Structural and Biological Characterization of Chromofungin, the Antifungal Chromogranin A-(47–66)-derived Peptide*

Vasostatin-I, the natural fragment of chromogranin A-(1–76), is a neuropeptide able to kill a large variety of fungi and yeast cells in the micromolar range. We have examined the antifungal properties of synthetic vasostatin-I-related peptides. The most active shortest peptide, named chromofungin, corresponds to the sequence Arg47–Leu66. Extensive1H NMR analysis revealed that it adopts a helical structure. The biophysical mechanism implicated in the interaction of chromofungin with fungi and yeast cells was studied, showing the penetration of this peptide with different lipid monolayers. In order to examine thoroughly the antifungal activity of chromofungin, confocal laser microscopy was used to demonstrate the ability of the rhodamine-labeled peptide to interact with the fungal cell wall, to cross the plasma membrane, and to accumulate in Aspergillus fumigatus, Alternaria brassicola, and Candida albicans. Our present data reveal that chromofungin inhibits calcineurin activity, extending a previous observation that the N-terminal region of chromogranin A interacts with calmodulin in the presence of calcium. Therefore, the destabilization of fungal wall and plasma membrane, together with the possible intracellular inhibition of calmodulin-dependent enzymes, is likely to represent the mechanism by which vasostatin-I and chromofungin exert antifungal activity.

The need for effective antifungal agents increases in parallel with the expanding number of immunocompromised patients at risk for invasive fungal infections. Currently available antifungal compounds act on targets also found in mammalian cells (1), which may result in toxicity. Thus, it becomes imperative to find antifungal compounds with no toxicity toward mammalian cells.
We have recently discovered new natural antimicrobial peptides derived from the processing of chromogranins A (CGA) 1 and B (CGB) and of proenkephalin-A, which are secreted with catecholamines upon stimulation of chromaffin cells in adrenal medulla (2)(3)(4)(5)(6)(7). These peptides have been hypothesized to play a role in stress situations, acting as an immediate protective shield against pathogens (6). Among these active neuropeptides, we have characterized the antibacterial and antifungal activities of vasostatin-I, CGA-(1-76), the highly conserved N-terminal fragment resulting from the natural processing of CGA (8). Our previous work has also shown that vasostatin-I is secreted from human polymorphonuclear neutrophils and that the C-terminal moiety of vasostatin-I CGA-(41-70) is essential for its antifungal activity (7).
The interaction of chromofungin with fungal membranes was analyzed using biophysical measurements of the penetration of these peptides within lecithin monolayers (12). Confocal laser microscopy on several fungi and yeast cells using rhodamine-labeled chromofungin showed the ability of this peptide to cross through cell wall and plasma membrane and to accumulate in microorganisms. Since previous studies have shown that the CGA N-terminal region interacts with calmodulin (CaM) in the presence of calcium (13), the sequence of chromofungin was compared with those of several CaM-binding enzymes and found to display common features. This property led us to study the effect of chromofungin on CaM-dependent enzymes choosing calcineurin (CaN) phosphatase activity.
Antifungal Activity-Filamentous fungi were grown on a five-cereal medium, and spores were harvested as previously described (18). The following filamentous fungi strains were used: Neurospora crassa (CBS 327-54), Aspergillus fumigatus, Alternaria brassicola (MUCL 20297), Nectria hematococca (160.2.2), Fusarium culmorum (MUCL 30162), Fusarium oxyporum (MUCL 909), and Trichophyton mentagrophytes. Yeast cells were precultured on a Sabouraud medium, and four strains were tested: Candida albicans, Candida tropicalis, Candida glabrata, and Cryptococcus neoformans. In order to test the antifungal activity of CGA-(47-66), spores (final concentration, 10 4 spores/ml) were suspended in a growth medium containing potato dextrose broth (Difco; in half-strength) and yeast cells in Sabouraud medium, with starting absorbance at 620 nm of 0.001. These media were supplemented with tetracycline (10 g/ml) and cefotaxime (0.1 g/ml). Fungal growth was assessed after an appropriate incubation period (24 or 40 h at 30°C). Aliquots of peptide extract (20 l) were incubated in microtiter plates with 80 l of fungal spores or yeast cultures. Growth of fungi was monitored microscopically and evaluated by measuring the culture absorbance at 595 nm (A 595 ) using a microplate reader. Yeast cells growth was assessed by the increase of A 620 .
All 1 H NMR measurements were carried out at 288 K at 500 MHz on a Bruker AMX500 spectrometer. Standard two-dimensional proton scalar and dipolar coupling experiments were carried out in phase-sensitive mode using the STATES-TPPI (time-proportional phase incrementation) scheme (19), with a spectral width of 6009 Hz in both dimensions and a relaxation delay of 3 s. Water signal suppression was achieved using a WATERGATE sequence (20), together with a very low power presaturation pulse on the water frequency. Amino acid proton spin systems were mostly assigned using a homonuclear Hartman Hahn (21) experiment with a 60-ms DIPSI2 (decoupling in the presence of scalar interaction) isotropic mixing pulse sequence. Sequential assignment was achieved using two NOESY spectra recorded with mixing times of 150 and 250 ms.
Processing was performed on an IBM RS/600 computer using the program FELIX version 2.10 (Biosym Inc.) and on an SGI INDY R5000 computer using UXNMR software (Bruker). Data were zero-filled and processed by apodization with a 90°-shifted sine squared bell window in both dimensions before Fourier transformation. The final size of matrices was 2048 ϫ 1024 points for all spectra. The computer model was obtained using the Molmol program (22).
Monolayer Measurements-Penetration experiments of peptides within artificial monolayer membrane (12) were performed at constant area (19.6 cm 2 ) in a thermostated (T ϭ 20 Ϯ 0.2°C) glass vessel equipped with a short vertical tube and containing 100 mM NaCl, 20 mM Tris-HCl, pH 7, as a subphase (volume 12 ml). The vessel was put on a miniature Variomag stirring system, and the whole device was enclosed in a plexiglass box. The lipids were spread at the air/water interface from an ethanol solution to give the desired initial surface pressure. Aliquots of a peptide stock solution (concentration in water: 10 Ϫ5 M res , expressed in amino acid residue/liter) were successively injected in the subphase under the lipid film through the vertical tube, and the bulk phase was gently stirred for a few minutes. The surface tension was measured with a Wilhelmy platinum plate hung to a Q11 balance (Schenk SA). The accuracy of surface pressure measurements was Ϯ0.2 mN/meter.
Fluorescence staining was monitored with a Zeiss laser scanning microscope (LSM 410) equipped with a planapo oil (ϫ 63) immersion lens (numerical aperture ϭ 1.4). Rhodamine emission was excited using the helium/neon laser 543-nm line, and the emission signal was filtered with a Zeiss long pass 595-nm filter. Fungi were subjected to optical serial sectioning (0.2-0.3 M) to produce images in the x-y plane. Each optical section was scanned eight times to obtain an average image. Images were recorded digitally in a 768 ϫ 576-pixel format.
Calcineurin Activity Assay-To measure the calcineurin phosphatase activity, we have used the BIOMOL GREEN TM assay kit (TEBU), including a 96-well microtiter plate format with all reagents supplied by the manufacturer. The RII phosphopeptide substrate (DVPIPGRFDRRVpS-VAAE), is the most efficient peptide substrate known for calcineurin (23,24). The detection of free phosphate released is based on the classic malachite green assay.
1 H NMR Spectroscopy of Chromofungin-The first step was to determine the three-dimensional structure of this peptide in water/trifluoroethanol solution (50/50) using 1 H NMR spectroscopy. The folding of peptides in the presence of trifluoroethanol is often encountered for peptides able to interact with membranes and has been widely used to study their three-dimensional properties. Among antibacterial peptides reported so far, trifluoroethanol has been used to study the structure of cecropin A (25), pardaxin P-2 (26), magainin 2 (27), and ranalexin (28). In addition, comparative study of 1 H NMR structures of magainin 2 and ranalexin in the presence of phospholipid micelles confirmed the helical structure previously found in trifluoroethanol/water solution (28,29).
Two regions of the 150-ms NOESY spectrum are shown in Fig. 1. They correspond to the fingerprint region, which contains the dipolar correlations between amide and H␣ protons (Fig. 1A) and correlations between amide protons of different residues (Fig. 1B). In addition, to sequential NOEs between the H␣ of one residue and the amide proton of the following one, several medium range NOEs could be observed between the H␣ of a residue and the amide proton 3-4 residues toward the C terminus ␣n(i ϩ 3), ␣n(i ϩ 4) for several residues along the sequence. This feature is a clear indication that the peptide adopts an ␣-helical structure in solution. The analysis of the amide-amide proton region of the NOESY map (Fig. 1B) shows many strong sequential and medium range HN-HN crosspeaks (nn(i ϩ 1), nn(i ϩ 2), nn(i ϩ 3)), which is expected in helical conformations.
All sequential and medium range NOEs, which have been observed for the backbone protons of the CGA-(47-66) peptide are summarized in Fig. 2. The distribution of ␣n(i ϩ 3) contacts along the sequence shows that the helical conformation extends through the entire peptide. The gap observed for residues located around Leu 61 is due to severe overlap between the H␣ protons resonance frequencies of residues 59, 61, and 62 (Fig.  1A). However, the observation of NOEs between the H␣ of the aspartate 63 and the amide proton of the C-terminal Leu 66 indicates that the C terminus of the peptide is folded into an ␣-helical conformation.
Three-dimensional Model of Chromofungin-A ribbon diagram of the average structure of the CGA-(47-66) peptide is shown in Fig. 3. This representation shows the amphipathic helical character of the C-terminal part of the sequence (Arg 53 -Leu 66 ). It is shown that hydrophobic leucine residues in positions 57, 58, 61, 64, and 66 are located on one side (blue ribbon), while four polar residues (Arg 53 , Lys 59 , Glu 60 , and Asp 63 ) are aligned on the other side (red ribbon). This feature is less pronounced in the N-terminal side of the ␣-helix, where a rather hydrophobic segment (Ile 48 , Leu 49 , Ile 51 ) is followed by a continuous stretch of hydrophilic residues from arginine 53 to asparagine 56. In Fig. 4, the representation of the helical wheel corresponding to the N-and C-terminal fragments of chromofungin (CGA-(47-56) (A) and CGA-(53-66) (B), respectively) emphasizes the importance of the region 53-66 for the amphipathic character.
Importance of the Amphipathic Helical Fragment CGA- (53)(54)(55)(56)(57)(58)(59)(60)(61)(62)(63)(64)(65)(66) for the Antifungal Activity of Chromofungin-Chromofungin, in the concentration range of 1-10 M, is highly active against a variety of filamentous fungi, and it completely inhibits the growth of N. crassa at 5 M (Table I). In order to examine the importance of this amphipathic C-terminal sequence, we have tested the antifungal activity of the chromofungin-derived peptide with two proline residues in place of Leu 61 and Leu 64 . These two proline residues disrupt the helical structure and the amphipathic character (Fig. 4). This control peptide inhibits the growth of N. crassa by 81% at 100 M, whereas it was inactive at the concentration of 5 M. This result shows that the amphipathic helix is crucial for the antifungal activity.
Many antimicrobial peptides have ␣-helical structures; the majority are cationic and amphipathic, but some hydrophobic peptides have also been described. Most antimicrobial peptides interact with inner and outer membranes and may be able to reach intracellular targets. Thus, it was important to study the interaction of chromofungin with fungal membranes.
Penetration of Chromofungin into Lipid Monolayers-Penetration experiments were performed at peptide concentrations ranging from 5 ϫ 10 Ϫ9 to 2 ϫ 10 Ϫ8 M res so that the adsorption to a clean air/water interface (i.e. in the absence of lipid film) did not induce any surface pressure increase. The initial pressure of the lipid films (20 Ϯ 0.5 mN/meter) was near the value generally considered as the internal pressure in cellular membranes (30). The addition of chromofungin (5 ϫ 10 Ϫ9 M res in the subphase) beneath lecithin film resulted in a slow increase, ⌬, of the surface pressure, which reached its maximum value, ⌬ M ϭ 1 mN/meter, after 3 h (Fig. 5, inset). This modest surface pressure increase is indicative of the penetration of chromofungin into the lipid monolayers, because agents interacting only with the lipid head groups without penetration do not modify the surface pressure (31). When ergosterol, the main sterol in yeast (32) and fungus (33)  The helical structure was built by assigning a and angle value of Ϫ57 and Ϫ47°, respectively for all residues. The residues are colored according to their hydropathy: red has been assigned to the arginine residue, which is the most hydrophilic (Ϫ4.5), and blue represents isoleucine (ϩ4.5). The computer model was obtained using the Molmol program (22).

FIG. 4. Helical wheel projection of the N-and C-terminal fragments of chromofungin according to the PLOT.A/HEL program.
A, CGA-(47-56); B, CGA-(53-65). Hydrophobic residues Ile and Leu are in boldface type. Underlined Leu 61 and Leu 64 were changed in proline residues to test the importance of the amphipathic character of chromofungin for the antifungal activity. lecithin/ergosterol monolayers, respectively) (Fig. 5), indicating that CGB-(602-626) is able to interact with lipid membranes. However, since CGB-(602-626) is inactive toward fungi, this peptide is unable to interact with the plasma membrane, probably because of the presence on the cell wall of complex structures as ␤-1,3-glucan and chitin (linear polymer of ␤-1,4-N-acetylglucosamine) preventing its targeting. At this stage, it was important to examine with confocal laser microscopy the interaction of chromofungin with various fungal microorganisms.
Confocal Laser Microscopy Analysis of the Interaction of Chromofungin with A. fumigatus, A. brassicola, and C. albicans-We used confocal laser microscopy to analyze the interaction of chromofungin with fungal membranes using the synthetic rhodamine-labeled CGA-(47-66)R peptide. The fluorescent rhodamine chromogranin B-derived peptide CGB-(602-626)R was used as a control, since it was inactive against fungi and yeasts. Prior the confocal microscopy analysis, we verified that the rhodamine-labeled peptides possess similar antifungal activities as those of unlabeled peptides. In these experiments, a concentration of 10 or 30 M peptide was used. Fungal spores of A. fumigatus, A. brassicola, and C. albicans were incubated at 30°C first during 24 or 40 h and then for 1, 16, or 21 h with fluorescent peptides.
Fungal spores of A. fumigatus were incubated during 24 h prior to treatment for 1 h at 30°C with fluorescent peptides (Fig. 6). By comparison with a control experiment (no peptide; Fig. 6A), CGA-(47-66)R 10 M was visible at the level of the cell wall, where it seemed to accumulate, but also in the inner part of A. fumigatus (Fig. 6B). In contrast, the labeled inactive peptide CGB-(602-626)R was undetectable in filamentous fungi (A. fumigatus; Fig. 6C) and yeast (C. albicans; data not shown) at any treatment time.
In a second series of experiments, we have tested the activity of chromofungin on fungal spores of A. fumigatus prepared by a first incubation for 40 h (Fig. 7A), prior to treatment for 1 h at 30°C with CGA-(47-66)R 30 M (Fig. 7B). The fluorescent active peptide was partly visible on the cell wall and the inner part of the distal growing end. This result showed that chromofungin penetrated through the cell wall at the fungus weakest point. After incubation for 40 h and then for 21 h at 30°C with CGA-(47-66)R 10 M, we observed that the fluorescent peptide invaded the numerous hyphae of A. fumigatus, inducing distortion of the wall and formation of numerous vacuoles (Fig. 7C). By phase-contrast microscopy, fungi regions invaded with the labeled peptide were visible as darkened areas.
In addition, we also examined the spores of A. brassicola (Fig. 8A) and C. albicans (Fig. 8B) prepared by a first incubation for 24 h followed by a second incubation for 1 h at 30°C with CGA-(47-66)R 30 M or CGA-(47-66)R 10 M, respectively. As compared with control experiments, the fluorescent peptide was detectable in cytoplasm and cell wall of both microorganisms. Fluorescence was absent, however, in the septum of A. brassicola (Fig. 8A, lower panel).
The treatment of A. fumigatus or C. albicans for 1 h with higher concentrations of unlabeled chromofungin (100 M) together with the inactive CGB-(602-626)R peptide (10 M) revealed an intense fluorescence in A. fumigatus (Fig. 6D) and in C. albicans (data not shown). This result indicated that chromofungin at high concentrations destabilizes cell wall, allowing inert peptide to penetrate into microorganisms.
Interaction of Chromofungin with Calmodulin-Besides destabilization of microorganism cell wall, chromofungin may also exert activity on intracellular targets. It has previously been shown that CGA is able to interact with CaM (13) and that the domain responsible for this interaction is probably located in the N-terminal region. CaM binds a variety of peptides, hormones, toxins, and enzymes in a calcium-dependent manner. The affinity of these peptides (16 -25 residues) for CaM appeared to correlate with their abilities to form positively charged amphiphilic helices named BAA helices (basic amphipathic ␣-helices) (34). The amino acid sequence of a large number of CaM-binding proteins has been determined and the location of the binding domains has been mapped (35). The comparison of the sequence of chromofungin with the CaMbinding regions from various target enzymes (Fig. 9) revealed a high 60% analogy between chromofungin (14) and the model peptide (13). A common feature to the sequences is the large number of positively charged residues (Arg, His, Lys) and the almost complete lack of acidic residues Asp or Glu. This alignment suggested to us that chromofungin could represent the region of CGA able to interact with CaM, thus affecting the activity of several CaM-dependent enzymes. This possibility was tested by measuring the effect of chromofungin on the activity of a calmodulin-dependent enzyme. Recently, it has been established that CaN, the CaM-activated phosphatase B, plays a crucial role in hyphal growth, morphology, and maintenance of the apical Ca 2ϩ gradient in the filamentous fungi N. crassa (36). This observation prompted us to examine the effect of chromofungin on CaN phosphatase activity. The experimental data reported in Fig.  10 showed that CaN phosphatase activity was completely inhibited by chromofungin at a concentration of 250 M. At a concentration of 5 M, corresponding to minimal inhibitory concentration against the filamentous fungi growth, the inhibition of CaN was evaluated to be 32%. The ability of chromofungin to inhibit CaN phosphatase activity at the same concentration range as that inhibiting fungal growth suggests that this molecular mechanism might be implicated. However, in addition to CaN, other CaM-dependent enzymes may be the target to chromofungin, which implies intracellular effects of this peptide. DISCUSSION The secretory granules of chromaffin cell from adrenal medulla synthesize, store, and, upon cell stimulation occurring during stress, exocytotically release catecholamines into circulation to activate target organs. Besides catecholamines, chromaffin granules contain a large number of proteins and peptides, including neuropeptides. The water-soluble proteins named chromogranin A and B and their derived peptides represent the major components. Furthermore, these proteins are widely distributed in endocrine, neuroendocrine, and nerve cells but also in immune cells (7). Sequences of CGA from a variety of species have now been reported (37), and from them the position of splicing sites and the size of the seven introns show a strong phylogenic conservation (38). It is important to point out that the N-terminal end of CGA is characterized by the presence of a disulfide loop (7) (which is also present on the N-terminal end of CGB) and that the sequences of CGA-(1-78) from several species are highly conserved, exhibiting 71% homology between frogs and humans (39) and up to 97% between bovine and humans (7).
Studies on CGA and CGB processing have established that in chromaffin granules these two proteins are attacked at both N-and C-terminal sides by proteolytic enzymes generating numerous peptides (2,8,40). In addition, the processing of CGA and CGB has been shown to be cell-and tissue-specific (41,42), giving rise to the release of multiple peptides recovered in blood and lymph circulations but also in biological fluids such as salivary (43), synovial, milk, and cerebrospinal fluids. 2 Vasostatin-I (CGA-(1-76)) is a natural N-terminal fragment of CGA, released from adrenal medulla (44) and sympathetic nerve terminals (45) in response to stimulation and first shown to exert vasoconstriction inhibitory activity on isolated human blood vessels (46) 40) with a unique affinity constant close to 50 nM (47). The role of vasostatin-I in stress response could represent an important biological function in addition to that of catecholamines on regulating blood circulation. Vasostatin-I also inhibits parathyroid hormone secretion (48), is neurotoxic in neuronal/ microglial cell cocultures (49), and induces cell adhesion (50,51). Our recent work has shown that vasostatin-I possesses antifungal activity (7), killing filamentous fungi and yeasts in the micromolar range. Furthermore, we reported that this new antibacterial and antifungal peptide is released from polymorphonuclear neutrophils, suggesting that it may act as a defensive molecule in infectious sites (7).
The past decade has witnessed a dramatic increase in knowledge of natural antimicrobial peptides, but most data concern antibacterial molecules. The emergence of fungal pathogens resistant to current therapies requires the discovery of new nontoxic antifungal molecules. Among all synthetic vasostatinderived peptides, chromofungin actually displayed the highest antifungal activity on both fungi and yeast. The 1 H NMR structural analysis of chromofungin demonstrates that, in our experimental conditions, chromofungin adopts an ␣-helical structure composed of two moieties, Arg 47 -Leu 52 and Arg 53 -Leu 66 , corresponding to a short N-terminal hydrophobic region and an amphipathic C-terminal fragment. This C-terminal region is crucial for the antifungal activity, as demonstrated here when two leucine residues have been replaced by proline residues. This helical structure is in accordance with secondary structure predictions of CGA, as early hypothesized by Yoo (13). Amphipathic ␣-helices were first described as structure/ function designs involved in lipid interaction (52). Such regions have been described for polypeptide hormones (endorphins) (53,54), venoms (bombolitin) (55), certain complex transmembrane proteins such as bacteriorhodopsin (56), and antibiotics (magainins) (27).
Antifungal peptides are classified according to their mode of action. A first peptide group acts by lysis which occurs via several mechanisms (e.g. destabilization of membrane, formation of aqueous pores, and intracellular mechanisms), while a second group interferes with cell wall synthesis or with the biosynthesis of glucan or chitin (1,57). Membrane penetration studies with chromofungin have demonstrated that this peptide was able to interact with lipid monolayers. The increase of the surface pressure in presence of ergosterol indicated a specificity for the fungal membranes. In addition, chromofungin induced membrane destabilization with the possible formation of holes, since the inactive peptide CGB-(602-626) could be recovered within fungi and yeast after preincubation with chromofungin. Thus, the specific antifungal activity of chromofungin should be dependent on a molecular mechanism located at the cell wall rather than at the sole plasma membrane. In addition, chromofungin penetrates into the cytoplasm, as evidenced by by confocal laser microscopy. Thus, within the microorganism it may interact with intracellular targets.
The cell wall is a physically rigid layer that protects the fungal cell from its environment, mediates cell-cell interaction, and is responsible for the shape of the cell. The structural skeleton of the cell wall is composed of ␤-1,3-glucan and chitin (linear polymer of ␤-1,4-N-acetylglucosamine) (58). Cell wall biosynthesis is a crucial process in the formation, growth, and morphogenesis of fungal cells. In addition, it has been indicated that Ca 2ϩ and CaM may activate chitin synthase and that CaM-dependent phosphorylation of microsomal proteins may be crucial for the activation of this enzyme (59). Thus, a possible mechanism for the destabilization of the cell wall induced by chromofungin may be its binding to CaM in the presence of Ca 2ϩ and thus the inactivation to the chitin synthase. Because of the homology of chromofungin with CaM-binding peptides, various CaM-binding proteins could be affected by this peptide. This is illustrated by the present observation that CaN is inhibited by chromofungin in in vitro experiments.
The interaction of chromofungin with CaM and CaM-binding proteins requires careful future investigation, because it opens the way for a new mechanism concerning antifungal molecules.
We have to mention here that although chromofungin possesses antifungal activity at micromolar range, it is a nonhemolytic peptide, unable to penetrate into astrocytes, oligodendrocytes, and neurons, 3 suggesting that this peptide is nontoxic at this concentration. Furthermore, it has recently been reported that the CGA-(47-57) peptide (RILSIL-RHQNL), included in chromofungin, contains a cell adhesion site for fibroblasts and smooth muscle cells (51), pointing out multifunctional roles for this ␣-helical CGA sequence, which is 100% conserved along evolution (38).  (47-66)). The calcineurin phosphatase activity was examined using the BIOMOL GREEN TM calcineurin assay kit (TEBU) including the RII phosphopeptide substrate (DVPIPGRFDRRV-pSVAAE) as calcineurin substrate. The detection of free phosphate released is based on the classic malachite green assay at 620 nm.
Vasostatin I appears as a peptide with multifunctional potency. Besides its property as a vascular modulator (46), it is able to kill bacteria and fungi (7) in in vitro tests. At this stage, it is not known whether it exerts this activity in vivo. The concentration of circulating vasostatin-I is in the 100 nM range, and its half-life in blood remains to be determined. However, as has been shown for enkelytin, a chromaffin granule peptide with antibacterial activity (5), antibacterial and antifungal peptides may concentrate in infectious sites and display synergistic activities.
To conclude, vasostatin-I sequence contains a highly conserved region of 14 amino acids; the corresponding peptide displays a potent antifungal activity due to its ability to cross microorganism cell membranes and perhaps to target and inhibit calmodulin-binding enzymes. Although some important questions concern the expression of its activity in physiological and pathological situations, chromofungin represents a prototypic lead molecule useful for the development of new therapeutic agents.