390 Volume 57, Number 4, 2003 APPLIED SPECTROSCOPY0003-7028 / 03 / 5704-0390$2.00 / 0q 2003 Society for Applied Spectroscopy

Spectroscopic Evaluation of the Effect of a Red MicroalgalPolysaccharide on Herpes-Infected Vero Cells

MAHMOUD HULEIHEL,* MARINA TALYSHINSKY, YELENA SOUPRUN, andVITALY ERUKHIMOVITCHThe Institute for Applied Biosciences, Ben-Gurion University of the Negev, Beer-Sheva, 84105 Israel

The sulfated polysaccharide obtained from a species of red microal-ga has proved to be a potent antiviral agent against various mem-bers of the herpes family. In the present study, we used microscopicFourier transform infrared spectroscopy (FT-IR) to investigate dif-ferences between normal cells, those infected with herpes viruses,and infected cells treated with red microalgal polysaccharide. FT-IR enables the characterization of cell or tissue pathology based oncharacteristic molecular vibrational spectra of the cells. The advan-tage of microscopic FT-IR spectroscopy over conventional FT-IRspectroscopy is that it facilitates inspection of restricted regions ofcell cultures or tissue. Our results showed signi� cant spectral dif-ferences at early stages of infection between infected and nonin-fected cells, and between infected cells treated with the polysaccha-ride and those not treated. In infected cells, there was an impressivedecrease in sugar content and a considerable increase in phosphatelevels in conjunction with the infection progress. Our results alsoproved that sugars penetrated and accumulated inside cells treatedwith the red microalgal polysaccharide. These could have been sug-ar fragments of low molecular weight present in the polysaccharidesolution, despite puri� cation by dialysis. Such sugar accumulationmight be responsible for a breakdown in the internal steps of theviral replication cycle.

Index Headings: Spectroscopy; Herpes virus; Fourier transform in-frared microscopy; FT-IR microscopy; Varicella zoster virus; Redmicroalgal polysaccharide.

INTRODUCTION

The Herpes family of viruses is responsible for a va-riety of severe infections in humans and animals. Includ-ed in this family are Herpes simplex types 1 and 2 (HSV-1, HSV-2) and Varicella zoster (VZV), which may infectskin, eyes, genitals, lungs, and the brain. In most cases,it is dif� cult to differentiate between herpetic infectionsand irritations or infections caused by agents such as bac-teria and fungi, particularly at early stages of the infec-tion.

Our previous data1,2 showed the potent antiviral activ-ity of the sulfated polysaccharide obtained from the redmicroalgae Porphyridium sp., which prevents the devel-opment of herpes virus infection in vitro (cell culture)and in vivo (laboratory animals). These results indicatedthat polysaccharide prevents the viral infection, either byblocking absorbance of the virus into the host cells and/or by blocking a step in the virus replication cycle. This� nding is dif� cult to explain since the polysaccharide hasa high molecular mass (MW ; 6 3 106 da), which shouldprevent its penetration into the infected cell.

In an attempt to clarify this apparent contradiction, we

Received 9 October 2002; accepted 22 November 2002.* Author to whom correspondence should be sent. E-mail:

mahmoudh@bgumail.bgu.ac.il.

applied FT-IR to the study of herpes-infected cells treatedwith the polysaccharide. FT-IR has enabled chemists andbiologists to reveal the biomolecular structure of proteins,nucleic acids, and lipids.3–6 The technique has been ap-plied in medicine,7 and initial efforts in the diagnosis ofcancers such as cancer of the cervix,8 colon,9 breast,10 andlung,11 as well as other disorders,12 have been encourag-ing. In addition to the use of FT-IR in tissue diagnosis,its role in the examination of body � uids has gained im-portance in the past few years. Mantsch’s group reportedthe successful diagnosis of arthritis based on NIR (nearinfrared) analysis of synovial � uid obtained from arthriticpatients.13 The middle-infrared (MIR) region was shownto be useful in the identi� cation of the pattern of disease,using the FT-IR spectrum of human sera.14 Quanti� cationof serum components such as glucose, total protein, cho-lesterol, and urea has been achieved with FT-IR spec-troscopy.15 Measurements of concentrations of urea, glu-cose, protein, and ketones in human urine have been per-formed using NIR spectroscopy.16 Our previous results17

proved the potential of FT-IR microscopy for detectionof cells infected with herpes viruses at very early stagesof the infection.

In the present study, we examined the FT-IR spectraof normal cells in culture and cells infected with herpesviruses in the presence or absence of the red microalgalpolysaccharide. Our results showed signi� cant spectraldifferences between infected and noninfected cells, andbetween infected cells treated with the polysaccharideand those not treated. There was an interesting inversecorrelation in treated and nontreated infected cells be-tween carbohydrate levels and phosphate levels.

MATERIALS AND METHODS

Cells and Viruses. Monkey kidney (Vero) cells weregrown at 37 8C in RPMI medium supplemented with 10%new-born calf serum (NBCS) and the antibiotics penicil-lin, streptomycin, and neomycin. HSV-1 and VZV stockswere used for infecting the cells.

Preparation and Puri� cation of Microalgal Polysac-charide. The polysaccharide produced from Porphyri-dium sp. was used in this study. The polysaccharideswere collected and puri� ed as previously described. Brief-ly, these polysaccharides are produced and secreted intothe growth medium of the red microalgae. The mediumwas collected, cells were removed by centrifugation, andthe supernatant containing the polysaccharides was dia-lyzed and lyophilized.18

Cell Infection and Estimation of Viral Infection.Monolayers of cells grown in 9 cm 2 tissue culture plateswere incubated at 37 8C for 2 h with HSV or VZV at

APPLIED SPECTROSCOPY 391

FIG. 1. FT-IR spectra in the region of 600–2000 cm21 of noninfectedVero cells, cells infected with 1 moi of HSV-1, and infected cells treatedwith 100 mg/mL of Porphyridium sp. polysaccharide immediately p.i.Spectra were obtained at 48 h p.i. Results are means of 5 different andseparate experiments for each cell culture. The SD for these means was#0.001. moi 5 multiplicity of infection.

FIG. 2. FT-IR spectra in the region of 1200–1400 cm21 of noninfectedVero cells, cells infected with 1 moi of HSV-1, and infected cells treatedwith 100 mg/mL of Porphyridium sp. polysaccharide immediately p.i.Spectra were obtained at 48 h p.i.

various multiplicities of infection (moi) in RPMI mediumcontaining 2% NBCS. The unabsorbed virus particleswere removed, and fresh medium containing 2% NBCSwas added. The monolayers were incubated at 37 8C. Atvarious times, post-infection (p.i.), the infected cells wereexamined by FT-IR microscopy. The cells were examinedunder an inverted light microscope for the appearance ofthe cytopathic effect (CPE), de� ned as areas of completedestruction of cells or of morphologically modi� ed cellsin the inspected � elds. The CPE was expressed as thepercentage of damaged cells in the inspected � eld.

Preparation of Slides. Since ordinary glass slides ex-hibit strong absorption in the range of wavelengths ofinterest to us, we used zinc sellenide crystals, which arehighly transparent to IR radiation. Normal cells from pas-sage 3–5 or cells infected with herpes viruses were pickedup from the tissue culture plates after treatment with tryp-sin (0.25%) for 1 min. The cells were centrifuged at 1000rpm for 5 min. Then each pellet was washed twice withsaline and resuspended in 100 mL of saline. The cellswere counted with a hematocytometer, and all tested sam-ples were centrifuged again and resuspended in saline ata concentration of 1000 cells/mL. A 1 mL drop of eachsample was placed on the sellenide crystal, air dried, andexamined by FT-IR microscopy. The radius of such a 1mL drop was about 1 mm.

Fourier Transform Infrared Spectra. The FT-IRmeasurements were performed in transmission mode witha liquid nitrogen cooled mercury cadmium telluride(MCT) detector of the FT-IR microscope (Bruker IRScope II), coupled to the FT-IR spectrometer (BRUKEREQUINOX model 55/S, OPUS software). The spectrawere obtained in the wavelength range 600–4000 cm21

in the MIR region. A spectrum was taken as an averageof 128 scans in order to increase the signal-to-noise ratio.Spectral resolution was 4 cm21. The aperture used in thisstudy was 100 mm. We found this aperture to yield thebest signal-to-noise ratio. The spectra were vector nor-malized after a rubber-band baseline correction. For each

cell type, at least � ve different measurements at varioussites were taken.

Statistical Analysis. The spectral differences betweennoninfected and infected cells in the region 1000–1480cm21 were analyzed statistically by the t-test.

RESULTS

Fourier Transform Infrared Spectra of Cells In-fected with Herpes Viruses. The FT-IR spectra of cellsinfected with 1 moi of herpes viruses differed markedlyfrom those of normal noninfected cells. In general, thecells infected with HSV-1 showed higher intensities ofabsorbance in most of the scanned region (Fig. 1). TheFT-IR spectra obtained 48 h p.i. of cells that were treatedwith 100 mg/mL of Porphyridium sp. polysaccharide im-mediately p.i. and with 1 moi of HSV-1 showed signi� -cantly lower intensities than the spectra of infected non-treated cells, and slightly higher intensities than those ofnormal noninfected cells (Fig. 1). In the region 1000–1200 cm21, the intensities of cells treated with polysac-charide were signi� cantly higher than those of nontreatedcells, whether infected or not. This is evidence of car-bohydrate accumulation in the treated cells.

Furthermore, the intensities of the absorbance in theregion 1200–1400 cm21, which includes assymetric phos-phate vibrations at 1235 cm21,19,20 were signi� cantly high-er ( P # 0.01) in infected cells than in noninfected cells.In infected cells treated with polysaccharide, the phos-phate levels were slightly higher than those in the non-infected nontreated control cells, but signi� cantly lowerthan levels in the HSV-infected nontreated cells (Fig. 2).

The dominant bands at 1655 and 1546 cm21 can beattributed to amide I and II vibrations, respectively.19 Thelevel of amide I was signi� cantly higher in normal non-infected cells than in cells infected with HSV withoutpolysaccharide treatment. There was no signi� cant dif-ference in the level of amide I between noninfected cellsand those in infected polysaccharide-treated cells.

Time Course of Carbohydrate Peak Behavior inCells Infected With Herpes Viruses. Vero cells wereinfected with 1 moi/cell of HSV-1 and examined by FT-

392 Volume 57, Number 4, 2003

FIG. 3. FT-IR spectra in the region of 1000–1200 cm21 of noninfectedVero cells and cells infected with 1 moi of HSV-1 at 48 and 94 h p.i.

FIG. 5. Phosphate and carbohydrate levels measured in HSV-1 infectedcells. (Upper ) Phosphate levels were calculated as peak area at 1235cm21; (lower ) carbohydrate levels were calculated as the peak area at1023 cm21. The results in this � gure are presented as a ratio betweenphosphate and carbohydrate levels in infected cells and the levels in thecontrol noninfected cells. Data are mean 6 SD (n 5 5).

FIG. 4. FT-IR spectrum of Porphyridium sp. polysaccharide at a con-centration of 100 mg/mL.

FIG. 6. A shift in the position of the FT-IR spectral peak at 862 cm21

as a result of infection of the Vero cells with HSV-1, compared tononinfected cells. Data are means 6 SD (n 5 5).

IR microscopy at various times p.i. Our data showed agradual disappearance of the peaks at 1023.3, 1053.8, and1156.6 cm21 over time, which correlated with the devel-opment of the viral infection (Fig. 3). These spectralpeaks can be attributed to carbohydrates, as previouslyreported.20 Peaks at positions 1036, 1061, and 1163 cm21

in the FT-IR spectrum of Porphyridium sp. polysaccha-ride (Fig. 4) appear to contribute to the peaks at 1023,1054, and 1156 cm21, respectively, in the spectra of thetreated vero cells (Fig. 3).

Our results also showed a good inverse correlation be-tween the phosphate peak area at 1235 cm21 and the car-bohydrate level (peak area at 1023 cm21) in Vero cellsinfected with HSV-1 (Fig. 5). In these infected cells, therewas a gradual decrease in carbohydrate levels (Fig. 5,upper plot) and a gradual increase in the phosphate levels(Fig. 5, lower plot), in correlation with the progress ofthe viral infection. Furthermore, our results showed agradual shift in the position of the peak at 862 cm21 toabout 854 cm21, in correspondence with the developmentof the viral infection (Fig. 6). This peak is attributed toN-type sugars.21,22 In contrast, there was no change in theposition of this peak in noninfected Vero cells during thetest period (Fig. 6). It is important to mention that in the

APPLIED SPECTROSCOPY 393

FIG. 7. Phosphate and carbohydrate levels measured in Vero cells in-fected with HSV-1 and treated with 100 mg/mL of Porphyridium sp.polysaccharide immediately p.i. for 72 h only or up to the end of theexperiment at 120 h p.i. As a control, the phosphate and carbohydratelevels were measured in noninfected Vero cells treated with 100 mg/mLof Porphyridium sp. polysaccharide up to the end of the experiment.(Upper ) Carbohydrate levels; (lower ) phosphate levels. Both carbo-hydrate and phosphate levels were determined as detailed in Fig. 5.Data are means 6 SD (n 5 5).

FIG. 8. Effect of Porphyridium sp. polysaccharide on the shift in po-sition of the peak at 862 cm 21 observed in HSV-1 infected cells. Verocells were infected with 1 moi HSV-1 and treated immediately p.i. with100 mg/mL of Porphyridium sp. polysaccharide for only 72 h or up tothe end of the experiment at 120 h. Data are means 6 SD (n 5 5).

noninfected control, Vero cells showed stable levels ofphosphates and carbohydrates during the test period (datanot shown).

Effect of Polysaccharide Treatment on Carbohy-drate and Phosphate Levels in HSV Infected Cells.Vero cells were infected with 1 moi of HSV-1 and treatedwith 100 mg/mL of Porphyridium sp. polysaccharide im-mediately after infection. The treatment with the poly-saccharide was either terminated at 72 h p.i. or continuedup to the end of the experiment at 120 h p.i. At varioustimes p.i., the cells were examined by FT-IR microscopy.

Our data showed a rapid increase and accumulation ofcarbohydrate levels in polysaccharide-treated cells duringthe � rst 72 h p.i., as con� rmed by an increase in absorp-tion in the region 1023 cm21 (Fig. 7, upper plot). Duringthis period, there was a moderate increase in phosphatecontents from 24 to 72 h p.i. After termination of treat-ment at 72 h p.i., there was a rapid decrease in the car-

bohydrate content with a concomitant rapid increase inphosphate levels (Fig. 7, upper and lower plots).

In cells treated with the polysaccharide continuouslyup to the end of the experiment (120 h p.i.), the carbo-hydrates reached saturation levels at 48 h p.i., while thephosphate levels continued to increase moderately (Fig.7). In noninfected control cells, the carbohydrate levelsstarted rising rapidly during the � rst 24 h of treatmentwith Porphyridium sp. polysaccharide, reaching satura-tion at about 36 h of treatment. However, the phosphatelevels in these cells remained stable during the experi-ment.

Our results, presented in Fig. 8, showed that in thecells that were infected with HSV and treated with Por-phyridium sp. polysaccharide up to the end of the exper-iment, there was no shift in the position of the peak at860 cm21. In contrast, when the treatment with the poly-saccharide was terminated at 72 h p.i., there was a rapidshift in the position of this peak to about 853 cm21.Changes in carbohydrate and phosphate contents in theinfected cells correlated with the development of the viralinfection (cytopathic effect). As can be seen in Fig. 9,during the � rst 72 h of infection and treatment with Por-phyridium sp. polysaccharide, the infection progressedslightly. When treatment was terminated at 72 h p.i., arapid increase in the development of infection occurredat 96 h p.i. In cells treated until 120 h p.i., there was acontinuous but moderate increase in infection.

DISCUSSION

In the present study we used FT-IR microscopy to an-alyze and study molecular changes in cells in culture afterinfection with herpes viruses. We also followed the mo-lecular changes resulting from treatment of these infectedcells with the sulfated polysaccharide from a red microal-gae, which had been previously proven to be a potentantiviral agent against various herpes viruses.1,2,23 FT-IRmicroscopy has been shown to be a sensitive and reliablemethod for detecting and tracking molecular changes incells as a result of viral infection17 or malignant transfor-

394 Volume 57, Number 4, 2003

FIG. 9. Kinetics of Porphyridium sp. polysaccharide activity on thedevelopment of HSV-1 cytopathic effect. Vero cells were infected with1 moi of HSV-1. At the end of the infection period, nonabsorbed virusparticles were removed by washing, and cell monolayers were fed withfresh medium ( m ) or medium containing 100 mg/mL polysaccharide( l , v ). The treatment with the polysaccharide either continued up tothe end of the experiment ( v ) or was terminated 3 days (72 h) ( l )after infection. Data are means 6 SD (n 5 5).

mation.9,10,24–26 Previously, we showed that by using FT-IR microscopy, we were able to detect early molecularchanges resulting from herpes infection of cells in cultureeven before any morphological changes could be seen.17

In the present study, we have shown the ability of FT-IR-microscopy to discriminate between cells infectedwith the herpes viruses with or without treatment withantiviral agents, such as a microalgal polysaccharide.

Our results showed a signi� cant accumulation of thecarbohydrates in the cells treated with the polysaccharide(Fig. 1). In all previous studies with such polysaccha-rides, investigators excluded the possibility of polysac-charide penetration of cells due to their high molecularweight. We concluded, however, that this carbohydrateaccumulation was most likely due to the entrance of thepolysaccharide into the cells. It seems that our polysac-charide solutions did, indeed, contain sugar fragmentswith low molecular weights, although the solutions hadbeen puri� ed by dialysis. The accumulation of carbohy-drate molecules (probably these small fragments) insidethe treated cells could be partially responsible for the an-tiviral activity of these polysaccharides against herpes vi-ruses. These results possibly yield a key to understandingthe mechanism of the antiviral activity of such polysac-charides.

It was found that the level of carbohydrates inside thetreated cells reached saturation after 36 h of treatment incontrol noninfected cells, as compared to 48 h in infectedcells (Fig. 7). The long period of time that elapsed untilsaturation levels of carbohydrates in infected-treated cellswere attained could support the possibility that the virus-es consumed some of the cellular carbohydrates.

Our results also showed a good inverse correlation be-tween carbohydrate levels and phosphate levels in cellsinfected with herpes virus (Fig. 5). A change in carbo-hydrate levels can be expected to have an impact on

phosphate levels since cell metabolism governs the syn-thesis of various metabolites containing the phosphategroup (such as ATP and GTP) and other biomolecularcomponents, including phospholipids, nucleic acids, andphosphorylated proteins. The explanation for our resultscould be that the infecting virus completely controls thelife cycle of the host cells. It consumes all the availablecellular enzymes and energy for its own rapid replication.These processes require high quantities of energy, mostof which come from host cell energy stores and produc-ers, mainly by oxidation of cellular carbohydrates.

In infected cells treated with Porphyridium sp. poly-saccharide immediately p.i., there was a rapid increase inthe carbohydrate levels, reaching saturation after 48 h oftreatment (Fig. 7, upper plot). The high levels of accu-mulated carbohydrates in these cells were accompaniedby a slight and moderate increase in phosphate levels(Fig. 7, lower plot) and a very low level of HSV infectiondevelopment (Fig. 9). After termination of the polysac-charide treatment at 72 h p.i., there was a rapid decreasein carbohydrate levels and a rapid increase in phosphatelevels (Fig. 7), accompanied by development of the cy-topathic effect (as con� rmed by morphological observa-tions) two days after the termination of the polysaccha-ride treatment (Fig. 9).

These results show a good correlation between the de-velopment of herpes virus infection, the decrease in car-bohydrate levels, and the increase in phosphate levels. Itseems that the antiviral activity of the polysaccharide isreversible and could depend on the accumulated carbo-hydrates in these cells. When treatment was terminatedat 72 h p.i., before the end of the experiments (120 hp.i.), phosphate levels rapidly increased—an indication ofinfection development (Fig. 7).

Although this reversible anti-herpes viral activity ofPorphyridium sp. polysaccharide had been observed inour previous studies,1,2,23 this is the � rst time we haveseen a direct correlation between the development of viralinfection and the levels of carbohydrates and phosphatesin cells, as con� rmed by FT-IR microscopy. Finally, ourresults showed a shift in the position of the peak at 860cm21 (attributed to N-type sugar21,22) as a result of infec-tion with HSV-1. This shift did not occur in infected cellstreated with Porphyridium sp. polysaccharide.

The results of this study once again proved FT-IR mi-croscopy to be a sensitive method for detection and in-vestigation of molecular and biochemical changes in cellsinfected with herpes viruses. FT-IR can be employed inthe study of the effect on infected cells of antiviral agents,such as polysaccharides.

1. M. Huleihel, V. Ishanu, J. Tal, and S. Arad, J. Appl. Phycol. 13,127 (2001).

2. M. Huleihel, V. Ishanu, J. Tal, and S. Arad, J. Biochem. Biophys.Methods 50, 189 (2002).

3. B. Schrader, Infrared and Raman Spectroscopy (VCH, Weinheim,1995).

4. H. Mantsch and D. Chapman, Infrared Spectroscopy of Biomole-cules (John Wiley and Sons, New York, 1996).

5. E. A. Cooper and K. Knutson, Pharm. Biotechnol. 7, 101 (1995).6. K. Brandenburg and U. Seydel, Chem. Phys. Lipids 96, 2340

(1998).7. M. Diem, S. Boydston-White, and L. Chiriboga, Appl. Spectrosc.

53, 148 (1999).

APPLIED SPECTROSCOPY 395

8. M. A. Cohenford and B. Rigas, Proc. Natl. Acad. Sci. U.S.A. 95,15327 (1998).

9. S. Argov, J. Ramesh, A. Salman, I. Sinelnikov, J. Goldstein, H.Guterman, and S. Mordechai, J. Biomed. Opt. 7, 1 (2002).

10. R. K. Dukor, M. N. Liebman, and B. Johnson, Cell. Mol. Biol. 44,211 (1998).

11. H. P. Wang, H. C. Wang, and Y. K. Huang, Sci. Total Environ. 204,283 (1997).

12. L. P. Choo, J. R. Mans� eld, N. Pizzi, R. L. Somorjai, M. Jackson,W. C. Halliday, and H. H. Mantsch, Biospectroscopy 2, 141 (1995).

13. R. A. Shaw, S. Kotowich, H. H. Eysel, M. Jackson, G. T. Thomson,and H. H. Mantsch, Rheumatol. Int. 15, 159 (1995).

14. W. Petrich, B. Dolenko, and J. Fruh, Proc. SPIE-Int. Soc. Opt. Eng.3918, 91 (2000).

15. R. A. Shaw, S. Kotowich, M. Leroux, and H. H. Mantsch, Ann.Clin. Biochem. 35, 624 (1998).

16. R. A. Shaw, S. Kotowich, H. H. Mantsch, and M. Leroux, Clin.Biochem. 29, 11 (1996).

17. A. Salman, V. Erukhimovitch, M. Talyshinsky, and M. Huleihel,Biopolymers (Biospec.) 67, 406 (2002).

18. O. Dubinsky, O. Simon, Y. Karamanos, S. Geresh, Z. Barak, andS. Arad (Malis), Plant Physiol. Biochem. 30, 409 (1992).

19. M. Huleihel, A. Salman, V. Erukhimovitch, R. Jagannathan, Z.Hammody, and S. Mordechai, J. Biophys. Biochem. Methods 50,111 (2002).

20. R. K. Dukor, in Handbook of Vibrational Spectroscopy, J. M. Chal-mers and P. R. Grif� ths, Eds. (John Wiley and Sons, New York,2001), p. 3335.

21. A. E. Taillandier and J. Liguier, in Handbook of Vibrational Spec-troscopy, J. M. Chalmers and P. R. Grif� ths, Eds. (John Wiley andSons, New York, 2001), p. 3465.

22. B. K. Brandenburg and U. Seydel, in Handbook of VibrationalSpectroscopy, J. M. Chalmers and P. R. Grif� ths, Eds. (John Wileyand Sons, New York, 2001), p. 3335.

23. M. Witvrouw and E. De Clercq, Gen. Pharmacol. 29, 497 (1997).24. S. Mordechai, J. Mordechai, J. Ramesh, C. Levi, M. Huleihel, V.

Erukhimovitch, A. Moser, and J. Kapelushni, Proc. SPIE-Int. Soc.Opt. Eng. 4491, 243 (2001).

25. M. Huleihel, A. Salman, V. Erukhimovitch, J. Ramesh, Z. Ham-mody, and S. Mordechai, J. Biochem. Biophys. Methods 50, 111(2002).

26. M. Huleihel, V. Erukhimovitch, M. Talyshinsky, and M. Karpasas,Appl. Spectrosc. 56, 640 (2002).