Logo of nihpaAbout Author manuscriptsSubmit a manuscriptHHS Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
J Invest Dermatol. Author manuscript; available in PMC 2009 Oct 5.
Published in final edited form as:
PMCID: PMC2757066

The Neuroendocrine Peptide Catestatin Is a Cutaneous Antimicrobial and Induced in the Skin after Injury


Epithelia establish a microbial barrier against infection through the production of antimicrobial peptides (AMPs). In this study, we investigated whether catestatin (Cst), a peptide derived from the neuroendocrine protein chromogranin A (CHGA), is a functional AMP and is present in the epidermis. We show that Cst is antimicrobial against relevant skin microbes, including Gram-positive and Gram-negative bacteria, yeast, and fungi. The antimicrobial mechanism of Cst was found to be similar to other AMPs, as it was dependent on bacterial charge and growth conditions, and induced membrane disruption. The potential relevance of Cst against skin pathogens was supported by the observation that CHGA was expressed in keratinocytes. In human skin, CHGA was found to be proteolytically processed into the antimicrobial fragment Cst, thus enabling its AMP function. Furthermore, Cst expression in murine skin increased in response to injury and infection, providing potential for increased protection against infection. These data demonstrate that a neuroendocrine peptide has antimicrobial function against a wide assortment of skin pathogens and is upregulated upon injury, thus demonstrating a direct link between the neuroendocrine and cutaneous immune systems.


Epithelia such as the skin impede the penetration of microbes by its structural integrity, pH, and biochemical constituents such as lipids, peptides, and proteins. Antimicrobial peptides (AMPs) are major components of this defense system and include multiple gene families encoding these “natural antibiotics.” These AMPs are essential for function at the boundary with the external environment as a rapid first line of defense against invasion by pathogens including bacteria, viruses, and fungi (Stolzenberg et al., 1997; Bals et al., 1998; Nizet et al., 2001; Bardan et al., 2004; Murakami et al., 2004; Schauber et al., 2007). Some AMPs, for instance, HBD-2 (human β-defensin 2) (Harder et al., 1997), HBD-3 (Harder et al., 2001), and cathelicidin (Dorschner et al., 2001; Schauber et al., 2007), are expressed at low levels constitutively, but are inducible following infection or injury. The mechanism of action for such cationic AMPs relies upon the interaction between the peptide and the microbial membrane to promote membrane destabilization, ultimately leading to cell lysis (Devine and Hancock, 2002). The relevance of the antimicrobial activity of cationic peptides has been shown in mice, where an absence of cathelicidin increases the susceptibility to several forms of infection, including skin invasion by Group A Streptococcus (Nizet et al., 2001), Citrobacter growth in the colon (Iimura et al., 2005), and the development of Escherichia coli in urinary tract infections (Chromek et al., 2006).

Peptides derived from the chromogranin/secretogranin (or “granin”) family, as well as several other neuropeptides, including α-melanocyte-stimulating hormone, calcitonin gene-related peptide, and neuropeptide Y, have been reported to have antimicrobial activity in vitro (Fox et al., 1997; Shimizu et al., 1998; Cutuli et al., 2000; Metz-Boutigue et al., 2003b), but the significance of this activity in vivo remains elusive. The granin family is distributed in secretory granules of endocrine, neuroendocrine, and neuronal cells (O’Connor, 1983; Taupenot et al., 2003), but its presence is not known at sites of microbial contact. The granin family includes chromogranin A (CHGA), isolated from chromaffin cells of the adrenal medulla (Banks and Helle, 1965), CgB (chromogranin B), characterized from a rat pheochromocytoma cell line (Lee and Huttner, 1983), and SgII (secretogranin II), originally described in the anterior pituitary (Fischer-Colbrie et al., 1995). Granins originate as pro-proteins stored in secretory granules with numerous cleavage sites for endopeptidases, such as serine endoproteases or plasmin (Dillen et al., 1993; Eskeland et al., 1996). Proteolytic processing of granins generates peptides that are released into the extracellular space upon stimulation.

Chga designates the gene encoding for the protein CHGA, which is processed at dibasic sites to give rise to several bioactive peptides, including the 21 amino acid catestatin (Cst). Cst was originally characterized as a catecholamine release inhibitory peptide acting as an antagonist of the neuronal nicotinic cholinergic receptor (Mahata et al., 1997,1998; Taupenot et al., 2003). The amino-acid sequence of human Cst is conserved across species (SSMKLSFRARAYGFRGPGPQL—human CHGA352–372) (RSMRLSFRTRGYGFRDPGLQL—mouse CHGA364–384) (Mahata et al., 1997,2004). Its calculated pl is ~11.72, rendering Cst as a highly basic peptide, and it exhibits an amphiphilic β-sheet structure characteristic of other AMPs (Kennedy et al., 1998; Preece et al., 2004). Recently, single nucleotide polymorphisms have been identified in Cst and are estimated to occur in ~4% of the human population (Wen et al., 2004). These include peptide variants Gly364Ser and Arg374Gly, which exhibit a loss of potency, or Pro370Leu, which exhibits higher potency as a nicotinic antagonist compared to wild-type Cst. Interestingly, Cst was previously shown to be expressed in peripheral mononuclear cells and exhibited antimicrobial activity against filamentous fungi, yeast, and bacteria in vitro (Briolat et al., 2005). However, its effectiveness against pathogens frequently encountered by the skin was not previously assessed.

Proteolytic processing of CHGA also results in the generation of peptides other than Cst with AMP activity, including prochromacin (bovine CHGA79–431), Chromacin I and II (bovine CHGA173–194 and bovine CHGA195–221), chromofungin (bovine CHGA47–66), and vasostatin I (bovine CHGA1–76). As granins are present within secretory chromaffin granules and are co-secreted together with catecholamines during the stress response, release of such peptides could, such as cathelicidins, provide antimicrobial activity to counteract the suppression of cell-mediated immunity resulting from glucocorticoid release (Elenkov et al., 1999).

In this study, we evaluated the hypothesis that Cst exhibits antimicrobial activity against skin pathogens and may participate in antimicrobial defense of the skin. Our findings demonstrate that CHGA and Cst may have a direct role in the establishment of an innate antimicrobial barrier in the skin. These observations support a previously undefined relationship between the granins and cutaneous antimicrobial defense.


Cst is directly antimicrobial against skin pathogens

Peptides derived from CHGA were previously shown to exhibit antimicrobial activity against microbial pathogens in vitro, although many were not typical of those encountered during cutaneous infection (Lugardon et al., 2000; Metz-Boutigue et al., 2003a; Briolat et al., 2005). We first evaluated the antimicrobial activity of Cst and other peptides derived from granins to determine which would exhibit maximal antimicrobial activity against common skin pathogens. Synthetic peptides derived from CHGA, CgB, and SgII (Table 1) were tested against relevant skin microbes, including Gram-positive bacteria, (Staphylococcus aureus and Group A Streptococcus (GAS)), Gram-negative bacteria (E. coli and Pseudomonas aeruginosa), a yeast (Candida albicans), and filamentous fungi (Aspergillus niger, Aspergillus fumigatus, and Trichophyton rubrum). Of the peptides tested, only Cst had antimicrobial activity at less than 100 µm (Table 2). MIC (minimum inhibitory concentrations) for Cst ranged from 5 to 50 µm. Lethal concentrations ranged from 30 to 75 µm, with lowest activity against S. aureus ATCC 25923. Peptides based on naturally occurring human polymorphisms of Cst that contain non-synonymous amino-acid substitutions (Gly364Ser and Pro370Leu) (Wen et al., 2004) exhibited higher potency against GAS and S. aureus ΔmprF (Table 2). Thus, Cst and its naturally occurring polymorphisms exhibited a wide range of microbial targets.

Table 1
Synthetic peptides derived from human ChgA, ChgB, or SgII
Table 2
Cst and naturally occurring human variants G364S and P370L are antimicrobial against skin pathogens

To evaluate the mechanism of antimicrobial action by Cst, membrane activity of the peptide was examined directly. Membrane permeability was assessed over time using E. coli strain ML-35p, which constitutively expresses a plasmid-encoded periplasmic β-lactamase to assess disruption of the outer membrane and a cytoplasmic β-galactosidase reporter protein to assess disruption of the inner membrane. These bacteria showed rapid inner membrane disruption in the presence of 1 µg ml−1 Polymyxin B (Figure 1a) or 20 µm human Cst (Figure 1b) that was comparable to the human cathelicidin LL-37. In E. coli, killing accompanied disruption of the inner and outer membrane as an increase in nitrocefin and a decrease in the red product was seen in cells treated with Cst (data not shown). Carbonate is a critical ionic factor found in mammalian tissues that enhances microbial sensitivity to AMPs at physiological NaCl concentrations (Dorschner et al., 2006). Membrane permeability to Cst was affected by the presence of NaHCO3 in a manner similar to cathelicidin (Figure 1a and b). Electron microscopy confirmed rapid disruption of E. coli by Cst (Figure 1c and f) within 10 minutes, with visible membrane blebbing compared to untreated cells (Figure 1d and g). These membrane changes were similar to that induced by incubation with LL-37 (Figure 1e and h). Further membrane disruption of E. coli by Cst was also observed in the presence of carbonate (data not shown). The membrane permeating effects of Cst were selective for bacterial membranes over mammalian plasma membranes, as Cst induced less than 1% hemolysis of human red blood cells at concentrations up to 100 µm. Additionally, the antimicrobial activity of wild-type Cst and its polymorphisms was enhanced in the presence of carbonate against both Gram-positive and Gram-negative bacteria (Table 3).

Figure 1
Cst acts as an antimicrobial through membrane disruption
Table 3
The presence of carbonate increases the antimicrobial activity of wild-type Cst and its naturally occurring human variants G364S and P370L

CHGA is expressed in the skin by keratinocytes

Given that CHGA-derived peptides were antimicrobial against a variety of skin pathogens, we sought to determine whether CHGA was also present in the epidermis similar to other AMPs (Koljonen et al., 2005). In normal human skin, Cst staining was seen in the suprabasal and granular keratinocytes within the epidermis, but not in the stratum corneum, and to a lesser extent in the dermis (Figure 2 a and b). The specificity of the Cst antibody was demonstrated by incubation with pre-immune serum (data not shown) and pre-incubation of the antibody with synthetic Cst, which reduced the immunoreactivity to levels observed with IgG control antibody (Figure S1). To confirm the resident skin cells responsible for immunofluorescent staining in skin, we performed a quantitative analysis of mRNA from whole human skin and cultured cells (Figure 2c). Chga transcript was abundantly present in whole skin, and detectable in epidermal keratinocytes at lower levels, but more abundantly than in fibroblasts. Cultured mouse connective tissue mast cells and human lipid-rich sebaceous cells were also evaluated for the presence of CHGA by PCR and immunostaining but were negative (data not shown). High-Chga-expressing SK-N-SH cells derived from a human neuroblastoma served as a positive control. Thus, these data represent the first identification of CHGA in keratinocytes, and supports the potential role for CHGA in cutaneous immune defense.

Figure 2
Epidermal keratinocytes express Chga

CHGA is processed to Cst in skin

We next asked whether CHGA was processed into the Cst peptide in skin and could thus, by virtue of its antimicrobial activity, partially explain its presence in the epidermis as a first line of defense against potential skin pathogens. Further purification of skin extracts subjected to HPLC analysis identified a small peak with mobility in HPLC identical to synthetic Cst (Figure 3a). These fractions also had maximal reactivity to an antibody against Cst (Figure 3b). ESI-MS (Electrospray ionization mass spectrometry) of this fraction defined its mass as 2,326 Da, exactly corresponding to the predicted molecular weight of human Cst. Based on the relative immunoreactivity against synthetic peptide standards, the abundance of Cst peptide in human skin was determined to be approximately 600 ng mg−1 of total protein. Quantitation of Cst in normal murine skin by immunoblot was determined to be around 20 µm within an approximate observed region of expression. Combined, these data support the conclusion that Cst, a presumptive AMP derived from the CHGA protein, is endogenously present in the skin.

Figure 3
CHGA is processed into Cst in skin

Cst is upregulated in skin following injury and infection

As cutaneous injury or infection promotes an increase in several known AMPs (Dorschner et al., 2001; Nizet et al., 2001; Ong et al., 2002; Fluhr et al., 2006), it was of interest to determine whether CHGA expression could also respond to injury. Murine skin was subjected to consecutive tape stripping and immunostained with an antibody against the CHGA-derived neuropeptide Cst. Skin sections from these mice exhibited expression of CHGA in the epidermis and dermis that rapidly increased after injury (Figure 4a–d). Cst staining was localized primarily within the suprabasal keratinocytes, the stratum corneum of the epidermis, and in dendritic appearing cells in the dermis. Occlusion applied immediately after barrier disruption blocked the increase in Cst immunoreactivity.

Figure 4
CHGA is upregulated following barrier disruption and infection

To determine if this increase in Cst immunostaining was preceded by an increase in AMP gene expression, normal skin and skin subjected to tape stripping (+/− occlusion) was excised and assessed for the expression of both Chga and the mouse cathelicidin, Camp, by quantitative PCR. An increase in gene expression of both Camp (Figure 4e) and Chga (Figure 4f) was observed, but this increase was not statistically significant, suggesting that the observed increase in protein expression seen in Figure 4a–d may reflect release of pre-stored peptide or amplification of a small increase in gene expression. To determine if Chga expression in the skin may increase in the presence of infection, normal and GAS-infected skin was excised and assessed for the expression of both Chga and Camp by quantitative PCR. In lesional skin, the expression of both Camp (Figure 4g) and Chga (Figure 4h) was significantly greater compared to normal, non-lesional skin. These data establish that Chga is inducible by injury or infection similarly to classical AMPs, potentially through different mechanisms.


CHGA is widely distributed within neuronal and neuroendocrine tissues and is a marker for neuroendocrine tumors (Weiler et al., 1988; Taupenot et al., 2003). The absence of CHGA leads to hypertension in mice, a response consistent with the role of Cst as a nicotinic cholinergic antagonist of catecholamine release (Mahapatra et al., 2005). Neuropeptides and the neuroendocrine system have also been hypothesized to be regulators of cutaneous immunity, although direct evidence for this is lacking (Fox et al., 1997; Ding et al., 2007). We now confirm that CHGA/Cst is expressed by keratinocytes and is upregulated following injury and infection. Furthermore, we show that CHGA is processed into the peptide Cst in skin, can act as an AMP against skin pathogens, and functions to kill bacteria in ways similar to other known AMPs in the skin. These findings suggest that a more direct link exists between antimicrobial defense of the skin and the neuroendocrine system.

CHGA/Cst was detected in the epidermis by immunostaining and increased in both the epidermis and the dermis following injury, suggesting that Cst could directly add to AMP activity in the skin. Expression of Chga was confirmed in isolated keratinocytes, but immunostaining suggested that neural elements in the dermis also contribute to the CHGA expression in whole skin. An additional cell of interest to consider is the poorly understood Merkel cell. These cells are known to express Chga, which is consequently used as a marker for detection of Merkel cell tumors (Koljonen et al., 2005). Our observation of CHGA function in skin also provides a new hypothesis for an immunological role for the Merkel cell. Thus, we speculate that the keratinocyte may act as a reservoir for CHGA storage and are induced to release CHGA into the extracellular environment following tissue damage or infection, whereas the Merkel cell may serve to impart a constitutive cutaneous immune surveillance.

In antimicrobial assays, human Cst (CHGA352–372) exhibited activity against multiple organisms and behaved similarly in vitro to other AMPs. For example, some AMPs kill optimally when carbonate is present, mimicking ionic conditions in vivo (Dorschner et al., 2006). Cst behaved similarly, as Gram-negative and Gram-positive bacteria become more susceptible to Cst in the presence of carbonate. In addition, enhanced antimicrobial activity of Cst was observed against the mutant S. aureus ΔmprF. The membrane of this mutant lacks lysylphosphatidylglycerol, which results in a more negative charge of the membrane surface (Peschel et al., 2001; Oku et al., 2004). As a consequence of this, S. aureus Δ mprF has an increased affinity for cationic peptides and displays a greater susceptibility to AMPs that require ionic interactions with the membrane (Peschel et al., 2001; Ruzin et al., 2003; Nishi et al., 2004). Thus, the current results suggest that Cst exerts its effect by a mechanism requiring electrostatic interaction with the membrane. Support for a membrane-active model of activity can also be derived from the increase in potency observed in the Cst variants. Homology modeling suggests that both substitutions distort the peptide backbone in the middle loop by 0.36Å or 0.59Å (Gly364Ser or Pro370Leu, respectively; Wen et al., 2004). This shift in the middle loop can increase the capacity of the peptide to interact with the membrane, explaining the increased activity of mutants compared to wild-type Cst. These observations of antimicrobial activity for Cst against skin pathogens are supported by prior in vitro analysis of its antimicrobial activity against other microorganisms (Briolat et al., 2005).

Whereas the minimum effective antimicrobial concentrations of Cst was 5µm or greater, the typical circulating plasma concentrations of CHGA (Takiyyuddin et al., 1990) and Cst (O’Connor et al., 2002) are in the nanomolar range. As the sequence of Cst is conserved across species, particularly between mouse and human, it is likely that the AMP activity of mouse Cst is comparable to human Cst. In these skin studies, the amount of Cst in murine epidermis was determined to be approximately 20 µm in its approximate observed area of expression, suggesting that there is sufficient amount of Cst locally to contribute to immune defense. However, two main factors make it difficult to determine the precise concentration of an active AMP within a designated tissue. First, the amount of the specific active proteolytic fragment, such as Cst, may be overestimated as the antibody recognizes many forms of CHGA containing the Cst sequence. Second, volume of the compartment in which it is distributed, and thus the denominator necessary to calculate the concentration of peptide, is difficult to accurately measure. Owing to these difficulties, it is not possible to definitively conclude that the local concentration of Cst exceeds that necessary to directly kill bacteria.

The abundant increase in CHGA/Cst following injury and infection supports its role to directly act as an AMP after injury. The abundance of mRNA of both Chga and Camp was increased following infection, but not significantly increased following barrier disruption. However, an increase in protein expression following tape stripping was seen and this was blocked by Latex occlusion. Cst is stored in secretory granules of the adrenal medulla, whereas cutaneous AMPs, such as cathelicidin, are partially stored in lamellar bodies and Golgi apparatus within keratinocytes (Eskeland et al., 1996; Braff et al., 2005). Both CHGA and cathelicidin are processed and then released from these secretory granules during periods of stress, for example, during infection and/or injury. Furthermore, the artificial restoration of the epidermal barrier using Latex occlusion immediately following barrier disruption by tape stripping inhibited the increase in Cst, indicating that barrier integrity regulates the release of Cst. Thus, injury may induce sufficient amounts of pre-stored Cst to be released into the microenvironment, whereas in the presence of infection, the increase in gene expression may provide a secondary defense mechanism to ensure that sufficient AMP activity by Cst is made available following depletion of pre-stored reservoirs. After injury, Cst would be available to provide an immediate defense against microbial pathogens, making the antibacterial concentrations observed in vitro physiologically relevant. However, it clearly appears that the abundance of Cst in skin is relatively low and its antimicrobial potency is not as great as other cutaneous AMPs. Therefore, it remains to be determined if the expression of Cst in skin provides a considerable increase in cutaneous protection related to other innate immune defense mechanisms.

CHGA has been reported to regulate the endothelial barrier following inflammation to protect against vascular leakage caused by tumor necrosis factor-α (Ferrero et al., 2004). In the skin, expression of CHGA may play a similar role during pathological states, such as infection, when tumor necrosis factor-α levels are elevated. Other studies have demonstrated that stress results in the mobilization of immune cells (Viswanathan et al., 2005; Atanackovic et al., 2006) and CHGA is known to be expressed in response to stress (Jiang et al., 2001). Furthermore, recent observations have identified direct links between glucocorticoid production and innate immune function of the skin barrier. (Aberg K, Gallo RL, Mauro TM, Feingold KR, Elias PM. Society for Investigative Dermatology, Meeting Abstract, 2006). Thus, further analysis is warranted to determine whether Cst may function in skin in a way other than acting as a direct AMP.

In conclusion, the current findings suggest that granins and their fragments contribute to host innate defenses against microbes, with Cst possibly acting directly as an AMP. The observation of expression of a secreted neuroendocrine molecule by keratinocytes, its induction within the epidermis following injury, its dependence upon barrier integrity, and its ability to act as both a natural antibiotic and a neuropeptide provides new insight into the integral communication between the neural and cutaneous immune system. These results indicate that CHGA/Cst may be a key factor of the local innate defense of skin. Further investigation is essential to determine how CHGA/Cst participates in cutaneous immunity. Potentially, the stress response associated with the adaptive fight-or-flight state could influence the cutaneous immune system through CHGA to alter the risk of infection. Ultimately, these previously unidentified observations indicate that the skin is a functioning neuroendocrine tissue that can modulate inherent antimicrobial activity.



Peptides were from Imgenex (San Diego, CA) and 95% purity confirmed by mass spectrometry.


S. aureus Rosenbach ATCC 25923 (ATCC, American Type Culture Collection, Manassas, VA), S. aureus ΔmprF, the enteroinvasive E. coli O29, E. coli ATCC 25922, P. aeruginosa, and GAS were used. Yeast used was C. albicans ATCC 14053. Fungi used were clinical isolates of A. niger, A. fumigatus, and T. rubrum.

Bacteria were grown at 37 °C in tryptic soy broth (TSB; Sigma, St Louis, MO) to stationary phase and a 10 × suspension at 1–3 × 106 colony-forming units (CFU) ml−1 prepared in 1 mm NaPB (sodium phosphate buffer), pH 7. Yeast was grown at 37 °C in a modified Dixon made of 4% malt extract (Fluka Biochemika, Steinheim, Germany), 0.6% Bacto Peptone (Becton Dickinson, Sparks, MD), 1% glucose (Sigma-Aldrich Co., St Louis, MO) and 1% Tween 80 (Sigma-Aldrich Co.) and a 10 × suspension at 3–6 × 105 CFU ml−1 was prepared in 1 mm NaPB. Fungi were routinely cultured on potato dextrose agar (Becton Dickinson, Sparks, MD) plates for 5–10 days at 30 °C. Conidia were collected from agar, filtered, and titrated to a concentration of 106 conidia per ml in 1 mm NaPB.

In vitro antimicrobial activity assays

Microorganisms were grown in 100 µl in sterile 96-well microtiter plates (Corning Inc., Corning, NY). The assay mixture contained 1mm sodium phosphate buffer, 10 µl of a 10 × suspension of each microorganism, 20% of culture media (TSB for bacteria, modified Dixon media for yeast, or potato dextrose agar media for fungi as described above), and 10 µl of a 10 × solution of each synthetic peptide (final concentrations ranging from 5 to 100 µm). For fungi or yeast, the assay also contained 16 µg ml−1 chloramphenicol per well. The plates were incubated at 30 °C for yeast and fungi or 37 °C for bacteria, and the growth was determined by measure of A600 over time with a Spectra max PLUS 384 (Molecular Devices, Sunnyvale, CA) microplate reader. In each experiment, the blank mean A600 value from mock inoculations was subtracted from each well A600 measurement to calculate the mean and SD.

The MIC was defined as the lowest peptide concentration that showed at least 95% inhibition of growth at the end of the experiment (after 1 day for bacteria, 2 days for yeast or fungi). In the case of some fungi, the MIC was determined as the lowest peptide concentration that prevented visual growth. Aliquots of each pathogen-peptide combination were also plated in agar to determine the lethal concentration defined as the lowest peptide concentration enabling <1% of growth recovered after treatment.

Hemolysis assay

Hemolytic activity was determined from human whole blood, freshly obtained and washed and resuspended in phosphate-buffered saline (PBS) with peptides at different concentrations. Blood was obtained following informed consent and as approved by the UCSD Human Research Protection Program protocol #031263. Samples were incubated at 37 °C for 2 hours and hemolysis determined by measurement of absorbance at 410 and 578nm. The hemolytic activity of each peptide was expressed as the percentage of total hemoglobin released compared with that released by incubation with 0.1% Triton X-100 as a positive control.

Membrane permeability

E. coli strain ML-35p was used to measure inner and outer membrane permeability. This strain constitutively expresses a plasmid-encoded periplasmic β-lactamase and cytoplasmic β-galactosidase, but lacks lactose permease. After overnight incubation at 37°C in TSB, cultures were re-inoculated in buffer A (1 mm NaH2PO4, 20% TSB) or buffer B (1 mm NaH2PO4, 20% TSB and 25 mm NaHCO3) for 3 hours. The subculture was then adjusted to 2.5–5 × 107 CFU ml−1 (A600 = 0.03) in buffer A or B.

To test permeabilization of the inner membrane, formation of ONP (o-nitrophenol) as the hydrolytic product of β-galactosidase on the colorless substrate ONPG (o-nitrophenyl-β-d-galactosidase) was determined by detecting in increase in absorbance at 420 nm (A420). To study permeabilization of the outer membrane, we followed a decrease in A390 due to the substrate nitrocefin and an increase in A486 due to the red product. The assay mixture contained 43 µl of bacteria from stock cultured in buffer A or B, with final concentrations of 4 µm LL-37, 20 µm Cst, 0.3 mg ml−1 ONPG, or 25 µg ml−1 nitrocefin. The concentrations of LL-37 and Cst used in this assay was determined as the optimal concentration which resulted in little to no toxicity while retaining maximal activity with this strain of E. coli. To observe bacterial growth, the A600 was also monitored. For the assay, samples were incubated at 37 °C in sterile 96-well microtiter plates (Corning Inc.) and the kinetics of β-galactosidase or β-lactamase activity calculated by the measurement of the A420 or A390/486, respectively. Absorbance was determined using a SpectraMax PLUS 384 device (Molecular Devices). As a positive control, bacteria were permeabilized with 1 µg ml−1 polymyxin B (Calbiochem, La Jolla, CA) and 1% Triton X-100 in buffer A or B.

Electron microscopy

Electron microscopy was performed on E. coli (ATCC 25922) grown to log phase in 20% TSB, 1 mm NaH2PO4. LL-37 (32 µm, positive control), Cst (50 µm), or an equal volume of water (negative control) was added to 108 bacteria and cultured for 10 minutes. Bacteria were pelleted at 1,000 g for 2 minutes at 4 °C, immersed in 4 °C 5% phosphate-buffered glutaraldehyde, pH 7.8, rinsed in cold phosphate buffer, post-fixed in 4% phosphate-buffered osmic acid for 2 hours, dehydrated in acetone, and embedded. Samples were processed for electron microscopy on a Zeiss EM transmission electron microscope.

Immunohistochemistry and three-dimensional imaging on human skin

Immunohistochemistry was performed on healthy human skin that was immediately embedded in OCT (optimal cutting temperature) compound (Sakura Finetechnical Co., Tokyo, Japan). Sections (10 µm) were fixed in 10% buffered formalin and then washed with PBS. Sections were blocked with 2% goat serum/3% BSA (Sigma) in PBS and incubated with a polyclonal rabbit anti-Cst antibody (1:1,000 in PBS containing 3% BSA). As a negative control, normal rabbit IgG was diluted with PBS at the same concentration as primary antibody. Slides were washed in PBS and then incubated with goat anti-rabbit IgG/Tetra Methyl Rhodamine Isothiocyanate (TRITC)-conjugated IgG/(Sigma, work solution 1:200 in PBS containing 3% BSA).

Images were captured on a DeltaVision deconvolution microscopy system (Applied Precision, Issaquah, WA) operated by SoftWoRx software (Applied Precision) on a Silicon Graphics O2 workstation using × 60 (NA 1.4) or × 100 (NA 1.4) oil immersion objectives. The fluorescent data sets were deconvoluted and analyzed by DeltaVision SoftWoRx proGrams (Applied Precision) on a Silicon Graphics Octane workstation to generate optical sections or three-dimensional images of the data sets.

Cell culture

Normal human keratinocytes isolated from foreskin were cultured in EpiLife medium (serum-free keratinocyte medium; Cascade Biologics, Portland, OR) supplemented with 0.06 mm calcium, EDGS (EpiLife defined growth supplement), and penicillin/streptomycin. Fibroblasts ATS F12 were cultured in DMEM—high glucose, no potassium azida medium (Cambrex, Walkersville, MD) supplemented with 10% bovine serum, 1% l-glutamine, 1% MEM (minimum essential medium)—non-essential amino acids and 1% penicillin/streptomycin (all purchased from Invitrogen Corporation, Carlsbad, CA). Propagation of human neuroblastoma SK-N-SH cells was performed according to the provider instructions (ATCC).

Quantitative real-time PCR for human skin, keratinocytes, and fibroblasts

Skin was biopsied and homogenized directly in Trizol reagent. RNA from skin was extracted per the manufacturer’s instructions. Total RNA from cells was isolated using Qiagen RNeasy Mini Kit (Qiagen, Valencia, CA). Reverse transcription of RNA isolated from skin tissue or cells into cDNA was achieved using Retroscript kit (Ambion, Austin, TX), following manufacturer’s instructions.

Quantitative real-time PCR was performed using an Applied Biosystems 7000 Sequence Detection System (Foster City, CA). For human Chga, primers were designed for the target gene human Chga (hChga) and GAPDH (glyceraldehyde-3-phosphate dehydrogenase) as an endogenous control. The sequences of the hChga primers were forward TGTAGTGCTGAACCCCCACC, reverse CTCTCGCCTTTCCGGATCT (product size 57 bp). These primers bind in the base 966 for the forward and 1,004 for the reverse of the cDNA NM_001275. The sequences of the GAPDH primers were forward CTTAGCACCCCTGGCCAAG, reverse TGGTCATGAGTCCTTCCACG (product size 62 bp). Five microliters of cDNA was added to 12.5 µl SYBR Green PCR Master Mix (Applied Biosystems), 0.25 µl of each 20 µm primer, and 7 µl of RNase/DNase-free H2O per reaction. Thermal profile is as follows: 50 °C 2 minutes, 95 °C 10 minutes, 40 × (94 °C 15 seconds, 60 °C 1 minute). Results were analyzed using the comparative Ct method (User Bulletin no. 2, Applied Biosystems).

HPLC purification and immunoblot of human skin extracts

Human skin biopsies obtained following informed consent and as approved by the University of California, San Diego Human Research Protection Program, protocol no. 031263, were extracted and fractionated by HPLC as described below.

Peptide separation was performed using an AKTA purification system (Amersham Pharmacia Biotech, Pistacaway, NJ) on a Sephasil peptide C18 column (12 µm, ST 4.6/250) (Amersham Pharmacia Biotech). HPLC was equilibrated in 0.1% trifluoroacetic acid at a flow rate of 0.5 ml minute−1 and eluted using a gradient of 0–100% acetonitrile, and monitored at 214, 230, and 280 nm. All collected fractions (1 ml) were lyophilized and then resuspended in 15 µl of sterile double-distilled H2O (Life Technologies, Carlsbad, CA).

For quantification of Cst, fractions were diluted in TBST (Tris-buffered saline-1% Tween), pH 7.0, transferred to a polyvinylidene fluoride membrane, then incubated with anti-rodent Cst antibody (1:1,000) or anti-human Cst antibody (1:500) in TBST/5% milk/5%BSA overnight at 4 °C. The membrane was washed and incubated with horseradish peroxidase-conjugated goat anti-rabbit antibody for 1 hour at room temperature and developed with ECL reagent (Amersham Pharmacia Biotech). The relative immunoreactivity was determined by ChemiImager version 5.5 software.


For all experiments, age-matched (8–10 weeks) and sex-matched adult littermates were used. All procedures performed in accordance with the Veterans Affairs San Diego and Veterans Affairs San Francisco Healthcare System subcommittee on animal studies.

Barrier disruption

Epidermal barrier disruption was induced by sequential cellophane tape stripping on the flanks of female Skh-1 mice as described previously (Wood et al., 1992). In some experiments, animals were occluded with a tightly fitted water vapor-impermeable latex membrane to artificially restore barrier function. Briefly, skin biopsies were taken from treated animals at baseline and 30 minutes, 1 hours, and 3 hours after barrier disruption and were immediately embedded in OCT compound and snap-frozen in liquid nitrogen (Sakura Finetechnical Co.).

Immunohistochemistry on mouse skin following barrier disruption

Tissue sections (10 µm) were fixed in 100% acetone and rinsed in PBS. Sections were incubated for 30 minutes in blocking buffer (4% BSA, 0.5% cold water fish gelatin in PBS) and incubated over night at 4 °C with a 1:3,000 dilution of primary antibody in blocking buffer. Sections were washed and incubated for 40 minutes at room temperature with an Alexa Fluor 488-conjugated goat anti-rabbit secondary antibody. Tissue sections were counterstained with propidium iodide and visualized on a Leica TCS-SP confocal microscope. Photos were taken at × 40 magnification.

GAS infection

GAS infection was performed as described previously (Betschel et al., 1998). Hair was removed from the back of anesthetized male 129/SvJ mice by plucking. Skin was then injected with 50 µl of a mid-logarithmic growth phase (A600 = 0.8, 4.8 × 108 CFU ml−1) of GAS NZ131 conjugated with 50 µl of sterile Cytodex beads. Skin from noninfected mice was used as a control. Normal or infected skin was excised using a 8 mm biopsy punch.

Quantitative real-time PCR for normal, tape-stripped, or GAS-infected mouse skin

Normal, tape-stripped (+/− occlusion), or lesional skin was excised and RNA isolated as described above. Quantitative PCR for mouse cathelicidin gene (CRAMP) was analyzed as described above using a FAM-CAGAGGATTGTGACTTCA-MGB probe with primers 5′-CTTCACCAGCCCGTCCTTC-3′ and 5′-CCAGGACHGACACAGCAGTCA-3′ and normalized to GAPDH using primers described above. A predeveloped Taqman assay probe (ABI, Foster City, CA) was used for the analysis of mouse Chga.

Measurement of Cst in murine epidermis

Normal murine skin was excised, minced, and incubated in dispase (Boehringer-Manheim, Manheim, Germany) at a concentration of 5mg ml−1 in 20 mm HEPES (4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid) buffer overnight at 4 °C. Epidermis was removed, homogenized in 1 × RIPA (radioimmunoprecipitation assay) buffer, and centrifuged at 1,000 g for 15 minutes at 4 °C. Epidermal supernatant was then blotted onto a polyvinylidene fluoride membrane along with Cst standards. The membrane was blocked and incubated with anti-Cst antibody (1:1,000) at room temperature. Blot was then incubated with goat anti-rabbit horseradish peroxidase and developed with ECL (enhanced chemoluminescence) reagent (Perkin-Elmer, Boston, MA). Epidermal extract reactivity was determined by comparing to reactivity of known concentrations of Cst ranging from 500 nm to 100 mm. To estimate the concentration of peptide in the epidermis, the calculated molar mass of peptide reacting to anti-Cst antibody was divided by an estimated volume of tissue identified to contain Cst by immunohistochemistry. For the area of skin excised, this volume was 18 mm3.

Statistical analysis

Data were analyzed using GraphPad Prism, version 2.1. (GraphPad Software Inc., San Diego, CA). The means and SD were calculated for each data set. Data were analyzed by an unpaired Student’s t-test. P-values <0.05 were considered significant.

Supplementary Material



Materials and Methods


Figure S1. Synthetic Cst blocks immunoreactivity of anti-Cst antibody.

Click here to view.(4.3K, NIHMS88591-supplement-FigS1)


This work was supported by the Department of Veterans Affairs and the National Institutes of Health Grants NIH/NIAID HHSN26620040029C, ADB contract nos. N01-AI-40029AI48176, AI052453, and AR45676. Belen Lopez-Garcia was recipient of a post-doctoral fellowship from Fundacion Ramon Areces (Spain). We thank Dr Marilyn Farquhar of the Electron Microscopy facility of the Department of Cellular and Molecular Medicine, UCSD for their assistance with the electron microscopy. We also sincerely thank Robert Dorschner and Kenshi Yamasaki for their technical advice and expertise.


antimicrobial peptide
chromogranin A
Group A Streptococcus
phosphate-buffered saline
tryptic soy broth



The authors state no conflict of interest.


  • Atanackovic D, Pollok K, Faltz C, Boeters I, Jung R, Nierhaus A, et al. Patients with solid tumors treated with high-temperature whole body hyperthermia show a redistribution of naive/memory T-cell subtypes. Am J Physiol Regul Integr Comp Physiol. 2006;290:R585–R594. [PubMed]
  • Bals R, Wang X, Zasloff M, Wilson JM. The peptide antibiotic LL-37/hCAP-18 is expressed in epithelia of the human lung where it has broad antimicrobial activity at the airway surface. Proc Natl Acad Sci USA. 1998;95:9541–9546. [PMC free article] [PubMed]
  • Banks P, Helle K. The release of protein from the stimulated adrenal medulla. Biochem J. 1965;97:40C–41C. [PMC free article] [PubMed]
  • Bardan A, Nizet V, Gallo RL. Antimicrobial peptides and the skin. Expert Opin Biol Ther. 2004;4:543–549. [PubMed]
  • Betschel SD, Borgia SM, Barg NL, Low DE, De Azavedo JC. Reduced virulence of group A streptococcal Tn916 mutants that do not produce streptolysin S. Infect Immun. 1998;66:1671–1679. [PMC free article] [PubMed]
  • Braff MH, Di Nardo A, Gallo RL. Keratinocytes store the antimicrobial peptide cathelicidin in lamellar bodies. J Invest Dermatol. 2005;124:394–400. [PubMed]
  • Briolat J, Wu SD, Mahata SK, Gonthier B, Bagnard D, Chasserot-Golaz S, et al. New antimicrobial activity for the catecholamine release-inhibitory peptide from chromogranin A. Cell Mol Life Sci. 2005;62:377–385. [PubMed]
  • Chromek M, Slamova Z, Bergman P, Kovacs L, Podracka L, Ehren I, et al. The antimicrobial peptide cathelicidin protects the urinary tract against invasive bacterial infection. Nat Med. 2006;12:636–641. [PubMed]
  • Cutuli M, Cristiani S, Lipton JM, Catania A. Antimicrobial effects of alpha-MSH peptides. J Leukoc Biol. 2000;67:233–239. [PubMed]
  • Devine DA, Hancock RE. Cationic peptides: distribution and mechanisms of resistance. Curr Pharm Des. 2002;8:703–714. [PubMed]
  • Dillen L, Miserez B, Claeys M, Aunis D, De Potter W. Posttranslational processing of proenkephalins and chromogranins/secretogranins. Neurochem Int. 1993;22:315–352. [PubMed]
  • Ding W, Wagner JA, Granstein RD. CGRP, PACAP, and VIP modulate Langerhans cell function by inhibiting NF-kappaB activation. J Invest Dermatol. 2007;127:2357–2367. [PubMed]
  • Dorschner RA, Lopez-Garcia B, Peschel A, Kraus D, Morikawa K, Nizet V, et al. The mammalian ionic environment dictates microbial susceptibility to antimicrobial defense peptides. FASEB J. 2006;20:35–42. [PubMed]
  • Dorschner RA, Pestonjamasp VK, Tamakuwala S, Ohtake T, Rudisill J, Nizet V, et al. Cutaneous injury induces the release of cathelicidin antimicrobial peptides active against group AStreptococcus. J Invest Dermatol. 2001;117:91–97. [PubMed]
  • Elenkov IJ, Webster EL, Torpy DJ, Chrousos GP. Stress, corticotropin-releasing hormone, glucocorticoids, and the immune/inflammatory response: acute and chronic effects. Ann N Y Acad Sci. 1999;876:1–11. discussion 11–13. [PubMed]
  • Eskeland NL, Zhou A, Dinh TQ, Wu H, Parmer RJ, Mains RE, et al. Chromogranin A processing and secretion: specific role of endogenous and exogenous prohormone convertases in the regulated secretory pathway. J Clin Invest. 1996;98:148–156. [PMC free article] [PubMed]
  • Ferrero E, Scabini S, Magni E, Foglieni C, Belloni D, Colombo B, et al. Chromogranin A protects vessels against tumor necrosis factor alpha-induced vascular leakage. FASEB J. 2004;18:554–556. [PubMed]
  • Fischer-Colbrie R, Laslop A, Kirchmair R. Secretogranin II: molecular properties, regulation of biosynthesis and processing to the neuropeptide secretoneurin. Prog Neurobiol. 1995;46:49–70. [PubMed]
  • Fluhr JW, Feingold KR, Elias PM. Transepidermal water loss reflects permeability barrier status: validation in human and rodent in vivo and ex vivo models. Exp Dermatol. 2006;15:483–492. [PubMed]
  • Fox FE, Kubin M, Cassin M, Niu Z, Hosoi J, Torii H, et al. Calcitonin gene-related peptide inhibits proliferation and antigen presentation by human peripheral blood mononuclear cells: effects on B7, interleukin 10, and interleukin 12. J Invest Dermatol. 1997;108:43–48. [PubMed]
  • Harder J, Bartels J, Christophers E, Schroder JM. Isolation and characterization of human beta-defensin-3, a novel human inducible peptide antibiotic. J Biol Chem. 2001;276:5707–5713. [PubMed]
  • Harder J, Siebert R, Zhang Y, Matthiesen P, Christophers E, Schlegelberger B, et al. Mapping of the gene encoding human beta-defensin-2 (DEFB2) to chromosome region 8p22-p23.1. Genomics. 1997;46:472–475. [PubMed]
  • Iimura M, Gallo RL, Hase K, Miyamoto Y, Eckmann L, Kagnoff MF. Cathelicidin mediates innate intestinal defense against colonization with epithelial adherent bacterial pathogens. J Immunol. 2005;174:4901–4907. [PubMed]
  • Jiang Q, Taupenot L, Mahata SK, Mahata M, O’Connor DT, Miles LA, et al. Proteolytic cleavage of chromogranin A (CgA) by plasmin. Selective liberation of a specific bioactive CgA fragment that regulates catecholamine release. J Biol Chem. 2001;276:25022–25029. [PubMed]
  • Kennedy BP, Mahata SK, O’Connor DT, Ziegler MG. Mechanism of cardiovascular actions of the chromogranin A fragment catestatin in vivo. Peptides. 1998;19:1241–1248. [PubMed]
  • Koljonen V, Haglund C, Tukiainen E, Bohling T. Neuroendocrine differentiation in primary Merkel cell carcinoma—possible prognostic significance. Anticancer Res. 2005;25:853–858. [PubMed]
  • Lee RW, Huttner WB. Tyrosine-O-sulfated proteins of PC12 pheochromocytoma cells and their sulfation by a tyrosylprotein sulfotransferase. J Biol Chem. 1983;258:11326–11334. [PubMed]
  • Lugardon K, Raffner R, Goumon Y, Corti A, Delmas A, Bulet P, et al. Antibacterial and antifungal activities of vasostatin-1, the N-terminal fragment of chromogranin A. J Biol Chem. 2000;275:10745–10753. [PubMed]
  • Mahapatra NR, O’Connor DT, Vaingankar SM, Hikim AP, Mahata M, Ray S, et al. Hypertension from targeted ablation of chromogranin A can be rescued by the human ortholog. J Clin Invest. 2005;115:1942–1952. [PMC free article] [PubMed]
  • Mahata SK, Mahata M, Wen G, Wong WB, Mahapatra NR, Hamilton BA, et al. The catecholamine release-inhibitory “catestatin” fragment of chromogranin a: naturally occurring human variants with different potencies for multiple chromaffin cell nicotinic cholinergic responses. Mol Pharmacol. 2004;66:1180–1191. [PubMed]
  • Mahata SK, Mahata M, Yoo SH, Taupenot L, Wu H, Aroda VR, et al. A novel, catecholamine release-inhibitory peptide from chromogranin A: autocrine control of nicotinic cholinergic-stimulated exocytosis. Adv Pharmacol. 1998;42:260–264. [PubMed]
  • Mahata SK, O’Connor DT, Mahata M, Yoo SH, Taupenot L, Wu H, et al. Novel autocrine feedback control of catecholamine release. A discrete chromogranin a fragment is a noncompetitive nicotinic cholinergic antagonist. J Clin Invest. 1997;100:1623–1633. [PMC free article] [PubMed]
  • Metz-Boutigue MH, Goumon Y, Strub JM, Lugardon K, Aunis D. Antimicrobial chromogranins and proenkephalin-A-derived peptides: antibacterial and antifungal activities of chromogranins and proenkephalin-A-derived peptides. Ann N Y Acad Sci. 2003a;992:168–178. [PubMed]
  • Metz-Boutigue MH, Kieffer AE, Goumon Y, Aunis D. Innate immunity: involvement of new neuropeptides. Trends Microbiol. 2003b;11:585–592. [PubMed]
  • Murakami M, Lopez-Garcia B, Braff M, Dorschner RA, Gallo RL. Postsecretory processing generates multiple cathelicidins for enhanced topical antimicrobial defense. J Immunol. 2004;172:3070–3077. [PubMed]
  • Nishi H, Komatsuzawa H, Fujiwara T, McCallum N, Sugai M. Reduced content of lysyl-phosphatidylglycerol in the cytoplasmic membrane affects susceptibility to moenomycin, as well as vancomycin, gentamicin, and antimicrobial peptides. Staphylococcus aureus. Antimicrob Agents Chemother. 2004;48:4800–4807. [PMC free article] [PubMed]
  • Nizet V, Ohtake T, Lauth X, Trowbridge J, Rudisill J, Dorschner RA, et al. Innate antimicrobial peptide protects the skin from invasive bacterial infection. Nature. 2001;414:454–457. [PubMed]
  • O’Connor DT. Chromogranin: widespread immunoreactivity in polypeptide hormone producing tissues and in serum. Regul Pept. 1983;6:263–280. [PubMed]
  • O’Connor DT, Kailasam MT, Kennedy BP, Ziegler MG, Yanaihara N, Parmer RJ. Early decline in the catecholamine release-inhibitory peptide catestatin in humans at genetic risk of hypertension. J Hypertens. 2002;20:1335–1345. [PubMed]
  • Oku Y, Kurokawa K, Ichihashi N, Sekimizu K. Characterization of the Staphylococcus aureus mprF gene, involved in lysinylation of phosphatidylglycerol. Microbiology. 2004;150:45–51. [PubMed]
  • Ong PY, Ohtake T, Brandt C, Strickland I, Boguniewicz M, Ganz T, et al. Endogenous antimicrobial peptides and skin infections in atopic dermatitis. N Engl J Med. 2002;347:1151–1160. [PubMed]
  • Peschel A, Jack RW, Otto M, Collins LV, Staubitz P, Nicholson G, et al. Staphylococcus aureus resistance to human defensins and evasion of neutrophil killing via the novel virulence factor MprF is based on modification of membrane lipids with l-lysine. J Exp Med. 2001;193:1067–1076. [PMC free article] [PubMed]
  • Preece NE, Nguyen M, Mahata M, Mahata SK, Mahapatra NR, Tsigelny I, et al. Conformational preferences and activities of peptides from the catecholamine release-inhibitory (catestatin) region of chromogranin A. Regul Pept. 2004;118:75–87. [PubMed]
  • Ruzin A, Severin A, Moghazeh SL, Etienne J, Bradford PA, Projan SJ, et al. Inactivation of mprF affects vancomycin susceptibility in Staphylococcus aureus. Biochim Biophys Acta. 2003;1621:117–121. [PubMed]
  • Schauber J, Dorschner RA, Coda AB, Buchau AS, Liu PT, Kiken D, et al. Injury enhances TLR2 function and antimicrobial peptide expression through a vitamin D-dependent mechanism. J Clin Invest. 2007;117:803–811. [PMC free article] [PubMed]
  • Shimizu M, Shigeri Y, Tatsu Y, Yoshikawa S, Yumoto N. Enhancement of antimicrobial activity of neuropeptide Y by N-terminal truncation. Antimicrob Agents Chemother. 1998;42:2745–2746. [PMC free article] [PubMed]
  • Stolzenberg ED, Anderson GM, Ackermann MR, Whitlock RH, Zasloff M. Epithelial antibiotic induced in states of disease. Proc Natl Acad Sci USA. 1997;94:8686–8690. [PMC free article] [PubMed]
  • Takiyyuddin MA, Cervenka JH, Hsiao RJ, Barbosa JA, Parmer RJ, O’Connor DT. Chromogranin A. Storage and release in hypertension. Hypertension. 1990;15:237–246. [PubMed]
  • Taupenot L, Harper KL, O’Connor DT. The chromogranin–secretogranin family. N Engl J Med. 2003;348:1134–1149. [PubMed]
  • Viswanathan K, Daugherty C, Dhabhar FS. Stress as an endogenous adjuvant: augmentation of the immunization phase of cell-mediated immunity. Int Immunol. 2005;17:1059–1069. [PubMed]
  • Weiler R, Fischer-Colbrie R, Schmid KW, Feichtinger H, Bussolati G, Grimelius L, et al. Immunological studies on the occurrence and properties of chromogranin A and B and secretogranin II in endocrine tumors. Am J Surg Pathol. 1988;12:877–884. [PubMed]
  • Wen G, Mahata SK, Cadman P, Mahata M, Ghosh S, Mahapatra NR, et al. Both rare and common polymorphisms contribute functional variation at CHGA, a regulator of catecholamine physiology. Am J Hum Genet. 2004;74:197–207. [PMC free article] [PubMed]
  • Wood LC, Jackson SM, Elias PM, Grunfeld C, Feingold KR. Cutaneous barrier perturbation stimulates cytokine production in the epidermis of mice. J Clin Invest. 1992;90:482–487. [PMC free article] [PubMed]
PubReader format: click here to try


Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...


  • Compound
    PubChem Compound links
  • Gene (nucleotide)
    Gene (nucleotide)
    Records in Gene identified from shared sequence links
  • Nucleotide
    Published Nucleotide sequences
  • PubMed
    PubMed citations for these articles
  • Substance
    PubChem Substance links

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...