Background
Human noroviruses are the most prevalent cause of food-borne illness in the US with an estimated overall infection rate that may be as high as one in six persons annually [
36]. These viruses are known to be highly resilient, persisting in the environment and foods for periods of a month or more [
18]. Nonthermal intervention options that are able to inactivate these noroviruses and other enteric viruses in foods and pharmaceuticals are limited. Sunlight has been implicated as a means by which viruses are environmentally inactivated [
10,
11,
14,
21,
22,
37,
38]. Indeed human norovirus was first known as the winter vomiting disease since its person-to-person transmission peaks during fall and winter months [
13,
52] and its elevated presence in sewage and shellfish is typically observed in the winter season [
4].
Atmospheric sunlight at sea level contains visible light (400–700 nm) but also contains a UV component, that is predominately 300–400 nm [
3,
12,
25] which does not substantially penetrate water since UV is strongly absorbed. Recent research has indicated that sunlight can interact with dissolved organic molecules in aqueous settings to catalyze formation of O
2 (a
1∆
g), commonly referred to as singlet oxygen, as well as other reactive oxygen species [
7,
9,
22,
35]. These complex organic molecules act as sensitizers that become energetically excited by UV–Vis light and then transfer electronic energy to oxygen molecules producing singlet oxygen. In fact this basic mechanism, by which light interacts with sensitizer molecules, is commonly utilized to inactivate enveloped viruses within plasma and blood products [
5,
6,
8,
23,
29‐
31,
34,
40]. It has also been shown that blue (405 nm) light can inactivate bacteria via its interactions with porphyrins (5 and 6 carbon multi-ring contain alternating single and double bonds) within the membrane of bacteria [
28], and is actively being investigated as a means of disinfection of medical and food surfaces ([
2,
17,
19,
27,
39]). Complex compounds such as methylene blue, rose bengal (a.k.a. red food dye 105) and even riboflavin (a.k.a vitamin B2) can also function as sensitizers to produce singlet oxygen [
15,
24,
50].
Ultrashort pulse laser (USPL) light treatments were previously shown to be capable of inactivating murine norovirus (MNV) and other viruses [
42,
45,
47‐
49]. Impulsive stimulated Raman scattering (ISRS) was the postulated inactivation mechanism [
43,
44,
46]. Essentially the ISRS hypothesis was that high frequency resonance vibrations are potentially induced by the 425 nm USPL, with a bandwidth of 420–430 nm, that may be capable of causing vibrations of sufficient strength that the capsid is destroyed after nonthermal treatments of an hr. or more [
43]. Here we evaluate whether the mechanism of laser-induced virus inactivation is strictly wavelength-dependent and pulse width-dependent using a tunable Ti:Sapphire USPL, having an 76 MHz repetition rate and a pulse width of 120 fs, and a continuous wave (CW) 408 nm wavelength laser.
Discussion
Previous research indicated that a 425 nm USPL with an average power of 250 mW was able to non-thermally inactivate nonenveloped viruses and bacteriophages [
42‐
46,
48,
49]. Essentially, vibrational resonance induced through an ISRS mechanism was hypothesized in which low frequency vibrational resonance modes of the virus capsid are induced that damage or destroy the virus capsid [
43]. To test this hypothesis, experiments were carried out to investigate the dependence of pulse width and wavelength on virus inactivation, using a tunable USPL and a CW laser.
It was conceivable that wavelengths within the 420–430 bandwidth of the 425 nm laser light delivered in femtosecond pulses would be required to produce this virus inactivation effect. We report here that both a visible USPL ranging from 400 to 450 nm, and a CW laser (408 nm) are capable of inactivating MNV. We demonstrate that addition of photosensitizer molecules, known to generate singlet oxygen in the presence of light, greatly enhance inactivation of MNV, indicating that the virus is sensitive to singlet oxygen. This observation is essentially consistent with a recently published research investigation on the effects of 405 nm blue light on Tulane virus [
19]. Further, the addition of sodium bisulfate, an oxygen scavenger, substantially reduced CW laser inactivation, strongly implicating singlet oxygen as the cause of visible laser light-induced virus inactivation.
Results of this study may conflict with a previous study which postulated that USPL induced impulsive stimulated Raman scattering (ISRS; [
43] causing virus inactivation. The hypothetical ISRS inactivation mechanism would involve using a USPL to excite resonant vibrational modes capable of shaking the virus capsid apart. As the original laser used in the previous study was not tunable, it could not be determined whether observed inactivation was specific to 425 nm light, or was not specifically wavelength-dependent. Here, virus inactivation experiments were conducted with MNV using a wavelength-tunable USPL to demonstrate that inactivation is not wavelength-specific, since inactivation occurs with a range of USPL wavelengths between 400 and 450 nm. In these experiments, the optical band is approximately 10 nm in spectral width, approximately 5 nm above and below the tuned wavelength setting. ISRS theory dictates that the optical band must be broad enough to cover both a pump and Stokes wavelength, and the pulse width has to be shorter than the period of the vibrational mode to cause coherent excitement. Thus, a priori, one would not expect that vibrational resonance induced by ISRS to necessarily be wavelength-dependent. However, due to pulse-width dependence, it is well understood that ISRS cannot be induced using CW lasers since they do not generate light pulses. Here we clearly demonstrate that CW lasers can inactivate MNV, which potentially conflicts with the hypothesis that ISRS is the mechanism of inactivation as proposed by Tsen et al. [
43] although it remains formally conceivable that both ISRS and singlet oxygen mechanisms could both contribute to laser inactivation observed by the USP laser. Furthermore, previous experiments indicated limited MNV inactivation when 425 nm USPL intensity was below 80 MW/cm
2 [
43]. Based on a focal length of 10 cm and a beam diameter of 0.1 cm, we estimate the average intensity of the CW laser at a maximum power of 500 mW to be only 24 KW/cm
2 which is more than three orders of magnitude smaller than the previously observed 425 nm USPL threshold of 80 MW/cm
2. Also, it is difficult to envision a scenario in which the low concentrations of sodium bisulfite that reacts with, and sequesters, dissolved oxygen molecules within the MNV sample would inhibit a laser-induced vibrational mechanism.
In the previous study using the USPL 425 nm laser [
43], inactivation via a one photon absorption process [
41] was discounted based on the lack of absorption of MNV at wavelength 425 nm and inactivation of a gradient centrifugation-purified virus sample which should have removed most extraneous cellular proteins and complex organic molecules that might absorb blue light and potentially act as singlet oxygen photoenhancers. Indeed it was noted that purified MNV and MNV from a cell lysate had roughly equivalent inactivation rates [
43] suggesting that singlet oxygen enhancers are not substantially present in virus stocks, which are derived from virus-infected cell lysate. This suggests that either the virus capsid itself may function as an endogenous enhancer molecule, or that an enhancer may not be strictly required to produce singlet oxygen when interacting with intense blue light. To the latter point, published experiments with pressurized liquid oxygen identify several absorption peaks within the visible spectra that could directly excite oxygen [
16] suggesting this may be possible.
Previously, Tsen et al. [
43] noted highly damaged capsids via electron micrographs after 425 nm USPL treatments. It was suggested that hydrogen bonds and hydrophobic interactions critical for maintaining capsid integrity were overcome by laser-induced vibrational forces. However, these observations are also compatible with a mechanism involving photo-oxidation of amino acids such as cysteine, histidine, tyrosine, methionine and tryptophan within individual capsomer proteins [
32]. Formation of peptide aggregates due to crosslinking induced by photo-oxidation chains has been described previously [
7,
32]. Indeed when laser-treating mouse cytomegalovirus (MCMV), a herpesvirus, Tsen et al. [
44] noted “formation of large strongly bound aggregates of MCMV capsid and matrix proteins that did not readily disassociate under dissociating or reducing conditions.” This observation would also be consistent with a photo-oxidative mechanism induced by reactive oxygen in which covalent crosslinks are formed between oxidized amino acids from adjacent polypeptide chains. Lack of demonstrable damage to viral RNA by the USPL is consistent with previous reports [
42,
43,
48]. In H
2O, the lifetime of singlet oxygen is short, estimated to be approximately 3.5 microseconds [
51], providing limited time for diffusion into the virus capsid. For example, an estimate of O
2 (a
1∆
g) diffusion produced intracellularly was estimated to be only 10 to 160 nm [
33]. Thus singlet oxygen’s site of production must be very close to its site of action. Presumably, singlet oxygen generated outside the capsid would not have sufficient time to defuse within the virus capsid to substantially damage the viral RNA.
Overall the implication that reactive oxygen species are induced by light within the visible spectra has substantial bearing on the field of environmental virology, as well as food production. This finding offers a potential explanation as to why norovirus illness was originally termed the “winter vomiting disease.” It has been thought that cooler environmental temperatures permitted the virus to remain viable for longer periods [
1,
26]. Results here would seem to support the idea that shorter, less-intense solar irradiation in the winter may also contribute substantially to environmental persistence of noroviruses.
Acknowledgments
We wish to thank Gary. P. Richards (USDA, Dover, DE) and Brendan Niemira (USDA, Wyndmoor, PA) for critical reading of the manuscript. Mention of trade names or commercial products in this publication is solely for the purpose of providing specific information and does not imply recommendation or endorsement by the U.S. Department of Agriculture (USDA). The USDA is an equal opportunity provider and employer.