyl phosphate) by NucA, however, was not inhibited by NuiA suggesting that small molecules gain access to the active site of NucA in the enzyme–inhibitor complex under conditions where cleavage of DNA is completely inhibited [119]. Similarly, the extracellular nuclease (Nuc) from S. marcescens was inhibited by the signal peptide obtained from the N-terminal portion of Nuc [120].

Unlike single-strand-specific nucleases, namely S1 nuclease from Aspergillus oryzae [121], nuclease β from Ustilago maydis [122] and nuclease Bh1 from Basidiobolus haptosporus [123], auto-retardation due to end-product inhibition has not been reported in sugar non-specific endonucleases. On the contrary, staphylococcal nuclease showed auto-acceleration during the hydrolysis of denatured DNA, native DNA and RNA. The phenomenon, best observed with polynucleotides intermediate in size between nucleic acids and short oligonucleotides, was attributed to the presence of substantial ‘breathing spaces’ in the intermediate size substrates [124].

8 Substrate specificity

Sugar non-specific nucleases are multifunctional enzymes and exhibit both endo- and exonuclease activities (Table 3). Moreover, endonucleases from potato tubers [41], tea leaves [42], barley aleurone layers [44] and rye germ ribosomes [47] exhibit 3′-phosphomonoesterase activity.

View this table:
Table 3

Substrate specificity and mode of action of sugar non-specific endonucleases

EnzymeSubstratessDNase:dsDNase ratioProductsMode of actionReference
Azotobacter nucleaseDNA, RNAss≥ds5′-oligonucleotidesendo[125]
Sm nucleaseDNA, RNAss=ds5′-mono-, di-, tri- and tetranucleotidesendo[23, 126]
Neurospora nucleaseDNA, RNAss≥ds5′-mono-, di-, tri- and tetranucleotidesendo[11]
Neurospora nucleaseDNA, RNAss≥ds5′-di-, tri-, tetra- and pentanucleotidesendo–exo[24]
Bacillus nucleasesDNA, RNA1.33′-mononucleotidesendo–exo[25]
Streptomyces nucleases[39]
18 kDaDNA, RNAss>ds5′-mono- and dinucleotidesendo/exo
34 kDaDNA, RNAss<ds5′-mono- and dinucleotidesendo/exo
C1 nucleaseDNA, RNAss=ds5′-oligonucleotidesendo[38]
Aspergillus nucleaseDNA, RNA55′-di-, tri-, tetranucleotidesendo[32]
Yeast nucleaseDNA, RNA25′-productsendo–exo[31]
Yeast nucleaseDNA, RNAss≥ds5′-productsendo[29]
Yeast nucleaseDNA, RNAss≥ds5′-mono-, di- and trinucleotidesendo[28]
Staphylococcal nucleaseDNA, RNA23′-mononucleotidesendo[7]
BAL 31 nuclease[82]
F (fast) formDNA, RNA35′-mononucleotidesendo–exo
S (slow) formDNA, RNA335′-mononucleotidesendo–exo
Leishmania nucleaseDNA, RNA25′-mononucleotides and oligonucleotidesendo[37]
Anabaena nucleaseDNA, RNAss=dsendo[33]
Yeast nuclease (influenced by RAD52)DNA, RNA85′-mononucleotides and oligonucleotidesendo[30]
Lysobacter nucleaseDNA, RNAss≥ds5′-productsendo[27]
Vibrio nucleaseDNA, RNAss≥ds5′-productsendo[26]
Schizosaccharomyces nucleaseDNA, RNAss≥ds5′-productsendo[35]
Nuclease RsnDNA, RNA25′-di-, tri- and tetranucleotidesendo[40]
Rye germ nuclease-IDNA, RNA, 3′-monoribonucleotides1.145′-oligoribonucleotides, 3′-oligodeoxyribonucleotidesendo[47]
Petunia pollen nucleaseDNA, RNA45′-oligonucleotidesendo[45]
Wheat nucleaseDNA, RNA153′-oligoribonucleotides, 5′-oligodeoxyribonucleotidesendo[46]
Rye germ nuclease-IIDNA, RNA1.145′-oligonucleotidesendo[48]
Barley nucleaseDNA, RNA, 3′-monoribonucleotidesss≥ds5′-oligonucleotidesendo[102]
Potato nucleaseDNA, RNA, 3′-monoribonucleotides35′-mononucleotidesendo[116]
Tea leaves[42]
nuclease ADNA, RNA, 3′-monoribonucleotides15′-mononucleotides and oligonucleotidesendo
nuclease BDNA, RNA, 3′-monoribonucleotides15′-mononucleotides and oligonucleotidesendo
Tobacco pollen nucleaseDNA, RNA35′-mono- and oligonucleotidesendo–exo[43]
Barley microspore nucleaseDNA, RNAss≥ds5′-productsendo[47]
Drosophila nucleaseDNA, RNAss=ds5′-mono- and oligonucleotidesendo–exo[53]
Drosophila embryo nucleaseDNA, RNAds>ss5′-productsendo[54]
Shrimp nucleaseDNA, RNAds>ssendo[52]
Sr nucleaseDNA, RNAds>ssendo[34]
Rat liver nucleaseDNA, RNA7>5′-tetranucleotidesendo[50]
Endonuclease GDNA, RNA (DNA:RNA)ds≥ss5′-productsendo[77]

Sugar non-specific endonucleases act on DNA, RNA and 3′-mononucleotides but the rate of hydrolysis of these substrates varies depending on the source of the enzyme. Thus, potato tuber nuclease showed higher activity on 3′-AMP and RNA [41], whereas rye germ ribosomal nuclease I preferred DNA to RNA and 3′-mononucleotides [47]. Nucleases from tea leaves [42] and barley aleurone layers [44] showed preference for RNA in comparison to DNA and 3′-mononucleotides. The substrate specificity of potato tuber nuclease falls in the order 3′-AMP=RNA>ssDNA>dsDNA [41] while that of rye germ ribosomal nuclease I is ssDNA>dsDNA>RNA>3′-AMP [47]. Tea leaf nucleases [42] and barley aleurone nuclease [102] showed substrate preferences in the order of RNA>ssDNA=dsDNA>3′-AMP.

The endonucleases from S. marcescens [110, 126], Anabaena sp. [33], N. crassa [11] and S. racemosum [34] hydrolyzed ssDNA, dsDNA and RNA at a similar rate while staphylococcal nuclease [105] and nuclease Rsn from R. stolonifer [40] hydrolyzed both DNA and RNA with a slight preference for ssDNA. Similar preferences for ssDNA were also shown by endonuclease M (Endo M) from Leishmania sp. [37] and yeast mitochondrial endonuclease [31]. In addition, Endo M degraded ssRNA rapidly but a RNA:DNA hybrid was resistant to cleavage. With increasing concentrations of Endo M, the unlabelled single-stranded overhang of DNA from the RNA:DNA hybrid was cleaved to give the perfect duplex RNA:DNA hybrid. However, in the presence of a 10-fold excess enzyme, the resulting RNA:DNA hybrid was also cleaved [37]. Similarly, endonuclease G from B. taurus showed RNase H activity in addition to DNase and RNase activities [61]. Interestingly, nucleases from L. enzymogens [27], shrimp hepatopancreas [52], S. racemosum [34] and S. antibioticus [39] showed preference for dsDNA. Although 18- and 34-kDa nucleases from S. antibioticus were reported to show preference for dsDNA [39], subsequently it was shown that 18-kDa nuclease prefers ssDNA (J. Sanchez, unpublished results).

Synthetic substrates like deoxythymidine 3′,5′-bis-(p-nitrophenyl phosphate) have been employed for the elucidation of the kinetics and mechanisms of a number of phosphodiesterases [127, 128]. Exonucleases, e.g. snake venom phosphodiesterase [129] and spleen phosphodiesterase [130], and endonucleases, e.g. pancreatic DNase [131], could cleave the synthetic substrate. Cuatrecasas et al. [132] studied the hydrolysis of various p-nitrophenyl ester derivatives of deoxythymidine 5′-phosphate by staphylococcal nuclease and demonstrated that p-aminophenyl, p-nitrophenyl or methyl derivatives are hydrolyzed rapidly. Based on these experiments it was proposed that staphylococcal nuclease requires R-pdT-R′ as the basic structural unit to be recognized as a substrate and the R′ group at C-3′ significantly influences the binding of the substrate or the inhibitor. Additionally, in the absence of R (i.e. free 5′-phosphate), the activity is completely inhibited as evidenced by strong inhibition with pdTp [105]. Friedhoff et al. [133] demonstrated that Serratia nuclease hydrolyzes the artificial minimal substrate, deoxythymidine 3′,5′-bis-(p-nitrophenyl phosphate), at a lower rate than DNA and RNA. The cleavage sites of staphylococcal nuclease, Serratia nuclease and pancreatic DNase I are shown in Fig. 3. Interestingly, with Serratia nuclease, the presence of phosphate 3′ to the bond to be cleaved is essential for hydrolysis [133, 134]. Similar observations were made in the case of Anabaena nuclease [119].

Figure 3

Cleavage specificities of different nucleases (reproduced with permission from Friedhoff et al. [133]).

9 Mode of action

Although sugar non-specific endonucleases recognize and hydrolyze a wide spectrum of substrates, they primarily cleave the internucleotide phosphodiester linkage. Based on the requirement for a free terminus, these enzymes can be classified as endonucleases, exonucleases or endo–exonucleases.

9.1 Endonucleases

This class of enzymes cleave the internal phosphodiester bonds of nucleic acids with or without free termini. Their ability to convert covalently closed circular DNA to a linear form, either directly or through a nicked circular DNA intermediate, is still considered as one of the criteria for inclusion of these enzymes in this group. They show a distributive mode of action and the products of hydrolysis are mononucleotides and/or oligonucleotides. Restriction endonucleases, structure-specific enzymes like flap endonuclease, Holliday junction resolvase, sequence-specific homing endonucleases, and nucleases from S. marcescens, N. crassa and S. aureus belong to this group. However, except for S. marcescens, N. crassa and staphylococcal nucleases, all the other enzymes mentioned above act mainly on DNA (especially dsDNA) and hence they will not be elaborated in this compilation.

9.2 Exonucleases

These enzymes generally require a free terminus for their action and are incapable of hydrolyzing covalently closed circular substrates. However, an exonuclease like snake venom phosphodiesterase is capable of nicking supercoiled DNA [135]. The products of hydrolysis are predominantly mononucleotides and the mode of attack is processive. Spleen phosphodiesterase and A. sydowii nuclease belong to this category of enzymes. Though these enzymes are sugar non-specific, they are strict exonucleases and hence will not be discussed in detail.

9.3 Endo–exonucleases

The enzymes belonging to this category exhibit both exo and endo modes of action.

Sugar non-specific nucleases hydrolyze both DNA and RNA, either endonucleolytically or exonucleolytically but some enzymes exhibit different modes of action on these substrates (Table 3). The majority of these enzymes degrade nucleic acids to oligonucleotides and some mononucleotides predominantly by their endonucleolytic action. Nucleases from N. crassa [24], tobacco pollen [43], S. cerevisiae [30, 31], A. nidulans [32] and D. melanogaster [53] cleave nucleic acids endo–exonucleolytically with the formation of oligonucleotides and a small amount of mononucleotides, whereas the nucleases from A. espejiana [12] and B. subtilis [25] degrade nucleic acids in an exo–endo fashion with mononucleotides as the main product of hydrolysis. BAL 31 nuclease hydrolyzes ssDNA endonucleolytically and shortens the linear duplex DNA from both 3′- and 5′-ends. However, the enzyme isolated from N. crassa mitochondria showed distributive endonuclease activity towards ssDNA but processive exonuclease activity towards dsDNA [24]. In addition, an endo–exonuclease from S. cerevisiae, influenced by the RAD52 gene, showed a non-processive endo mode of action on ssDNA and weakly processive exonucleolytic activity on dsDNA resulting in the formation of oligonucleotides of varying chain length [30]. In contrast, A. nidulans nuclease showed endo and exo modes of action on both ss and dsDNA [32]. Tobacco pollen nuclease, apart from exhibiting an endo mode of action on ssDNA and exo mode with dsDNA, also showed endonucleolytic activity on dsDNA leading to the formation of a typical 58-bp long fragment [43]. The exo mode of action of N. crassa endo–exonuclease [24] is different from those exhibited by typical exonucleases like snake venom phosphodiesterase [136], spleen phosphodiesterase [137] and A. sydowii nuclease [138], in that the hydrolytic products of N. crassa enzyme consist of di-, tri- and tetranucleotides.

The end-products of hydrolysis of DNA and RNA, by sugar non-specific endonucleases, are oligonucleotides and/or mononucleotides with 5′- or 3′-phosphoryl termini. With the exception of S. aureus [7] and B. subtilis nucleases [25], which produce 3′-mononucleotides, all the other nucleases produce 5′-mononucleotides and/or oligonucleotides with 5′-PO4 and 3′-OH termini. However, wheat chloroplast nuclease hydrolyzes ssDNA endonucleolytically, liberating oligonucleotides with 3′-OH and 5′-PO4 termini while oligonucleotides liberated from RNA hydrolysis have 3′-PO4 and 5′-OH termini [46]. Rye germ ribosomal nuclease I, on the other hand, liberates oligonucleotides ending in 3′-OH and 5′-PO4 from RNA and 3′-PO4 and 5′-OH from ssDNA and dsDNA [47].

10 Mechanism of action

Mechanisms of hydrolysis of dsDNA by endonucleases have been studied by light-scattering, viscometry and sedimentation analysis [139141]. In the ‘double-hit’ mechanism, nicks are made at random sites in each strand at points away from each other and the complete fragmentation of duplex DNA does not occur until the two nicks are opposite to each other (multiple encounter). However, in the ‘single-hit’ mechanism, the nicks on either strand are made at points close or opposite to each other, resulting in the complete scission of the duplex DNA in a single encounter [142].

Melgar and Goldthwait [143] studied the effect of metal ions on the hydrolysis of DNA by pancreatic DNase I and demonstrated that, in the presence of Mg2+, the DNA is cleaved predominantly by a double-hit mechanism, whereas with Mn2+, Ca2+ or Co2+ the mechanism of hydrolysis shifts to single-hit mode. On the contrary, metal ions had no effect on the mechanism of action of nuclease Rsn and it cleaved dsDNA through a single-hit mechanism in the presence of Mg2+, Mn2+ and Co2+[40]. In the case of nucleases from yeast mitochondria [28] and S. marcescens [144], it has been suggested that the cleavage of the phosphodiester linkage of native DNA occurs at the same or nearby sites, suggesting a single-hit mechanism.

In general, sugar non-specific endonucleases show exo- and endonucleolytic activities on both DNA and RNA (Table 3). BAL 31 nuclease (both S and F forms) degrades ssDNA exonucleolytically, whereas dsDNA degradation occurs in a terminally directed manner, in which the removal of nucleotides takes place from both the ends of dsDNA. The ratio of the turnover number for the exonuclease activity of the F form, to shorten the duplex DNA, is approximately 27±5 times higher than that of the S form. Apart from terminally directed exonuclease activity, some endonuclease activity was also found to be associated with both F and S species against 5′-terminated single-stranded tails generated by the exonuclease action [82, 145]. The exonuclease activity of the S form, on duplex DNA, decreased with increasing G+C content, whereas the action of F form was not very dependent on the base composition [146]. BAL 31 nuclease F species degraded linear duplex RNA in a terminally directed manner, resulting in the removal of nucleotides from both 3′- and 5′-ends. In comparison, the S species showed very little activity against duplex RNA [147]. Subsequently, Lu and Gray [148] showed that both forms degrade ssDNA in a processive manner from the 5′-end as against the 3′-termini for the duplex DNA. Endo–exonucleases from B. subtilis [149], N. crassa mitochondria [24], S. cerevisiae [31], A. nidulans [32], and D. melanogaster [53] also showed 5′→3′-directed exonuclease action on duplex DNA.

11 Action on polynucleotides

Action of sugar non-specific endonucleases on synthetic polynucleotides revealed that the rate of hydrolysis varies with the source of the enzyme. Wheat chloroplast nuclease hydrolyzed various synthetic polynucleotides in the order of poly(A)>poly(U)>poly(C) and poly(dA)>poly(dT)>poly(dC) but poly(G) and poly(dG) were resistant to cleavage [46]. Staphylococcal nuclease also exhibited a similar preference for the homopolyribonucleotides to that of wheat chloroplast nuclease [105]. However, nuclease I from rye germ ribosomes showed high specificity for poly(C), while other ribopolynucleotides were hydrolyzed in the order of poly(A)≥poly(U)=poly(G). Although rye germ ribosomal nuclease I hydrolyzed the double-stranded deoxyriboheteropolymer poly(dT).poly(rA) at a very slow rate, it failed to hydrolyze the riboheteropolymer poly(A).poly(U), suggesting its preference for single-stranded nucleic acids [47]. Similarly, barley nuclease hydrolyzed the polynucleotides in the order of poly(C)>poly(U)>poly(A)>poly(A).poly(U)>poly(G)=poly(G).poly(C) [102]. In contrast, rye germ ribosomal nuclease II hydrolyzed the double-stranded polymer poly(I).poly(C) at a higher rate than poly(A).poly(U) followed by single-stranded polymers in the order of poly(U)>poly(A)>poly(C)>poly(I). However, poly(G) and poly(dI).poly(dC) were resistant to cleavage [48]. S. cerevisiae nuclease could degrade poly(dT), poly(U), poly(A) and poly(C) but not poly(G), poly(A).poly(U), poly(AU) and poly(A).poly(dT) [31]. Endonuclease from potato tubers showed preference for polynucleotides in the order of poly(U)>poly(A)>poly(I)>poly(C), whereas the enzyme from rat liver nuclei [50] hydrolyzed these substrates in the order of poly(U)>poly(C)>poly(A). A. agilis nuclease hydrolyzed only poly(A) whereas poly(U), poly(C) and poly(G) were resistant to cleavage. While a 10-fold and 30-fold excess enzyme was required for significant hydrolysis of poly(U) and poly(C), respectively, poly(G) was resistant to cleavage even at higher enzyme concentrations [21]. Similarly, nuclease Rsn hydrolyzed only poly(A) while poly(U), poly(C) and poly(G) were resistant to cleavage. However, detectable cleavage of poly(U) and poly(C) could be observed only on prolonged incubation with excess enzyme [106]. S. marcescens kums 3958 nucleases hydrolyzed poly(A) slowly but poly(I).poly(C) was degraded at a higher rate followed by poly(A).poly(U) and poly(dG).poly(dC). However, poly(G), poly(C), poly(U), poly(dT), poly(dA) and poly(dA).poly(dT) were resistant to hydrolysis [23]. On the contrary, Serratia nuclease cleaved the duplex form in the order of poly(dG).poly(dC)>poly(A).poly(U)>poly(dI).poly(dC)>poly(I).poly(C) but poly(dA).poly(dT) was resistant to hydrolysis [126]. Similarly, Anabaena nuclease could readily hydrolyze poly(dG).poly(dC) but poly(dA).poly(dT) was resistant to cleavage, suggesting the preference of the enzyme for the former [118].

12 Action on plasmid DNA

Cleavage of closed circular DNA is considered as one of the main characteristics of endonucleases. In plasmid and phage, closed circular duplex DNA exists in a supercoiled form as a consequence of torsional strain, which at sufficiently high negative superhelical density promotes unwinding of helical twists [150]. Non-specific endonucleases nick supercoiled DNA (Form I) to give rise to linear duplex DNA (Form III) via the formation of nicked circular DNA (Form II). Thus, N. crassa (mitochondrial and vacuolar) nucleases could readily convert Form I DNA to Form II DNA but its subsequent conversion to linear duplex DNA (Form III) occurred at a slow rate [24]. Similar observations were made with nucleases from A. espejiana [12], barley aleurone layers [102], S. cerevisiae [30], A. nidulans [32], Leishmania sp. [37] and S. antibioticus [39]. In the case of N. crassa mitochondrial nuclease, the enzyme action can be controlled by adjusting the concentrations of Mg2+ in the reaction mixture. At low enzyme concentrations, in the presence of 0.1 mM Mg2+, it exhibits strict endonuclease activity and high specificity for Form I DNA. However a 4–8-fold excess enzyme, in the presence of 10 mM Mg2+, accelerated the conversion of Form II DNA to Form III DNA followed by exonucleolytic degradation of Form III DNA [24]. D. melanogaster endonuclease converted Form I DNA rapidly to Form III DNA with very little accumulation of Form II DNA. The appearance of low amounts of Form II DNA was attributed to the preferential nicking on one strand before it is linearized [53]. Similarly, low concentrations of nuclease Rsn converted Form I DNA to Form III DNA via Form II DNA. However, with increase in the incubation time, degradation of Form III DNA was observed. Additionally, the conversion of Form I DNA to Form III DNA and its further degradation in the presence of Co2+ was slow in comparison to Mg2+ and Mn2+, suggesting the relatively low preference of the enzyme for dsDNA in the presence of Co2+[40].

Leishmania nuclease, at low enzyme concentrations, generates single base nicks that can be ligated by T4 DNA ligase to yield covalently closed circular DNA [37]. Similarly, S. pombe endonuclease also produces nicks which are resealable by T4 DNA ligase [35]. In contrast, the nicks generated in supercoiled DNA by BAL 31 nuclease (F and S forms) could not be ligated back to covalently closed DNA, since they were extended into gaps by the exonuclease action of these enzymes. However, the ligation of the gaps could be achieved by carrying out the reaction using T4 DNA ligase in combination with T4 DNA polymerase [151]. BAL 31 nuclease exhibited enhanced endonucleolytic activity on netropsin-bound supercoiled plasmid DNA with no detectable exonucleolytic activity [152].

It is known that intercalating agents change the superhelical density of plasmid DNA in the order of less negatively supercoiled→relaxed→positively supercoiled. Moreover, negatively supercoiled DNA is known to form stably unwound DNA conformations including Z-DNA, cruciform and homopurine–homopyrimidine structures. BAL 31 nuclease cleaves very highly supercoiled DNA prepared from covalently closed relaxed DNA (Form I°) with EtBr. Initial nicking rates of PM2 Form I DNA by BAL 31 nuclease are readily measurable at superhelical densities as low as −0.02 and the nicking activity on positively supercoiled DNA becomes detectable at superhelical densities between 0.15 and 0.19 [81]. It was also noted that non-stressed (i.e. non-supercoiled), non-modified duplex DNAs are resistant to endonucleolytic action [153].

13 Cleavage preference

Staphylococcal nuclease hydrolyzed denatured DNA at a faster rate than native DNA, suggesting that the rate of cleavage depends on the conformation of the substrate [154]. Based on the findings of Felsenfeld and Sandeen [155] that dA–dT-rich regions of native DNA undergo denaturation at somewhat lower temperatures (i.e. below the average Tm) than regions containing higher proportions of dG–dC nucleotide pairs, von Hippel and Felsenfeld [108] postulated that at low temperatures dA–dT regions undergo local opening and closing or strand separation reaction (‘breathing’) to a greater extent than dG–dC-rich regions of DNA. This ‘breathing’ phenomenon facilitates the formation of functional enzyme–substrate complexes and accounts for the relative predominance of deoxyadenylic and thymidylic acid residues in the early digestion products of DNA by staphylococcal nuclease [108]. Additionally, Cuatrecasas et al. [105] observed that the 5′-mononucleotides dA and dT are strong inhibitors of staphylococcal nuclease and suggested it to be the basis of strong binding of the enzyme to dA–dT-rich regions, thereby accounting for the release of deoxyadenylic and thymidylic acid residues in the early part of hydrolysis. This led Wingert and von Hippel [109] to suggest that the specificity of the enzymatic attack, on native DNA, depends on the conformational motility (‘breathing’ or ‘dynamic’ structure) of DNA molecule and that denatured DNA is preferentially attacked when both denatured and native DNA are present. This is supported by the observation that staphylococcal nuclease binds to an exposed single-stranded region and nicks transiently melted base pairs in duplex DNA [15]. Moreover, the formation of single-stranded regions during the course of digestion of native DNA was correlated to ss-cut-triggered local unwinding of the double helix. The above phenomenon has been observed in DNA degrading enzymes capable of producing single-strand nicks [156]. The different pathways for the formation of ssDNA regions are depicted in Fig. 4. The occurrence of gaps in plasmid DNA by the action of BAL 31 nuclease also supports the formation of single-strand regions in native DNA by DNA degrading enzymes. Similar observations were also made with N. crassa endo–exonuclease [24].

Figure 4

Mechanisms of generation of ssDNA regions. The possible modes (a–e) of generation of ssDNA regions are depicted. The arrows indicate the cleavage sites of single-strand-specific enzymes (reproduced with permission from Galcheva-Gargova et al. [156]).

Meiss et al. [144] studied the cleavage of PCR-generated oligonucleotides by Serratia nuclease and demonstrated that the enzyme cleaves preferentially the GC-rich regions (particularly d(G).d(C) tracts) in dsDNA and avoids d(A).d(T) tracts. It was also shown that, compared to DNase I, Serratia nuclease is more non-specific and attacks a particular substrate more evenly under standard reaction conditions. Though the non-specific nature of cleavage is maintained in the presence of high ionic strength or DMSO, addition of urea made the enzyme more selective and this led the authors to suggest that Serratia nuclease could be sensitive to global features like the width of the minor groove [144]. Additionally, experiments with synthetic model substrates (oligonucleotides) showed that Serratia nuclease prefers purine-rich oligonucleotides, probably because it adopts a helical structure with pronounced base stacking. However, cleavage specificities with homo- and heteroduplexes indicated that the A-form of nucleic acid is preferred over the B-form [157]. Similarly, endonuclease from Anabaena sp. also preferred A-form DNA [118]. In the case of endonuclease G from B. taurus, it was shown that the enzyme preferentially cleaves (dG)n.(dC)n tracts in DNA [51, 77]. Synthetic polynucleotides, namely poly(dC)n.poly(dG)n and the oligomer d(GGGGCCCC), adopt a right-handed structure similar to classical A-form DNA [158, 159]. The susceptibility of these oligonucleotides to endonuclease G led Ruiz-Carrillo and Renaud [51] to conclude that the enzyme recognizes A-type DNA. Like Serratia [157] and Anabaena [118] nucleases, endonuclease G does not cleave d(A).d(T) and (dG.dA)34.(dT.dC)34. However, endonuclease G hydrolyzes (dG)24.(dC)24 tracts in supercoiled DNA, in a bimodal way every 9–11 nucleotides, with the maxima in one strand corresponding to minima on the opposite strand, suggesting that the enzyme binds preferentially to one side of the double helix [51]. The segments of (dC–dG) in DNA restriction fragments and in recombinant plasmids adopt a left-handed conformation in high salt solution, while the neighboring regions of natural sequences remain in right-handed helices [160]. Kilpatrick et al. [161] showed that BAL 31 nuclease cleaves the B–Z junction in the presence of high salt concentration but it does not cleave DNA under conditions where (dC–dG)n blocks exist in the B conformation.

14 Base specificity

The majority of sugar non-specific endonucleases reported so far are base non-specific, but the enzymes (Sm1 and Sm2) from S. marcescens [162] and tea leaves [42] showed some base specificity during the initial stages of hydrolysis. Nucleases Sm1 and Sm2 from S. marcescens preferred uracil and guanosine residues respectively in the polynucleotide chain [162]. This difference in the base specificity was correlated to the difference in the N-terminal tripeptide fragment [163]. Tea leaf nuclease hydrolyzed RNA with preferential liberation of 5′-AMP and 5′-GMP, but 5′-UMP and 5′-CMP were liberated after a considerable lag period. Although hydrolysis of DNA gave similar results, 5′-dCMP was detected only after exhaustive digestion of the substrate [42]. These observations indicated the preference of the enzyme for purine nucleotides. Interestingly, nuclease Rsn did not exhibit any base preference and cleaved both ss and dsDNA in a non-specific manner [88] but it showed high preference for adenylic acid linkages in RNA [106]. Action of barley aleurone layer nuclease on dinucleoside monophosphates revealed a strong preference for purine nucleosides as the 5′-residue followed by uridine as the 3′-residue. Similarly, staphylococcal nuclease hydrolyzed various dinucleoside monophosphates in the order of d-TpT>d-CpC>d-GpG. It was observed that the presence of 5′-phosphate in the dinucleotides brought about a 25-fold increase in the activity, compared to 5-fold when the phosphate was present at the 3′-position [164].

15 Associated phosphomonoesterase activity

Among sugar non-specific endonucleases, only those from potato tubers [41], barley aleurone layers [102], tea leaves [42] and rye germ ribosomes [47] exhibit 3′-phosphomonoesterase activity. Interestingly, all the aforementioned enzymes, except rye germ ribosomal nuclease I [47], are 3′-monoribonucleotides-specific. In addition, some enzymes exhibit a base preference for the hydrolysis of various 3′-monoribonucleotides. Thus, tea leaf nucleases (A and B) hydrolyzed various 3′-monoribonucleotides in the order of A>G>U>C [42], whereas potato tuber nuclease hydrolyzed them in the order of A>U>G>C [41]. In contrast, barley aleurone layer nuclease readily hydrolyzed 3′-AMP and 3′-GMP while 3′-CMP and 3′-UMP were resistant to cleavage [102].

16 Structure and function

The amino acid sequences of nine sugar non-specific endonucleases, derived from their gene sequences, showed that most of them are made up of a single polypeptide chain (Fig. 5). Based on the structural and functional similarity, endonucleases from Anabaena sp. [33], S. cerevisiae [165], S. racemosum [34], C. echinulata [38], S. pombe [35] and B. taurus [61] were shown to belong to the Serratia family of non-specific endonucleases. In all the enzymes, except staphylococcal nuclease, the active site residues are conserved (Fig. 5), including the DRGH motif which contains the active site histidine [119, 166, 167]. Among them, C1 nuclease from C. echinulata showed 44% and 42% sequence homology with S. cerevisiae and S. pombe nucleases, respectively [38].

Figure 5

Sequence alignment of non-specific endonucleases. Sma –S. marcescens, Asp –Anabaena sp., Bos –B. taurus, Sce –S. cerevisiae, Sra –S. racemosum, Spo –S. pombe, Cun –C. echinuclata, Shr – shrimp hepatopancreas and Sta –S. aureus. The numbers on the right give the last residue of each line. Identical amino acid residues (excluding those of S. aureus) having ≥50% sequence similarity are shaded and regions exhibiting substantial homology are boxed. The DRGH motif is depicted in bold face. The conserved active site residues are marked with an asterisk. The sequence alignments were made based on the data from Ho et al. [38], Friedhoff et al. [166, 167] and Wang et al. [168].

Staphylococcal nuclease consists of a single polypeptide chain of 149 amino acids with no disulfide bonds or free cysteine [83]. Shrimp nuclease is also a single polypeptide containing an open reading frame encoding a putative 21-residue signal peptide and a 381-residue mature protein. The polypeptide is cross-linked by five disulfide bonds [168]. C1 nuclease from C. echinulata is a monomeric protein containing 252 amino acids cross-linked by a single disulfide bond [38]. Similarly S. racemosum nuclease, a glycoprotein, is made up of a single polypeptide chain of 250 amino acids with a single disulfide bond and the carbohydrate moiety is attached to Asn134 [169]. Serratia nuclease is produced as a 266-amino acid pre-protein with a signal peptide consisting of first 21 residues [170, 171]. Among the two major isoforms produced by S. marcescens [23, 96], with near identical biochemical properties, the 245-amino acid mature nuclease (Sm2), with a Mr of 26.7 kDa, is the result of a typical signal sequence processing. The second isoform (Sm1) lacks the first three N-terminal amino acids [96, 163, 172, 173].

Although a large number of sugar non-specific endonucleases have been reported to date, structural studies have been limited to staphylococcal nuclease [174176] and Serratia nuclease [85, 166, 177, 178]. The crystal structure of staphylococcal nuclease, in the presence of Ca2+ and pdTp at 1.5–1.6 Å resolution, showed the involvement of 30 residues in three separate sections of the helix and, in addition, about 24 residues form a three-stranded section of anti-parallel β-sheet. Residues 44–53 were found to be somewhat disordered and formed a highly solvent-exposed large loop (omega-loop). Moreover, the structure showed the presence of an inhibitor pocket which is predominantly neutral or hydrophobic. In addition, the inhibitor pocket also contained several residues, which specifically participate in binding of calcium ion and the nucleoside diphosphate [174176]. The crystal structure of Serratia nuclease, a dimeric enzyme, was first solved at 2.1 Å by Miller et al. [177], and it was then refined at 1.7 Å and 1.1 Å resolution by Lunin et al. [178] and Shlyapnikov et al. [179], respectively. The crystal structure revealed that the dimer consists of two identical monomers and each monomer contains the active site. Moreover, each monomer consists of a central six-stranded anti-parallel β-sheet flanked on one side by a helical domain and on the opposite side by a dominant helix and a very long coiled loop. The crystal structure also showed that the cleft between the long helix and the loop, near His89, may contain the active site [177].

16.1 Active site

Considerable work has been done on the substrate specificity and mode of action of sugar non-specific endonucleases but very little information is available regarding the nature of their active site. S. aureus nuclease bound 5′-mononucleotides in a ratio of 1:1 and was completely dependent on the presence of Ca2+. Moreover, binding of 5′-mononucleotides and pdTp to the enzyme was affected by low amounts of DNA and RNA, suggesting the presence of a common catalytic site for the hydrolysis of both the substrates [180]. Studies using amino and nitrotyrosyl derivatives of Tyr85 and Tyr113 indicated that they are in stereochemical proximity and act in a concerted manner [181]. Through site-directed mutagenesis it was established that Tyr85 has a more direct role in the active site than Tyr113 or Tyr115 [182]. Enzyme crystals, obtained in the presence of Ca2+ and pdTp, revealed the involvement of Lys84 and Tyr85 in the catalytic activity and that they form hydrogen bonds with the 3′-phosphate of the inhibitor, whereas Arg35 and Arg87 form hydrogen bonds with the 5′-phosphate. Carboxylate ions of Glu43, Asp21 and Asp40 serve as ligands for the binding of Ca2+[174]. Cotton et al. [175] suggested that pdTp, in particular its 5′-phosphate, and the activating Ca2+ are precisely and rigidly locked into the position at the hydrolytic site. It was also proposed that while Glu43 acts as a general base, Arg35 and Arg87 are involved in the neutralization of the charges on the phosphates and Ca2+ performs the dual role of aiding catalysis and stabilization of the trigonal bipyramid transition state. Crystal structure studies of staphylococcal nuclease at 1.6 Å resolution confirmed the above observations [176].

Based on the sequence homology of related nucleases [20, 166], mutational analysis of the conserved amino acid residues [166] and crystal structure [85, 177], Friedhoff et al. [167] proposed that the active site of Serratia nuclease comprises at least four residues, namely: Arg57, His89, Asn119 and Glu127. Since the mutants of active site residues showed a similar effect on the hydrolysis of both DNA and RNA, the authors suggested the existence of a common catalytic site for the hydrolysis of both substrates. Furthermore, it was observed that among all mutants of His89, only H89N showed measurable enzyme activity in addition to a decrease in the kcat, suggesting the probable role of His89 in catalysis. However, the similar substrate binding efficiency of the wild-type enzyme and mutant H89A, of the non-cleavable modified substrate (a decamer in which phosphates are replaced by phosphorothioates), suggested that His89 is not involved in substrate binding. It was also demonstrated that His89 has to be deprotonated to function as a general base in the catalytic activity [167]. The close proximity of Arg57 and Arg87 to the phosphate backbone in the substrate [177], coupled with the decrease in the kcat of the Arg57 mutants and with no change in the Km, suggested the involvement of Arg57 in catalysis. Moreover, the significant decrease in the kcat of the Arg57 mutant was correlated to its involvement in positioning and polarizing the phosphate of the scissile phosphodiester bond and/or stabilization of the transition state. Asn119 was implicated in catalysis since mutants of Asn119 showed a decrease in their kcat and not in the Km [166, 167]. Subsequently Miller et al. [183] established the involvement of Asn119 in metal binding. Compared to the native enzyme, the alanine mutants of Glu127 showed reduced activity while the mutants E127D and E127Q showed similar activities to that of the native enzyme, indicating that Glu127 might be participating indirectly in the hydrolysis of the substrates [167]. Non-specific endonuclease from Anabaena sp. showed nearly 30% sequence homology with Serratia nuclease and this was restricted mainly to the active site regions and the central six-stranded β-sheet of the enzyme. Based on the three dimensional structure of S. marcescens nuclease and mutational analysis of the catalytically active residues, a structural model for the Anabaena nuclease was proposed. Accordingly, it was suggested that His124, Asn155 and Glu163 (corresponding to His89, Asn119 and Glu127 of S. marcescens nuclease) are involved in the catalytic activity of the enzyme. Additionally, through mutational analysis it was demonstrated that Asn155 and Asp121 are involved in the coordination of the metal ion cofactor. Subsequently cleavage studies with deoxythymidine 3′,5′-bis-(p-nitrophenyl phosphate) revealed the role of His124 as a general base and that of Glu163 in assisting the general acid [119].

Chemical modification studies on nuclease Rsn, from R. stolonifer, showed the involvement of two histidine, one tryptophan and two carboxylate residues in the catalytic activity of the enzyme. Substrates of nuclease Rsn, viz. DNA and RNA, could not protect the enzyme against DEP- and EDAC-mediated inactivation, whereas substrate protection was observed in the case of NBS-mediated inactivation of the enzyme. Km and kcat values of the partially inactivated enzyme samples suggested that while histidine and carboxylate are involved in catalysis, tryptophan is involved in substrate binding. Furthermore, fluorescence quenching studies on native and modified nuclease Rsn, using metal ions, indicated the involvement of carboxylate in metal binding [184].

Based on the sequence similarity with S. marcescens and site-directed mutagenesis, the involvement of His87, His85 and His211 in the catalytic activity of endonucleases from C. echinulata [38], S. racemosum [169] and shrimp hepatopancreas [168], respectively, was demonstrated. Moreover, the probable involvement of Asn241 in catalysis and Lys210 and Arg253 in substrate binding was postulated for shrimp hepatopancreas nuclease [168]. In the case of the mitogenic factor secreted by S. pyogenes, His122 was shown to be the residue essential for the nuclease activity [36].

16.2 Coordination and function of metal ions

Staphylococcal nuclease is a Ca2+-dependent non-specific enzyme, while S. marcescens nuclease requires Mg2+ for its activity. Extensive studies on these enzymes have provided insights into the substrate binding, metal interaction and catalytic mechanism. In case of staphylococcal nuclease, Ca2+ is coordinated directly to Asp21 and Asp40, while Glu43 is coordinated to the calcium ion through a water molecule. Additionally, the calcium ion was found to coordinate with two additional water molecules (Fig. 6) [174, 176, 185]. The roles of Asp21 and Asp40 were further delineated by electron-spin paramagnetic resonance studies, in which Asp21 was found to influence the binding of Ca2+ more strongly than Asp40 [186]. Through the crystal structure at 1.95 Å resolution of the D21E ternary complex, where D21E is bound to both Ca2+ and pdTp (transition state analogue), it was demonstrated that Ca2+ in the active site binds to Glu21 via bidentate coordination and to Glu43 through an inner sphere coordination. The cooperativity of the metal ion and the inhibitor (pdTp) in the ternary complex was also demonstrated [187].

Figure 6

Schematic view of the binding site of staphylococcal nuclease in the enzyme–pdTp–Ca2+ complex. The heavy dots represent regions of electron density assigned as water molecules. Hydrogen bond and ionic interactions are shown as dotted lines. The interaction between Ca2+ and Val41 is through the carbonyl oxygen of the peptide bond (reproduced with permission from Cotton et al. [175]).

On the other hand, in Serratia nuclease the catalytic magnesium binding site is located between the long helix (residues 116–135) and the loop extending from residues 50 to 114, which contains the catalytically essential His89. It was shown that Mg2+ bound to Asn119 (the only protein ligand) is associated with five water molecules to complete an octahedral coordination complex (Fig. 7). Glu127 and His89 are located nearby and each is hydrogen-bonded to water molecules in the coordination sphere. Kinetic and site-specific mutagenesis studies suggested that this metal–water cluster contained the catalytic metal ion. The residues involved in the formation of metal–water cluster coordination are conserved in enzymes homologous to Serratia nuclease. Additionally, it was demonstrated that Serratia nuclease is a unique enzyme of the endonuclease family in which the amide side-chain acts as the sole protein ligand for magnesium coordination [183]. Though Lunin et al. [178] proposed that Asp86 is involved in Mg2+ coordination, Miller et al. [183] demonstrated that Asp86 interacts indirectly with the metal binding site by forming hydrogen bonds with the amide nitrogen of Asn119 and Gln120. Similarly, Meiss et al. [119] showed that the metal cofactor is bound to Asp121 in addition to Asn155, which is the primary ligand for binding the cofactor.

Figure 7

Schematic view of the magnesium binding site of Serratia nuclease (reproduced with permission from Miller et al. [183]).

16.3 Enzyme–substrate interactions

In staphylococcal nuclease, binding of either Ca2+ or the substrate/inhibitor to the enzyme is not mutually exclusive [180, 188]. Cuatrecasas et al. [180] showed that Ca2+ does not bind to the enzyme in the absence of nucleotide or substrate and hence suggested that Ca2+ and nucleotide bind to the enzyme in a unique complex formed only when all three components are present. Before the crystal structures of staphylococcal nuclease were solved, Cuatrecasas et al. [189] postulated that the active site of the enzyme consists of three phosphate binding ‘subsites’ (namely P1, P2 and P3) and the binding of oligonucleotides occurs predominantly by ionic interactions. The contribution of subsites in the binding of substrates/inhibitors to form a complex was shown to be in the order of P1>P2>P3. Also, the cleavage specificity of the hydrolytic site, visualized as an integral part of the P1 subsite, was shown to be influenced by the presence of nucleoside on the 5′-side rather than the 3′-side [189]. Weber et al. [190] postulated that the binary enzyme–dTdA complex is one of the intermediates leading to the active ternary enzyme–metal–substrate complex. Based on the conformational changes between the binary and ternary complexes it was demonstrated that the metal-induced changes, both at the attacked phosphorus and at the leaving group of the enzyme-bound substrate, may contribute to catalysis.

Stanczyk and Bolton [191], using crystal structures of the ternary complexes of staphylococcal nuclease with Ca2+ and pdTp, showed that the 5′-phosphate of pdTp interacts with the solvent-inaccessible Arg35 and Arg87, whereas the 3′-phosphate and Ca2+ interact with the solvent-accessible ?-ammonium group of Lys84 and the phenolic hydroxyl group of Tyr85. It was also observed that the interaction of a terminal 3′-phosphate for exonucleolytic activity differs slightly from the interaction with an internal phosphodiester linkage required for the endonucleolytic cleavage. Subsequently, using the ternary complexes of the enzyme with pdTp, pdGp and p-nitrophenyl-pdTp, the above authors [191] demonstrated that the conformational features of the bound nucleic acid determine the differences in the observed catalysis rather than the nucleotide itself. Similar observations were also reported by Grissom and Markley [192].

In the case of Serratia nuclease, the minimum substrate recognized is a pentanucleotide containing phosphate at the 5′-position [193]. Friedhoff et al. [133] demonstrated that the 5′-phosphorylated pentamers are cleaved predominantly at the second phosphodiester bond and to a lesser extent (5%) at the third phosphodiester position. Additionally, at high enzyme concentrations, cleavage of the tetramer was observed at the first and second phosphodiester positions with no detectable cleavage at the third phosphodiester bond. Based on the above observations, the authors concluded that a phosphate 3′ to the scissile phosphodiester bond is important for the cleavage. This conclusion is supported by the cleavage of the artificial substrate deoxythymidine-3′,5′-bis-(p-nitrophenyl phosphate) by S. marcescens nuclease [134]. Furthermore, Serratia nuclease is not active against methylphosphonate-substituted pentanucleotides suggesting that the negative charges and/or the hydrogen bond acceptors on the phosphate to be attacked, as well as on the neighboring phosphates, are required for binding and correct orientation of the substrate in the active site of the enzyme [133, 194]. The requirement of phosphate groups for the effective binding of the substrate to the active site of the enzyme was found to be similar for both S. aureus and S. marcescens nucleases. Moreover, cleavage experiments with polydeoxyadenylates, containing a single phosphorothioate at different positions, indicated that Serratia nuclease hydrolyzed the Rp-diastereomer at a higher rate than the Sp-diastereomer. Hence, the authors opined that the Sp orientation prevents cleavage at the phosphorothioate group and retards the cleavage at a position 5′ to the substitution, whereas the Rp-phosphorothioate group enhances the cleavage at the 5′-position adjacent to the site of substitution [133]. Interestingly, the role of Trp123 in substrate binding was established by kinetic studies involving mutant and wild-type enzymes [157]. Similarly, in the case of the Anabaena nuclease Meiss et al. [119] demonstrated the involvement of Trp159 in substrate binding. Moreover, studies on the mechanism of inhibition of Anabaena nuclease (NucA) by its inhibitor (NuiA), using several active site mutants of NucA, suggested that Glu163 of NucA might be participating in cofactor binding or protonation of the leaving group as the target amino acid involved in the inhibition of the enzyme. Based on these observations, it was concluded that Glu163 of the enzyme interacts with the positively charged amino acids of the inhibitor, while Arg93, Arg122 and Arg167 of the enzyme interact with the negatively charged amino acids of the inhibitor [119].

16.4 Water-assisted metal ion catalysis

The crystal structures of staphylococcal nuclease [174176] and Serratia nuclease [178, 183] revealed the presence of water molecules adjacent to the metal ions. In both cases, the reaction is triggered by the water molecule which acts as the nucleophile and attacks the scissile phosphodiester bond. The reaction mechanism, for staphylococcal nuclease, based on the crystal structure [175, 176], kinetic studies [195, 196] and site-directed mutagenesis [186, 197201], is depicted in Fig. 8.

Figure 8

Reaction mechanism of staphylococcal nuclease (reprinted with permission from Weber et al. [190], Copyright 1991, Americal Chemical Society).

The reaction mechanism involves the Ca2+ activator bound in a septacoordinate complex [176] to Asp21, Asp40 and the amide carbonyl group of Thr41 as well as the 5′-phosphate of the competitive inhibitor (pdTp), with the remaining three ligands probably being the water molecules. The attacking water molecule is presumed to be near the Ca2+, either in direct coordination or in the second sphere of Glu43, to permit its carboxylate group to function as a general base [199, 200]. Upon nucleophilic attack, the phosphodiester linkage of the substrate is converted into a trigonal–bipyramid intermediate stabilized by Arg87 [186, 201], which also acts as the general acid to protonate the 5′-oxygen of the leaving nucleoside.

Carboxylate groups are known to be effective nucleophilic catalysts in the intramolecular hydrolysis of phosphate esters [202]. The mechanism of staphylococcal nuclease suggested by Cotton et al. [175] requires that the γ-carboxylate group of Glu43 acts as a general base, facilitating the attack of water on the scissile phosphodiester bond. Judice et al. [203], while probing the mechanism of staphylococcal nuclease with unnatural amino acids, suggested that Glu43 plays a more complex structural role in catalysis rather than acting as a general base. To ascertain the role of Glu43 as a general base, Loll et al. [204] solved the crystal structure of the ternary complex of staphylococcal nuclease in the presence of Co2+, as the structure is very similar to the structure obtained with Ca2+ with little or no distortions of the groups involved in the catalytic activity. It was observed that Co2+ occupies the site identical to the Ca2+ binding site and both the metals show a similar interaction with the inner sphere of Glu43, indicating that Glu43 is still fully capable of activating the water molecule to act as a general base. Additionally, it was noted that the inner sphere interaction of Ca2+ with Asp21 was missing in the Co2+ structure, thus establishing the fact that the loss of catalytic activity of Co2+-substituted nuclease is probably due to its inability to bind the inner sphere water [204]. As mentioned earlier, Glu43 in staphylococcal nuclease is situated at the base of the solvent-exposed and conformationally mobile omega-loop in the active site [174]. High resolution X-ray structures revealed that the substitution of Glu43 by aspartic acid significantly changes the structure of the omega-loop and reduces the interaction of Ca2+ with its active site ligands, resulting in a diminished hydrogen-bonded network of the water molecules [205]. Mehdi and Gerlt [206] showed that the hydrolysis of substrates by staphylococcal nuclease proceeds with the inversion of configuration at phosphorus, aided by Ca2+, which assists in the proper positioning of the carboxylate group via an intervening water molecule and neutralizes the phosphate ester charge by a direct ionic interaction. Similar observations were made by Potter et al. [207, 208] with the S1 and P1 nucleases. The rate determining step for the hydrolysis of DNA/RNA by staphylococcal nuclease at pH 9.5 was shown to be substrate binding and product dissociation rather than cleavage of the phosphodiester bond. However, under similar conditions with the E43D mutant, in which the putative active site general base catalyst Glu43 is replaced by aspartic acid, cleavage of the phosphodiester bond becomes the rate determining step [209].

As mentioned earlier, Serratia nuclease also catalyzes the hydrolysis of phosphodiester bonds through a metal ion-mediated mechanism with a water molecule acting as the nucleophile. Using synthetic substrates [134], site-directed mutagenesis [166, 167] and crystal structures [177, 183], Friedhoff et al. [210] proposed a ‘general base model’ for the reaction mechanism of Serratia nuclease (Fig. 9A). A similar reaction mechanism has also been postulated by Meiss et al. [119] for the endonuclease from Anabaena sp. (Fig. 9B).

Figure 9

Reaction mechanism of Serratia (A) and Anabaena (B) nucleases (reproduced with permission from Meiss et al. [119]).

In the ‘general base mechanism’ postulated for Serratia nuclease (Fig. 9A), His89 acts as a general base by abstracting a proton from the water molecule thereby activating it for a nucleophilic attack on the phosphorous atom adjacent to the scissile bond, followed by cleavage of the 3′ O–P bond [134, 177]. Magnesium ion acts as the Lewis acid to stabilize the negative charge on the pentacoordinate phosphate transition state and the leaving group. It has been postulated that Asn119 is probably involved in the binding and correct positioning of the metal cofactor. Arg57 has been shown to be involved in the transition state stabilization [210]. Interestingly, homing endonuclease I-PpoI from Physarum polycephalum [210, 211], colicin E9 endonuclease [212] and junction resolvase endonuclease (Endo VII) from phage T4 cells [213] shows an identical active site and has been postulated to have a similar mechanism. Anabaena nuclease follows a similar reaction mechanism, where His124 acts as the general base, Asn155 in binding and correct positioning of the cofactor, and Glu163 in protonation of the leaving group (Fig. 9B). In contrast to Arg57 in Serratia nuclease, the transition state stabilization in Anabaena nuclease is effected by Asp95, either by hydrogen bonding or by binding to a second metal ion [119]. Active site characterization of nuclease Rsn from R. stolonifer showed the involvement of similar residues to those of S. marcescens nuclease in the catalytic activity of the enzyme. Hence the authors suggested that, like S. marcescens nuclease, nuclease Rsn may also follow the metal–water cluster-mediated mechanism for the hydrolysis of DNA and RNA [184].

Miller et al. [183] postulated two schemes for the hydrolysis of phosphodiester bonds by Serratia nuclease, where an unligated water molecule may be directly activated by His89, or the magnesium-bound water is activated by His89. In this the metal ion may alter the pKa of the bound water to produce a more nucleophilic hydroxide ligand, and this activated water molecule may mediate the cleavage of the phosphodiester bond. In the former, the cluster polarizes the phosphate atom, stabilizes the phosphorane and helps protonate the leaving group, whereas in the latter, the cluster polarizes the phosphate atom, stabilizes the phosphorane and helps activate the attacking hydroxide.

Although Lunin et al. [178] initially proposed a general acid model for Serratia nuclease, subsequently based on the observation of Kolmes et al. [134] that the E127A mutant, and not H89A, cleaves deoxythymidine 3′,5′-bis-(p-nitrophenyl) phosphate (as the protonation of the leaving group is not necessary), Shlyapnikov et al. [179] proposed a consensus mechanism similar to the one proposed by Friedhoff et al. [210]. In the case of Serratia nuclease, the rate determining step for the hydrolysis of oligonucleotides is substrate binding and/or phosphodiester bond cleavage but not product dissociation [133].

The reaction mechanisms of staphylococcal and Serratia nucleases suggest that the hydrolysis of the phosphodiester bond is very much dependent on the proper orientation of the residues so as to initiate an in-line attack of the water molecule on the scissile phosphodiester bond. Also, the proper orientation is achieved by the concerted action of Asn21, Asn40 and Ca2+ in the case of staphylococcal nuclease [175, 176] and Asn119 and Mg2+ in Serratia nuclease [177, 183]. It is interesting to note that though both enzymes could efficiently perform the dual roles of proper orientation and transition state stabilization of the intermediate or the leaving group through a single metal atom, P1 nuclease from P. citrinum, a zinc metalloprotein, required three metal ions for performing a similar role [214]. Accordingly, the scissile phosphate of the substrate binds close to ZnII and the base 5′ to the bond to be cleaved stacks against phenylalanine (Phe61) and forms hydrogen bonds with aspartic acid (Asp63). ZnI and ZnIII bridge the water molecule, which acts as the nucleophile attacking the phosphate in-line with the P–O3′ bond [215]. The carboxylate group of aspartic acid (Asp45), which also serves as a ligand of ZnI, helps to orient the hydroxide properly for the attack. An arginine residue (Arg48) stabilizes the resulting pentacoordinate transition state and the attacking hydroxide ion along with the leaving oxyanion (O3′) occupy apical positions in this transition state. ZnII plays a crucial role in activating the phosphate and stabilizing the leaving O3′ oxyanion. Thus, all the three zinc ions are important for catalysis [214].

17 Biological role

As mentioned earlier, nucleases play an important role in replication, recombination, restriction and repair. Extracellular enzymes have been implicated in scavenging of nucleotides and phosphates for the growth and metabolism. Nucleases have been shown to have a role in the formation of nicks during meiotic recombination. The reduced amounts of endo–exonuclease in rad52 mutants of S. cerevisiae were correlated to its probable involvement in repair and recombination in both mitotic and meiotic cells [30]. The ability of N. crassa endo–exonuclease to act on supercoiled DNA and its ability to initiate single-strand and double-strand breaks in DNA along with the processive exonucleolytic action on circular DNA in the presence of ssDNA binding proteins, led Chow and Fraser [24] to suggest that it might be involved in the production of stable lengths of ssDNA needed for exchanges during recombination and recombinational repair of mitochondrial DNA. In the case of N. crassa the repair deficient and UV-sensitive mutants, uvs-2, uvs-3, uvs-6 and nuh-4, could not secrete endo–exonucleases. These mutants had a higher level of endo–exonuclease precursor than the wild-type, indicating that they may have some defect either in the protease(s) that control the nuclease level or in the regulation of protease(s). The above mutants were also sensitive to various mutagens and mitomycin C and exhibited a high frequency of spontaneous, recessive lethal mutations and deletions, indicating the involvement of N. crassa nuclease in repair [216]. Moreover, it has been suggested that N. crassa mitochondrial endonuclease may have a role in replication [11]. Siwecka et al. [47] postulated the interaction between rye germ nuclease-I and ribosomes to be a part of the regulatory mechanism of cell metabolism. Owing to the striking correlation between genetic instability and the presence of poly(dG).poly(dC) tracts in DNA (the preferred substrate of endonuclease G), Ruiz-Carrillo and Renaud [51] suggested a probable role for endonuclease G in recombination. Pollen has been used as a promising host for introducing DNA into higher plants [217, 218]. The pollen nucleases from P. hybrida [45] and tobacco [43] prefer ssDNA thereby enhancing the chances of the uptake and expression of native DNA.

Recently, Nicieza et al. [39] isolated two extracellular nucleases from S. antibioticus, with Mr values of 18 and 34 kDa, which are nutritionally regulated and reach their maximum activity during aerial mycelium formation and sporulation. Their role appeared to be DNA degradation in the substrate mycelium and supply of building blocks for macromolecular biosynthesis in the aerial mycelium, and they acted in concert with the periplasmic nuclease. Of the two extracellular nucleases, the 18-kDa nuclease appeared to be reminiscent of NUC-18, a thymocyte nuclease assumed to have a key role in glucocorticoid-stimulated apoptosis [219, 220]. Interestingly, the N-terminal sequence of the 18-kDa protein showed striking similarity to proteins of the cyclophilin family which degrade DNA in a Ca2+/Mg2+-dependent manner and their role in apoptosis has been reviewed [221, 222]. The mitogenic factor of S. pyogenes has been implicated in pathogenesis similar to other pyrogenic exotoxins, but the role of the nuclease activity exhibited by the mitogenic factor is still not clear [36].

DNA/RNA non-specific nucleases from S. aureus and S. epidermidis are found in a variety of clinical and food infections [223228]. Similarly, nucleases from Vibrio cholerae [229] and S. marcescens [230] have been postulated to play a role during invasion or establishment of an infection. Matousek et al. [231] showed that pollen RNases, owing to their ability to degrade dsRNA, may function as defense proteins against viral infection as components of a degradation complex which participates in the apoptosis of the tapetal cell layer and for nucleoside re-utilization by the developing pollen.

The main role of the associated nucleotidase activity of nucleases is to scavenge nucleotides and phosphates for growth and survival of the organism under environmentally stressed conditions. Brown and Ho [102] opined that barley nuclease, in concert with other acid phosphatases, might be involved in the hydrolysis of remnant nucleic acid in the endosperm for providing the valuable nutrients for heterotrophic embryonic growth. A similar role has been suggested for potato tuber nuclease [41].

18 Applications

Since their discovery, sugar non-specific endonucleases have been extensively used as analytical tools for the determination of nucleic acid structure. The N. crassa nuclease was employed for the isolation of pure lac operon [232], isolation of tRNA and rRNA gene hybrids [233, 234] and detection of sequence heterology [235]. Staphylococcal nuclease has been used to monitor hybridization reactions with DNA [14] and as a probe for drug binding sites on DNA [19]. The ability of S. aureus and N. crassa nucleases to discriminate between uridine and cytidine under defined conditions has been exploited for rapid RNA sequencing [236]. Wheat chloroplast nuclease, due to its preference for pyrimidines in non-base-paired regions, has been used as a tool for the structural analysis of RNA [16]. The ssDNase activity of BAL 31 nuclease was used for probing regions of altered secondary structures in negatively and positively supercoiled closed circular DNA [18]. Additionally, the dsDNase activity of BAL 31 nuclease which progressively shortens the duplex DNA, from both ends, was exploited for ordering restriction endonuclease-generated DNA fragments [17]. Serratia nuclease, under the trade name of ‘Benzonase’, is used to eliminate nucleic acid contamination from purified recombinant proteins and single cell protein preparations. The ability of Serratia nuclease to degrade DNA has paved way for its utilization as a killer gene for the self-destruction of microorganisms released into the environment in addition to its use as an anti-viral agent [58].

19 Conclusions

The ability of sugar non-specific endonucleases to recognize a wide variety of nucleic acid structures has led to considerable efforts to evaluate their role in different cellular processes as well as application as analytical tools to study nucleic acid structure. The majority of these enzymes share common properties like multiple activity and metal ion requirement. Although these enzymes have been extensively studied with respect to their catalytic properties, very little attention was paid to their structure–function relationship. The amino acid sequences of typical sugar non-specific endonucleases from S. marcescens, S. cerevisiae, B. taurus, C. echinulata var. echinulata, S. racemosum, S. pombe, Anabaena sp. and shrimp hepatopancreas showed significant homology in their primary structure. The comparison of primary sequences, conserved sequences and residues involved in the active site of sugar non-specific endonucleases will advance our understanding on the evolutionary aspect of these enzymes. Until now, the crystal structures of only two enzymes, namely staphylococcal and Serratia nucleases, have been solved. Solving the three dimensional structures of different sugar non-specific enzymes will be of great importance as it will add valuable information regarding the structural homology of these enzymes at the tertiary level. Moreover, the majority of the sugar non-specific endonucleases show the requirement of more than one metal ion for their activity. Some enzymes exhibit synergism when metal ions are used in combination. Interestingly, the sugar non-specific endonuclease from R. stolonifer shows a high preference for ssDNA in the presence of Co2+. Hence, site-specific mutagenesis in combination with the crystal structures of these enzymes, in the presence and absence of metal ions, will not only yield information regarding the residues involved in the catalytic activity but also the catalytic mechanism of this class of enzymes. In contrast, sugar non-specific endonucleases from plants do not show an obligate requirement for metal ions for their activity. The nature of the active site and the three dimensional structure of these enzymes will be helpful in comparing the mechanisms of action of metal requiring and metal non-requiring endonucleases.


This work was supported by a grant from the Department of Science and Technology, Government of India, to V.S. This is communication No. 6613 from the National Chemical Laboratory, Pune, India.


  1. [1].
  2. [2].
  3. [3].
  4. [4].
  5. [5].
  6. [6].
  7. [7].
  8. [8].
  9. [9].
  10. [10].
  11. [11].
  12. [12].
  13. [13].
  14. [14].
  15. [15].
  16. [16].
  17. [17].
  18. [18].
  19. [19].
  20. [20].
  21. [21].
  22. [22].
  23. [23].
  24. [24].
  25. [25].
  26. [26].
  27. [27].
  28. [28].
  29. [29].
  30. [30].
  31. [31].
  32. [32].
  33. [33].
  34. [34].
  35. [35].
  36. [36].
  37. [37].
  38. [38].
  39. [39].
  40. [40].
  41. [41].
  42. [42].
  43. [43].
  44. [44].
  45. [45].
  46. [46].
  47. [47].
  48. [48].
  49. [49].
  50. [50].
  51. [51].
  52. [52].
  53. [53].
  54. [54].
  55. [55].
  56. [56].
  57. [57].
  58. [58].
  59. [59].
  60. [60].
  61. [61].
  62. [62].
  63. [63].
  64. [64].
  65. [65].
  66. [66].
  67. [67].
  68. [68].
  69. [69].
  70. [70].
  71. [71].
  72. [72].
  73. [73].
  74. [74].
  75. [75].
  76. [76].
  77. [77].
  78. [78].
  79. [79].
  80. [80].
  81. [81].
  82. [82].
  83. [83].
  84. [84].
  85. [85].
  86. [86].
  87. [87].
  88. [88].
  89. [89].
  90. [90].
  91. [91].
  92. [92].
  93. [93].
  94. [94].
  95. [95].
  96. [96].
  97. [97].
  98. [98].
  99. [99].
  100. [100].
  101. [101].
  102. [102].
  103. [103].
  104. [104].
  105. [105].
  106. [106].
  107. [107].
  108. [108].
  109. [109].
  110. [110].
  111. [111].
  112. [112].
  113. [113].
  114. [114].
  115. [115].
  116. [116].
  117. [117].
  118. [118].
  119. [119].
  120. [120].
  121. [121].
  122. [122].
  123. [123].
  124. [124].
  125. [125].
  126. [126].
  127. [127].
  128. [128].
  129. [129].
  130. [130].
  131. [131].
  132. [132].
  133. [133].
  134. [134].
  135. [135].
  136. [136].
  137. [137].
  138. [138].
  139. [139].
  140. [140].
  141. [141].
  142. [142].
  143. [143].
  144. [144].
  145. [145].
  146. [146].
  147. [147].
  148. [148].
  149. [149].
  150. [150].
  151. [151].
  152. [152].
  153. [153].
  154. [154].
  155. [155].
  156. [156].
  157. [157].
  158. [158].
  159. [159].
  160. [160].
  161. [161].
  162. [162].
  163. [163].
  164. [164].
  165. [165].
  166. [166].
  167. [167].
  168. [168].
  169. [169].
  170. [170].
  171. [171].
  172. [172].
  173. [173].
  174. [174].
  175. [175].
  176. [176].
  177. [177].
  178. [178].
  179. [179].
  180. [180].
  181. [181].
  182. [182].
  183. [183].
  184. [184].
  185. [185].
  186. [186].
  187. [187].
  188. [188].
  189. [189].
  190. [190].
  191. [191].
  192. [192].
  193. [193].
  194. [194].
  195. [195].
  196. [196].
  197. [197].
  198. [198].
  199. [199].
  200. [200].
  201. [201].
  202. [202].
  203. [203].
  204. [204].
  205. [205].
  206. [206].
  207. [207].
  208. [208].
  209. [209].
  210. [210].
  211. [211].
  212. [212].
  213. [213].
  214. [214].
  215. [215].
  216. [216].
  217. [217].
  218. [218].
  219. [219].
  220. [220].
  221. [221].
  222. [222].
  223. [223].
  224. [224].
  225. [225].
  226. [226].
  227. [227].
  228. [228].
  229. [229].
  230. [230].
  231. [231].
  232. [232].
  233. [233].
  234. [234].
  235. [235].
  236. [236].
View Abstract