INTRODUCTION

Top Abstract Introduction Materials & Methods Results Discussion References

Microbes which are native to boiling acid hot springs include five genera assigned by 16S rRNA sequencing, G+C mole percent composition, metabolic characteristics, and protein sequence conservation studies to the order Sulfolobales (8). The order Sulfolobales is placed within the crenarchaeotal subdivision of the archaea (49). Sulfolobus is the largest genus in this order, comprising at least six species. These include S. acidocaldarius (5), S. solfataricus (13), S. shibatae (15, 52), S. metallicus (23), S. icelandicus (54), and S. hakonensis (44), all of which are obligate aerobes and either facultative chemoheterotrophs or strict lithoautotrophs. Most physiological and biochemical studies, however, focus on only three species, S. solfataricus, S. acidocaldarius, and S. shibatae.

S. solfataricus exhibits diverse modes of metabolism in batch culture at temperatures ranging between 70 and 90°C. It grows lithoautotrophically by oxidizing sulfur (5, 25, 50) and chemoheterotrophically on reduced-carbon compounds (13, 16, 19). Despite this metabolic flexibility, the utilization of reduced-carbon nutrients such as plant-derived starch or cellulose is poorly understood. Input of such carbon is typically rare in acidic hot springs and depends upon external factors such as fire or wind. Since polysaccharide hydrolysis and sugar carmelization are active processes in hot acid environments, successful competition for carbohydrates should necessitate mechanisms for rapid assimilation. Endogenous reserves of starch in plants and exogenous starch utilization in microbes often depend upon an -amylase which generates linear maltodextrins as well as an -glucosidase (maltase) which converts maltose and maltodextrins to glucose (26). In animals, however, -glucosidases are also critical for utilization of intracellular stores of glycogen. For example, glycogen storage disease (Pompe's disease) in humans is a direct consequence of -glucosidase deficiency (21). Many -glucosidase genes from eukaryotic and eubacterial organisms have been cloned and characterized (20). Although such enzymes also occur among the archaea (10, 40), none of their corresponding genes have yet been characterized, precluding an analysis of their intracellular functions or evolutionary origins.

Members of the genus Sulfolobus utilize starch as the sole carbon and energy source (16) and have both -amylase and -glucosidase activities (4). It has been reported, however, in surveys of Sulfolobus species that carbohydrate utilization profiles are distinct (16). For example, S. solfataricus contains a -glycosidase (17, 34) which appears to be largely absent in S. acidocaldarius (17). We recently reported the purification and characterization of the major soluble -glucosidase (maltase) and the secreted -amylase from S. solfataricus (19, 40). To further explore archaeal mechanisms for carbohydrate utilization and to examine their relationship with those of eukaryotes and eubacteria, the -glucosidase gene from S. solfataricus was cloned and characterized, and its distribution and associated activity in common Sulfolobus cultivars were examined.

	MATERIALS AND METHODS

Top Abstract Introduction Materials & Methods Results Discussion References

Strains and cultivation. The identity of S. solfataricus 98/2 (22, 40) was confirmed by DNA sequence analysis of a cloned PCR fragment spanning residues 99 to 626 of the 16S rRNA gene. The GenBank accession numbers for this sequence and that determined for S. solfataricus P2 (DSM 1617) are L36990 and L36991, respectively. Comparison of the resulting sequences with previously published citations for S. shibatae (GenBank accession no. M32504) and S. acidocaldarius (30, 35, 53) (GenBank accession no. X03235) confirmed the identity of strain 98/2 as S. solfataricus. Cells were cultured at 80°C in a minimal salts medium (1), modified as described previously (5), at pH 3.0 with various carbohydrates at 0.2% (wt/vol) as sole carbon and energy sources. All manipulations of Escherichia coli were as described previously (38).

Molecular biology methods. Restriction digestion and ligation of DNA were performed as described previously (3). Plasmid transformation was performed with DH5 cells as described previously (18). Isolation of plasmid DNA was performed by the alkali lysis procedure (2). DNA sequence analysis was as described previously (39), and DNA alignment and analysis were performed with the fragment assembly programs of the Wisconsin Genetics Computer Group software package, version 8.1.

Recovery of a malA PCR product. The -glucosidase was purified, denatured, and resolved by SDS-polyacrylamide gel electrophoresis (SDS-PAGE) as described previously (40). Protein was transferred electrophoretically to Problot membranes (Applied Biosystems), and the mature N-terminal protein sequence was determined by gas phase sequencing. Determination of internal sequence composition was accomplished by chemical cleavage of the purified protein with cyanogen bromide as described previously (32). The resulting peptides were fractionated by SDS-PAGE, transferred to membranes, and sequenced as for the mature protein. The resulting sequence information was used to design degenerate oligonucleotides, using inosine to reduce primer degeneracy. The oligonucleotides used were 5'-AA(A/G)AT(T/A/C)TA(T/C)GA(A/G)AA(T/C)AG(A/G)GG-3' as the upper primer and 5'-GC(G/A)TAIA(A/G)IA(A/G)(G/A)TA(C/T)TTICC-3' as the lower primer, where parentheses indicate additional nucleotides at positions of degeneracy. The resulting PCR product was blunted with Klenow enzyme and ligated to pACYC184 at the EcoRV site.

Northern blot analysis. S. solfataricus total RNA was extracted as described previously (9) from 100-mg amounts of wet cell paste derived from mid-exponential-phase maltose-grown cells. Nuclease activity was minimized by harvesting and processing cells within 2 min. Electrophoresis of RNA samples was as described previously (6), and the RNA was then electrophoretically transferred to Nytran extra-strength or Hybond N membranes and cross-linked by UV irradiation. Two types of probes were used for Northern blot analysis: double-stranded DNA (dsDNA) probes were generated as described for Southern blot analysis, and RNA probes (riboprobes) were generated by using the riboprobe buffer kit (Promega) and the manufacturer's protocol. The riboprobe template was the malA region comprising bp 141 to 2265 cloned into plasmid pT7T3 18U (Pharmacia). Northern hybridizations with riboprobes were performed at 55°C with 50% formamide. Northern hybridizations with DNA probes were performed as described for Southern hybridizations. Washed membranes were used to prepare autoradiograms with Kodak X-Omat film. Molecular weight standards were RNA transcripts (United States Biochemicals).

Primer extension analysis. Primer extension was performed essentially as described previously (46). The extension primer was 5'-ATGGTTCTCCTATAACTACTTTGTAAACGC-3' and was end labeled by using phage T4 polynucleotide kinase (New England Biolabs). The labeled oligonucleotide was purified by C₁₈ chromatography as described previously (41). Total RNA for primer extension analysis was prepared as described for Northern blot analysis. Forty micrograms of RNA was hybridized to 100,000 cpm of labeled primer for 90 min at 65°C. The resulting products were analyzed on 6% (wt/vol) Long Ranger (FMC Bioproducts) acrylamide gels containing 7 M urea, using DNA sequencing reaction products as size standards. Gels were then dried and used to prepare autoradiograms.

Heterologous expression. A 2.1-kb region spanning bp 141 to 2265, encompassing the malA coding region and 30 bp of flanking sequence, was blunted and cloned into the SmaI site of pUC19. This construct was then digested with KpnI and PstI, and the resulting 2.1-kb fragment was subcloned into the KpnI and PstI sites of plasmid pLITMUS 29 (New England Biolabs). The pLTIMUS 29 derivative was then digested with StuI and PvuII and religated to itself to remove a T7 promoter located 3' to the malA sequence. The resulting plasmid was introduced into E. coli DH5 for production of the recombinant protein by using stationary-phase cell suspensions, typically in 0.5 liter amounts. Cells were recovered by centrifugation and resuspended in 100 mM sodium acetate (pH 4.5) to a final density of 4 × 10¹⁰ cells/ml. Cells were broken by sonication as described previously (29). Assays for -glucosidase activity with p-nitrophenyl--D-glucopyranoside were performed as described previously (40). In enzymatic assays for hydrolysis of maltose and glycogen, release of glucose was monitored with a glucose oxidase assay kit (Sigma) as described previously (40). E. coli DH5 containing pLITMUS 29 without the malA insert was processed and assayed in an identical manner for use as a negative control in the assays. All assays were performed in duplicate, and the results were averaged. Proteins were resolved by SDS-PAGE under reducing conditions with unstained low- and high-molecular-weight markers (Bio-Rad). Prior to electrophoresis, samples were adjusted to 2% (wt/vol) SDS and 3 mM -mercaptoethanol and boiled for 10 min. Complete denaturation of the maltase required pretreatment with 6 M guanidine hydrochloride as described previously (40). SDS-polyacrylamide gels were stained with Coomassie blue R250 to visualize protein.

Phylogenetic analysis. Phylogenetic analyses (distance and parsimony) were performed with PHYLIP 3.57c (14). Maximum-likelihood analysis utilized the program PUZZLE (43). A multiple-sequence alignment was made by using the program CLUSTAL W (45). SEQBOOT was used to generate 100 bootstrapped data sets. Distance matrices were calculated with PROTDIST with the Dayhoff PAM matrix option. One hundred unrooted trees were inferred by neighbor-joining analysis of the distance matrix data by using NEIGHBOR. Bias introduced by the order of sequence addition was minimized by randomizing the input order. The most frequent branching order was determined with CONSENSE. The most parsimonious tree was determined with PROTPARS and CONSENSE for analysis of the SEQBOOT bootstrapped data sets.

Gel mobility shift analysis. Gel mobility shift assays were performed generally as described previously (7). The labeled probe was generated by the incorporation of [-³²P]dATP into the 5' overhanging end of a purified restriction fragment. Cell extracts used as a source of DNA binding proteins were prepared as described previously (24). Briefly, mid-exponential-phase cultures of S. solfataricus 98/2 were harvested by centrifugation, resuspended, and lysed by addition of Triton X-100. The resulting lysate was clarified by centrifugation, glycerol was added to a final concentration of 20% (vol/vol), and the extract was rapidly frozen and stored at 80°C. Labeled probe with or without added unlabeled competitor DNAs was incubated with cell extract at 70°C for 30 min, and the samples were then analyzed by nondenaturing PAGE. The resulting gel was dried and used to prepare autoradiograms.

Nucleotide sequence accession number. The sequence resulting from analysis of regions lying 3' to the malA coding region has been deposited in GenBank under accession no. AF042494.

	RESULTS

Top Abstract Introduction Materials & Methods Results Discussion References

Generation times and -glucosidase activities of Sulfolobus species on sole carbon sources. To better understand the metabolic relatedness between several of the better-characterized Sulfolobus species, the ability to utilize starch and its degradation products, maltose and glucose, as sole carbon and energy sources was evaluated in batch culture (Table 1). Both S. solfataricus 98/2 and S. shibatae utilized all three carbon sources with various efficiencies, as indicated by their respective generation times. S. acidocaldarius, however, exhibited a more limited carbon source utilization pattern. No growth was evident for S. acidocaldarius on maltose; however, contrary to previous reports (16, 27) growth was observed on glucose. -Glucosidase activities in crude cell extracts of S. solfataricus and S. shibatae were typically 10-fold greater than those seen with S. acidocaldarius during growth on all carbon sources (Table 1). Activities varied nearly twofold for S. solfataricus (comparing all carbon sources), while in S. shibatae, -glucosidase activity was six- to sevenfold higher during growth on maltose or starch than during growth on glucose. -Glucosidase in S. acidocaldarius was undetectable during growth on glucose and was 28-fold lower during growth on starch relative to that of either of the other two species.

View this table: [in this window] [in a new window]   TABLE 1.   Generation times and -glucosidase activities of Sulfolobus species during growth on different sole carbon and energy sources

Cloning and characterization of the S. solfataricus malA gene. To further investigate the apparent differences in polysaccharide utilization of maltose by these Sulfolobus species, the gene encoding the -glucosidase from S. solfataricus was cloned and characterized. Cloning of the gene was accomplished by using gene-specific PCR primers derived from the S. solfataricus -glucosidase protein sequence. Amino acid sequencing indicated that the mature N-terminal maltase sequence was MQTIKIYENLGVYLWIGEP. Since the purified S. solfataricus -glucosidase is generally resistant to proteolytic degradation (40), protein fragments for internal N-terminal sequence analysis were generated by chemical cleavage with cyanogen bromide. A fragment of 19 kDa was selected and yielded the N-terminal sequence VGKYLLYAPI. The resulting amino acid sequence information was used to design degenerate oligonucleotides which were then used to amplify a DNA fragment of 1.6 kb by PCR. A 731-bp HindIII-EcoRV fragment derived from the resulting 1.6-kb PCR product was used to generate a radiolabeled probe for Southern hybridization to verify the origin of the amplification product. This probe cross-hybridized with single DNA fragments of 1.2, 1.7, and 1.4 kb in HindIII, XbaI, and HincII genomic digests, respectively, of S. solfataricus DNA. The probe was used to screen a genomic S. solfataricus phage  library consisting of 672 individually propagated recombinant phages by Southern blot analysis. A single isolate was identified (-7F7), which contained a 15.1-kb insert of S. solfataricus DNA. Southern blot analysis of restriction digests of the -7F7 phage was performed with the PCR-derived probe. Cross-hybridizing restriction fragments which were identical in size to those observed previously with genomic DNA were observed and indicated that the -glucosidase-coding region is contained in a 4.3-kb BamHI fragment of the -7F7 insert. This 4.3-kb BamHI fragment was subcloned and sequenced. The sequence located immediately 5' to the malA coding region was subcloned from -7F7 as a 2.3-kb SacI-HindIII fragment. A 344-bp SacI-BamHI fragment from the extreme 5' end of this fragment was used as a Southern blot probe to identify the next upstream overlapping  clone from the genomic S. solfataricus 98/2 library. This isolate was named -1H4. A 4.1-kb HindIII-SacI fragment which cross-hybridized to the same SacI-BamHI 344-bp fragment was subcloned from -1H4. A 2-kb segment of this 4.1-kb HindIII-SacI fragment was sequenced to complete the analysis of putative genes lying 5' to malA. Analysis of regions lying 3' to the malA coding region was done with a 1.8-kb 3' overlapping HindIII-HindIII fragment derived from -7F7. The resulting sequence, comprising a nearly 7-kb DNA contig, has been deposited in GenBank (see Materials and Methods).

View larger version (64K): [in this window] [in a new window]   FIG. 1.   DNA sequence analysis of malA. Glycosyl hydrolase and ATP/GTP binding motifs are indicated by double and single underlining, respectively. Putative promoter and termination sequences are in boldface. The start of transcription is indicated as +1.

View larger version (29K): [in this window] [in a new window]   FIG. 2.   Northern blot analysis of the malA region. (A) Schematic diagram of the malA region. Numbers indicate G+C mole percent compositions. (B, C, and D) Northern blots of the malA region. (B) Lane 1, probe P1; lane 2, probe P2. (C) Lane 1, probe P2; lane 2, probe P3; lane 3, probe P4. (D) Riboprobes. Lane 1, probe P5; lane 2, probe P6.

Northern blot analysis of the malA region. Northern blot analysis was performed to evaluate the expression of malA and its surrounding regions during growth of S. solfataricus on maltose as the sole carbon and energy source. Probes P1 to P4 were dsDNA probes, and probes P5 and P6 were RNA probes (riboprobes). A probe derived from a 682-bp ClaI-HincII fragment from the 3' end of the malA coding region (Fig. 2A, probe P3) cross-hybridized to a single transcript of approximately 2.4 kb, indicating that malA is expressed during growth on maltose (Fig. 2C, lane 2). Probes derived from regions either 5' or 3' to the malA coding region were used to assess gene expression of immediately flanking sequences. These included an EcoRV-EcoRI fragment of 499 bp located 150 bp 5' to malA (Fig. 2A, probe P2) and an EcoRI-BamHI fragment of 231 bp located 192 bp 3' to malA (Fig. 2A, probe P4). No cross-hybridization was evident with either of these probes (Fig. 2C, lanes 1 and 3). An additional probe (Fig. 2A, probe P1) was used to examine expression of ORF1, located approximately 1 kb 5' of malA. This probe was derived from a SacI-BamHI fragment located in the center of ORF1. An approximately 2.4-kb transcript was detected with this probe (Fig. 2B, lane 1), while again no transcript was evident with probe P2 (Fig. 2B, lane 2). Riboprobes (Fig. 2A, probes P5 and P6) were used to determine the direction of transcription of malA. The 2.4-kb transcript was evident with the antisense riboprobe P6 (Fig. 2D, lane 2), while no transcript was detected with the sense riboprobe P5 (Fig. 2D, lane 1). These results indicate that the malA gene is transcribed from a site immediately adjacent to the gene and away from the large noncoding intergenic region.

Characterization of the malA regulatory region. There is a potential archaeal promoter sequence located 32 bp 5' to the start codon of the malA gene (Fig. 1). The putative promoter (TTTATA) closely matches the consensus promoter sequence for Sulfolobus (37). A box B motif (TGA) (37) is also evident 7 bp 5' to the malA start codon. Primer extension analysis indicated that malA transcription initiates on the guanine of the putative box B motif (Fig. 3). The mapped start site is 8 bp 5' to the malA start codon. Although there is a potential ribosome binding site spanning positions 3 to +3, which are complementary to the six 3'-terminal bases of the 16S rRNA of S. solfataricus (30, 35, 53), this sequence overlaps in part the site of malA transcription initiation. The utilization of this sequence for the initiation of translation is therefore unclear. The malA mRNA is only slightly larger than the coding region of the gene (Fig. 2). Since transcription initiates very close to the start of the coding region, termination of transcription of the gene must occur close to the end of malA. The near-consensus terminator sequence (TTTTTCA) (11) located immediately 3' to the stop codon of malA may play a role in this process.

View larger version (59K): [in this window] [in a new window]   FIG. 3.   Primer extension analysis of malA. Lanes 1 to 4, DNA sequencing reactions (T, C, G, and A, respectively); lane 5, primer extension product.

View larger version (44K): [in this window] [in a new window]   FIG. 4.   Gel shift analysis of the malA promoter region. Lane 1, probe (malA p-L); lane 2, probe and 5 µg of cell extract; lane 3, probe, extract, and unlabeled probe; lane 4, probe, extract, and unlabeled competitor (malA p-S); lane 5, probe, extract, and unlabeled pUC19 DNA. A and B indicate the positions of retarded complexes. At the bottom schematic diagrams of malA fragments used as probe and competitor are shown.

malA distribution among Sulfolobus species. Southern blot analysis with a malA gene probe was performed to analyze the distribution of this gene among the three commonly cultivated Sulfolobus species. Two isolates of S. solfataricus were included in the analysis, strain 98/2 from Yellowstone National Park and strain P2 (DSM 1617) from Italy. Genomic digests prepared with EcoRV (Fig. 5A) or HindIII (Fig. 5B) were then probed under stringent hybridization conditions with a 731-bp EcoRV-HindIII malA gene fragment encompassing nucleotides 714 to 1445 of the malA coding region (Fig. 1). Both strains of S. solfataricus exhibited strongly hybridizing bands of 2.9 kb following EcoRV digestion and 1.2 kb following HindIII digestion (Fig. 5, lanes 3 and 4), in agreement with the Southern blot results obtained previously with the 731-bp HindIII-EcoRV probe fragment derived from the initial malA PCR product. For S. shibatae, single weakly hybridizing bands of 0.65 kb following EcoRV digestion (Fig. 5A, lane 1) and 3.7 kb following HindIII digestion (Fig. 5B, lane 1) were also observed. No cross-hybridization was observed, however, between the S. solfataricus malA gene and S. acidocaldarius genomic DNA digests (Fig. 5, lanes 2).

View larger version (29K): [in this window] [in a new window]   FIG. 5.   Distribution of malA determined by Southern blot analysis. The probe comprised bp 714 to 1445 of malA. (A) EcoRV digests. (B) HindIII digests. Lanes: 1, S. shibatae DNA; 2, S. acidocaldarius DNA; 3, S. solfataricus 98/2 DNA; 4, S. solfataricus P2 DNA.

Phylogenetic analysis of -glucosidase sequences. Amino acid sequences of -glucosidases and the related sucrase isomaltases were retrieved from the Swiss-Prot and EMBL/GenBank/DDBJ databases. A multiple sequence alignment of 6 bacterial and 11 eukaryotic sequences in addition to the S. solfataricus sequence was made. The region of the S. solfataricus -glucosidase used for the alignment included 569 amino acid residues spanning positions 50 to 618. The S. solfataricus -glucosidase is the only representative of the archaea, since no other archaeal -glucosidases were found in the databases. A conserved stretch of amino acids located in the middle of the three fungal sequences was deleted to minimize sequence gaps in the alignment. The alignment of sequences then was analyzed by distance, parsimony, and maximum-likelihood methods. The E. coli malZ gene product was used as the outgroup. The sequences clustered into two groups typically of either eubacterial or eukaryotic affiliation by all three methods of analysis. Nearest-neighbor distance analysis and parsimony analysis indicate that the S. solfataricus -glucosidase is most closely related to mammalian enzyme homologs (Fig. 6). Maximum-likelihood analysis gave similar results (data not shown).

View larger version (34K): [in this window] [in a new window]   FIG. 6.   Phylogenetic analysis of -glucosidase and sucrase isomaltase sequences. A neighbor-joining distance tree is shown. Distances are indicated by the bar in the lower left corner, which represents 10 substitutions per 100 residues. Percent occurrence is given for nodes with values of >30%. Lower values are not shown or are indicated by a dash. Left-hand values are measures of distance; right-hand values are measures of parsimony. GenBank accession numbers are shown to the right of the names of the source organisms.

Recombinant S. solfataricus -glucosidase activity. To prove that malA encodes a hyperthermophilic -glucosidase, the malA gene was overexpressed in E. coli and the recombinant enzyme was purified and characterized. Purification of the recombinant enzyme to apparent homogeneity employed heat fractionation of clarified cell sonicates followed by anion-exchange fast protein liquid chromatography and gel filtration fast protein liquid chromatography as described previously (40). The recombinant S. solfataricus -glucosidase exhibited significant recalcitrance to denaturation as indicated by its behavior during denaturing SDS-PAGE. Despite boiling in the presence of 2% (wt/vol) SDS for 10 min, the -glucosidase failed to enter the separating gel and instead migrated in significant amounts (representing 45% of the total observed protein) in the stacking gel (data not shown). However, 95% of the natural enzyme treated in an identical manner was observed in the multimeric form (40), suggesting that the recombinant enzyme dissociates more readily under these conditions. Complete denaturation of the recombinant -glucosidase required additional treatment with 6 M guanidine hydrochloride, resulting in exclusive formation of the 80-kDa monomer (data not shown). The purified recombinant enzyme hydrolyzed p-nitrophenyl--D-glucopyranoside with a K_m of 2.16 mM and a V_max of 3.08 µmol of p-nitrophenol/min at 85°C. It exhibited a pH optimum for maltose hydrolysis of 4.5 (Fig. 7A). In contrast to its apparent greater tendency to dissociate during SDS-PAGE, the recombinant -glucosidase exhibited greater thermostability than the natural enzyme, with a half-life of 39 h at 85°C at a pH of 6.0. 

View larger version (14K): [in this window] [in a new window]   FIG. 7.   pH optima for maltose and glycogen hydrolysis. (A) Maltose hydrolysis; (B) glycogen hydrolysis. Buffers were as follows: pH 2.0 to 5.0, 100 mM sodium acetate (closed circles); pH 3.5 to 9.0, 100 mM sodium phosphate (open circles).

	DISCUSSION

Top Abstract Introduction Materials & Methods Results Discussion References

We report here the identification and characterization of the gene (malA) encoding the major soluble -glucosidase (maltase) of S. solfataricus. This is the first report of an -glucosidase sequence from the archaeal domain. The presence of an S. solfataricus malA homolog and corresponding -glucosidase activity in S. shibatae suggests that these Sulfolobus species have similar pathways for the utilization of maltose and maltooligomers. Lack of a malA homolog or significant -glucosidase activity in S. acidocaldarius may explain the inability of S. acidocaldarius to utilize maltose as a sole carbon and energy source. An -glucosidase thus may be essential for utilization of maltose among certain members of the genus Sulfolobus and represents a distinguishing physiological feature for Sulfolobus species identification. Such metabolic divergence lends further support to the suggestion that the Sulfolobus genus be revised (8).

Maltose utilization by these Sulfolobus species necessitates mechanisms for assimilation of maltose or maltodextrins, and specific transport systems have been identified recently in S. shibatae (51). However, the purified S. solfataricus enzyme also uses glycogen as a substrate. S. solfataricus accumulates glycogen as the major intracellular storage polysaccharide (28); thus, glycogen utilization may require the S. solfataricus -glucosidase. This is further supported by the observation that unlike maltose hydrolysis (40), glycogen hydrolysis by the S. solfataricus -glucosidase exhibits a more neutral pH optimum approximating that of the intracellular environment of this organism (31). Perhaps a dual role for the -glucosidase in the utilization of endogenous and exogenous polysaccharides can explain the apparent -glucosidase activity observed in both S. solfataricus and S. shibatae during growth on glucose. Constitutive expression of malA may be necessary to balance catabolic and anabolic metabolic needs. Since eubacterial -glucosidases lack glycogen-hydrolytic activity (33, 47, 48), the results presented here further distinguish archaeal -glucosidases from those of eubacteria.

The large intergenic sequence located 5' to malA is a distinguishing feature of the malA region. Northern blot analysis indicates that there is a lack of apparent transcripts encoded on either strand in the region covered by the dsDNA probe P2 (bp 651 to 152) produced during chemoheterotrophic growth on maltose. The entire intergenic region is also largely devoid of sequences encoding proteins; there is only one deduced sequence in excess of 39 residues (a protein of 88 residues) encoded in the region subjected to Northern analysis. However, this DNA sequence lacks a consensus promoter sequence and does not produce a detectable transcript during growth on maltose. Such noncoding regions are relatively rare in prokaryotic genomes, which are typically dense with genes. This is also true for the S. solfataricus P2 genome (42). It is therefore possible that the region 5' to malA provides some additional function to the S. solfataricus genome. Genomic measurements of G+C mole percent compositions lend additional support to this idea. The G+C content of the S. solfataricus genome is 38 mol% (13, 15). However, the approximately 1-kb intergenic region lying 5' to malA is distinctly lower in its G+C composition, with an average of 30 mol%, in contrast to the two flanking coding regions, which exhibit values of 38 mol%. It must therefore be assumed that intergenic regions such as this are rare in the S. solfataricus genome, as they would otherwise reduce global measurements of base composition.

Transcription of malA appears to utilize a consensus archaeal promoter motif. This sequence has been previously described as the box A motif TTTATA (11, 12). Gel shift analysis directly supports a role for the TATA box A region in the 5' flanking region for malA. Complex formation was dependent upon the 3'-terminal 42 bp, of which the most 5'-terminal region comprises the TATA box A sequence. Additionally, primer extension analysis indicates that the point of transcription initiation is at the conserved guanine located within the so-called box B region. As this residue lies only 8 bp 5' to the malA start codon, the mechanism employed for translation initiation of malA must operate within significant sequence constraints. Similar observations have been made for other archaeal genes, and as yet the mechanisms employed for translation initiation of these types of genes remain obscure.

Phylogenetic analysis of the -glucosidase and sucrase isomaltase sequences by three methods (distance, parsimony, and maximum likelihood) yielded similar trees with nearly identical branching topologies. These methods place the S. solfataricus enzyme with those of eukaryotes, specifically mammals, rather than with eubacterial orthologs. Recent studies on the relatedness of archaea to eubacteria and eukaryotes have suggested that archaeal central metabolic enzymes exhibit greatest relatedness to those of eubacteria. The results presented here, however, indicate that at least some archaeal metabolic pathways, such as those associated with carbohydrate metabolism, may have an evolutionary origin more in common with those of certain eukaryotes.