NEW MICROBIOLOGICA, 36, 257-265, 2013

Latent Herpesvirus 8 infection improves both insulin and glucose uptake in primary

endothelial cells

Angela Ingianni, Enrica Piras, Samuela Laconi, Fabrizio Angius, Barbara Batetta, Raffaello PompeiDepartment of Biomedical Sciences, University of Cagliari, Cagliari, Italy

INTRODUCTION

Human Herpesvirus 8 (HHV8) has a characteris-tic tropism for B lymphocytes and is known asthe causative agent of the Kaposi sarcoma (KS),as well as of some B cell lymphoproliferative dis-eases, namely primary effusion lymphoma (PEL)and multicentric Castleman disease (reviewed byAblashi et al., 2002; Hengge et al., 2002;Dourmishev et al., 2003; Ganem, 2010; Wen andDamania, 2010). KS lesions are characterized by neoangiogenesisand the production of typical spindle cells of en-dothelial origin. In vitro HHV8 infection of en-dothelial cells causes dramatic changes in the cel-lular phenotype resembling the spindle shape ofKS lesion cells (Ablashi et al., 2002). On the oth-er hand, the effects of lytic and latent HHV8 in-

Corresponding authorDr. Raffaello PompeiMicrobiology Section, University of CagliariVia Porcell, 4 - 09124 Cagliari, ItalyE-mail: rpompei@unica.it

fection on endothelial cell functions and trigger-ing of inflammatory processes are still largely un-known (Caselli et al., 2007; Gregory et al., 2012). HHV8 infection induces profound modificationsin the behaviour of both primary and immortali-zed endothelial cells. HHV8 causes an intense transcriptional repro-gramming in HUVEC cells (Wang et al., 2004),stimulates the Warburg effect in latently infectedTIME endothelial cells with an increase in gly-colysis and glucose consumption (Delgado et al.,2010) and activates hypoxia-induced factors(Carroll et al., 2006). In a recent work, Rose et al.(2007) found that in HHV8-infected dermal mi-crovascular cells (E-DMVEC) the expression ofthe insulin receptor (IR) was strongly induced inlatently infected cells. Binding of ligands to theIR triggered a signal cascade that led to a similaractivation of downstream effector molecules,which regulated cell growth and survival(Ottensmeyer et al., 2000; Raggo et al., 2005;McAllister & Moses, 2007). Over-expression of the IR in KS tissue comparedto normal skin was reported by Wang et al. (2004)

Human Herpesvirus 8 (HHV8), the causative agent of Kaposi sarcoma, induces a profound modification of infectedcell behaviour, with reprogramming of gene expression and changes in physiological properties, over-expression ofthe insulin receptor, increased resistance to stress conditions and prolonged cell survival in conditions of serum dep-rivation.This paper shows that HHV8 infection induces a strong enhancement of both insulin and glucose uptake in primaryendothelial cells (HUVEC). The increase in insulin uptake is already evident in the lytic phase of the viral infectiouscycle, and reaches a maximum of up to 71% during the latent phase, whilst glucose uptake is slightly depressed dur-ing the lytic viral infection, but significantly enhanced compared with the control during the latent phase of viral in-fection, with an average increase of about 37% 25 days after cell infection.

KEY WORDS: Cell-host interaction, Latent viral infection, Cell metabolism, Insulin uptake.

SUMMARY

Received December 6, 2012 Accepted March 23, 2013

and Rose et al. (2007). They independentlyshowed that IR gene expression was induced inHHV8-infected primary DMVEC, AIDS-KS tissueand immortalized endothelial (TIME) cells. HHV8 is believed to establish persistent infectionfor the duration of the host’s existence, with alimited gene expression program, which main-tains the circularized viral DNA genome duringcellular replication, with only occasional switch-ing to the lytic phase infection (Parravicini et al.,2000; Douglas et al., 2010). In addition, Watanabeet al. (2003) and Wang and Damania (2008) foundthat the HHV8 latency-associated nuclear anti-gen (LANA) prolonged the life span of primaryHUVEC cells and augmented cell survival in thepresence of apoptotic inducers and under condi-tions of serum deprivation. Many other worksstress that HHV8 infection induces intense anddurable changes in the physiological propertiesof HHV8-infected cells (Dourmishev et al., 2003;Carroll et al., 2006; Ganem, 2010; Guilluy et al.,2011), but very little is known about the meta-bolic modifications underlying these altered cellbehaviours. This work performed a series of experiments de-signed to verify whether overexpression of the IRcould affect insulin and glucose uptake in HHV8-infected HUVEC cells. The results demonstratedthat HHV8 induced a marked and significant en-hancement of both insulin and glucose uptake ininfected HUVEC cells during the latent phase ofviral infection.

MATERIALS AND METHODS

Cells and virusesHUVEC cells were purchased from Invitrogen(Life Technologies, UK) and grown in M200medium (Gibco, Life Technologies, UK) enrichedwith low serum growth supplement (LSGS,Invitrogen). HUVEC cells were always kept in asemi-confluent state and were sub-cultured atleast once a week. For all the experiments HU-VEC cells from sub-cultures 2 to 5 were used.HHV8 permanently infected BC3 cells were kind-ly donated by Caselli et al. (2007) and were grownin RPMI-1640 medium supplemented with 10%foetal calf serum (Invitrogen). BC3 cells wereused to produce 100x concentrated stocks ofHHV8, as described by Cerimele et al. (2001). The

virus pellet was suspended in RPMI, filteredthrough a 0.22 micron filter and kept at -80°C un-til used. The quantitative analysis of virusgenomes present in the stock preparation was ob-tained by a real-time polymerase chain reaction(rtPCR), with a Step-One Real Time System(Applied Biosystems, Life Technologies, UK) us-ing primers to amplify the orf 26 gene. The stan-dard HHV8 DNA curve was generated by a seri-al dilution of a plasmid (pBlueScript SK-ORF26)containing a cloned fragment of HHV8 DNA(ORF26, nucleotides 47261 to 47550). The puri-fied cell-free inoculums contained an averagenumber of 4.7x105 copies of viral DNA/mL. Forcell infection, about 5x104/mL HUVEC were seed-ed in 12 multiwell plates. The cells were infected with concentrated HHV8at a multiplicity of at least 10-20 genomes per cellin M200 medium containing 2 μg/mL of poly-brene for 2 h at 37°C. After 24-48 h the infectedcells were observed with a light microscope to de-tect the typical spindle cell morphology. Cell in-fection was examined by a cytofluorometricmethod. Only cell monolayers with at least 60%of HHV8 infected cells were used for the experi-ments.

PCR and RT-PCRThe presence and level of transcription of HHV8in HUVEC cells were analyzed by PCR and RT-PCR for the amplification of the orf 26 and orf73 genes (LANA). Genomic DNA was isolated us-ing the Easy-DNA kit (Invitrogen, UK) followingthe manufacturer’s instructions. Total RNA wasextracted with a PureLink RNA Mini kit (LifeTechnologies) and treated with TURBO DNase(Applied Biosystems) before the synthesis ofcDNA by random hexamers and Superscript II(Invitrogen, Carlsbad, CA, USA). PCR amplifica-tion was performed using, respectively, 100 ng oftotal DNA or 200 ng of total RNA extracted fromthe infected cells. More specifically, primers andconditions for PCR amplification were as follows:orf 26 took place with the primers orf26 fw 5’-GCCGAAAGGATTCCACCATTGTGCT-3’ andorf26 rev 5’- GGGCCCCGGCCGATATTTTGG-3’for 40 cycles (15’’ at 95° C, 1’ at 60° C and 15’’ at72°C) plus 10’ at 72°C of extension; orf 73 ampli-fication took place with primers orf73 fw 5’-ATC-CTCGGGAAATCTGGTCT-3’ and orf73 rev 5’-TTCAGCGTTTCAGTGTCTGC-3’ for 40 cycles

258 A. Ingianni, E. Piras, S. Laconi, F. Angius, B. Batetta, R. Pompei

(15’’ at 95° C, 1’ at 60° C and 15’ at 72°C) plus 10’at 72°C of extension. The amplification of thehousekeeping β-actin gene (ACTB) was used as acontrol. PCR products were run in 2% agarosegels and visualized by ethidium bromide stain-ing. Acquisition and processing of images wereperformed by PhotoDoc-It Imaging SystemDigital (UVP, Cambridge, MA, USA) andPhotoPaint expression (Corel, Ottawa, Canada).

qPCR and ELISA test for insulin receptorRT-PCR was done on a Step-One rtPCR System(Applied Biosystems). To normalize gene expres-sion between the control and HHV8-infected HU-VEC, β-actin was assessed. The total RNA isolat-ed by PureLink RNA Mini kit (Life Technologies)was treated with Turbo DNase (AppliedBiosystems); 2 μg of total RNA were used for thesynthesis of cDNA using a Superscript VILOcDNA Synthesis kit (Invitrogen). The followingprimers were selected by using Primer Expresssoftware (Applied Biosystems): act fw 5’-CAC-CATTGGCAATGAGCGGTTC-3’, act rev 5’-AG-GTCTTTGCGGATGTCCACGT-3’ for the expres-sion of the β-actin gene, and ins fw 5’-GAG-GCAGGGCTGCCAAT-3’, ins rev 5’-TC-CTGGGCAGACGTCACCGA-3’ for the expressionof the IR gene (IRα). The reactions were run us-ing cDNA 20-fold diluted with Power Sybr GreenPCR Master mix (Applied Biosystems) followingthe manufacturer’s instructions. The relative ex-pression values between the control and HHV8-infected HUVEC were calculated by the compar-ative cycle threshold (ct) method using RelativeQuantitation Study software (Applied Bio -systems). The specificity of the amplicons wasmonitored by analysing the dissociation curve ofthe amplified product. For the ELISA assay, theHUVEC cells were seeded in 24 multiwell platesat a multiplicity of 3x104 cells per well and in-fected or mock infected with HHV8 after 24-48h, as described previously. Cells were washedtwice with PBS at the scheduled times and keptin a serum-free medium for 2 h at 37°C. Theywere then fixed in iced methanol for 4 min at 4°C.After washing, the plates were treated for 2 h with1% BSA to neutralize any aspecific uptake of an-tibodies. The wells were washed again and a pri-mary anti-insulin receptor antibody, diluted 1:200in PBS, was added for 1 h at 37°C (rabbit anti-IRβ, Santa Cruz Biotechnology CA, USA). The

cells were washed and the secondary diluted1:300 antibody was added for 45 min at 37°C(goat anti-rabbit IgG, Sigma-Aldrich, Milan, It).Finally, the cells were washed and the peroxidasesubstrate was added for 15 min. Sample ab-sorbance was read at a WL of 370 in a Pharmaciaultrospec III spectrophotometer.

Direct insulin-peroxidase assayThe direct insulin uptake to HUVEC cells was an-alyzed with peroxidase-conjugated bovine insulinused at a dilution of 1:1000 in PBS, according tothe manufacturer’s instructions (SIGMA, 200-300IU/mg of protein). The assay was run as describedin the previous paragraph for the ELISA test. Theinsulin-peroxidase was kept on cells for 1 h at37°C, then the cells were washed 3 times and thesubstrate for peroxidase was added. After 15 min,the reaction was stopped and read in absorbanceat a WL of 370 (all the samples were processed intriplicate).

Insulin-gold assay, microscopy and imagingIn order to detect and visualize the insulin boundto the cell membrane receptors in situ, 10 nm in-sulin-gold (I-0391, Sigma Chemical Co., MO,USA) were used. Both controls and HHV8-in-fected HUVEC cells were diluted from stock cul-tures and seeded (in triplicate) at a density of2x105 in 35 mm glass-bottomed dishes (MatTek,Ashland, MA, USA) and cultured at 37°C in a 5%CO2 incubator in growth medium. 3, 14 and 25days after viral infection, the cells were washedand serum-free M200 medium was added for twohours. After this, cells were thoroughly washedwith PBS and fixed with iced methanol for 4 minat 4°C. After washing, the cells at the centre ofthe dish were covered with 100 µL of insulin-gold,diluted 1:20 according to the manufacturer’s in-structions. After one hour of incubation, the cellswere washed and silver enhancer (Silver en-hancer kit SE100, SIGMA, USA) was added to en-large colloidal gold labels, as suggested by themanufacturer. Silver enhancer activity wasstopped after 5-10 min with 2.5% sodium thio-sulfate and the cells were washed with PBS. Allobservations were performed with an OlympusIX 71 inverted wide-field microscope (Olympus,Tokyo, Japan) using 20x/0.7 and 60x/1.3 planapochromatic objectives (Olympus UPlanSAposeries). Twelve-bit images were captured using a

Insulin and glucose uptake in HHV8-infected cells 259

cooled CCD camera (Sensicam PCO, Kelheim,Germany). Nominal resolutions of images were0.3 μm/pixel and 0.1 μm/pixel for the 20x and 60xobjectives, respectively. In order to simplify imageanalysis and data reporting, the images were con-trast-inverted to obtain positive values (as in flu-orescence microscopy). Image analysis was per-formed with the ImagePro Plus package (MediaCybernetics, Silver Springs, MD, USA). At least10 microscopic fields and 200 cells were individ-ually selected and measured for each experi-mental group. Calculations were made with Excel(Microsoft Co., Redmond, WA, USA). Normalizeddata (reported in Figure 3) represent the averagepercentage of the density value (intensity per pix-el) ± standard error (SE). Statistical analysis was performed with GraphPadPrism (GraphPad Software Inc. La Jolla, CA,USA) software and Statistica (StatSoft, Tulsa, OK,USA). Comparisons between HHV8 and controldata were made by the nonparametric Mann-Whitney U test. Significance was set at p<0.05and p<0.01.

2-deoxy-D-glucose uptakeHUVEC cells were seeded in 6 multiwell platesat a concentration of 1.5x105/mL in M200 medi-um with LSGS. After 24 h, the cells were washedand infected for 2 h at 37°C with HHV8 at a mul-tiplicity of 10-20 viruses per cell, as described pre-viously. Then the plates were washed and incu-bated in complete M200 medium for 1-25 days.Both infected and mock-infected cells were pre-pared in triplicate for each sample. Glucose up-take was detected by the method described byNguyen et al. (2005). Briefly, the cells werewashed with Hank’s solution and glucose-freemedium was added for 1 h at 37°C on the sched-uled days. After this, 17 IU of insulin were added

260 A. Ingianni, E. Piras, S. Laconi, F. Angius, B. Batetta, R. Pompei

FIGURE 1 - Lytic and latent HHV8 infection of HUVECcells. A and B: Morphologic appearance of control andHHV8-infected sub-confluent HUVEC cells at a magni-fication of 10x10. The cells were always kept in a sub-confluent state and were sub-cultured at least once aweek. After 24 h from infection more than 80% of cellsshowed the typical spindle cell shape of HHV8-infectedendothelial cells. C) Detection of ORF26 and ORF73transcripts during lytic and latent HUVEC infection byHHV 8. M: markers; C+: ORF26 and ORF73 positivecontrols. The amplification of the housekeeping β-actingene was used as a control.

A

B

C

to each well for 10 min (recombinant human in-sulin, Sigma 91077C). Then 0.2 μCi/mL of [3H]2-deoxy-D-glucose (Perkin-Elmer, MA, USA) wereadded to each well and left for 10 min. The cellmonolayers were washed 3 times with cold PBS,dissolved in 1N NaOH and analyzed in a liquidscintillation counter.

RESULTS

HHV8 infection of HUVEC cellsAfter infection with HHV8, within 24 h almost allthe HUVEC cells underwent a strong shape mod-ification, with formation of the typical spindle cellaspect (Figure 1A, B). Analysis of the viral genomedemonstrated that during the first 7 days, only theorf 26 gene was expressed, which is representa-tive of a lytic infection, and it remained slightlyevident until the 14th day (Figure 1C). From day 14onwards, orf 73 expression (LANA), representa-tive of a latent infection, was clearly detectableand this was the only gene expressed on day 25.By cytofluorometric analysis the number of HHV8positive cells at day 25 after infection was foundto be still higher than 50%.

Production of insulin receptor in HHV8-infectedcellsThe production of the IR was studied by two dif-ferent methods. An ELISA assay was used to de-tect the IRβ-subunit, whilst a RT-PCR at days +3(lytic phase) and +25 (latent phase) was used toidentify and quantify the IRα-subunit. Both testsrevealed that IRs were over-expressed after HHV8infection in HUVEC cells. During lytic infectionat day 3, IR production appeared to be unaltered(as well as at days 7-14 in Figure 2A, data notshown) or slightly increased (Figure 2B). Amarked increase in the IRβ was evident on day14 after infection (Figure 2B), whilst a dramaticover-expression of both IRs subunits was detect-ed on the 25th day, which reached an increase ofbetween 90% (IRα, Figure 2A) and 135% (IRβ,Figure 2B) in infected cells compared to themock-infected ones. The differences observed be-tween infected and control cells on the 25th daywere found to be statistically highly significantfor both IRα and IRβ (p<0.01). All the tests wererepeated at least twice and, when necessary, theywere corrected to the number of cells.

Insulin and glucose uptake in HHV8-infected cells 261

FIGURE 2 - Expression of insulin receptor in HHV8-in-fected HUVEC cells. A: Quantitative RT-PCR of HUVECcell IRα cDNA. Empty column: mock infected cells; graycolumn: HHV-8 infected HUVEC cells 3 and 25 days af-ter infection. B: ELISA test for detection of InsulinReceptor (IRβ) in HUVEC cells infected with HHV8. M:Mock-infected cells (in triplicate); HHV8: HumanHerpesvirus 8-infected cells from days 1 to 25 after in-fection.

A

B

INS

ULI

N R

EC

EP

TO

R E

XP

RE

SS

ION

%IN

SU

LIN

RE

CE

PT

OR

%

262 A. Ingianni, E. Piras, S. Laconi, F. Angius, B. Batetta, R. Pompei

FIGURE 3 - Insulin binding and uptake in HHV8-infected HUVEC cells. A: Distinctive microscopic field portionsof insulin bound to cell membrane receptors in HHV8 infected and control HUVEC cells after 3, 14 and 25 daysfrom infection; the bar at the bottom left is 30 µm. Images were contrast inverted to analysis and visualization, bythis way (as in fluorescence microscopy) cells appear brilliant on dark background. Moreover, to highlight the dif-ferences between HHV8-infected and control cells, we used a pseudo-colour algorithm with the ImagePro Plus pack-age (Media Cybernetics, Silver Springs, MD, USA), that clearly identifies and distinguishes the mere silver enhancersingle or aggregated particles (blue) bound to the insulin-gold, from the physiological cellular contrast (green) typi-cal in wide-field microscopy. The algorithm sets up a “gate”, working as a “watershed”, thereby visually depicting thetrimmed data in the image. B: Normalized imaging data obtained by insulin gold assay are reported as the averagepercent of the density value (intensity per pixel) ± standard error (bars); significance was set up at p<0.05 (*) andp<0.01 (**). C: Direct insulin-peroxidase assay of HHV8-infected and mock-infected HUVEC cells from 1 to 25 daysafter infection. The data indicate the absorbance values at 370 WL. All the tests were done in triplicate and the val-ues were always normalized to the number of cells.

Insulin binding and uptake in HHV8-infectedHUVEC cellsTo verify if the over-expression of IRs could in-fluence the uptake of insulin, two different meth-ods were used on the HUVEC cells. The insulin-gold assay method was employed for detectingand visualizing the insulin bound to the cell mem-brane receptors in situ (Figure 3A). Imagingmethods were used to highlight the differencesin density and concentration of insulin-gold par-ticles, which were additionally enhanced by silverprecipitation. In this case, insulin binding start-ed increasing in the lytic phase of infection(+23.6% on day 3, p<0.01), augmented on day 14(+40.6%, p=0.02) and, finally, reached a maxi-mum during the latent phase on day 25 with anincrease of 52.5% (p<0.001) as compared to thecontrol. Insulin binding at day 7 did not show sig-nificant differences from day 3 (data not shown).Differences between HHV8-infected and controlcells were found to be statistically significant ona daily basis (Figure 3B). No significant differ-ence was detected between uninfected and mock-infected cells. According to these results, the di-rect insulin-peroxidase assay (Figure 3C) showedan improved insulin uptake by HHV8-infectedcells against the control starting from day 3,reaching a maximum value at days 14 and 25.Although a gradual decrease of total insulin bind-ing was observed from day 1 to 25, at day 25 af-

ter infection the insulin uptake by HHV8-infect-ed cells was about 71% higher than in controlcells.

Uptake of [3H]2-deoxy-D-glucose in HHV8-infected cellsThe over-expression of IRs and the rise in insulinuptake by HHV8-infected cells also had a signif-icant influence on glucose uptake (Figure 4).During the lytic infection, glucose uptake ap-peared to be slightly depressed until day 3, witha decrease of 16-22% compared to the control;from day 7 post infection onwards, the HHV8-in-fected cells showed a progressively increased up-take of glucose up to 23±7% on day 14, reachinga maximum of 37±9% on day 25. These valuesare the mean of 3 experiments and in all casesthe scored data were normalized for the numberof cells. The differences observed between HHV8-infected and control cells on the 25th day post-in-fection were statistically significant (p=0.026).

DISCUSSION

The results presented in this paper show thatHHV8 infection has a profound impact on bothinsulin and glucose uptake in primary HUVECcells. It is known that HHV8 is responsible fordramatic changes in the aspect and behaviour ofinfected endothelial cells. Spindle cell formationis a constant modification of cell morphology.Infected cells are more resistant to stress and tox-ic substances (Wang and Damania, 2008), presentan increase in hypoxia-induced factors (Caroll etal., 2006), and show increases in vascular per-meability (Guilluy et al., 2011). Moreover, Roseet al. (2007) demonstrated that HHV8-infectedcells undergo an intense modification of cellularinsulin receptors. In this paper Rose et al. (2007)data on IRs over-expression were confirmed. Inaddition, the insulin uptake by infected cells wasdemonstrated to be highly enhanced, reaching amaximum during the latent phase of HHV8 in-fection. The enhanced uptake of insulin led to anaugmented entry and accumulation of deoxy-glu-cose inside the infected HUVEC cells. After 25days of infection, the infected cells presented aglucose uptake that was around 40% higher thanthe controls. During the lytic infection, a slightdecrease of glucose uptake was observed in our

Insulin and glucose uptake in HHV8-infected cells 263

FIGURE 4 - [3H]2-Deoxy-D-glucose uptake in HHV8infected HUVEC cells. Black squares = Deoxy-glucoseuptake by HHV8-infected cells from days 1 to 25 afterinfection +/- SD. The deoxy-glucose uptake is reportedas percentage of the control. The values are the mean of3 experiments and in all cases the scored data were nor-malized for the number of cells.

experiments until the 3rd day; this fact could bedue to a temporary depression of cell metabolismcaused by the viral infection, but after 4-7 daysthe glucose income into the cells constantly in-creased as compared to the control, reaching amaximum on the 25th day after infection. There are many demonstrations that HHV8-in-fected cells acquire new stable properties and aremetabolically more efficient, but the biochemicalbasis of these modifications is still not understood.Wang and Damania (2008) found that HHV8 in-fection caused the activation of the prosurvivalphosphatidylinositol 3-kinase (PI3K)/Akt/mammalian target of rapamycin pathway, butwhich viral product was responsible for thesechanges has never been clarified. During the la-tent phase, HHV8 is known to express only a verylimited number of genes. LANA (orf 73) is a nu-clear antigen, necessary for viral latency (Stuberet al., 2007); during latency, vFLIP (orf 71) andvCyclin are produced in small amounts, while ka-posins (A-C, K12) are produced in large amountsand are mainly located in the cell membrane (Wenand Damania, 2010). Work is in progress to as-certain whether one or more of the viral factorsproduced during latency are involved in the cellmembrane modifications and metabolic changes. Whether these findings on increased insulin andglucose uptake in HHV8-infected cells can be re-ferred to general cell metabolism, requires muchmore investigation. It is known that HHV8, likemost other Herpesviruses, can persist in infect-ed persons throughout their life-span (Douglas etal., 2010) and, while clinical Kaposi sarcoma is arare disease, HHV8 infection shows a high preva-lence in some populations (Ingianni et al., 2007,2009). Recently Guilluy et al. (2011) and Gregoryet al. (2012) reported that HHV8 infection can ac-tivate the PI3K/AKT/mTOR pathway in endothe-lial cells and that the PI3K/AKT/mTOR signallingpathway is essential for the control of cell prolif-eration and also regulates anabolic activities with-in the cell. Since these in vitro growth character-istics mimic early-stage tumorigenesis, Rose etal. (2007) proposed that IRs up-regulation con-tributes to Kaposi sarcoma tumour development.Also Caselli et al. (2007) suggested that HHV8 in-fection may play an important role in promotingin vivo inflammation and pathogenic angiogene-sis typical of HHV8-associated lesions. In conclusion, over-expression of IRs, enhance-

ment of insulin binding and increase glucose up-take can give HHV8-infected cells a metabolic ad-vantage, that is required for tumour cell prolifer-ation and pathological neo-angiogenesis, as foundin clinical Kaposi sarcoma.

ACKNOWLEDGEMENT This work was performed with the contribution ofthe Fondazione Banco di Sardegna 2011-12.

REFERENCES

ABLASHI D.V., CHATLyNNE L.G.,WHITMAN G.E. JR,CESARMAN E. (2002). Spectrum of Karposi’s sarco-ma-associated Herpesvirus, or Human Herpesvirus8, diseases. Clin. Microbiol. Rev. 15, 439-464.

CARROLL P.A., KENERSON H.L., yEUNG R.S., LAGUNOFF

M. (2006). Latent Kaposi’s sarcoma-associatedHerpesvirus infection of endothelial cells activatehypoxia-induced factors. J. Virol. 80, 10802-10812.

CASELLI E., FIORENTINI S., AMICI C., DI LUCA D., CARUSO

A., SANTORO M.G. (2007). Human Herpesvirus 8acute infection of endothelial cells induces mono-cyte chemoattractant protein 1-dependent capil-lary-like structure formation: role of the IKK/NF-kBpathway. Blood 109, 2718-2726.

CERIMELE F., CURRELI F., ELy S., FRIEDMAN-KIEN A.E.,CESARMAN E., FLORE O. (2001). Kaposi’s sarcoma-as-sociated Herpesvirus can productively infect pri-mary human keratinocytes and alter their growthproperties. J. Virol. 75, 2435-2443.

DELGADO T., CARROLL P.A., PUNJABI A.S., MARGINEANTU

D., HOCKENBERy D.M., LAGUNOFF M. (2010).Induction of the Warburg effect by Kaposi’s sarco-ma Herpesvirus is required for the maintenance oflatently infected endothelial cells. PNAS. 107 (23),10696-10701.

DOUGLAS J.L., GUSTIN J.K., VISWANATHAN K., MANSOURI

M., MOSES A.V., FRUH K. (2010). The great escape:viral strategies to counter BST-2/tetherin. PLoSPathog. 6, 1-12.

DOURMISHEV L.A., DOURMISHEV A.L., PALMERI D.,SCHWARTz R.A., LUKAC D.A. (2003). Molecular ge-netics of Kaposi’s sarcoma-associated Herpesvirus(Human Herpesvirus 8) epidemiology and patho-genesis. Microbiol. Mol. Biol. Rev. 67, 175-212.

GANEM D. (2010). KSHV and the pathogenesis ofKaposi’s sarcoma: listening to human biology andmedicine. J. Clin. Invest. 120, 939-949.

GREGORy S.M., WANG L., WEST J.A., DITTMER D.P.,DAMANIA B. (2012). Latent KSHV infection of mono-cytes downregulates expression of adaptive immu-nity response costimulatory receptors and proin-flammatory cytokine expression. J. Virol. 86, 3916-3923.

264 A. Ingianni, E. Piras, S. Laconi, F. Angius, B. Batetta, R. Pompei

GUILLUy C., zHANG z., BHENDE P.M., SHAREK L., WANG

L., BURRIDGE K., DAMANIA B. (2011). Latent KSHVinfection increases the vascular permeability of hu-man endothelial cells. Blood. 118, 5344-5354.

HENGGE U.R., RUzICKA T., TyRING S.K., STUSCHKE M.,ROGGENDORF M., SCHWARTz R.A., SEEBER S. (2002).Update on Kaposi’s sarcoma and other HHV8 as-sociated diseases. Part 1: epidemiology, environ-mental predispositions, clinical manifestations, andtherapy. Lancet. Infect. Dis. 2, 281-292.

INGIANNI A., CARTA F., REINA A., MANAI M., DESOGUS A.,POMPEI R. (2007). Prevalence of Herpesvirus 8 in-fection in type 2 diabetes mellitus patients. Am. J.Infect. Dis. 3, 123-127.

INGIANNI A., MADEDDU M.A., CARTA F., REINA A., LAI C.,POMPEI R. (2009). Epidemiology of HumanHerpesvirus type 8 infection in cardiopathic pa-tients. Online J. Biol. Sci. 9, 36-39.

MCALLISTER S.C., MOSES A.V. (2007). Endothelial cell-and lymphocyte-based in vitro systems for under-standing KSHV biology. Curr. Top. Microbiol.Immunol. 312, 211-244.

NGUyEN M.T., SATOH H., FAVELyUKIS S., BABENDURE J.L.,IMAMURA T., SBODIO J.L., zALEVSKy J., DAHIyAT B.I.,CHI N.W., OLEFSKy J.M. (2005). JNK and tumornecrosis factor-alpha mediate free fatty acid-in-duced insulin resistance in 3T3-L1 adipocytes. J.Biol. Chem. 280, 35361-35371.

OTTENSMEyER F.P., BENIAC D.R., LUO R.z., yIP C.C.(2000). Mechanism of transmembrane signaling:insulin binding and the insulin receptor.Biochemistry. 39, 2103-2112.

PARRAVICINI C., CHANDRAN B., CORBELLINO M., BERTI E.,PALLI M., MOORE P.S., CHANG y. (2000). Differentialviral protein expression in Kaposi’s sarcoma-associ-

ated herpesvirus-infected diseases: Kaposi’s sarco-ma, primary effusion lymphoma, and multicentricCastleman’s disease. Am. J. Pathol. 156, 743-749.

RAGGO C., RUHL R., MCALLISTER S., KOON H., DEzUBE

B.J., FRUH K., MOSES A.V. (2005). Novel cellulargenes essential for transformation of endothelialcells by Kaposi’s sarcoma-associated herpesvirus.Cancer Res. 65, 5084-5095.

ROSE P.P., CARROLL J.M., CARROLL P.A., DEFILIPPIS V.R.,LAGUNOFF M., MOSES A.V., ROBERTS C.T. JR, FRUH

K. (2007). The insulin receptor is essential for virus-induced tumorigenesis of Kaposi’s sarcoma.Oncogene. 26, 1995-2005.

STUBER G., MATTSSON K., FLABERG E., KATI E., MARKASz

L., SHELDON J.A., KLEIN G., SCHULz T.F., SzELEKy L.(2007). HHV-8 encoded LANA-1 alters the high or-ganization of the cell nucleus. Mol. Cancer. 6, 28-33.

WANG H.W., TROTTER M.W., LAGOS D., BOURBOULIA D.,HENDERSON S., MAKINEN T., ELLIMAN S., FLANAGAN

A.M., ALITALO K., BOSHOFF C. (2004). Kaposi sarco-ma herpesvirus-induced cellular programming con-tributes to the lymphatic endothelial gene expres-sion in Kaposi sarcoma. Nat. Genet. 36, 687-693.

WANG L., DAMANIA B. (2008). Kaposi’s sarcoma-associ-ated herpesvirus confers a survival advantage to en-dothelial cell. Cancer Res. 68, 4640-4648.

WATANABE T., SUGAyA M., ATKINS A.M., AQUILINO E.A.,yANG A., BORRIS D.L., BRADy J., BLAUVELT A. (2003).Kaposi’s sarcoma-associated herpesvirus latency-associated nuclear antigen prolongs the life spanof primary umbilical vein endothelial cells. J. Virol.77, 6188-6196.

WEN K.W., DAMANIA B. (2010). Kaposi’s sarcoma-asso-ciated herpesvirus (KSHV): molecular biology andoncogenesis. Cancer Lett. 28, 140-150.

Insulin and glucose uptake in HHV8-infected cells 265