Properties of a genetically reconstructed Prevotella ruminicola endoglucanase

APPLIED AND ENVIRONMENTAL MICROBIOLOGY, Nov. 1992, p. 3593-3597

Vol. 58, No. 11

0099-2240/92/113593-05$02.00/0 Copyright © 1992, American Society for Microbiology

Properties of a Genetically Reconstructed Prevotella ruminicola Endoglucanase GIUSEPPE MAGLIONE,'t OSAMU MATSUSHITA,1 JAMES B. RUSSELL,2'3 AND DAVID B. WILSON"* Section of Biochemistry, Molecular and Cell Biology' and Section of Microbiology, Division of Biological Sciences, Cornell University, and Agricultural Research Service, U.S. Department of Agriculture, 3 Ithaca, New York 14853 Received 26 May 1992/Accepted 24 August 1992

A pUC19-derived plasmid was constructed that coded for a hybrid cellulase with the Thermomonosporafusca E2 cellulose-binding domain at its C terminus joined to the PrevoteUla ruminicola 40.5-kDa carboxymethyl cellulase (CMCase). The hybrid enzyme was purified and characterized enzymatically. It bound tightly to cellulose, and its specific activities on carboxymethyl cellulose, amorphous cellulose, and ball-milled cellulose were 1.5, 10, and 8 times that of the 40.5-kDa CMCase, respectively. Furthermore, the modified enzyme gave synergism with an exocellulase in the degradation of filter paper, while the 40.5-kDa CMCase did not.

MATERIALS AND METHODS Bacteria. The bacterial strains and plasmids used in this study are listed in Table 1. E. coli DH5a was used as the host strain for all plasmids except pC34, where E. coli JM109 was used. Strains containing plasmids were grown in LB medium containing 100 ,ug of ampicillin per ml. DNA manipulations. Restriction endonucleases and T4 DNA ligase were purchased from New England Biolabs. All recombinant DNA procedures were carried out as described by Maniatis et al. (14). Electroporation was used to introduce plasmids into E. coli strains (1). CMCase and E3 cellulase. The P. ruminicola 40.5-kDa CMCase and T. fusca cellulase E3 were purified as described previously (15, 24). Enzyme assays. French-pressed extracts prepared in 0.05 M KP1 (pH 6.5) or purified enzymes were added to a final volume of 0.4 ml containing 1% CMC (low viscosity; Sigma) in 0.05 M KPi (pH 6.5) and incubated at 37°C for 1 h. Reducing sugar was measured by the DNS reagent (17). Activities on ball-milled cellulose, swollen cellulose, or filter paper were assayed identically. Assays were carried out at several enzyme concentrations, and the results were plotted to determine the amount of enzyme required to give 4% hydrolysis of the substrate as recommended by Ghose (5). CMC overlay assay. Transformation plates were overlaid with 8 ml of 0.7% agarose-0.05% CMC (4HIF Hercules) in 0.01 M Tris-HCl (pH 6.5). The plates were incubated at 37°C for 5 h, stained for 1 h with 0.1% Congo red, and destained with 1 M NaCl for at least 15 min. Purification of the reconstructed enzyme (RCMCase). E. coli DH5ct(pGF7) was grown for 8 h at 37°C. The cells were harvested by centrifugation (10,000 x g for 10 min at 4°C). The cell pellet was resuspended in 0.05 M KP, (pH 6.5), and the cell sus ension was passed through a French press at 20,000 lb/in . The lysate was centrifuged at 10,000 x g for 10 min, and the supernatant was dialyzed against 1 mM KPi (pH 6.5). The dialyzed extract was chromatographed on a hydroxyapatite column (1.5 by 10 cm) as described by Matsushita et al. (15). The fractions showing the highest CMCase activity were pooled and concentrated by using a Centricon 30 filter (Amicon, Danvers, Mass.). Ammonium sulfate (final concentration, 1 M) was added, and the sample was loaded on a phenyl Sepharose CL-4B column (Sigma

Cellulose is the most abundant polymer in nature, but mammals do not produce enzymes which can degrade this material. Ruminant animals have developed the capacity for cellulose digestion through a symbiotic relationship with ruminal microorganisms. The animal provides a habitat for the microbial growth, and the microbes convert low-quality feeds into organic acids that the animal can utilize (11). The rumen microflora is very complex, but only a few species of ruminal bacteria are cellulolytic (2, 9). A variety of noncellulolytic ruminal bacteria can utilize cellodextrins (20), and some of these bacteria produce endoglucanases which can degrade carboxymethyl cellulose (CMC). Prevotella ruminicola (Bacteroides ruminicola) secretes two immunologically cross-reacting carboxymethyl cellulases (CMCases) into the culture supernatant (16). These enzymes also degrade xylan, but they have very low activity on native cellulose. The lack of cellulose digestion may be related to the inability of these enzymes to bind tightly to cellulose. Both CMCases are encoded in the same structural gene, and this gene has been cloned and sequenced (15, 16). The larger CMCase (88 kDa) is encoded in two overlapping reading frames that are one base out of frame, while the smaller (82-kDa) CMCase is encoded in the second reading frame (16). P. ruminicola produces only the 82- and 88-kDa CMCases, but Escherichia coli transformants carrying the entire P. ruminicola gene produce large amounts of a 40.5kDa CMCase in addition to an 84-kDa CMCase and the 88-kDa CMCase. The 40.5-kDa CMCase initiates at nucleotide 2185 (15, 16) so that it is encoded by the 3' end of the second reading frame. In this article, we describe the production, isolation, and activity of a reconstructed cellulase that is encoded in a gene that was constructed by joining the 40.5-kDa CMCase gene in frame to the region (3' end of the Thermomonospora fusca E2 gene) encoding the cellulosebinding domain (4, 13).

*

Corresponding author.

t Permanent address: Istituto Ricerca Adattamento Bovini e Bufali all'Ambiente del Mezzogiorno, Consiglio Nazionale delle Ricerche, 80147 Naples, Italy. 3593

3594

APPL. ENvIRON. MICROBIOL.

MAGLIONE ET AL. TABLE 1. Bacterial strains and vectors used in this study

Bacterial strain

Genotype and relevant features

or vector

Reference

Strains DH5a

F-(480dlacZAM15) A(lacZYA-argF) U169 deoR recAI endA hsdRl7 supE44 X thi-1

10

JM109

gyrA96 rel41 recAl supE44 endAl hsdR17gyrA96 reLA1 thiA(lac-proAB) F'[traD36 proAB+laclq

25

lacZAM15] Vectors pC39

pCBD2 pCBD3 pC34 pGF7

pUC18 containing a 960-bp EcoRI-HindIII fragment encoding the C-terminal end of the T. fusca E2 structural gene pC39 modified by insertion of a synthetic adaptor (5' ATAAGCTTATGC) pUC19 carrying the HindIII-PstI fragment encoding the cellulase E2-binding domain pUC18 carrying a 1.160-kb HindIII-HindIII fragment encoding the CMCase structural gene and its promoter from P. ruminicola B14 pUC19 carrying a fusion of the coding regions for the CMCase- and E2-binding domains

Chemical Co., St. Louis, Mo.). The column was washed with 1 volume of 0.6 M (NH4)2SO4-0.01 M NaCI-0.005 M KPi (pH 6), 3 volumes of 0.3 M (NH4)2SO4-0.005 M NaCl0.005 M KPi (pH 6.0), and finally with 1 volume of 0.005 M KPi (pH 6.5) which eluted the RCMCase in a single peak. Binding assay. Ten micrograms of purified RCMCase was added to different amounts of Avicel type pH-102 (FMC Corp., Philadelphia, Pa.) in 0.05 M KPi (pH 6.0) (final volume, 1 ml), and the mixtures were incubated and rotated for 1 h at 50°C. The samples were centrifuged, and the supernatants were assayed for CMCase activity to determine the amount of unbound enzyme. Western blots (immunoblots). Cells of the indicated strain were harvested in log phase by centrifugation, resuspended in 1/20 volume of 0.05 M KPi (pH 6.5), and disrupted in a French pressure cell at 20,000 lb/in2. The extracts were mixed with loading dye, boiled for 3 min, and electrophoresed on a sodium dodecyl sulfate (SDS)-12% polyacrylamide gel as described by Laemmli (12). Proteins were electrophoretically transferred to a sheet of nitrocellulose membrane (BA85; Schleicher & Schuell) (23). Both the CMCase and RCMCase were identified by using rabbit antibody raised against either the purified CMCase or cellulase E2. Goat anti-rabbit immunoglobulin G-alkaline phosphatase conjugate from Bio-Rad (Richmond, Calif.) was used as the second antibody as described by the supplier. RESULTS Plasmid pGF7 (Fig. 1) contained a hybrid cellulase gene which was a fusion of the 40.5-kDa CMCase gene, including its regulatory sequences, joined in frame to the region coding for the cellulose-binding domain from the T. fusca E2 gene. This construction took place in three steps (Fig. 1). First, a synthetic adaptor (5' ATAAGCTTATGC 3') containing a HindIlI site was inserted into the unique SacII site of plasmid pC39 to produce pCBD2. pC39 is pUC18 with a 0.96-kDa EcoRI-HindIII fragment from double-stranded M13 E2 plasmid DNA carrying the 3' end of the T. fiusca E2 gene (13). Second, a 414-bp HindIII-PstI fragment coding for the E2 cellulose-binding domain was excised from pCBD2 and ligated into pUC19 (cut at the same restriction sites) to produce pCBD3. Third, a 1,160-bp HindIII-HindIII fragment encoding the 40.5-kDa CMCase was excised from pC34 and ligated into HindIII-cut pCBD3 to produce pGF7. Recombinants carrying the fragment in the correct orientation to produce a fusion protein were identified by the CMC overlay

This study This study This study 16 This study

assay, and their structure was confirmed by restriction mapping. The purification procedure developed for the RCMCase (see Materials and Methods) gave a final yield of 25%, and the purified enzyme had a 10-fold-higher specific activity than the crude extract (61 versus 6.2 pumol/min/mg of protein). Purified RCMCase gave a single band on SDS-polyacrylamide gel electrophoresis (Fig. 2), and its molecular mass was 54 kDa versus the 40.5 kDa found for the CMCase. RCMCase reacted with antibodies which had been prepared against either the 40.5-kDa CMCase or cellulase E2 (Fig. 3). RCMCase bound to Avicel as expected, while the CMCase did not (Fig. 4). Thus, the RCMCase was not present in lanes B to E (Fig. 4) since it bound to the Avicel, while the intensity of the CMCase bands in lanes I to 0 was equal to the sample with no Avicel in lane H. The enzymatic activity of purified RCMCase was determined on CMC, acid-swollen cellulose, and ball-milled cellulose (Table 2). RCMCase had a slightly higher specific activity on CMC than the CMCase and was 10 times more active on acid-swollen cellulose. The CMCase was not able to hydrolyze 4% of the ball-milled cellulose, but RCMCase was able to hydrolyze this substrate at a rate of 1.2 p,mol/ min/mg of protein. T. fusca produces an exocellulase, E3, which degrades filter paper (24), and this enzyme showed synergism with RCMCase but not with CMCase (Table 3). Neither the CMCase nor RCMCase acting alone was able to give 4% digestion of filter paper, but a mixture of RCMCase and E3 was twice as active on filter paper as E3 was alone. A mixture of the CMCase and E3 had slightly less activity than E3 alone. DISCUSSION At least three species of ruminal bacteria are able to digest cellulose at a rapid rate (Fibrobacter succinogenes, Ruminococcus albus, and Ruminococcus flavefaciens), but there has been little progress in the purification of rumen bacterial cellulases that have a high activity on crystalline cellulose. Yu and Hungate (26) isolated a crude crystalline cellulase from R. albus, and the activity was found in a very-highmolecular-weight fraction. When attempts were made to purify the enzyme, virtually all of its activity was lost. F. succinogenes degrades crystalline cellulose faster than ruminococci (2), but its cellulases are very tightly cell associated and cell extracts have little if any activity (7, 8).

VOL. 58, 1992

~ ~ ~mal

PROPERTIES OF A P. RUMINICOLA ENDOGLUCANASE

3595

,indlll

/ /bla

pCBD2

STEP I1* Digested SacliI inserted 12mer adaptor

l

E2

3.65 Kb

c

Digested Hindl l-Pstl

l dndlll

o~~~~~~~RI Digested Hindi I I-PstI Isolated 414 bp fragment

STEP 2

Digested Hindl l I

Digested Hindl 1 isolated 1 160 bp fragment

Ncol Hindill

|

FIG. 1. Construction of plasmid pGF7. Symbols:

E3, T. fiusca E2 DNA; _, P. ruminicola CMCase DNA; O, pUC19 DNA.

3596

~ 43.0

APPL. ENvIRON. MICROBIOL.

MAGLIONE ET AL.

97.4

! SS l --~68.0 _ S , |w _ w 43.0

18.4

FIG. 4.

SDS-polyacrylamide gel electrophoresis of RCMCase centrifugation of

and CMCase. Lanes: A to E, supernatant after

FIG. 2. SDS-polyacrylamide gel electrophoresis of RCMCase. Lanes: A, purified RCMCase; B, size markers (numbers on the right indicate sizes in kilodaltons).

purified RCMCase incubated with 5, 10, 25, 50, and 100 mg of Avicel, respectively; F, purified RCMCase; G, molecular weigt markers; H, purified CMCase; I to 0, supernatant after centrifugation of A to

Gardner et al. (3) purified an exocellulase from R flavefaciens FD-1 that had about the same activity (0.019 p,mol/ min/mg) on filter paper as our reconstructed enzyme. A variety of noncellulolytic ruminal bacteria produce a ,B-glucanase that can degrade cellodextrins and CMC, and some of these bacteria are acid tolerant (21). On the basis of these criteria, it appeared that P. ruminicola might be used

:s200.0

~~~~~9.4 68.0

25.7 18.4

154

purified

CMCase incubated with Avicel

as

described for lanes

E, respectively.

to create a ruminal bacterium that digests cellulose at low pH. We originally planned to move a cellulase gene from T. fusca into P. ruminicola B14 (22). While such a plan may still be fruitful, there was no guarantee that this foreign gene would be expressed. Since P. ruminicola was already able to produce a CMCase, the following question then arose: can the CMCase be modified to increase its activity on native cellulose? Cellulomonas fimi produces crystalline cellulases which bind tightly to cellulose, and these enzymes can be cleaved by trypsin to produce two polypeptides. One fragment retains activity on CMC but loses its ability to bind to cellulose (6). Similar results were found for T. fusca cellulase E2 (4). These and other studies along with an analysis of the amino acid sequences of cellulases show that many cellulases have separate catalytic and binding domains and that both sites are needed for optimum activity against crystalline cellulose (18). Recent work by Ong et al. (18) showed that the addition of a cellulose-binding domain to E. coli alkaline phosphatase and to a ,B-glucosidase immobilized the enzymes on cellulose with retention of activity. On the basis of these earlier results, it appeared that the P. ruminicola CMCase, which did not bind cellulose, might be improved by the addition of a cellulose-binding domain. When the 40.5-kDa CMCase gene was fused to DNA encoding the E2 cellulose-binding site, the hybrid cellulase had 10-fold more activity on insoluble cellulose. Therefore, cellulose binding was indeed beneficial, but further work is needed to assess the potential impact on P. ruminicola. TABLE 2. Activity of the native and modified enzymes Molecular

FIG. 3. Western extract of E.

DH5a

coli

blot of RCMCase

and

CMCase.

Lanes:

JM109 harboring pC34; B and D, extracts of E.

harboring pGF7; C, kilodaltons).

A,

coli

(numbers on the right containing samples A and B was treated with rabbit antisera prepared against homogeneous 40-kDa P. ruminicola CMCase; the filter containing sample D was treated with antisera prepared against homogeneous T. E2. indicate sizes in

size

Enzyme

Sp act CMC (5%

(kDa)

digestion)

Ball-milled cellulose (4% digestion)

Acid-swollen cellulose (4% digestion)

54 40.5

61 46

1.2 0.17a

17 1.6

markers

The filter

fisca

RCMCase CMCase

(jmol/min/mg of protein) on:

mass

a Four percent digestion not achieved.

VOL. 58, 1992

PROPERTIES OF A P. RUMINICOLA ENDOGLUCANASE

TABLE 3. Synergism of the native and modified enzymes

Sp act (Lmol/min/mg of protein) on filter paper (4% digestion)

Enzyme or enzyme mixture

RCMCase ................................... CMCase ................................... E3 .................................. RCMCase + E3 ................................... CMCase + E3 .................................. a

0.026a 0.013a 0.040 0.097 0.034

7.

8. 9.

Four percent digestion not achieved.

10.

However, our results appear to differ from the results of Poole et al. (19), who constructed fusion proteins carrying a Pseudomonas fluorescens xylanase-binding domain joined to the N terminus of Clostridium thennocellum endoglucanase and to the N terminus of R albus endoglucanase A. Both fusion proteins bound to cellulose, but their enzymatic activities were unchanged. There are two important differences between our work and that described by Poole et al. One is that we used the cellulose-binding domain from a cellulase while Poole et al. used one from a xylanase. The other is that our enzyme had some activity on both swollen and ball-milled celluloses while Poole et al. reported no activity on cellulosic substrates for their original enzymes. Since it is known that the presence of a cellulose-binding domain does not significantly affect the activity of cellulases on soluble substrates like CMC and barley 1-glucan, it is not surprising that Poole et al. did not see an increase in the activity of their fusion cellulases on these substrates.

11.

12. 13. 14. 15. 16. 17. 18.

ACKNOWLEDGMENTS This work was partly supported by grant 85-CRCR-1-1880 from the U.S. Department of Agriculture and by the United Dairy Forage Research Center, Madison, Wis. G.M. was supported by a fellowship from the Istituto Ricerca Adattamento Bovini e Bufali all'Ambiente del Mezzogiorno, Consiglio Nazionale delle Ricerche, Naples, Italy.

19.

20. 1.

2. 3. 4.

5.

6.

REFERENCES Ausubel, F. M., R. Brent, R. E. Kingston, D. D. Moore, J. G. Seidman, J. A. Smith, and K. Struhl (ed.). 1989. High-efficiency transformation by electroporation, p. 1.8.4. In Current protocols in molecular biology, vol. 1. John Wiley & Sons, Inc., New York. Bryant, M. P. 1973. Nutritional requirements of the predominant rumen cellulolytic bacteria. Fed. Proc. 32:1809-1813. Gardner, R. M., K. C. Doerner, and B. A. White. 1987. Purification and characterization of an exo-3-1,4-glucanase from Ruminococcus flavefaciens FD-1. J. Bacteriol. 169:4581-4588. Ghanges, G. S., and D. B. Wilson. 1988. Cloning of the Thermomonospora fusca endoglucanase E2 gene in Streptomyces lividans: affinity purification and functional domains of the cloned gene product. Appl. Environ. Microbiol. 54:2521-2526. Ghose, J. K. 1987. Measurement of cellulase activities. Pure Appl. Chem. 59:257-268. Gilkes, N. R., R. A. J. Warren, R. C. Miller, Jr., and D. G. Kilburn. 1988. Precise excision of the cellulose binding domains

3597

from two Cellulomonas fimi cellulases by a homologous protease and the effect on catalysis. J. Biol. Chem. 263:10401-10407. Groleau, D., and C. W. Forsberg. 1983. Partial characterization of the extracellular carboxymethylcellulase activity produced by the rumen bacterium Bacteroides succinogenes. Can. J. Microbiol. 29:504-517. Groleau, D., and C. W. Forsberg. 1981. Celluloytic activity of the rumen bacterium Bacteroides succinogenes. Can. J. Microbiol. 27:517-530. Halliwell, G., and M. P. Bryant. 1963. The celluloytic activity of pure strains of bacteria from the rumen of cattle. J. Gen. Microbiol. 32:441-448. Hanahan, D. 1983. Studies on transformation of Escherichia coli with plasmids. J. Mol. Biol. 166:557-580. Hungate, R. E. 1966. The rumen and its microbes. Academic Press, Inc., New York. Laemmli, U. K. 1970. Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature (London) 227:680-685. Lao, G., G. S. Ghanges, E. D. Jung, and D. B. Wilson. 1991. DNA sequences of three 3-1,4-endoglucanase genes from Thermomonospora fusca. J. Bacteriol. 173:3397-3407. Maniatis, T., E. F. Fritsch, and J. Sambrook. 1982. Molecular cloning: a laboratory manual. Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y. Matsushita, O., J. B. Russell, and D. B. Wilson. 1990. Cloning and sequencing of a Bacteroides ruminicola B14 endoglucanase gene. J. Bacteriol. 172:3620-3630. Matsushita, O., J. B. Russell, and D. B. Wilson. 1991. A Bacteroides ruminicola 1,4-p-D-endoglucanase is encoded in two reading frames. J. Bacteriol. 173:6919-6926. Miller, G. L., R. Blum, W. E. Glennon, and A. L. Burton. 1960. Measurement of carboxymethylcellulase activity. Anal. Biochem. 1:127-132. Ong, E., J. M. Greenwood, N. R. Gilkes, G. Kilburn, R. C. Miller, and R. A. J. Warren. 1989. The cellulose-binding domains of cellulases: tool for biotechnology. Trends Biotechnol. 7:239-243. Poole, D. M., A. J. Durrant, G. P. Hazelwood, and H. S. Gilbert. 1991. Characterization of hybrid proteins consisting of the catalytic domains of Clostridium and Ruminococcus endoglucanases, fused to Pseudomonas non-catalytic cellulose binding domains. Biochem. J. 279:787-792. Russell, J. B. 1985. Fermentation of cellodextrins by cellulolytic and noncellulolytic rumen bacteria. Appl. Environ. Microbiol.

49:572-576. 21. Russell, J. B., and D. B. Dombroski. 1980. Effect of pH on the efficiency of growth by pure cultures of rumen bacteria in continuous culture. Appl. Environ. Microbiol. 39:604-610. 22. Russell, J. B., and D. B. Wilson. 1988. Potential opportunities and problems for genetically altered rumen microorganisms. J. Nutr. 118:271-279. 23. Towbin, H., T. Staehelin, and J. Gordon. 1979. Electrophoretic transfer of proteins from polyacrylamide gels to nitrocellulose sheets: procedure and some applications. Proc. Natl. Acad. Sci. USA 78:4350-4354. 24. Wilson, D. B. 1988. Cellulases of Thermomonospora fuisca. Methods Enzymol. 160:314-323. 25. Yanisch-Perron, C., J. Vieira, and J. Messing. 1985. Improved M13 phage cloning vectors and host strains: nucleotide sequences of the M13mpl8 and pUC19 vectors. Gene 33:103-119. 26. Yu, I., and R. E. Hungate. 1979. The extracellular cellulases of Ruminococcus albus. Ann. Rech. Vet. 10:251-254.