INTRODUCTION

Top Abstract Introduction Materials and Methods Results and Discussion References

Fungal infections are an increasingly important health threat as the number of immunocompromised individuals continues to rise. A limited repertoire of effective antifungal drugs with acceptably low host toxicity has motivated the search for novel antifungal drug targets, but the identification of targets unique to fungi has been a challenge, given the remarkable similarity between fungal and mammalian metabolic and signal transduction pathways. There are, however, several promising antifungal target evaluations under active investigation. Enzymes involved in the biosynthesis of the fungal cell wall, lipid composition of the plasma membrane, and DNA and protein synthesis have been targeted, with various degrees of success (8, 9). Recently, the list of potential antifungal targets has expanded to include histidine protein kinases (HPKs), a class of signal transduction molecules once thought to be restricted to the procaryote kingdom but now known to exist in fungi and other lower eucaryotes and plants (24).

HPKs and two-component regulation are used extensively in procaryotic signal transduction processes that allow bacteria to adapt and respond to the environment. Two-component signal transduction systems are so named because at their core are two modules, the histidine kinase and a response regulator. In response to an environmental signal, the kinase autophosphorylates a conserved His residue. The His-phosphate is then transferred to an aspartyl residue on the response regulator, thereby altering its activity. Recently, it has been shown that two-component systems are essential for the growth and viability of Bacillus subtilis, establishing HPKs as important new targets for antibacterial drug design (6). Natural product and synthetic inhibitors of bacterial HPKs have been described that include hydrophobic tyramines and substituted salicylanilides (11, 15, 22, 23). These compounds exhibit MICs in the range of 1 to 10 µg/ml and inhibit bacterial HPKs 50% inhibitory concentrations (IC₅₀s) in the low micromolar range (4). However, despite the correlation between bactericidal activity and inhibition of HPKs, it has not been established that HPK inhibition accounts for the bactericidal activity.

Although two-component signal transduction pathways have been described in bacteria, fungi, slime molds, and plants (reviewed in reference 24), HPKs and two-component pathways have not been found in vertebrates, despite considerable effort. Saccharomyces cerevisiae encodes one membrane-associated HPK (Sln1) and two response regulators, Ssk1 and Skn7, that regulate osmotic and oxidative stress response (12-14, 19, 20, 25). A second family of eucaryotic histidine kinase represented by NIK1 has been isolated from Neurospora crassa (1). Pathogenic fungi, Candida albicans and Aspergillus nidulans, appear to encode both NIK1 and SLN1 types of histidine kinases (2, 17, 21). Although they differ in the overall domain organization, there is a very high degree of sequence conservation between individual HPK domains, presumably indicating a conserved catalytic mechanism. Deletion of S. cerevisiae SLN1 causes derepression of the HOG1 osmosensing pathway, resulting in either cell death or a severe growth defect, depending on the strain background (16). Deletion of NIK1 in C. albicans results in a defect in hyphal formation which may affect virulence (2, 17, 21). Based on these observations, inhibitors of two-component signaling pathways might be effective antifungal agents. However, this idea has not been tested.

In this report we show that a set of antibacterial HPK inhibitors also have antifungal properties. A subset of these compounds also inhibits Sln1 histidine kinase activity in vitro; however, the inhibition of HPK activity does not correlate with the antifungal MICs. Finally, the antifungal properties of this group of compounds are unaffected by deletion of SLN1, the only known HPK in S. cerevisiae. Thus, the antifungal properties of these compounds do not appear to be due to inhibition of two-component HPKs.

	MATERIALS AND METHODS

Top Abstract Introduction Materials and Methods Results and Discussion References

Yeast strains and growth conditions. S. cerevisiae strains used in this study were derived from laboratory stocks (J. S. Fassler) using standard yeast crossing and transformation techniques (3). JF1735 (MATa sln1::TRP1 ssk1::LEU2 his4-917 leu2-1 lys2-128 trp11 ura3-52) was derived from JF1331 (MATa his4-917 leu2-1 lys2-128 trp11 ura3-52) by carrying out the disruption of the SLN1 and SSK1 as described previously (14, 25). JF1331 is a descendant of S288c. Yeast cultures were grown at 30°C in rich medium (yeast extract-peptone-dextrose [YEPD]: 1% Bacto yeast extract, 2% Bacto Peptone) with glucose (2% [wt/vol]) as a primary carbon source. C. albicans CAF2-1 (ura3::imm434/URA3) is derived from SC5314 (7).

Inhibitors. RWJ compounds 49815, 49968, and 61907 were provided by the R. W. Johnson Pharmaceutical Research Institute (4, 22).

Yeast growth inhibition assays. Compounds were added at the indicated concentrations to exponentially-growing yeast cultures (5 × 10⁶ cells/ml), and the growth was monitored by measuring the optical density at 660 nm (OD₆₆₀). Cultures were incubated at 30°C with shaking.

Sln1 histidine kinase assay. Expression and affinity purification of glutathione S-transferase (GST)-Sln1 kinase has been described previously (14). Briefly, glutathione beads containing approximately 10 µg of GST-Sln1 (50% slurry) were added to buffer C (50 mM Tris [pH 7.5], 50 mM KCl, 5 mM MgCl₂, 0.1% -mercaptoethanol, 50% glycerol). Inhibition experiments were performed by adding the compound to be tested in buffer C containing 5% DMSO. The kinase reaction (final volume, 50 µl) was initiated by the addition of 3.5 µl of [-³²P]ATP (3,000 Ci/mmol; Amersham) followed by incubation at room temperature for 60 min. Unincorporated [-³²P]ATP was removed by washing four times with 1 ml of buffer C. The beads were resuspended to obtain a volume of 80 µl by using buffer C, and then 20 µl of 5× protein loading buffer (30 mM Tris [pH 6.7], 10% sodium dodecyl sulfate [SDS], 60% glycerol, 0.8 M -mercaptoethanol, 0.001% bromophenol blue) was added. Samples were incubated at room temperature for 15 min and then resolved by SDS-polyacrylamide gel electrophoresis (PAGE) (10% polyacrylamide). Phosphorylation of the kinase (³²P) was determined by measuring the emission of radioactivity by using a Packard instant imager or the gel was transferred to nitrocellulose for autoradiography.

	RESULTS AND DISCUSSION

Top Abstract Introduction Materials and Methods Results and Discussion References

Bacterial HPK inhibitors also inhibit yeast cell growth. Inhibition of two-component signaling pathways has been identified as a promising target for antibacterial drug design (reviewed in reference 5). For example, a family of hydrophobic tyramines have been described that inhibit two-component histidine kinases and growth of gram-positive bacteria in the micromolar concentration range (4). However, despite excellent correlation between growth inhibition and HPK inhibition, it has not been possible to conclude that the bactericidal activity resulted from kinase inhibition, because of the large number of two-component kinases in bacteria. Two-component systems have recently been described in S. cerevisiae (16, 19), Neurospora crassa (1), and Candida albicans (17, 21). In S. cerevisiae, loss of Sln1 function is lethal, due to a hyperstimulation of the HOG1 mitogen-activated protein (MAP) kinase pathway (Fig. 1). Based on these observations, we reasoned that antibacterial histidine kinase inhibitors might also possess antifungal properties.

View larger version (22K): [in this window] [in a new window]   FIG. 1.   Model for the SLN1 osmosensing pathway illustrating how deletion or inhibition of Sln1 kinase may lead to cell death or growth inhibition. (A) Under physiological conditions, the histidine kinase activity of Sln1 is controlled by the osmotic balance. Hyperosmotic conditions shift the equilibrium to the unphosphorylated form of Sln1, and the Hog1 MAP kinase is activated. (B) Deletion of the SLN1 gene or inhibition of Sln1 kinase activity leads to an accumulation of unphosphorylated Ssk1 which, in turn, causes hyperstimulation of the HOG MAP kinase pathway and cell death.

View larger version (12K): [in this window] [in a new window]   FIG. 2.   Chemical structures of compounds reported to inhibit bacterial two-component HPKs (4, 19).

Sln1 histidine kinase inhibition. Next, an in vitro kinase reaction was used to determine directly if RWJ-49815, RWJ-49968, and RWJ-61907 inhibit the HPK activity of yeast Sln1. The autokinase activity of purified GST-Sln1 was tested with and without incubation with the inhibitors (Fig. 3). Two of the compounds, RWJ-49815 and RWJ-49968, inhibit Sln1 kinase, IC₅₀s of approximately 10 µM. In contrast, RWJ-61907 failed to inhibit Sln1 kinase at concentrations up to 100 µM despite an MIC of 2.5 to 5.0 µg/ml (Table 1). This was somewhat surprising, given the presumption that the antibacterial and antifungal properties of these compounds were due to HPK inhibition. This raises the possibility that the kinase is not the target of inhibition. However, we could not rule out that RWJ-61907 is metabolically converted in yeast to a form that inhibits HPK in vivo. One way to address this possibility is to compare the abilities of these compounds to inhibit growth of wild type and yeast cells in which the SLN1 gene has been deleted (sln1).

View larger version (18K): [in this window] [in a new window]   FIG. 3.   Inhibition of GST-Sln1 kinase by RWJ-49815 (A), RWJ-49968 (B), or RWJ-61907 (C). Results are expressed as a function of GST-Sln1 phosphorylation (Sln1-PO4) in the absence of drug after a 60-min incubation. The inset shows the autoradiogram of GST-Sln1-PO4 used to quantitate autokinase activity. This is a representative experiment; essentially the same results were obtained on three separate occasions.

Growth inhibition of SLN1 and sln1 yeast strains. Our current understanding of the Sln1 osmosensing pathway provides us with a unique opportunity to address the target specificity of putative HPK inhibitors (Fig. 1). Sln1 is the only two-component histidine kinase in S. cerevisiae. Deletion of the SLN1 gene results in a severe growth defect or cell death, depending on the strain (15). Lethality can be suppressed by deletion of the response regulator SSK1 or of the genes encoding components of the HOG1 MAP kinase pathway, SSK2 (MEK kinase), PBS2 (MEK), or HOG1 itself (16). For example, an sln1 strain is inviable, whereas an sln1 ssk1 strain grows normally. It follows that a true anti-HPK drug will inhibit a wild-type (SLN1) strain but will be ineffective against an sln1 ssk1 strain.

sln1 ssk1

sln1 ssk1

View larger version (18K): [in this window] [in a new window]   FIG. 4.   Growth inhibition of JF1331 (SLN1) (solid lines) and JF1735 (sln1ssk1) (dashed lines) by RWJ-49815 (squares) or RWJ-61907 (circles). The indicated concentration of drug was added to exponentially-growing cultures (5 × 106 cells/ml), and the OD660 was determined after incubation at 30°C for 5.5 h. The values plotted represent the averages of triplicate samples. Individual values varied less than 10%.