Isolation, characterization and regulation of moonlighting proteases from Candida glabrata cell wall

Candida glabrata (C. glabrata) cell wall proteins play a role in virulence and in initial host immune recognition and responses. We isolated and characterized C. glabrata cell wall proteases from a clinical hospital C. glabrata T1638 blood isolate and estimated the enzymatic activities and their ability to degrade gelatin and processing proMMP-8 and assess the regulation of these proteases with salt treatment, mercaptoethanol and fermented lingonberry juice from Vaccinium vitis idaea L. The cell wall proteases were enzymatically released from the cell wall and beta1,3bonded proteases were fractioned into 10–50 kDa and >50 kDa fractions with anionic DEAEsepharose ion-exchange chromatography and gel filtration. Proteins were monitored and analyzed with MDPFzymography, and five gelatinolytic bands were cut out from a parallel silver-stained gel for the LCMS/MS analysis. The proteases lacked a signal sequence, indicating that they are moonlighting proteases. Human proMMP-8 activation assays were performed with both fractions and verified by western-immunoblot using aMMP-8 specific antibody. Inhibition of proMMP-8 conversion to the lower molecular active enzyme species were demonstrated with fermented lingonberry juice. The results indicate that moonlighting proteases may play a role in the virulence of C. glabrata.


Introduction
Traditionally C. glabrata cell wall proteins are classified into GPIand-PIR-proteins [1,2]. C. glabrata has not been shown to secrete proteases, such as SAPs (secreted aspartic proteases) found from C. albicans, but certain studies show moonlighting proteins in the cell wall of C. albicans [3] and C. glabrata [4,5]. These intracellular proteins are secreted by a non-classical pathway in extracellular vesicles (ECVs) and are found in several microbial species attached to the cell wall [6][7][8]. The concept of intracellular proteins having an extracellular function has given an insight into protein molecules containing multiple functions compared to what is traditionally shown [9]. Studies with Saccharomyces cerevisiae also show cell wall proteins found both in the cytosol and fungal cell wall [10]. C. glabrata cell wall proteins are in the first contact line with the host and eventually can attribute to the virulence of the organism and host immune responses.

Discussion
Our results indicate, that C. glabrata possesses moonlighting proteases in the cell wall. Several moonlighting proteases have been identified from C. glabrata [21]. These findings corroborate with our results. Some findings indicate cell wall proteins of C. albicans having properties affecting virulence in challenged environments, such as starvation and temperature upshift [22]. This may be the case also in C. glabrata virulence. Glucose utilization is crucial for microbial growth and survival, e. g. the proper function of GDH (glutamate dehydrogenase) in Staphylococcus aureus gives it enhanced ability to use glutamate as an energy source in abscesses etc. with limited glucose availability [23].
The inactivation of the >50 kDa proteases with mercaptoethanol showed, that there are intra/intermolecular disulphide bridges crucial for the gelatinolytic activity. The salt treatment of the fraction indicates some ionic bond-related activation of the gelatinolysis.
Inactivation of the 10-50 kDa and >50 kDa fractions enzymatic activity with fermented lingonberry juice presents a potential therapeutical agent against C. glabrata virulence. Natural substances are being increasingly tested for antimicrobial properties. An interesting study has shown that mammalian glutamate dehydrogenase could be inhibited by green tea phenolics [24]. Fermented lingonberry juice also contains polyphenols, and further studies are needed to establish the inhibitory factor(s) from lingonberry. Recently, a clinical pilot study revealed the reduction of oral fluid aMMP-8 in the oral cavity by a lingonberry oral mouthwash intervention and management [25]. The gelatinolytic activity and activation of human proMMP-8 by C. glabrata cell wall associated moonlighting proteases may be directly involved in the local connective tissue breakdown as well as induction and modulation of the inflammatory immune responses in the affected tissues by the invading yeast. Inhibition of this activity by fermented lingonberry juice in vitro may have clinical implications in C. glabrata infection control. The anti-inflammatory properties of fermented lingonberry juice are to be further investigated.

Preparation of the C. glabrata cell wall fractions
C. glabrata T-1638 was is a blood isolate from a Helsinki University Hospital patient. By preliminary gelatin-zymographic scanning of several blood isolates T-1638 showed highest gelatinolytic enzymatic activity. The cell wall extract was prepared by the following methods.  (3095 g, RT). The supernatant was stored at − 20 • C and volume was restored with PBS. The cells were incubated a second time with lyticase similarly and the obtained supernatants were pooled (cell wall fraction). The cell wall fraction was centrifuged with 10 kDa and 50 kDa cutoff filters according to the manufacturer's instructions (Amicon® Ultra-15, Millipore, Billerica, MA), fraction volumes were restored with PBS and proteins were recovered from the filter to obtain 10-50 kDa and >50 kDa fractions.

Salt and mercaptoethanol (ME) treatments
To estimate ionic or disulfide bonds in the enzyme, 100 μl of the >50 kDa fraction was incubated in 1 ml 20 mM Tris-HCl, pH 8.2 + 0.5 M NaCl for 1 h, centrifuged with a Millipore 10-50 kDa cutoff device (3095 g, 10 min, RT) and washed twice. The flowthrough was also collected and concentrated 10 × with a Millipore 10 kDa cutoff device. Samples of the >50 kDa fraction and the salt-treated fractions were incubated with or without 2 μl of β-mercaptoethanol (final concentration = 1.5 mM) for 1 h (37 • C). Samples were run on 8% MDPF-gelatin zymography.

Ion-exchange chromatography
Tha buffer of the >50 kDa fraction was changed to 20 mM Tris-HCl, pH 8.2. Ion-exchange with diethylaminoethyl cellulose chromatography for the >50 kDa cell wall extract was performed according to the manufacturer's instructions (DEAE-FF, HiTrap® IEX Selection kit, GE Healthcare, Chicago, IL). The proteases were eluted with 0.5 M NaCl in 20 mM Tris-HCl, pH 8.2.

Gel filtration
A 15 cm column (V = 9.5 ml) was used for gel filtration. Sephadex G-100 filtration for both 10-50 kDa and >50 kDa fractions, and Sephadex G-200 filtration for the >50 kDa fraction were performed.
Parallel 8% SDS-PAGE with silver stain were run.

Western immunoblot
The detection of proMMP-8 conversion and inhibition was verified by a modified enhanced chemiluminesence (ECL) Western blotting kit according to the protocol recommended by the manufacturer (GE Healthcare, Buckinghamshire, UK). As specific primary and secondary antibodies were performed polyclonal anti-MMP-8 [27] and anti-rabbit IgG horseradish peroxidase-linked (GE Healthcare, Buckinghamshire, UK), respectively [12,28].

Mass analysis
Parallel 8% SDS-PAGE gels were silver-stained [29], protein bands  Mass analysis was performed as in El Omar et al. [30], but with a 6 μl sample size. Liquid chromatography coupled to tandem mass spectrometry (LC-MS/MS) analysis was carried out on an EASY-nLC1000 (Thermo Fisher Scientific, Darmstadt, Germany) connected to a Velos Pro-Orbitrap Elite hybrid mass spectrometer (Thermo Fisher Scientific, Germany) with nano electrospray ion source (Thermo Fisher Scientific, Germany analyzed. Full MS scan was acquired with a resolution of 60 000 at normal mass range in the orbitrap analyzer and followed with CID -MS2 top 20 most intense precursor ions within ion trap (energy 35). Data was acquired using LTQ Tune software. Acquired MS2 scans were searched against Uniprot Candida glabrata protein database using the Sequest search algorithms in Thermo Proteome Discoverer. Allowed mass error for the precursor ions was 15 ppm and for the fragment 0.8 Da. A static residue modification parameter was set for carbamidomethyl +57.021 Da (C) of cysteine residue. Methionine oxidation was set as dynamic modification +15.995 Da (M). Only full-tryptic peptides were allowed for maximum of 1 missed cleavages were considered.

Author contributions
The authors P⋅P, T.S, T.T. and P. N-I. participated in planning, analysis of the data and reviewing the manuscript. The author P.P. performed the sample preparation and other characterizations of C. glabrata cell wall proteases, except western immunoblots and proMMP-8 conversion and inhibition assays which were performed by the author T.T.

Funding
The study has been supported by grants from the Finnish Dental Society Apollonia, Finland; the Helsinki University Hospital Research Foundation (TYH 2016251, TYH 2017251, TYH 20188229, Y1014SLO17 and Y1014SLO18) Helsinki, Finland and Karolinska Institutet, Stockholm, Sweden.

Declaration of competing interest
The author P.P. is the inventor and holder of the patent EP 2 585 087 B1. The author T.S. is the inventor of US patents 5652227, 5866432, 6143476, 2017002357A1 and PI758USPC. The authors declare that there are no conflicts of interest.