Anabaena sp. Strain PCC 7120 Gene devH Is Required for Synthesis of the Heterocyst Glycolipid Layer

Abstract

In response to deprivation for fixed nitrogen, the filamentous cyanobacterium Anabaena sp. strain PCC 7120 provides a microoxic intracellular environment for nitrogen fixation through the differentiation of semiregularly spaced vegetative cells into specialized cells called heterocysts. The devH gene is induced during heterocyst development and encodes a product with characteristics of a trans-acting regulatory protein. A devH mutant forms morphologically distinguishable heterocysts but is Fox⁻, incapable of nitrogen fixation in the presence of oxygen. We demonstrate that rearrangements of nitrogen fixation genes take place normally in the devH mutant and that it is Fix⁺, i.e., has nitrogenase activity under anoxic conditions. The Fox⁻ phenotype was shown by ultrastructural studies to be associated with the absence of the glycolipid layer of the heterocyst envelope. The expression of glycolipid biosynthetic genes in the mutant is greatly reduced, and heterocyst glycolipids are undetectable.

Anabaena sp. strain PCC 7120 is a filamentous cyanobacterium in which, in response to deprivation for fixed nitrogen, vegetative cells at semiregular intervals differentiate into nitrogen-fixing cells called heterocysts. Nitrogenase, the enzyme that catalyzes the reduction of atmospheric dinitrogen (N₂) to ammonia, is rapidly and irreversibly inactivated in the presence of oxygen. The heterocyst provides a microoxic intracellular environment for nitrogenase through the development of a thick envelope, an alteration of photosynthetic activity, and an increased rate of respiration (26).

The heterocyst envelope is composed of an inner laminated glycolipid layer that greatly reduces penetration of oxygen and an outer polysaccharide layer that protects the glycolipid layer (17, 22). Studies of cyanobacterial mutants with heterocyst defects have led to the identification of a number of genes involved in the development of the heterocyst envelope. Biosynthesis of the polysaccharide layer is regulated by the products of devR (4) and hepK (28), which encode part or all of a two-component regulatory system (27). The hepA, hepB, and hepC genes are involved in the synthesis of the polysaccharide layer (14, 25, 28). Synthesis of the glycolipid layer involves a hglE gene (5) and the cluster of genes hglD, hglC, and hglB (hetM) (1, 3). The hglK (2) and devBCA gene cluster (8) are implicated in glycolipid transport and/or assembly.

The devH gene was identified in a screen for sequences up-regulated during heterocyst development (6). The steady-state level of devH transcripts increases fivefold upon nitrogen starvation of Anabaena sp. strain PCC 7120. In the absence of fixed nitrogen, a devH mutant strain (A57) forms heterocysts but is incapable of fixing nitrogen in the presence of oxygen (the Fox⁻ phenotype) (12). The DevH protein is most closely related to the cyanobacterial transcriptional regulator NtcA, with a high degree of identity in the helix-turn-helix motif presumed to be involved in DNA binding (12). The similarity of DevH and NtcA and the phenotype of the devH mutant suggest that DevH plays a role as a trans-acting regulatory protein with a role in heterocyst function. Here we provide initial characterization of DevH protein expression and show that the Fox⁻ phenotype of the devH mutant is associated with defective heterocyst envelopes that lack the glycolipid layer.

MATERIALS AND METHODS

Strains and culture conditions.

Strains used in this study are listed in Table 1. Liquid cultures of cyanobacteria for protein or nucleic acid isolations were grown in modified Kratz and Myers medium supplemented with 5 mM NH₄NO₃ (K&M, N⁺) as described previously (12). The devH mutant (A57) (12) was grown in 40 μg of neomycin sulfate/ml. For the developmental time course, cells were grown essentially as described previously (12) in medium lacking a source of combined nitrogen (K&M, N⁻). Cells collected at various times after the removal of fixed nitrogen (nitrogen stepdown) were used for total DNA or RNA isolation or for Western analysis. For nitrogenase assays, liquid cultures were grown in BG-11 medium as described previously (12) (see “Nitrogenase activity assays,” below). In preparation for electron microscopy, Anabaena sp. strains PCC 7120 and A57 were grown for 5 days in AA/8 + N (15) plus, for A57, 80 μg of neomycin sulfate/ml at 30°C in the light at ca. 140 μE m⁻² s⁻¹ (Li-Cor quantum radiometer/photometer, model LI-185A; Lincoln, Nebr.). Cells were then sedimented, washed three times with AA/8 (15), and plated on a Nuclepore Rec-85 membrane atop AA agar for 3 days.

TABLE 1.

Bacterial strains, plasmids, and oligonucleotides used in this study

Strain, plasmid, or oligonucleotide	Characteristicsa or sequence (5′ to 3′)a	Source
Anabaena sp. strains
PCC 7120	Wild type	R. Haselkorn
A57	PCC 7120 devH::pBN1-239B, Nmr	12
E. coli strains
DH5α	F′recA, p80dlacZdM15	Bethesda Research Laboratories
BL21(DE3)pLysS	F−omp T hsdSB(rB− mB−) gal dcm (DE3)/pLysS	Novagen
Plasmids
pBluescript KS(+); pBS	Cloning vector, Apr	Stratagene
pET30a	Expression vector, Kmr	Novagen
pET23930a	700-bp devH open reading frame in pET30a, Kmr	This study
pAn154.3	1.9-kb HindIII fragment containing nifH gene in pBR322, Apr Tetr	19
pAn256	3.0-kb HindIII fragment containing nifD element recombination site in pBR322, Apr Tetr	19
pRNAseP	602-bp EcoRI genomic PCR fragment of mpB in pBS, Apr	12
Oligonucleotides
239-W	TGGAATTCATGCAATCTCCATCCTC	This study
239-Xe	ATGAATTCTTAGCTAGCTCTGTTGA	This study
HgIB-F1	ATTCGGCACAGGCAATGG	This study
HgIB-R2	TTGTTCATCAGACCAGCG	This study
HgIC-F2	GGTATGGCATTGACGAGC	This study
HgIC-R2	CTCTAATAGGCTTGTCCG	This study
HgID-F1	GGCTTCCTACTCAACTGG	This study
HgID-R1	TCGTCAAACTCACCATCC	This study
HgIE-F1	TTAGCGGAGTTAGAAAGC	This study
HgIE-R1	TCTTGTAACGTCCATCGC	This study
HgIE-F5	GCCCTCAATCAAGTTTTG	This study
HgIE-R7	GAGTATTAAAGGTCACCC	This study
Fdx-F3	CTTACACTATCACCAGCC	This study
Fdx-R4	CAGCAACTTGTTCTCTCG	This study

^aAp^r, ampicillin resistance, Bm^r, bleomycin resistance; Km^r, kanamycin resistance; Sm^r, streptomycin resistance; Nm^r, neomycin resistance; Tet^r, tetracyline resistance.

Escherichia coli strains were grown as described previously (12). E. coli strain DH5α (Bethesda Research Laboratories, Gaithersburg, Md.) was the host for most plasmids, and E. coli strain BL21(DE3)pLysS was the host for protein expression.

Isolation and analysis of nucleic acids.

DNA was isolated as described previously (12). Total RNA was isolated with the RNAwiz reagent (Ambion) following the manufacturer's instructions. Southern, Northern, and slot blot preparations were performed as described previously (6, 12).

Probe fragments were obtained by PCR amplification of Anabaena sp. strain PCC 7120 genomic DNA or by isolation of plasmid inserts (Table 1). The probe fragments used were the following: (i) for nifD, a 3.0-kb HindIII fragment from pAn256; (ii) for nifH, a 1.9-kb HindIII fragment from pAn154.3; (iii) for rnpB, a 602-bp EcoRI fragment from pRNAseP; (iv) for hglB, a 1,225-bp fragment amplified with primers HglB-F1 and HglB-R2; (v) for hglC, a 1,327-bp fragment amplified with primers HglC-F2 and HglC-R2; (vi) for hglD, a 761-bp fragment amplified with primers HglD-F1 and HglD-R1; (vii) for hglE1 (alr5351), a 1,104-bp fragment amplified with primers HglE-F1 and HglE-R1; (viii) for hglE2 (all1646), a 1,360-bp fragment amplified with primers HglE-F5 and HglE-R7. The hglE probes were shown to be specific to each of the two copies of the hglE gene (all1646 or alr5351) by hybridization to Southern blots of Anabaena sp. strain PCC 7120 genomic DNA.

Alkaline phosphatase-labeled probes (all except nifH) were prepared using the AlkPhos Direct kit (Amersham Biosciences), and hybridizations were conducted according to the manufacturer's recommendations. Hybridization signals were detected by using ECF substrate (Amersham Biosciences), followed by scanning on a Storm instrument (Molecular Dynamics). Bands were quantitated using ImageQuant software (Molecular Dynamics). A ³²P-labeled, randomly primed nifH probe was prepared using a Rediprime kit (Amersham Biosciences) according to the manufacturer's instructions and hybridized as described previously (12). Rearrangement of the fdxN gene was assayed by amplification of DNA isolated from differentiated and undifferentiated filaments by using primers Fdx-F3 and Fdx-R4.

Expression and purification of His-tagged DevH recombinant protein and production of antibodies.

The devH gene was PCR amplified from Anabaena sp. strain PCC 7120 genomic DNA using oligonucleotides 239-Xe and 239-W and cloned into the EcoRI site of vector pET30A (Novagen, Madison, Wis.) to produce plasmid pET23930a. This plasmid encodes recombinant DevH (rDevH) bearing a hexahistidine (His₆) tag.

An overnight culture of E. coli BL21(DE3)pLysS containing pET23930a was inoculated into 10 ml of 2YT medium supplemented with chloramphenicol (50 μg/ml) and kanamycin (50 μg/ml) and incubated at 37°C with shaking (200 rpm). When the optical density at 600 nm (OD₆₀₀) of the culture reached 0.6 to 1.0, isopropylthiogalactopyranoside was added to a final concentration of 1 mM. Cells were incubated further for 3 h and harvested by centrifugation at 5,000 × g for 5 min. The pellet was resuspended in 600 μl of 10 mM Tris-HCl (pH 7.5), and the cells were lysed by cavitation with a Fisher Scientific Co. sonic dismembrator, model 300, for 4 min at 4°C. Soluble fractions were separated from cell debris and inclusion bodies by centrifugation at 12,000 × g for 10 min. The protein concentration in the supernatant solution was determined using the Bio-Rad protein assay, with bovine serum albumin as the standard. Soluble fractions were analyzed by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE).

For purification of rDevH, frozen pellets of 100 ml of induced E. coli BL21(DE3)pLysS cells carrying pET23930a were thawed at 4°C in binding buffer (5 mM imidazole, 0.5 M NaCl, 20 mM Tris-HCl; pH 7.9). Cells were disrupted by sonication for 4 min at 4°C with the sonic dismembrator, and cell debris was separated by centrifugation at 23,000 × g for 20 min. Further purification of rDevH from the soluble fraction was conducted with a His•Bind resin (His•Bind kit; Novagen), according to the manufacturer's instructions. Eluted fractions were analyzed by SDS-PAGE. The band corresponding to rDevH (35 kDa) was excised from an SDS-12% polyacrylamide gel and used for polyclonal antibody production in rabbits (Cocalico Biologicals, Inc., Reamstown, Pa.).

Western analysis.

Cells collected at 0, 24, and 48 h after nitrogen deprivation were resuspended in 10 mM Tris-HCl, 1 mM EDTA, 1 mM phenylmethylsulfonyl fluoride (PMSF; pH 8.0) and stored at −80°C until further analysis. Frozen samples were thawed, and cells were disrupted by cavitation for 7 min at 4°C, followed by vortexing with glass beads for 5 min. Cell debris was eliminated by centrifugation at 4°C, 10,000 × g, for 20 min. Total protein (40 to 50 μg) was fractionated by SDS-10% PAGE and electroblotted onto a nitrocellulose membrane (NitroBind; Osmonics) with transfer buffer (25 mM Tris-HCl, 192 mM glycine, 0.1% SDS, and 20% methanol) (21). Membranes were pretreated in TBST (20 mM Tris, 137 mM NaCl, 0.1% Tween; pH 8.0) containing 5% nonfat dry milk for 1 h at room temperature. Incubations with the DevH immune serum or a dinitrogenase reductase immune serum (provided by P. Ludden) were at a 1:5,000 dilution in TBST. A secondary hybridization with donkey anti-rabbit antibody linked to horseradish peroxidase (Amersham Biosciences) was at a dilution of 1:5,000. Detection was performed using the ECL Plus system (Amersham Biosciences) followed by scanning on a Storm instrument (Molecular Dynamics). Bands were quantitated using ImageQuant software (Molecular Dynamics). Triplicate analyses were conducted for wild-type and A57 cells.

Nitrogenase activity assays.

For comparison of nitrogenase activity under oxic conditions, wild-type and A57 cultures were grown in BG-11 N⁺ to an OD₆₀₀ of 0.4 and transferred to BG-11 N⁻ as described previously (12). The cultures were constantly shaken and bubbled with air. After nitrogen stepdown, 3-ml samples were periodically transferred to 25-ml tubes and incubated for 30 min at 30°C in the presence of 10% (vol/vol) acetylene in air. Ethylene in the gas phase was measured using gas chromatography as described previously (12). Chlorophyll a was analyzed by standard methods (16). Nitrogenase activity was expressed as the nanomoles of ethylene produced per microgram of chlorophyll a per hour.

The nitrogenase activity of strain A57 under oxic or anoxic conditions was compared following the methodology of Ernst et al. (7) with slight modifications. Cultures were grown in BG-11 N⁺ to an OD₆₀₀ of 0.4 and then incubated for 2 days in BG-11 N⁻ in air to ensure sufficient glycogen accumulation for nitrogenase expression under anoxic conditions (20). Following this step, cultures were concentrated approximately sixfold. For determination of nitrogenase activity under oxic conditions, 3-ml aliquots were transferred to 25-ml tubes fitted with serum stoppers. To the remainder of the concentrated cell suspension, 3-(3,4-dichlorophenyl)-1,1-dimethylurea (DCMU) was added to a final concentration of 20 μM. Samples (3 ml) of the DCMU-treated cell suspension were transferred to 25-ml tubes fitted with serum stoppers, and the gas phase was replaced by argon (anoxic conditions). After 0 and 22 h at 30°C, acetylene was added to a final concentration of 10% (vol/vol) to duplicate samples of the cell suspensions under oxic (air) or anoxic (argon) conditions. Ethylene chromatographic peaks were measured 4 h after acetylene addition and continued incubation at 30°C.

Electron microscopy.

Cultures were prepared for electron microscopy according to the method of Black et al. (2).

Glycolipid analysis.

Wild-type and A57 strains that had been deprived of fixed nitrogen for 0 or 48 h were extracted with chloroform-methanol (2:1, vol/vol). The organic phases were concentrated, and glycolipids were separated on a silica gel 60 plate, using as a mobile phase chloroform, methanol, acetic acid, and water in a ratio of 85:15:10:3.7 (by volume) (18). Plates were developed by charring with 25% sulfuric acid at 200°C for 5 min (24).

RESULTS AND DISCUSSION

DevH levels increase after nitrogen starvation.

To facilitate studies of DevH, the protein was overexpressed in E. coli as rDevH. A protein of the expected size was expressed (35 kDa) and used to raise polyclonal antibodies.

In cell extracts of Anabaena sp. strain PCC 7120 grown in air in the presence or absence of fixed nitrogen, rDevH antiserum detected a protein of 29 kDa (Fig. 1B). This molecular mass corresponds well with that predicted from the devH coding region. In extracts of devH mutant strain A57, a protein of about 33 kDa was detected (Fig. 1B). The 33-kDa protein is a DevH fusion protein resulting from the insertional inactivation used to generate the devH mutant (12), in which the last nine codons of the devH open reading frame are replaced by 43 codons derived from vector sequences. In wild-type DevH, the helix-turn-helix motif is 24 amino acids from the carboxy terminus of the protein; in the mutant DevH protein from A57 there are 58 amino acids between the helix-turn-helix domain and the carboxy terminus. Although the A57 fusion protein is primarily wild type in sequence and maintains the helix-turn-helix domain, the protein is apparently nonfunctional, perhaps because the extension interferes with DNA binding.

FIG. 1.

Western analysis of DevH and dinitrogenase reductase in Anabaena sp. strain PCC 7120 (WT) and its devH mutant, A57. Protein samples (50 μg) from the wild-type strain and strain A57 were collected after 0, 24, or 48 h of nitrogen deprivation under oxic conditions, fractionated by SDS-10% PAGE, and stained with Coomassie brilliant blue R250 (A) or blotted onto a nitrocellulose membrane and detected by immunostaining with antiserum against DevH (B) or dinitrogenase reductase (C).

In wild-type Anabaena sp. strain PCC 7120, a low level of devH transcripts is detected in undifferentiated filaments. By 24 h of nitrogen stepdown, there is a fivefold increase in the steady-state levels of devH transcripts (12). DevH protein is detectable in undifferentiated filaments (0 h) of wild-type Anabaena sp. strain PCC 7120 and induced by 24 h after nitrogen starvation, although the magnitude of induction (∼twofold) is less than that for the devH transcript (Fig. 1B). The DevH fusion protein from mutant A57 also shows a pattern of induction (Fig. 1B), although the protein level in undifferentiated filaments (0 h) on average is higher than that observed for the wild-type protein and the induction by 24 h after nitrogen starvation is lower. Because the mutant and wild-type proteins differ in sequence, differences in stability of the transcripts, or in translation or stability of the proteins may affect the relative levels of the proteins in the wild-type and mutant strains. Alternatively, the regulation of expression of DevH may be altered in the mutant. Preliminary studies have demonstrated that rDevH binds specifically to the devH promoter (unpublished data), suggesting that devH, like ntcA (reviewed in reference 13), may have autoregulatory activity.

The nifHDK operon rearranges normally in A57, but nifHDK transcripts are undetectable.

In response to nitrogen limitation, strain A57 forms morphologically distinguishable cells at semiregular intervals that bind a heterocyst-specific stain but lack the polar bodies characteristic of mature heterocysts (12). In the absence of fixed nitrogen, strain A57 is unable to fix nitrogen in the presence of oxygen (12) and, thus, has a Het⁺ Fox⁻ phenotype (7).

The inability of strain A57 to grow diazotrophically could result from impairment of some aspect of nitrogen fixation. Expression of nitrogen fixation (nif) genes in Anabaena sp. strain PCC 7120 is preceded by three programmed heterocyst-specific DNA rearrangements. One of these removes an 11-kb interruption in the nifD gene of the nifHDK operon that encodes nitrogenase and dinitrogenase reductase (11), and a second eliminates a 55-kb element from the heterocyst-specific ferredoxin gene, fdxN (10). Anabaena sp. strain PCC 7120 mutants that lack NtcA, the protein most closely related to DevH, are impaired in the nifD and fdxN gene rearrangements and lack nif gene expression (9, 23).

The nifD rearrangement occurs normally in strain A57 under oxic conditions (Fig. 2A). The fdxN rearrangement was also observed (data not shown); however, nifHDK transcripts were undetectable by Northern analysis (Fig. 2B). Consistent with this result, dinitrogenase reductase (NifH) was undetectable by immunostaining with antiserum against the enzyme in strain A57 after nitrogen starvation in the presence of oxygen (Fig. 1C).

FIG. 2.

Southern analysis of the nifD rearrangement (A) and Northern analysis of nifH transcripts (B) in Anabaena sp. strain PCC 7120 (WT) and its devH mutant, A57. (A) Genomic DNA isolated from the wild-type strain and strain A57 that had been collected at 0 and 48 h after nitrogen stepdown under oxic conditions was digested with HindIII, fractionated, blotted, and hybridized with a nifD probe. Sizes in kilobases are indicated at the left. The nifD rearrangement occurs only in heterocysts (∼10% of the cells in differentiated filaments; 48-h sample) and generates HindIII fragments of 1.8 and 2.1 kb. (B) Northern analysis of nifH transcripts during a time course of heterocyst development in wild-type and A57 strains. Samples of total RNA (5 μg) from the wild-type strain and strain A57 cells deprived of fixed nitrogen for 0, 6, 24, or 48 h under oxic conditions were fractionated, blotted, and hybridized with a nifH probe. Hybridization to an rnpB probe was used as a loading control. The sizes (in kilobases) of the three stable nifHDK operon transcripts are indicated on the left.

The devH mutant is Fix⁺.

The absence of nifHDK transcripts in strain A57 could result from a defect in the regulatory system that induces transcription of the operon. Alternatively, transcription or transcript stability could be impaired by oxygen that enters the cell as a consequence of heterocyst structural defects. Evaluation of nitrogenase activity by acetylene reduction assays performed in the presence or absence of oxygen can be used to distinguish between these possibilities.

Nitrogenase activities of the wild-type and A57 strains after nitrogen starvation were measured under oxic (bubbling air) conditions. Acetylene reduction was detected in the wild-type strain by 37 h after nitrogen stepdown and peaked at ca. 46 h after stepdown at 16.4 nmol of ethylene produced per μg of chlorophyll a per h. No acetylene reduction by strain A57 was detected. In the presence of fixed nitrogen, when the systems for heterocyst development are suppressed, neither strain reduced acetylene.

The ability of strain A57 to reduce acetylene was tested under oxic and anoxic conditions. Nitrogenase activity was detected in A57, but only in concentrated cell suspensions that were maintained under anoxic conditions. To increase the sensitivity of the assay, the incubation time with acetylene was increased from 30 min to 4 h. Even longer incubation times have been used to detect nitrogenase activity in some Fox⁻ mutants (7). When assayed starting after 22 h of incubation under argon, the A57 cell suspension produced a total of 5.7 nmol of ethylene, while no ethylene was detected for a similar cell suspension maintained under oxic conditions. The ability of strain A57 to reduce acetylene under anoxic conditions demonstrates that the strain can induce a functional nitrogenase and thus has a Fix⁺ phenotype.

The devH mutant is Hgl⁻.

The Fox⁻ Fix⁺ phenotype of A57 suggests that the strain might have heterocyst defects that facilitate the entry of oxygen into the cell. To determine if such defects could be discerned, the ultrastructure of A57 heterocysts was examined by electron microscopy. Figure 3 shows that the laminated layer of envelope glycolipids is absent in strain A57 heterocysts. These results were confirmed by thin layer chromatography (TLC) of lipids from differentiated filaments. The heterocyst glycolipids that were apparent in the wild type after heterocyst induction were absent from the strain A57 profile (Fig. 4).

FIG. 3.

Ultrastructural analysis of heterocysts of Anabaena sp. strain PCC 7120 (WT) and its devH mutant, A57. Shown are transmission electron micrographs of ultrathin sections of heterocysts of the wild-type strain (A and B) and of A57 (C and D). Boxed regions in panels A and C are shown at higher magnification in panels B and D, respectively. The laminated layer of glycolipids (GL), present only in the heterocysts of the wild-type strain, is indicated in panel B. Bar, 0.5 μm (A and C) or 0.1 μm (B and D).

FIG. 4.

TLC analysis of glycolipids from Anabaena sp. strain PCC 7120 (WT) and its devH mutant, A57, collected at 0 or 48 h after nitrogen stepdown. The arrows indicate the heterocyst-specific glycolipid fractions.

Expression of glycolipid biosynthetic genes is altered in the devH mutant.

The absence of heterocyst-specific glycolipids in strain A57 heterocysts suggests that the expression of glycolipid biosynthetic genes may be altered in the mutant. Slot blots of total RNA from wild-type and A57 strains grown in the presence or absence of fixed nitrogen were probed with the heterocyst glycolipid biosynthetic genes hglB, hglC, hglD, hglE1, and hglE2. Figure 5 shows that transcripts from hglC and hglE1 were induced in strain A57 under oxic conditions, but at 20 to 30% of the wild-type level. Similar results were obtained for hglB and hglD (data not shown). hglE2 was induced in the wild type but not detectably expressed in strain A57. If heterocyst envelope glycolipids are synthesized, it is at such a low level that they are undetectable by TLC and insufficient to form a glycolipid layer. Alternatively, there may be hitherto-undetected turnover of those glycolipids. Because DevH is similar to other transcriptional regulators, it could regulate expression of the network of genes required for heterocyst glycolipid formation by direct interaction with their promoters. The effect of DevH also could be indirect and exerted through other gene products. Identification of the genes with altered expression in devH mutants, as well as the identification of genomic sites to which DevH binds, should provide a clearer picture of the role of DevH in heterocyst development and function.

FIG. 5.

Slot blot analysis of hgl gene expression in Anabaena sp. strain PCC 7120 (WT) and its devH mutant, A57. Samples of total RNA (10 μg) from cells deprived of fixed nitrogen for 0 or 24 h under oxic conditions were transferred to a Hybond N⁺ membrane and hybridized with hglC, hglE1 (alr5351), or hglE2 (all1646) probes. Hybridization to an rnpB probe is presented as a loading control.

Differentiation of a functional heterocyst requires coordinated expression and assembly of components that comprise the polysaccharide and glycolipid layers of the heterocyst envelope. Biosynthesis of the polysaccharide layer is controlled by the HepK/DevR two-component regulatory system (27), and we have shown here that biosynthesis of the glycolipid layer is regulated by DevH. Elucidation of components in the pathways upstream of each of these distinct transcriptional regulators should allow identification of a common control element(s) that coordinates their activation during heterocyst development.