2786 Biotechnol. & Biotechnol. eq. 26/2012/1

Article DOi: 10.5504/bbeq.2011.0097 Mb

Biotechnol. & Biotechnol. eq. 2012, 26(1), 2786-2793Keywords: influenza virus, viral inhibition, protease inhibitor, StreptomycesAbbreviations: AcA: ε-aminocaproic acid; BAPNA: N-bensoil-DL arginine-p-nitroanilide; BCA: bicinchoninic acid; BSA: bovine serum albumin; CD: circular dichroism; CPE: cytopathic effect; DSC: differential scanning calorimetry; ec50: 50% effective concentration; EC90: 90% effective concentration; ELISA: enzyme-linked immunosorbent assay; EPTT: 50% end point titration technique; FCS: foetal calf serum; HA: haemagglutinin; HPLC: high performance liquid chromatography; i.n.: intranasal; KIU: Kunitz inhibitory units; LD50: 50% lethal doses; M: mortality; MAb: monoclonal antibody; MDCK: Madin-Darby canine kidney cells; MIC: minimal inhibitory concentration; PBS: phosphate buffered saline; MST: mean survival time; PI: protective index; PR: protective ratio; RPLC: reversed phase liquid chromatography; Rim: rimantadine hydrochloride; TC50: 50% toxic concentration; TCID50: 50% tissue culture infectious doses; SI: selectivity index; SSI: Streptomyces subtilisin inhibitor; s.c.: subcutaneously

IntroductionInfluenza remains one of the serious problems of public health. Despite of the recent achievements of antiviral chemotherapy, the need of new antiviral agents continues to exist alongside with the necessity of novel approaches and methods. Current anti-influenza drugs – M2 ion channel blockers and neuraminidase inhibitors – target viral components. Recently, cellular proteins are coming to light as potential targets for new anti-viral drugs. The principal idea is to affect the mechanisms underlying virus-cell interactions and favouring viral replication in virus-infected cells. This kind of strategy instigates interest as it could eliminate the selection of viral resistance and would be suitable for different influenza viruses independently of the viral type and strain. It is accepted that the virulence of a particular influenza virus strain depends on the ability of its HA precursor HA0 to be cleaved post translation to subunits HA1 and HA2 by trypsin-like proteases of the host (10, 15, 27, 33). Cleavage of influenza virus HA is essential to infectivity, allowing fusion of viral and host cell membranes prior to the release of nucleocapsid into the cytoplasm.

ANTIVIRAL POTENTIAL OF A PROTEOLYTIC INHIBITOR FROM STREPTOMYCES CHROMOFUSCUS 34-1

Julia Serkedjieva1, Michelle Dalgalarrondo2, Lidia Angelova-Duleva1,3, Iskra Ivanova3

1Bulgarian Academy of Sciences, Institute of Microbiology, Sofa, Bulgaria2French National Institute for Agricultural Research- BIA-FIPL, Nantes, France3Sofia University, Faculty of Biology, Department of Microbiology, Sofia, BulgariaCorrespondence to: Julia SerkedjievaE-mail: jserkedjieva@microbio.bas.bg

ABSTRACTThe novel proteinaceous protease inhibitor, isolated from the culture supernatants of Streptomyces chromopfuscus 34-1 (SS 34-1) demonstrated a specific and selective anti-influenza virus effect. By the use of complementary virological assays it was demonstrated that the expression of the viral haemagglutinin (HA) on the surface of infected cells, the virus-induced cytopathic effect (CPE) and the infectious virus yields (IVY), used as measures of A/Germany/34, strain Rostock (H7N1) (A/Rostock) virus growth, were all reduced at non-toxic concentrations of SS 34-1 in a dose-dependent way. In a single cycle of viral growth, the penetration of viral particles and the late stages of viral replication were the most susceptible to inhibition.The preparation protected mice from mortality in the experimental A/Aichi/2/68 (H3N2) influenza virus infection. The SS 34-1-treatment (0.16 mg/mouse/day) of virus-infected mice led to a considerable reduction of mortality rate (index of protection = 71.0 %) and marked prolongation of mean survival time (+ 5.5 days). The lesions of the lungs due to infection as well as lung infectious virus titres were significantly reduced – δ log10 TCID50/ml up to 4.5. The application of SS 34-1 induced a continuous rise of the anti-protease activity but did not cause substantial changes in the lung protease activity of healthy mice. The infection triggered a slight reduction of protease activity in the lungs at 5 and 48 h p.i. and a marked increase at 24 h and 6 day p.i. Protease inhibition in the lungs was reduced at 24 and 48 h p.i. and an increase was observed at 5 h, 6 and 9 day p.i. SS 34-1-treatment brought both activities to normal levels. It was assumed that SS 34-1 exerted its inhibitory effect indirectly, through an increase of the protease-inhibitory activity. Histopathological examination of the lungs showed that under SS 34-1-treatment pulmonary lesions were less severe than those of untreated controls.The isolated novel protease inhibitor was purified by anion-exchange chromatography and reversed phase-HPLC. Analysis by differential scanning calorimetry (DSC) revealed that at pH 7.5 the denaturation temperature was 75 °C and the denaturation itself was irreversible. Analysis by circular dichroism (CD) in the UV region 190-250 nm showed that at pH 7.5 the spectrum presented 2 minima at 208 and 222 nm, typical for an α-helical structure. At pH 2.5 and after heating the spectrum corresponded to an unordered structure.

2787Biotechnol. & Biotechnol. eq. 26/2012/1

Taking into account the important role of proteolytic processing for viral reproduction, one of the possible targets for chemotherapy of influenza virus infection is the blocking of proteolytic cleavage of viral proteins; it is believed that this would result in inhibition of subsequent rounds of viral replication and spread in the respiratory tract. For influenza viruses with monobasic HAs, exogenous inhibitors of serine proteases, including ε-aminocaproic acid (ACA) (15), aprotonin (17) and ambroxol (32) have been shown to reduce HA-cleavage and virus activation in cultured cells, in chick embryos and in the lungs of infected mice.

The bacteria of the genus Streptomyces produce proteinase inhibitors with interesting properties, belonging in large part to the family of Streptomyces subtilisin inhibitors (SSI). There are literature data that proteinase inhibitors of microbial origin could be used successfully as antiviral agents. Lozitski et al. (16) presented experimental evidence that the replication of influenza virus was efficiently inhibited by a proteinase inhibitor MI 0114, produced by Streptomyces sp. Our group (25) has found antiviral activity for a protease inhibitor produced by Streptomyces sp. 225b (SS 225b). The replication of human immunodeficiency virus type 1 (7) and cytomegalovirus (26) were also efficiently inhibited by proteinase inhibitors, produced by Streptomyces species.

One hundred Streptomyces strains were screened for the production of proteolytic inhibitors. The most active products were investigated further for antiviral activity. Influenza virus replication was inhibited most effectively by SS 225b and SS 34-1. The selectivity indices of viral inhibition reached high values, which corresponded to the elevated levels of their specific protease inhibitory activity (2).

Here we present a further characterization of the anti-influenza virus effect of SS 34-1 in vitro and in vivo together with some properties of the investigated inhibitor.

Materials and MethodsCompoundsRibavirin (Rib), rimantadine hydrochloride (Rim), ε-aminocaproic acid (ACA), trypsin, bovine serum albumin (BSA) and N-bensoil-DL-arginine-p-nitroanilide (BAPNA) were purchased from Sigma-Aldrich Chemie GmbH, Deisenhofen, Germany.

Microbial strain and cultivation conditionsStreptomyces chromofuscus 34-1 was isolated from a soil sample. In the antiviral experiments partially purified fermentation extract was used (SS 34-1). The dilutions were made in cell culture medium ex tempore. Media and cultivation conditions were as described by Angelova et al. (2).

Fermentation product purificationPartial purification was performed according to Hiraga et al. (8). After centrifugation, 40 min, 12 000 rpm/min at 4 °C, the supernatants were precipitated with 80% (NH4)2SO4 for two days at 4 °C, centrifuged for 40 min at 4 °C, 12 000 rpm. The precipitates were re-suspended in 20 mM Tris HCl (pH 7.0) and dialyzed against the same buffer for two days.

Protein determinationThe protein content was determined according to Bradford (5), using the Bio-Rad assay reagent (Bio-Rad, Munich, Germany) and BSA as the standard and was expressed in mg/ml.

Preparation of lung tissue homogenatesOn days 1, 2, 6 and 9 post infection (p.i.) 3 mice of each group were anaesthetized with ether and exsanguinated by section of the subclavian arteries. Lungs were removed aseptically, washed in cold phosphate buffered saline (PBS), blotted dry; tissue pieces of about 1 g were disintegrated mechanically in ice-cold PBS and subsequently by an ultrasound disintegrator (MSE, England) for 3 min (interrupting sonication every 15 sec). The homogenates were centrifuged (8 000 rpm, 30 min, 4 °С) and the supernatants were examined for protease and protease-inhibitory activities.

Measurement of protease activity and protease inhibitory activityProtease activity in the lungs was measured by the method utilizing BAPNA as the substrate. One unit of proteolytic activity (U) was defined as the amount of the enzyme, producing one µmol BAPNA for 1 min at 37 °С. The specific proteolytic activity was evaluated (U/mg protein).

Protease inhibitory activity in the lungs was determined by a method based on the suppression of the cleavage of BAPNA by trypsin. Trypsin inhibitory units (TIU) were defined as the amount of inhibitor necessary to decrease by 50% the activity of one trypsin unit which hydrolyses one µmol of BAPNA per minute. The specific protease-inhibitory activity was evaluated (TIU/mg protein).

Both activities were determined at 405 nm on UV-VIS spectrophotometer Shimadzu.

Circular dichroism (CD) spectraCD spectra were measured at 20 °C using a Jobin Yvon Mark 6 dichrograph. Spectra were the averages of three accumulated scans with subtraction of the baseline. The cylindrical cells used had a path length of 0.01 cm in the far-UV (190-250 nm) spectral regions. Protein concentrations were calculated as described in Angelova et al. (1) by the bicinchoninic acid (BCA) colorimetric assay, using BSA as the standard. Deconvolution of the far-UV spectra was done using the Dichroweb program.

DSC analysisCalorimetric measurements were carried out within the temperature range of 10-100 °C at a heating rate of 1 K/min and excess pressure of 2 atm with a MCS-DSC unit (Microcal, USA). The tryptic inhibitor was dissolved in potassium phosphate buffer 10mM pH 7.8 for studies at basic pH and in Glycine/HCl buffer 40 mM pH 2.5 for studies at acidic pH.

Virology

Viruses, cells and mediaMadin-Darby canine kidney (MDCK) cells were as in Angelova et al. (2). Avian influenza virus A/Rostock and human influenza virus A/Aichi were adapted to MDCK cells. In the animal experiments A/Aichi adapted to mice lungs was used

2788 Biotechnol. & Biotechnol. eq. 26/2012/1

(A/Aichi-a). The virus infectious titres were over the range of 106-107 TCID50/ml (50% tissue culture infectious doses/ml), the HA titres – 1024-2048. The virus stocks were stored at -70 °C. The strains are from the collection of the Institute of Microbiology, Bulgarian Academy of Sciences, Sofia.

Mice Inbred ICR mice (16-18 g) were obtained from the Experimental Animal Station of BAS in Slivnitza, Sofia. The experimental groups were of 10-12 animals each. At the end of the experiments, surviving mice were sacrificed by cervical dislocation. They were quarantined 24 h prior to use and maintained on standard laboratory chow and tap water ad libitum. The number of experimental animals was reduced as much as possible, depending on statistical significance. Refinement of the tests with animals was achieved by careful planning of multifactor experiments. The animals were bred under standard conditions approved by the Bulgarian Veterinary Health Service.

Antiviral assaysThe antiviral effect was studied in multicycle and single cycle experiments of viral growth and the following antiviral assays were used.

Cytopathic effect (CPE) reduction assay was as in Serkedjieva and Hay (23). In short – quadruplicate confluent monolayers in 96-well plates were overlaid with 2× drug-containing medium (0.1 ml) and an equal volume of virus suspension (100 TCID10/ml). The virus-induced CPE was scored after 72 h under an inverted microscope. Changes in cell morphology were scored: score 0 = 0% CPE, score 1 = 0 to 25% CPE, score 2 = 25 to 50% CPE, score 3 = 50 to 75% CPE, score 4 = 75 to 100% CPE. The dilution which reduced CPE by 50% (50% effective concentration, EC50) with respect to virus control was estimated. The selectiveity index (SI) was determined by the ratio TC50/ec50. SI ≥ 4 was considered to stand for a significant selective inhibition.

50% end point titration technique (EPTT) was as in Vanden Berghe et al. (29). The antiviral activity was determined by the difference between the virus titres of control and treated viruses (d log10 TCID50/ml). The selective anti-influenza drug Rim was used as a positive control.

Infectious virus yield (IVY) reduction assay was performed as follows. Triplicate monolayers in 24-well plastic plates were inoculated with drug-containing medium (pre-treatment of cells), washed twice with PBS and challenged with 100 TCID50/0.1 ml of the virus in the presence of SS 34-1. After 60 min adsorption at room temperature the monolayers were washed twice with PBS and were incubated at 37 °C in drug-containing medium for 24 h. Cells and supernatants were pooled after one freeze-thaw cycle and titrated by HA and CPE. The concentration that reduced virus infectivity by 90% (1 log10TCID50/0.1 ml; EC90) was determined.Time of addition study used single cycle virus growth conditions. Confluent monolayers in 24-well plates were incubated with plain or drug-containing medium for 1 h at 37 °C (effect of cell pre-treatment), washed twice with PBS and challenged with undiluted infectious allantoic fluid (MOI = 10-

50 TCID50/cell). After adsorption for 1 h at 4 °C or 37 °C in the presence or absence of SS 34-1, the cells were washed twice with PBS and were overlaid with plain or drug-containing medium for 1 h to allow viral penetration into cells (effect on penetration). Then plain or drug-containing medium was added and cells were cultivated for 8 h at 37 °C. At hourly intervals after infection, medium was replaced with SS 34-1-containing medium. At the 8th h after infection the cells and supernatants were pooled after one freeze-thaw cycle and HA and infectious titres were determined. Virus titers were determined by the CPE assay by endpoint titration (19). When multiple cycles virus growth conditions were employed, the preparation (125.0 µg/ml) was added at varying times relative to viral infection; the experiments were performed by the EPTT.

Enzyme-linked immunosorbent assay (ELISA) was carried out according to Belshe et al. (3). The monoclonal antibody (MAb) to viral HA HC58 was kindly provided by Dr. A.J. Hay, the WHO Collaborative Centre of Influenza, Mill Hill, London, UK.

Virus infection in miceThe experimental influenza virus infection (EIVI) was induced under light ether anaesthesia by intranasal (i.n.) inoculation of A/Aichi-a. To induce lethal infection, mice were challenged with 10 LD50 (50% lethal doses), corresponding to 103 TCID50/ml, in a volume of 0.05 ml PBS per mice.

Experimental designMice were treated subcutaneously (s.c.) 2 h before and 24, 48 and 72 hours after viral exposure with SS 34-1 (0.16 mg/mouse/day) in 0.1 ml PBS and were observed for death daily for 14 days. The known proteolytic inhibitor ACA and the antiviral drug Rib were used for the purpose of comparison and were inoculated orally according to the same treatment schedule in daily doses of 1 mg/mouse/day and 0.125 mg/mouse/day respectively. The protective effect was evaluated by the increase of the rate of survival (%) and the mean survival time (MST). The protective index (PI) was determined from the equation (PR - 1)/PR×100, where PR (protective ratio) is Mcontrol/Mexperiment and M is mortality as in Serkedjieva and Ivanova (25).

To determine the lung parameters 3 animals from each experimental group were sacrificed on day 6 after infection, lungs were removed aseptically, weighed and lung consolidation (score) was scored as described in Serkedjieva and Ivanova (25). Lungs were homogenized to a 10% suspension in PBS and ten-fold dilutions (0.2 ml) were assayed for infectivity in MDCK cells in the presence of 2 µg/ml trypsin. Virus-induced CPE was scored as described above. Infectious virus titres were evaluated according to Reed and Muench (19) and expressed in log10 TCID50 /ml.

For histopathological examination lungs were fixed in 10% formalin for 24 h, dehydrated in alcohol of increasing concentration, imbedded in paraffin for 24 h at 56 °C, thin sectioned (5 μm, microtome МС-2 У4.2) and stained with hematoxylin and eosin.

The results are mean values from 3 or 4 independent experiments. Increase of survival rates compared to placebo

2789Biotechnol. & Biotechnol. eq. 26/2012/1

controls was evaluated using Fisher’s exact test. Student’s t-test was employed to analyze differences in survival times and infectious titres, *P<0.05.

Results and DiscussionInfluenza is a highly contagious, febrile, acute infection of the nose, throat, bronchial tubes, and lungs caused by influenza virus. The infection is usually self-limiting, culminating in a local and systemic reaction. However, there remains a significant proportion of patients, who develop severe illness and complications, such as the elderly, the very young and the immunocompromised; hence this infection represents one of the serious problems of human health. Even with the development of killed virus vaccines the need for effective anti-influenza therapies still exists. WHO has recently announced that “another influenza pandemic is inevitable and possibly imminent”, an alarming prospect, considering the devastating previous pandemics (31). Indeed the emergence of a new strain of influenza A H1N1 (swine flu) has led to the first influenza pandemic since 1968. The first step in a potential pandemic is the ability of the virus to enter human cells which is mediated by the viral surface glycoprotein haemagglutinin (20). The ability of the influenza viruses HA to mediate fusion between viral and endosomal membranes during virus entry into the cell depends on cleavage of fusion-incompetent precursor HA0 into disulfide-linked subunits HA1 and HA2 by a host endoprotease. Cleavage of HA is essential for infection, determines viral pathogenicity and tissue tropism and is necessary for the virus to establish infection in the host as well as to spread within the host (4). The mammalian and the non-pathogenic avian influenza virus strains have HAs with monobasic cleavage site and are usually cleaved only in a restricted number of cell types so that these viruses cause local infection (13). In general the glycoproteins are activated by secreted proteases like serum plasmin, kallikrein, urokinase, thrombin, acrosin, tryptase Clara and mini-plasmin in rat lungs, mast cell tryptase and tryptase TC30, found in porcine lungs (24 and references cited in). Cleavage activation of influenza monobasic HAs by host proteases is generally thought to occur extracellularily on the surface and/or in the lumen of the respiratory tract. The induction of influenza virus infectivity by host cell proteases is strictly regulated by inhibitors of the serine proteases such as human mucus protease inhibitor in the upper respiratory tract (9) and pulmonary surfactant (12) in the lower respiratory tract. The host enzymes responsible for the cleavage event are believed to correspond to the pathogenicity of the virus and are determined based on the cleavage site sequence (6, 21, 30). The majority of HA subtypes possess a single arginine at their cleavage site which facilitates cleavage by trypsin, a protease mainly localized to the respiratory tract in humans and the gastrointestinal tract in birds (20). Since protease activities predominate over those of endogenous inhibitory compounds under normal airway conditions, administration of protease inhibitors in the early stage of infection significantly suppresses viral entry and viral multiplication (11). Protease inhibitors may normalize proteolytic balance and thus limit the inflammatory process and the development of a syndrome of haemorrhage and coagulation defects and they may suppress

proteolytic activation of virions to reduce virus multiplication and spread (33). On the other hand, the proteinase inhibitors are widely distributed in the plant and animal kingdoms as well as in bacteria and some viruses. They take part in the processes of control and regulation of cell metabolism by activation and deactivation of cell proteases. They are of great importance in medicine, agriculture and biotechnology (18). Proteinase inhibitors are frequently found in Streptomyces spp. and are classified as members of the Streptomyces subtilisin inhibitor (SSI) family on the basis of their similar structures and protease inhibitory specificities (14, 28).

In the light of these findings we investigated the potential virus-inhibitory effect of proteolytic inhibitors from Streptomyces. As a first approach an extensive screening study for producers of proteinase inhibitors was carried out on 100 strains. Eighteen of them and/or their variants produced inhibitors of proteinaceous nature (2); 15 strains did not show any inhibitory effect. The specific inhibitory activity of the remaining 60 strains varied from 62.7 to 1 280.0 KIU/mg protein. Further the most active 18 strains and their variants, 23 samples in total, were characterized with respect to their protein content, specific activity towards trypsin, cell-toxicity and virus-inhibitory effects. The results from Tricin-SDS-PAGE analysis of the most promising preparations (SIA > 500 KIU/mg) indicated the presence of proteins with molecular mass between 6.5 – 14.5 kDa, which is characteristic of the proteinaceous proteolytic inhibitors.

The most active protease inhibitors were SS 225, SIA 653.9 KIU/mg and SS 34-1, SIA 625.0 KIU/mg. The production of the proteinase inhibitors reached its maximum values at 72-96 h of cultivation and was closely related to mycelial growth. The dynamics of the proteinase inhibitory activity during the submerged cultivation with glucose and starch as carbon sources followed the same pattern (2).

VirologyTreatment with SS 34-1 at concentrations of 10-1 200 µg/ml did not show any visible changes in cell morphology or cell density; thus TC50 exceeded 1 200 μg/ml. It is important to note that the cell-toxic effect was cell-specific; CEF (chicken embryonic fibroblast) cells were more sensitive than MDCK and MDBK (Madin-Darby canine kidney) cells (2).

It has been found that in doses up to 1 200 μg/ml SS 34-1 did not exhibit any direct inactivating effect on A/Rostock, and there was no detectable reduction of infectivity or HA-production (2).

Effect of SS 34-1 on virus replicationIn all experiments non drug treated, mock infected cells were used as cell control and non drug treated, virus infected cells as virus control. Rim (2 µg/ml) was used as a positive control for inhibition of viral replication.

The susceptibility of representative influenza viruses to the inhibitory action of SS 34-1 was assessed by the reduction of virus-induced CPE. The virus-inhibitory effect of SS 34-1 was strain-dependent, the preparation inhibited viruses, activated both by subtilisin and trypsin proteinases. Most sensitive to

2790 Biotechnol. & Biotechnol. eq. 26/2012/1

inhibition was the replication of A/Rostock and A/Aichi and the least sensitive was A/PR8 (H1N1) (19).

In the present paper the in vitro anti-influenza virus effect of SS 34-1 was studied in detail in a number of assays on the replication of A/Rostock in MDCK. The results are presented in Table 1. A/Rostock was selected because of its high susceptibility to SS 34-1 inhibition and its ability to multiply in MDCK cells without trypsin. In experiments employing multiple cycles of virus replication the respective EC50-s varied from 44.3 to 52.6 µg/ml. The expression of HA on the surface of infected cells, determined by an ELISA with anti-HA MAb, was inhibited with EC50 of 44.3 µg/ml. SS 34-1 reduced the virus-induced CPE of A/Rostock in MDCK with an EC50 of 62.4 µg/ml. A similar result was obtained applying the EPTT assay (EC50 = 52.6 µg/ml). In the IVY reduction assay, the concentration of SS 34-1 which partially inhibited HA production and reduced the infectious titre by 1 log10 TCID50/ml (EC90) was 81.3 µg/ml; the reduction of infectivity increased up to 3.5 log10 at a concentration of approx 250 µg/ml (results not shown). All assays revealed a dose-dependence of the virus-inhibitory effect.

TABLE 1

Comparative sensitivity of A/Rostock to inhibition by SS 34-Assay ec50 (μg/ml)Contact assayа > 1 200c

ELISA 44.3CPE reduction 62.4EPTT 52.6IVY reductionb 81.3

a MIC, minimal inhibitory concentration;b ec90, 90% effective concentration;c previously determined in Serkedjieva et al. (22)

We tested 10 samples – crude and partially purified fermenatation extracts as well as extracts from different fermentations, for inhibitory activity on A/Rostock virus

replication. Although in general the partial purification of the crude fermentation extracts lead to an increase of the protease-inhibitory activity, the virus-inhibitory effect was diminished (results not shown).

0

1

2

3

4

5

6

7

8

VC

pretre

atmen

t

adso

rption

pene

tratio

n 1 2 3 4 5 6 7

h after infection

viru

s tit

re (l

og10

TC

ID 50/m

l)

Fig. 1. Inhibitory effect of SS 34-1 (250.0 μg/ml) on A/Rostock replication in MDCK cells (single cycle of viral replication). VC: virus control – virus-infected and cell culture medium-treated.

Time of addition studiesTo investigate the effect of SS 34-1 on different steps of viral replication the preparation (125.0 µg/ml) was added at varying times relative to viral infection. The pre-treatment of cells as well as the addition at the time of adsorption did not result in decrease of virus infectivity. A slight inhibition was observed when the adsorption was performed at 37 °C, allowing internalisation of viral particles. Virus replication was significantly reduced when SS 34-1 was added at the time of penetration, δ log10 TCID50/ml = 4.5, and markedly inhibited when the preparation was inoculated after virus infection, δ log10 TCID50/ml = 4.8. The results were comparable with the effect of ACA. More precisely the mode of action of SS 34-1 was investigated in experiments employing single cycle virus

TABLE 2Protective effect of SS 34-1 on lung parameters of mice, infected with influenza A/Aichi virus infection

Experimental group Day p.i. N survival/totalLung parameters

Infectious titre(log10 TCID50/ml) Scorea Lung indexb (%)

VC1 2 20/20 3.3±0.4 1.0±0.3 1.0VC+ SS 34-12 10/10 1.5±0.3 0.5±0.1 1.0VC 6 16/20 6.1±0.3 3.0±0.2 1.18VC+ SS 34-1 10/10 1.1±0.4 1.0±0.3 1.0VC 9 7/20 5.2±0.5 2.7±0.5 1.08VC+ SS 34-1 7/10 1.6±0.3 1.0±0.3 0.98

p.i.: post infection; a scores 0-4, assigned to % visible consolidation;b lung weight/body weight × 100; 1 virus control; 2 0.16 mg/mouse/day, -2, +24, +48, +72 h with respect to infection, s.c.

2791Biotechnol. & Biotechnol. eq. 26/2012/1

growth conditions (Fig. 1). The above observations were confirmed – the penetration of viral particles and the processes occurring on the 7th h of the replicative cycle were most sucseptible to inhibition with 250 µg/ml of SS 34-1.

The obtained results show that the fermentation extract does not inactivate influenza virus A/Rostock but is able to interfere with the penetration of viral particles into susceptible cells as well as with the late stages of viral replication (post-transnational processing of viral lipoproteins, assembly and release of newly formed viral particles from infected cells). Both the internalization and release of influenza virus particles depend on the post-translational cleavage of the HA precursor (HA0) to subunits HA1 and HA2 (27). Thus, the antiviral effect of SS 34-1 is presumably due to inhibition of this cleavage through inhibition of the host cells serine type proteases. Production of HA was also reduced significantly (results not shown).

Protective effect of SS 34-1 in the murine EIVIFurther we investigated the protective effect of SS 34-1 in the murine model of experimental influenza infection induced by A/Aichi-a virus. The intranasal inoculation of the virus to mice produced a damaging infection of the lungs, causing viral pneumonitis. PBS was used for a placebo control and Rib (5 × 10 mg/kg) – as a positive control. For the purpose of comparison the known proteolytic inhibitor ACA was used as well in the dosage of 5 × 50 mg/kg. ACA, an analogue of lisin, inhibits plasmin activity in vivo by competing for the interaction of fibrin, plasminogen and plasminogen activator. In any case this inhibitor interferes with the replication of viruses that are activated by secreted proteases (33). Applied according to a prophylactic-therapeutic schedule, SS 34-1 (0.16 mg/mouse/day) protected mice from mortality in the EIVI. The preparation was not toxic to the experimental animals and the protection surpassed that exhibited by Rib (PI = 62.4%) and by ACA (PI = 56.1%), both inoculated according to the same treatment schedule (2). Both the rate of survival (71.0%) and MST (+5.5 days) were increased; the effect was dose-related. The mice receiving SS 34-1 treatment showed minimal pathological lesions in the lungs, whereas control untreated animals had total hemorrhagic pneumonia. Lung parameters (consolidation and weights) were all markedly reduced in the SS 34-1-treated mice. Lung infectious virus titres were significantly decreased – δ log10 TCID50/ml reached 4.5 (Table 2).

The lung protease and protease-inhibitory activities were estimated at 5 h p.i. and on days 1, 2, 6 and 9 p.i. The application of SS 34-1 induced a continuous rise of the anti-protease activity but did not cause substantial changes in the lung protease activity of healthy mice. It could be assumed that in the course of the infection SS 34-1 exerted its inhibitory effect on the virus protein proteolysis and hence, on the activation of viral particles indirectly, through an increase of the protease-inhibitory activity. The infection triggered a slight reduction of protease activity in the lungs at 5 and 48 h p.i. and a marked increase at 24 h and 6 days p.i. Protease inhibition in the lungs was reduced at 24 and 48 h p.i. and an increase was

observed at 5 h, 6 and 9 days p.i. SS 34-1-treatment brought both activities to normal levels (Fig. 2 and Fig. 3).

The histopathological examination of the lungs showed that under SS 34-1-treatment pulmonary lesions were less severe than those of untreated virus controls (Fig. 4).

Clarifying the anti-influenza virus effects of the protease inhibitor SS 34-1 might open the way to utilize this substance in the control of influenza infection. Equally the information obtained from our investigation may facilitate the studies on the development of novel antiviral agents of bacterial origin.

50

70

90

110

130

150

170

190

5h 1 2 6 9days p.i.

% o

f con

trol (

U/m

g pr

otei

n)SS 34-1 VC VC+SS 34-1

Fig. 2. Protease activity in the lungs of SS 34-1-treated influenza virus-infected mice. Experimental groups: SS 34-1 - mock-infected and SS 34-1-treated; VC: virus control – virus-infected and PBS-treated; VC+SS 34-1 – virus-infected and SS 34-1-treated.

50

70

90

110

130

150

170

190

5h 1 2 6 9days p.i.

% o

f con

trol (

TIU

/mg

prot

ein)

SS 34-1 VC VC+SS 34-1

*

Fig. 3. Protease-inhibitory activity in the lungs of SS 34-1-treated influenza virus-infected mice. Experimental groups: as in Fig. 2.

Physicochemical properties of SS 34-1The novel protease inhibitor isolated from culture supernatants of Streptomyces chromofuscus was subjected to purification by a procedure including amonium sulfate precipitation, ion exchange chromatography on DEAE Sepharose, reversed phase liquid chromatography (RPLC) and analytical HPLC. The purified preparation was named PISC-2002. The inhibitory effect of PISC-2002 was determined on some serine-proteases (trypsin, α-chymotrypsin, subtilisin, proteinase K, and proteinase of S. albovinaceus) and aspartylproteases (pepsin and cathepsin D). The protease inhibitor strongly inhibited subtilisin, proteinase K, trypsin and proteinase of S.

2792 Biotechnol. & Biotechnol. eq. 26/2012/1

albovinaceus, weakly inhibited pepsin, but did not inhibit chymotrypsin and cathepsin D. According to the obtained Ki, PISC-2002 could be considered as a subtilisin type of inhibitor like plasminostreptin, 3 SSI, SIL 1, 2, 3 and 4 and SIL 8 (1). 10 mg of pure product were obtained from 2 g powdered dried supernatant containing 1.3 g protein. The final preparation was homogenous as judged by SDS-PAGE, analytical HPLC analysis and mass spectrometry, giving a molecular mass of 11.264 Da. The studied inhibitor was stable over a large range of pH (2 - 10) and at high temperatures (80 °C/30 min), mainly because of the presence of proline and of a high content in hydrophobic amino acids. Analysis by electrospray mass spectrometry estimated a protein with Mw 11 214 Da. Amino

acid analyses revealed very rich contents of hydrophobic residues: Ala (16.9%), Gly (10.8%), Leu (9.5%), Pro (9%) and Val (10%) (1). The amino acid composition was very similar to that of the plasminostreptin and SSI from S. albogriseolus. the N-terminal sequence of PISC-2002 showed high homology with various protease inhibitors produced by Streptomyces species through similarity searches. The consensus motif XXLXXXSALVLT of SSI was found also in the inhibitor PISC-2002. The N-terminal sequence of PISC-2002 gave two sequences differing by one residue (1).

Fig. 5. DSC analysis of tryptic inhibitor 44.78 µM in potassium phosphate buffer 10mM pH 7.8. Solid line: first scan, broken line: second scan.

Fig. 6. DSC analysis of tryptic inhibitor 116 µM in Glycine/HCl buffer 40 mM pH 2.5. Lines: as in Fig. 4.

Analysis by DSC established that at pH 7.5 the denaturation temperature was 75 °C and the denaturation was irreversible. The DSC thermogram of the protein at basic pH is shown in Fig. 5. A single peak of heat capacity corresponding to a thermal unfolding transition was detected. The maximum heat capacity was observed at 75 °C. The thermogram of the inhibitor obtained during the second scan of the solution differed markedly from that of the first scan. A decreasing intensity of the peak was observed indicating irreversible thermal unfolding of the protein. The DSC thermogram of protein at acidic pH is shown in Fig. 6. Again a single peak was found. The maximum heat capacity was observed at

a)

b)

c) Fig. 4. Histopathological examination of the lung tissue of SS 34-1-treated influenza virus-infected mice. (a) SS 34-1; (b), VC; (c), VC+SS 34-1 as in Fig. 2.

2793Biotechnol. & Biotechnol. eq. 26/2012/1

41 °C. A discrepancy between the first and the second scan was seen once more. The peak intensity decreased considerably, demonstrating a contribution of irreversible thermal unfolding of the protein. At pH 2.5 the endothermic peak was sharper. The thermal transition of the inhibitor was lowered from 75 °C to 41 °C, indicating a less stable conformation of the molecule at acidic pH. The thermogram of the inhibitor at pH 8 is very noisy with an exothermic process around 20 °C.

Analysis by CD in the far UV region (190 - 250 nm) showed that at pH 7.5 the spectrum presented 2 minima at 208 and 222 nm, typical for an α-helical structure. At pH 2.5 and after heating the spectrum corresponded to an unordered structure.

On the basis of its properties, the similarity of their structures and protease inhibitory specificities SS 34-1 was classified as a member of the SSI family.

ConclusionsThe virus-inhibitory effect of a fermentation product produced by Streptomyces chromofuscus 34-1 was found to be specific and selective, strain-related and dose-dependent. Its antiviral potential was demonstrated in cell cultures and in the experimental influenza virus infection in mice. It was assumed that in the course of the infection SS 34-1 exerted its inhibitory effect on the virus protein proteolysis and hence, on the activation of viral particles indirectly, through an increase of the protease-inhibitory activity.

The present results are in accordance with the findings that protease inhibitors of microbial origin could be used successfully as antiviral agents.

AcknowledgementsThe authors acknowledge the skilful technical assistance of Mrs. K. Todorova, Institute of Microbiology, Bulgarian Academy of Sciences. This study was supported by research grant D002-188 from the National Science Council, Bulgaria.

REFERENCES1. Angelova L., Dalgalarrondo M., Minkov I. et al. (2006)

BBA – General Subjects, 1760, 1210-1216.2. Angelova L., Remichkova M., Ivanova I. et al. (2006) J.

Basic Microbiol., 46, 504-512.3. Belshe R.B., Smith M.H., Hall C.B. et al. (1988) J. Virol.,

62, 1508-1512.4. Boettcher C., Ludwig K., Herrmann A. et al. (1999)

FEBS Lett., 463, 255-259. 5. Bradford M.M. (1976) Anal. Biochem., 72, 248-254.6. Chen J., Lee K.H., Steinhauer D.A. et al. (1998) cell,

95, 409-417.7. De Clercq E. (2001) Curr. Med. Chem., 8, 1543-1572.8. Hiraga K., Suzuki T., Oda K. (2000) J. Biol. Chem., 275,

25173-25179.9. Kido H., Beppu Y., Imamura Y. et al. (1999) Biopolymers,

51, 79-86.

10. Kido, H., Niwa, Y., Beppu, Y. et al. (1966) Adv. Enzyme Regul., 36, 325-347.

11. Kido H., Okumura Y., Yamada H et al. (2007) Curr. Pharm. Des., 13, 405-414.

12. Kido H., Sakai K., Kishino Y. et al. (1993) FEBS Lett., 322, 115-119.

13. Klenk H.D. and Garten W. (1994) Trends Microbiol., 2, 39-43.

14. Laskowski M. and Kato I. (1980) Annu. Rev. Biochem., 49, 593-627.

15. Lozitsky V.P., Fedchuk A.S., Puzis L.E. et al. (1987) Voprosi Virusologii, 32, 413-419. (in Russian)

16. Lozitsky V.I., Kashkin A.P., Shneider M.A. et al. (2002) In: Abstracts 15th ICAR, 17-21 March 2002, Prague, Check Republic, p. A75.

17. Ovcharenko A.V. and Zhirnov O.P. (1994) Antiviral Res., 23, 107-118.

18. Rawlings N.D., Tolle D.P., Barrett A.J. (2004) Biochem. J., 378, 705-716.

19. Reed L.J. and Muench H. (1938) Am. J. Hyg., 27, 493-498.

20. Rumshlag-Boom E., Guo Y., Wang J. et al. (2009) Virol J., 6, 76-84.

21. Russell C.J. and Webster R.G. (2005) cell, 123, 368-371.22. Serkedjieva J., Angelova L., Roeva I. et al. (2006) in:

Proceedings XIth Congress of Bulgarian microbiologists, 5–7 October 2006, Varna, Bulgaria, 25-29.

23. Serkedjieva J. and Hay A.J. (1998) Antiviral Res., 37, 121-130.

24. Serkedjieva J. and Ivanova E. (1997) Acta virol., 41, 65-70.

25. 25. Serkedjieva J. and Ivanova I. (2009) J. Pure Appl. Microbiol., 3, 441-452.

26. Shu Y.Z., Ye Q., Kolb J.M. et al. (1997) J. Nat. Prod., 60, 529-532.

27. Steinhauer D.A. (1999) Virology, 258, 1-20.28. Taguchi S., Kikuchi H., Suzuki M. et al. (1966) Appl.

Environ. Microbiol., 59, 4338-4341.29. Vanden Berghe D.A., Vlietink A.J., Van Hoof L. (1986)

Bull. Inst. Pasteur, 84, 101-147.30. Webster R.G. and Rott R. (1987) cell, 50, 665-666.31. World Health Organization. Pandemic Influenza

Preparedness and Response: A WHO Guidance Document. 2009.

32. Yang B., Yao D.F., Ohuchi M., et al. (2002) Eur. Resp. J., 19, 952-958.

33. Zhirnov O.P., Ikizler M.R., Wright P.F. (2002) J. Virol., 76, 8682-8689.

34. Zhirnov O.P., Ovcharenko A.V., Bukrinskaja A.G. (1985) Voprosi virusologii, 30, 207-214. (in Russian)