BO LR OI GO ID NCAOLM PA OR NT EI CN LT ES The effect of riboflavin and ultraviolet light on the infectivity of arboviruses Helen M. Faddy,1* Natalie A. Prow,2* Jesse J. Fryk,1 Roy A. Hall,2 Shawn D. Keil,3 Raymond P. Goodrich,3 and Denese C. Marks4

BACKGROUND: Arboviruses are an emerging threat to transfusion safety and rates of infection are likely to increase with the increased rainfall associated with climate change. Arboviral infections are common in Australia, where Ross River virus (RRV), Barmah Forest virus (BFV), and Murray Valley encephalitis virus (MVEV), among others, have the potential to cause disease in humans. The use of pathogen reduction technology (PRT) may be an alternative approach for blood services to manage the risk of arboviral transfusion transmission. In this study, the effectiveness of the Mirasol PRT (Terumo BCT) system at inactivating RRV, BFV, and MVEV in buffy coat (BC)-derived platelets (PLTs) was investigated. STUDY DESIGN AND METHODS: BC-derived PLT concentrates in additive solution (SSP+) were spiked with RRV, BFV, or MVEV and then treated with the Mirasol PRT system. The level of infectious virus was determined before and after treatment, and the reduction in viral infectivity was calculated. RESULTS: Treatment with PRT (Mirasol) reduced the amount of infectious virus of all three arboviruses. The greatest level of inactivation was observed for RRV (2.33 log; 99.25%), followed by BFV (1.97 log; 98.68%) and then MVEV (1.83 log; 98.42%). CONCLUSION: Our study demonstrates that treatment of PLT concentrates with PRT (Mirasol) reduces the infectious levels of RRV, BFV, and MVEV. The relevance of the level of reduction required to prevent disease transmission by transfusion has not been fully defined and requires further investigation. In the face of a changing climate, with its associated threat to blood safety, PRT represents a proactive approach for maintaining blood safety.

824 TRANSFUSION Volume 55, April 2015


merging transfusion-transmissible pathogens are a threat to blood safety.1 Arthropod-borne viruses (arboviruses) are a group of vector-borne viruses, many of which are considered to be emerging pathogens and have the potential to cause disease in humans. Given that arboviral infection often results in asymptomatic viremia, there is a risk of transfusion transmission.2 Indeed, arboviruses such as West Nile virus (WNV), chikungunya virus (CHIKV), and dengue virus (DENV) have already threatened the safety of the blood supply globally.2-6 Increased rainfall favors mosquito breeding, resulting in increased arboviral transmission,7-9 and climate change is predicted to increase the transmission of many vector-borne pathogens, representing a significant threat for the maintenance of a safe blood supply in the future.10

ABBREVIATIONS: BC = buffy coat; BFV = Barmah Forest virus; CHIKV = chikungunya virus; CPE = cytopathic effect; DENV = dengue virus; MVEV = Murray Valley encephalitis virus; PRT = pathogen reduction technology; RRV = Ross River virus; TCID50 = 50% tissue culture infectious dose; WNV = West Nile virus. From 1Research and Development, Australian Red Cross Blood Service, and the 2Australian Infectious Disease Research Centre, School of Chemistry and Molecular Biosciences, The University of Queensland, Brisbane, Queensland, Australia; 3Terumo BCT, Lakewood, Colorado; and 4Research and Development, Australian Red Cross Blood Service, Sydney, New South Wales, Australia. Address reprint requests to: Helen Faddy, 44 Musk Avenue, Kelvin Grove, Qld 4059, Australia; e-mail: [email protected] *These authors served as joint first authors. This work was supported by Terumo BCT, in the form of reagents, equipment, and grants. Received for publication June 26, 2014; revision received August 29, 2014, and accepted September 1, 2014. doi: 10.1111/trf.12899 © 2014 AABB TRANSFUSION **;**:**-**. 2015;55:824–831. Volume **, ** **





Arboviruses are relatively common in Australia, where they can cause disease in humans. The most prevalent arboviruses of human significance in Australia are Ross River virus (RRV) and Barmah Forest virus (BFV).7 Both belong to the genus Alphavirus (together with CHIKV) and have a single-stranded positive-sense RNA genome. Kangaroos and wallabies are the main reservoir hosts for these viruses,11,12 and they share similar mosquito vectors (Aedes vigilax and Culex annulirostris in Northern Australia13-15 and A. camptorhynchus in southern areas).11,16 Viral transmission to humans occurs in areas where infected mosquito vectors are present, with notified cases reported from all states and territories.7 RRV and BFV can cause nonfatal, debilitating arthritic symptoms in humans;17 however, subclinical infection can occur and is possibly more common than clinical infection.18,19 The average incubation period for RRV infection is 7 to 9 days, but up to 21 days has been observed in some cases.20 A similar incubation period is believed to occur for BFV infection.21 RRV and BFV display a clear seasonal trend: although notifications are reported yearround, the largest numbers of cases are diagnosed toward the end of summer and into autumn.7 Rainfall, temperature, tidal levels, and humidity correlate with changes in the incidence of RRV and BFV infection in humans.22-25 Rates of RRV infection were elevated nationally during the 2010 to 2011 southern hemisphere summer, a season with increased rainfall across much of the country compared to historical averages. Subsequently the number of BFV cases also increased at this time in specific regions across Australia.7 Another endemic Australian arbovirus, Murray Valley encephalitis virus (MVEV), is a member of the genus Flavivirus (along with WNV and DENV) and also has a positive-sense, single-stranded RNA genome. MVEV is transmitted through mosquito vectors, with C. annulirostris being the most common mosquito species and, like WNV, birds are the most common reservoir hosts.26 The incubation and presymptomatic periods in humans are believed to be similar to those observed for RRV infection.27 MVEV can cause encephalitis, with permanent neurologic sequelae or death, and has a high fatality rate in clinical cases.28,29 Subclinical infection can occur and can be up to 1000 times more common than clinical infection.29,30 In the north of Australia MVEV is enzootic.7 Transmission in the south east of the country is rare, with cases often occurring after extreme weather conditions, presumably a result of migration of infected wild birds into flood-affected areas.7,31 In the summer of 2010 to 2011 there was an increase in the number of MVEV cases (15 cases nationally, up from an average of 1.4 cases per annum for the previous 5 years).7 Given that RRV, BFV, or MVEV infection results in viremia and that subclinical infection is possible19,28 these arboviruses pose a risk to transfusion safety. To date there 2


Volume **, ** **

have been no known cases of transfusion-transmitted RRV, BFV, or MVEV; however, transfusion transmission has been documented for other arboviruses, including WNV and DENV.2 Studies suggest that asymptomatic viremia for RRV infection in mice can last from 5 to 9 days;32 however, little is known about the duration of asymptomatic viremia in humans infected with RRV, BFV, or MVEV. The risk of collecting a RRV-infected donation in high-risk areas of Australia after increased rainfall has been estimated to be in the range of 1 in 2500 to 1 in 60,000 donations.33 BFV may also threaten the safety of the blood supply,1 with viral transmission likely to be modulated under different climate change scenarios and varying across different coastal regions,34 suggesting that the future potential risk to blood safety will be regionally dependent. The presence of recent MVEV infection in blood donors has not been demonstrated to date despite MVEV cases in horses and sentinel chickens in the area at the time,9 suggesting that MVEV may pose a limited risk to blood safety. However, this study was based on a relatively small sample size. In Australia, donors are not permitted to donate if feeling unwell at the time of donation and blood components are quarantined or recalled from donors reporting any illness within 7 days of donation as an additional safety precaution. Individuals diagnosed with either RRV or BFV are ineligible to donate blood for 4 weeks after recovery. Pathogen reduction technology (PRT) refers to a group of technologies that attempt to permanently reduce or remove the ability of blood-borne pathogens to replicate, thereby reducing the risk of transfusion-transmitted infection.35 The Mirasol PRT system, developed by Terumo BCT (Lakewood, CO), requires the addition of a photosensitive agent (riboflavin) before the inactivation process.35,36 This system has been demonstrated to inactivate a range of bacteria, parasites, and viruses,37 including the arboviruses CHIKV,38 DENV,39 and WNV.37 RRV, BFV, and MVEV may pose a concern for transfusion safety given that infection results in viremia and that asymptomatic infection is possible. As such, we investigated the efficacy of the Mirasol PRT system to inactivate these arboviruses in buffy coat (BC)-derived platelets (PLTs) in additive solution (AS). This study did not examine the effect of the Mirasol PRT system on PLT quality after treatment, as this has previously been investigated.40

MATERIALS AND METHODS PLTs and PLT quality control This study had approval from the Australian Red Cross Blood Service Human Research Ethics Committee. All donations were from eligible, voluntary donors. Whole blood units (450 ± 45 mL) were collected (Day 0) into topand-bottom bags containing 63 mL of CPD (Fresenius Volume 55, April 2015 TRANSFUSION 825


Kabi, Bad Homburg v.d.h., Germany) and stored at 22°C between collection and processing. Pooled PLTs were prepared by pooling four ABO/D-matched leukoreduced BCs in AS (SSP+; Macopharma, Mouvaux, France) containing approximately 28% plasma carryover as previously described.40 For selected units, only BCs collected from younger donors (those under the age of 45 years at the time of donation) were used to minimize the chance of manufacturing a PLT concentrate from a donor who had been exposed to RRV, BFV, or MVEV, which could potentially contain neutralizing antibodies within their donation. As increased age is associated with an increased chance of being exposed to these viruses, selecting a cutoff of 45 years minimized the chance of obtaining a PLT concentrate from a donor exposed to these viruses, while still enabling adequate numbers of BCs to be available for pooling.

Assessment of PLT quality For each PLT concentrate, the following quality control (QC) variables were assessed: PLT count, unit volume, swirl, and pH. Briefly, the PLT count was determined using an automated hematology analyzer (CellDyn Ruby, Abbott Laboratories, Abbott Park, IL); pH was measured using a pH meter (Seven Multi, Mettler Toledo Ltd, Port Melbourne, Australia); and the unit volume was determined by weighing the unit, subtracting the weight of the empty bag, and dividing the weight by the specific gravity of PLTs (1.025 g/mL). PLT swirl was determined to be present or absent by visual inspection.

Cell culture African green monkey kidney (Vero) cells (ATCC CCL-81) were cultured in Dulbecco’s modified Eagle’s medium (DMEM; Thermofisher, Scoresby, Victoria, Australia) supplemented with 2% fetal bovine serum (FBS; Sigma, Castle Hill, New South Wales, Australia), penicillin (50 U/mL), streptomycin (50 μg/mL), and l-glutamine (2 mmol/L; Gibco, Mulgrave, Victoria, Australia) at 37°C and 5% CO2. Aedes albopictus salivary gland cells (C6/36, ATCC CRL-1660) were cultured in RPMI 1640 (Thermofisher, Scoresby, Victoria, Australia) supplemented with 10% FBS, penicillin (50 U/mL), streptomycin (50 μg/mL), and l-glutamine (2 mmol/L) at 28°C and 5% CO2.

Viruses and viral production RRV Strain T48, BFV Strain SW31081, and MVEV Strain 1-51 were used for this study and propagated in C6/36 cells. Virus stocks were produced by infecting a subconfluent monolayer of C6/36 cells with virus at a multiplicity of infection of 0.1 to 1 in RPMI containing 2% FBS (2% FBS/RPMI). Culture supernatant was harvested 826 TRANSFUSION Volume 55, April 2015


and clarified at 72 to 96 hours postinfection and stored at −80°C.

Viral inactivation PLT concentrates were treated with the Mirasol PRT system (Terumo BCT) within 8 hours of pooling as per the manufacturer’s instructions. Due to this requirement, all units were spiked and treated. Viral inactivation was only reported for units with no detectable neutralizing antibodies against the virus type investigated (see below). A sample (20 mL) was removed under aseptic conditions from the PLT concentrate for PLT QC testing (PLT count, pH), testing for virus-specific antibodies, and use as a negative control. The PLT concentrate was then spiked with virus (RRV, BFV, or MVEV; 20 mL), and a pretreatment sample (1 mL) was collected. The infected PLT concentrates were then transferred to a Mirasol illumination bag. Riboflavin (500 μmol/L; 35 mL in 0.9% sodium chloride) was added to give a final concentration of approximately 50 μmol/L and the products were illuminated with UV light in an illuminator (Mirasol, Terumo BCT) to 9.1 J/ mLPLASMA. The average illumination time was dependent on the final weight of the bag and percentage of plasma carryover (assumed to be 26% for all units after viral spiking). A posttreatment sample (1 mL) was collected.

Evaluation of infectious virus The concentration of infectious virus in the stocks, as well as in the pre- and postillumination samples, was determined by titration on Vero cells in 96-well plates by serial 10-fold dilutions in 2% FBS and DMEM. After 5 days of incubation, wells exhibiting cytopathic effect (CPE) were identified, and the viral titer was determined and expressed as a 50% tissue culture infectious dose (TCID50).41 To confirm the presence of virus infection in cultured wells, cells were fixed with phosphate-buffered saline (PBS) containing 20% (vol/vol) acetone and 0.2% (wt/vol) bovine serum albumin (BSA) and subjected to an enzyme-linked immunosorbent assay (ELISA) as described previously.42 Briefly, plates were blocked with blocking buffer (100 μL; 0.05 mol/L Tris/HCl [pH 8.0], 1 mmol/L EDTA, 0.15 mol/L NaCl, 0.05% Tween 20, 0.2% wt/vol casein) for 1 hour before probing of fixed antigen with in-house monoclonal antibodies (MoAbs) targeting each virus at a predetermined optimum dilution diluted in blocking buffer. Wells were washed four times with wash buffer (PBS with 0.05% vol/vol Tween 20), and bound antibodies were detected with horseradish peroxidase– conjugated goat anti-mouse preadsorbed for anti-human immunoglobulin (Jackson Immunoresearch, West Grove, PA). The plates were washed six times, and enzyme activity was visualized by the addition of substrate solution (100 μL; 1 mmol/L 2,2′-azino-bis(3-ethylbenzthiazoline6-sulfonic acid), 3 mmol/L H2O2 in a citrate-phosphate buffer, pH 4.2). Absorbance was measured at 405 nm. The Volume **, ** **





TABLE 1. Results of PLT QC testing, and the required specifications, for PLT concentrates used in inactivation studies Antibody* (positive/negative) Specifications Blood service Mirasol PRT system RRV Bag R3 Bag R4 Bag R5 Bag R8 Mean ± SD BFV Bag B1 Bag B3 Bag B4 Bag B5 Mean ± SD MVEV Bag M1 Bag M2 Bag M3 Bag M4 Mean ± SD

Positive swirl (pass/fail)


PLT count (×109/mL) >0.8 0.8-1.5

Unit volume (mL) >160 250-450



Negative Negative Negative Negative

Pass Pass Pass Pass

7.03 7.09 7.10 7.04 7.07 ± 0.04

0.923 0.712 0.882 1.000 0.879 ± 0.122

336.0 327.0 347.0 328.0 334.5 ± 9.3

Negative Negative Negative Negative

Pass Pass Pass Pass

7.12 7.00 7.10 7.11 7.08 ± 0.06

0.996 1.127 0.828 0.916 0.967 ± 0.127

344.0 321.0 329.0 324.0 329.5 ± 10.2

Negative Negative Negative Negative

Pass Pass Pass Pass

7.03 7.06 7.09 7.10 7.07 ± 0.03

0.993 0.937 0.738 0.750 0.855 ± 0.130

326.0 346.0 328.0 330.0 332.5 ± 9.1

* Indicates the presence of a neutralizing antibody against either RRV, BFV or MVEV.

criterion for specific recognition of antigen was defined as being at least twofold greater than that generated by probing uninfected control wells within the same plate. The level of viral inactivation was taken as the difference in TCID50 between the pre- and posttreatment samples and is shown as both a log and a percentage reduction.

preabsorbed secondary antibody conjugate to ascertain viral replication. Titers of not more than 20 were classified as being negative for neutralizing antibodies against either virus, whereas titers of 40 or more were classified as being positive.

RESULTS Detection of neutralizing antibodies in PLT supernatants Heat-inactivated samples (supernatants from PLT concentrates) were titrated in serial twofold dilutions from 1 in 20 to 1 in 2560 in DMEM without FBS in wells of a 96-well microtiter plate (Costar, Corning, Noble Park, Victoria, Australia). Approximately 100 infectious units (IU) of RRV, BFV, and MVEV diluted in DMEM with 2% FBS were added to each well containing diluted supernatant and plates were incubated at 37°C for 1 hour. Vero cells were then added to each well at a density of 2 × 105 cells/ plate and incubated for 5 days at 37°C in a humidified CO2 incubator before microscopic examination for CPE. Cells were also added to a 1-in-20 dilution of each supernatant in the absence of virus to ensure that the samples were not toxic to the cells. The neutralization titer was expressed as the reciprocal of the highest supernatant dilution where CPE had occurred. To confirm neutralization of the virus in each well, the cells were fixed with PBS containing 20% (vol/vol) acetone and 0.2% (wt/vol) BSA and subjected to an ELISA protocol using in-house MoAbs (anti-flaviviral E protein antibody [MoAb 4G2] for MVEV, anti-RRV protein antibody [MoAb G8], or anti-BFV protein [10E10-C2]) with 4


Volume **, ** **

All PLT concentrates showed an acceptable positive swirl and pH, and the volume was within the specifications of the Mirasol PRT system (Table 1). The PLT counts for the majority of concentrates were also within the specifications of the system. Three PLT concentrates (RRV Bag 4, MVEV Bag 3, and MVEV Bag 4) were slightly below specifications, but were within 11% of the lower limit. A similar level of viral inactivation was observed in these bags to what was seen for those bags that were within the specifications for PLT count. The presence of PLTs did not appear to interfere with the functioning of the viral infectivity assay (data not shown). Moreover, the presence of riboflavin did not affect viral infectivity nor did it create a toxic effect on the reporter cells (data not shown). The efficacy of Mirasol PRT treatment, at the optimal and validated dose,38,40 was investigated in pooled BC PLT concentrates in AS. Under these conditions, Mirasol treatment resulted in a mean of 2.33 log IU/mL, or 99.25%, reduction in RRV infectivity (Table 2). For BFV, the mean level of viral inactivation was 1.97 log IU/mL, or 98.68%, while for MVEV, infectivity was reduced by 1.83 log IU/mL, or a 98.42% reduction (Table 2). None of the PLT concentrates used for the inactivation studies described in Volume 55, April 2015 TRANSFUSION 827



TABLE 2. Reduction of RRV, BFV, or MVEV in PLT concentrates after Mirasol treatment TCID50 (log IU/mL) Before treatment After treatment RRV Bag R3 Bag R4 Bag R5 Bag R8 Mean ± SD BFV Bag B1 Bag B3 Bag B4 Bag B5 Mean ± SD MVEV Bag M1 Bag M2 Bag M3 Bag M4 Mean ± SD


5.13 5.92 5.88 5.92 5.71 ± 0.39

3.13 2.21 2.19 1.78 2.33 ± 0.57

5.87 6.13 7.01 6.13 6.29 ± 0.50

3.81 4.59 4.74 4.13 4.32 ± 0.43

2.06 1.54 2.27 2.00 1.97 ± 0.31

5.47 5.68 6.05 5.74 5.74 ± 0.24

3.86 3.86 4.06 3.86 3.91 ± 0.10

1.61 1.82 1.99 1.88 1.83 ± 0.16

* Virus detected below limit of detection (2 log IU/mL). IU = infectious units.

TABLE 3. The presence of viral neutralizing antibodies prevented the functioning of viral infectivity assays Spiked virus RRV



Reduction Percent

Unit number Bag R1 Bag R2 Bag R3* Bag R4* Bag R5* Bag R6* Bag R7* Bag R8* Bag B1 Bag B2 Bag B3* Bag B4* Bag B5* Bag M1* Bag M2* Bag M3* Bag M4*

Viral infectivity compromised Yes Yes No No No Yes Yes No No Yes No No No No No No No

Neutralizing antibody titer 160 80

The effect of riboflavin and ultraviolet light on the infectivity of arboviruses.

Arboviruses are an emerging threat to transfusion safety and rates of infection are likely to increase with the increased rainfall associated with cli...
96KB Sizes 5 Downloads 10 Views