Ctc

Fig. 3. (A) Inhibition of morphological development by the Cu(I)-specific chelator BCDA. Confluent lawns of S. lividans (bottom) and S. coelicolor (top) have developed aerial hyphae and spores after 7 d incubation at 30°C (left). On R2YE agar plates with 25 ^M, BCDA added, morphological development is blocked and does not proceed further than vegetative mycelium. The production of secondary metabolites is not inhibited. (B) Copper counteracts the inhibition of BCDA. S. coelicolor (left) and S. lividans (right) were grown on R2YE agar plates containing 25 ^M BCDA. The addition of copper sulfate via a cotton filter induces development of aerial hyphae and spores in a circular zone around the filter.

Fig. 3. (A) Inhibition of morphological development by the Cu(I)-specific chelator BCDA. Confluent lawns of S. lividans (bottom) and S. coelicolor (top) have developed aerial hyphae and spores after 7 d incubation at 30°C (left). On R2YE agar plates with 25 ^M, BCDA added, morphological development is blocked and does not proceed further than vegetative mycelium. The production of secondary metabolites is not inhibited. (B) Copper counteracts the inhibition of BCDA. S. coelicolor (left) and S. lividans (right) were grown on R2YE agar plates containing 25 ^M BCDA. The addition of copper sulfate via a cotton filter induces development of aerial hyphae and spores in a circular zone around the filter.

model, the effect of the copper ions would play a role higher up in the "bld-cascade." However, the bld mutants suffer from pleiotropic effects that could affect the putative "copper pathway" as well.

The mechanism responsible for the strong enhancement of the differentiation by the addition of copper and for the developmental block that occurs after the addition of copper-specific chelators is not well understood in streptomycetes. Considering the observations made in P. anserina, deregulation of cellular copper metabolism is obvious. Assuming that S. lividans is lagging behind in development

Fig. 4. Copper content of S. lividans and S. coelicolor. The copper content of mycelium of S. lividans and S. coelicolor grown in medium with and without extra copper added was determined by atomic absorption. Mycelium was concentrated by centrifugation and dissolved in nitric acid. The copper concentration (in ppm) is expressed per milligram of mycelium (wet weight). Samples were taken during vegetative growth (around 40 h), just before aerial hyphae started to appear (65 h), during full aerial hyphae development (100 h), and during sporulation (120 h).

Fig. 4. Copper content of S. lividans and S. coelicolor. The copper content of mycelium of S. lividans and S. coelicolor grown in medium with and without extra copper added was determined by atomic absorption. Mycelium was concentrated by centrifugation and dissolved in nitric acid. The copper concentration (in ppm) is expressed per milligram of mycelium (wet weight). Samples were taken during vegetative growth (around 40 h), just before aerial hyphae started to appear (65 h), during full aerial hyphae development (100 h), and during sporulation (120 h).

because of a malfunction in copper homeostasis, one would expect to see different intracellular copper levels in surface-grown cultures of S. lividans and S. coelicolor. Atomic absorption measurements of the cellular copper concentrations of S. lividans and S. coelicolor mycelium harvested at various moments during growth on solid medium showed the same constant copper content during growth for both strains. A particular increase at the onset of differentiation was not detected (Fig. 4). On growth media with extra copper added, the mycelium does contain higher copper levels that increase in time, but, again, there is no major difference between S. lividans and S. coelicor (Keijser, unpublished). This may rule out that copper needs to be imported in higher amounts during one specific stage of growth.

3.1. The ram Cluster

Complementation of morphological mutants with genomic libraries of the wild-type strain is a powerful and successful technique to clone the corresponding gene (32). Several clones of a genomic library of the closely related strain S. coelicolor complemented the copper-dependent phenotype of S. lividans. The S. lividans transformants produced aerial hyphae approx 24-36 h earlier than the parent strain. Extra copper addition to the medium could not achieve a further advancement of the onset of differentiation (51). Sequencing and Southern blot analysis showed that, without exception ,these complementing clones were derived from the same region of the S. coelicolor genome. They contain the ram cluster that was previously identified as a clone that resulted in rapid aerial mycelium formation (55). This locus consists of five genes encoding a putative membrane-bound serine/threonine kinase (ramC), a transcription factor of the two-component response regulator family (ramR), two ABC transporters (ramAB), and a small ORF (ramS) that does not have any homology to known sequences in the database (Fig. 5). The ramC gene is not required for the complementation because this ORF was not intact on all of the complementing clones. Integration of a single copy of the ramSABR genes in the genome of S. lividans was sufficient to get full complementation of the copper-dependent differentiation. Introduction of ramS or ramR alone had no effect (51).

3.2. The Role of Ram Proteins in the Onset of Aerial Hyphae Formation

The need of S. lividans for the S. coelicolor ram genes to get timely aerial hyphae formation suggests the absence of a (active) ram cluster in the former strain. Surprisingly, a complete ram

Fig. 5. Organization of the ram and amf clusters. The ram cluster consists of five genes: a membrane-bound serine/threonine kinase (ramC), a small peptide (ramS), two ABC transporters (ramA and ramB), and a transcription factor (ramR). The stop codon of ramA overlaps with the start codon of ramB, suggesting translational coupling of the two ORFs. The amf cluster consists of the same number of genes. The ramC ortholog is not cloned/sequenced completely, but the available C-terminal sequence suggests that the ORF upstream of ORF6 is similar to ramC. The amfC gene reported in the literature (56) does not belong to this cluster of genes.

Fig. 5. Organization of the ram and amf clusters. The ram cluster consists of five genes: a membrane-bound serine/threonine kinase (ramC), a small peptide (ramS), two ABC transporters (ramA and ramB), and a transcription factor (ramR). The stop codon of ramA overlaps with the start codon of ramB, suggesting translational coupling of the two ORFs. The amf cluster consists of the same number of genes. The ramC ortholog is not cloned/sequenced completely, but the available C-terminal sequence suggests that the ORF upstream of ORF6 is similar to ramC. The amfC gene reported in the literature (56) does not belong to this cluster of genes.

cluster is present in S. lividans and is shown to be fully capable of complementing the copper-dependent phenotype when present in at least two copies (51). In order to get better insight in the participation of the proteins encoded by these genes in development, ramABR disruption mutants were constructed and the transcriptional regulation and organization was analyzed. Introduction of an extra copy of ram dramatically increased the low level of transcription of S. lividans. Transcription, arising in part from promoters upstream of ramS, was found to be confined to the growth phase in which aerial hyphae are formed, underscoring the role of this cluster during this developmental stage (51). This was further illustrated by the phenotype of S. lividans ramABR mutants. The disruption of ramABR in S. lividans resulted in a loss of the ability to form aerial hyphae and spores, whereas disruption in S. coelicolor resulted in a severe delay of the onset of aerial hyphae formation but not a bld phenotype (Fig. 6). Vegetative growth and secondary metabolism were not affected by the mutation. Interestingly, normal development was restored when grown in the presence of excess copper. In addition, when grown in close proximity to differentiating wild-type strains, the development of ramABR mutants was also restored (Fig. 7). This is indicative for extracellular complementation. The ram disruption mutant is no longer capable of exporting a signal molecule whereas the wild type strain is and can provide the mutant with this signaling molecule. The role of RamR as a transcription factor could be either in the regulation of the production of this signal, the regulation of ramAB transcription, or the regulation of the expression of an unknown factor.

One of the bld mutants (bldA) does respond to the ram mutant when grown next to one another. All other mutants tested, bldD, bldF, and bldG, do not respond nor does the ram mutant react to the bld mutants (51). However, this analysis is not extensive enough to conclude whether ram is or is not part of the proposed "bld-cascade" (11).

The data do not strongly support a direct association between ram and copper. A model in which ram and copper are part of independent morphological pathways seems the simplest solution for the moment. In this model, the copper-dependent pathway is induced under high copper-ion conditions

S.lividans 1326 S.lividans 1326 A ramA/B/R

S.coelicolor M145 S.coelicolor Ml 45 A ramA/B/R

Fig. 6. Phenotype of ram disruption mutants. Wild-type strains (left) and ram disruption mutants (right) of S. lividans and S. coelicolor were grown on R2YE (0.2 pM copper). The S. lividans mutant has a bald phenotype, whereas the S. coelicolor mutants is still capable of aerial hyphae formation, albeit much slower than the wild type.

Fig. 6. Phenotype of ram disruption mutants. Wild-type strains (left) and ram disruption mutants (right) of S. lividans and S. coelicolor were grown on R2YE (0.2 pM copper). The S. lividans mutant has a bald phenotype, whereas the S. coelicolor mutants is still capable of aerial hyphae formation, albeit much slower than the wild type.

(>2 pM) and development then becomes ram independent. The observation that a strong reduction of the available copper results in a block of differentiation could be explained by the need of a specific copper protein during development that could operate during both the ram-dependent and the high-copper-dependent routes. This is supported by the fact that vegetative growth is not inhibited by the Cu(I)-specific chelator BCDA (i.e., enough copper is still available for the copper proteins active during this growth phase). The only condition when ram is essential for timely development is in the presence of "normal" copper levels. Under these conditions, the ram dependence is absolute for S. lividans, whereas S. coelicolor seems to have yet another route to its disposal that allows slow development even in the absence of ram (Fig. 8).

Fig. 7. Communication between the ram disruption mutant and the wild-type strain. Wild-type S. lividans (left) and the ramABR disruption mutant (right) grown on solid medium under conditions where the mutant alone does not form aerial hyphae and spores. The edge of patches of mutant hyphae growing closest to the wild-type mycelium show the typical white color of aerial hyphae. This is indicative for an extracellular factor that induces development in the mutant and is produced by the wild-type strain. Therefore, the ramABR mutant is either not capable of secreting this factor or unable to produce it. The former would be in agreement with the putative function of RamAB.

Fig. 7. Communication between the ram disruption mutant and the wild-type strain. Wild-type S. lividans (left) and the ramABR disruption mutant (right) grown on solid medium under conditions where the mutant alone does not form aerial hyphae and spores. The edge of patches of mutant hyphae growing closest to the wild-type mycelium show the typical white color of aerial hyphae. This is indicative for an extracellular factor that induces development in the mutant and is produced by the wild-type strain. Therefore, the ramABR mutant is either not capable of secreting this factor or unable to produce it. The former would be in agreement with the putative function of RamAB.

vegetative mycelium

Fig. 8. Copper and the onset of aerial hyphae formation. Binding of all copper ions to the copper-specific chelator BCDA blocks the switch from vegetative growth to aerial mycelium formation (left). The ram-dependent pathway operates during growth on R2YE (0.2 pM copper), whereas under the same conditions, a second pathway that is ram independent but slower (dotted line) seems to be active in S. coelicolor. Higher copper levels (>2 pM) render S. coelicolor and S. lividans ram independent.

Fig. 8. Copper and the onset of aerial hyphae formation. Binding of all copper ions to the copper-specific chelator BCDA blocks the switch from vegetative growth to aerial mycelium formation (left). The ram-dependent pathway operates during growth on R2YE (0.2 pM copper), whereas under the same conditions, a second pathway that is ram independent but slower (dotted line) seems to be active in S. coelicolor. Higher copper levels (>2 pM) render S. coelicolor and S. lividans ram independent.

3.3. The amf Cluster

In S. griseus, one of the important factors determining morphological differentiation as well as physiological differentiation is A-factor. This bacterial hormone belongs to the family of y-butyrolactones. A developmental mutant, HH1, that is deficient in A-factor production has a bald phenotype (no aerial hyphae formation) and no longer produces streptomycin (6). A gene cluster complementing this mutant for morphological development but not for A-factor and streptomycin production is amf. The organization of the S. griseus amf cluster and the S. coelicolor ram cluster are very similar (Fig. 5). Two genes (amfAB) encode ABC transporters and a third gene (amfR), transcribed in the opposite direction, encodes a response regulator of a two-component regulator system. The amf cluster also contains a ramS ortholog and an upstream ORF with similarity to ramC. The overall similarity of the gene products of ramAB and amfAB is sufficiently high to assume that these transporters have a similar function. This does not necessarily mean that they transport the same molecule and operate at the same point of the developmental cascades in S. griseus and S. coelicolor.

The participation of the membrane-bound transporters AmfAB in the onset of aerial hyphae formation in S. griseus is, in fact, dubious. The amfB gene was shown to be dispensable for complementation of strain HH1, whereas amfR and amfA were needed (57). Later, it was demonstrated in wild-type S. griseus that the disruption of only amfR results in a bald phenotype. This mutant could be complemented by amfR, including some upstream sequences but without amfA (58).

0 0

Post a comment