Article Text

Download PDFPDF

Original article
Inhibition of viral replication in vitro by antiviral-treated amniotic membrane. Possible use of amniotic membrane as drug-delivering tool
  1. R Mencucci1,
  2. I Paladini1,
  3. U Menchini1,
  4. J J Gicquel2,
  5. R Dei3
  1. 1Eye Clinic, Department of Oto-Neuro-Ophthalmological Surgical Sciences, University of Florence, Florence, Italy
  2. 2Department of Ophthalmology, Poitiers University Hospital, Poitiers, France
  3. 3Microbiology Unit, Department of Public Health, University of Florence, Florence, Italy
  1. Correspondence to Rita Mencucci, Eye Clinic, Department of Oto-Neuro-Ophthalmological Surgical Sciences, University of Florence, Viale GB, Morgagni 85, 50134 Florence, Italy; rita.mencucci{at}


Purpose To investigate if amniotic membrane (AM) incubated with antivirals can inhibit viral growth in vitro.

Methods AM samples were incubated with a solution of acyclovir or trifluridine. The treated AM was placed onto monolayers of Vero cells, a continuous cell line from monkey kidney, infected with herpes simplex virus. Viral growth was assessed in comparison to control infected cells by direct examination with an inverted microscope at low magnification for the presence and extension of the typical cytopathic effect, or by estimation of viral genomes.

Results AM soaked in acyclovir or trifluridine inhibited significantly the development of herpes simplex virus in cell cultures, based on the viral growth compared with controls. Non-treated AM did not significantly affect viral replication.

Conclusions Our preliminary in vitro data show that antiviral-treated amniotic membrane can inhibit viral replication. Therefore, the possibility to combine the previously published anti-inflammatory properties of AM with the capability to absorb antivirals and sustain drug release could be taken into consideration.

  • Amniotic membrane transplantation
  • antiviral-treated amniotic membrane
  • ocular surface
  • cornea
  • microbiology
View Full Text

Statistics from

The use of amniotic membrane (AM) in the treatment of epithelial defects dates around the first quarter of the 20th century; its use in the treatment of ophthalmological epithelial defect was first reported in 1946.1 Later, only a few reports were published.2 3 The breakthrough came with the development of modern preservation technology.4 5

The use of fresh AM is hindered by the risk of transmission of blood-borne infections; this made improvement of cryopreservation necessary, although differences between fresh and cryopreserved AM have been reported.6

AM, consisting of thick basement membrane and an avascular stromal matrix, is able to express multiple antiangiogenic factors, anti-inflammatory proteins, growth factors, and protease inhibitors. The recent surge of interest in the ophthalmic uses of AM has led to its application in a large number of conditions for a variety of indications such as, among others, conjunctival and corneal reconstruction, inflammatory disorders and ocular infections.4 5 7

Many mechanisms have been invoked in relation to its clinical application, such as reduction of inflammation, inhibition of vascularisation, anti-infective properties, promotion of epithelialisation and limitation or prevention of scarring.8–10

Amniotic membrane transplantation (AMT) has become somehow a routine intervention in a variety of ocular pathological conditions, infective or not.11 When used as a graft (epithelial side up), AM is expected to become incorporated in the recipient tissue. If it is used as a patch (epithelial side down), it works as a biological bandage affording a cover for a limited duration or a combination of these. The use of AM has been also suggested in the treatment of infectious keratitis because of its intrinsic anti-infective properties probably mediated by its anti-inflammatory effects and because AM may act as a long-term drug delivery system.12–14 In this respect, the utilisation of AM also in wound healing, skin lesions and burns has to be mentioned.15–18

Addressing the use of AM to deliver drugs, we had previously reported the antibacterial activity of antibiotic (Netilmicin)-treated AM in an in vitro model. Antibiotic uptake was dose dependent and occurred rapidly; our observations seemed promising in view of the possible clinical applications.19

AMT has been recently proposed for the management of ulcerative and necrotising herpetic stromal keratitis in conjunction with antivirals and corticosteroids.20 In Europe, nucleotide analogues such as acyclovir (ACV) and trifluridine (TFU) are often used topically for the treatment of stromal herpes keratitis. ACV is a synthetic purine nucleoside analogue with in vitro and in vivo inhibitory activities against herpes simplex virus (HSV) types 1 and 2 and varicella-zoster virus. The inhibitory activity of ACV is highly selective owing to its affinity for the enzyme thymidine kinase encoded by HSV and varicella-zoster virus. This viral enzyme converts ACV into ACV monophosphate, a nucleotide analogue. The monophosphate is further converted into diphosphate by cellular guanylate kinase and into triphosphate by a number of cellular enzymes. The block of viral DNA is accomplished in three ways: (1) competitive inhibition of viral DNA polymerase, (2) incorporation into and termination of the growing viral DNA chain and (3) inactivation of the viral DNA polymerase. TFU is a synthetic pyrimidine nucleoside analogue that inhibits enzymes of the DNA pathway and is incorporated into progeny viral DNA (but also cellular) causing imperfect transcription of late messenger RNA, hence the production of useless viral proteins. Its range of activity includes HSV-1, HSV-2, cytomegalovirus, vaccinia virus and possibly adenovirus. Because of its toxicity, topical preparations only are available.21 The aim of our study was therefore to evaluate AM as a delivering system for antiviral drugs. We report an in vitro study assessing the inhibition of viral replication by AM treated with ACV and TFU.

Materials and methods

Amniotic membrane

The Tuscan Regional Cornea Bank, Lucca, Italy, kindly donated the human material. One preserved AM was threphined into 10 AM samples, diameter 2.5 cm. Each sample was weighted.


A clinical isolate of HSV-1 from an orofacial lesion was used throughout the study. A stock suspension of the virus grown in cell cultures was aliquoted and stored frozen.


ACV, for intravenous use, 250 mg, and TFU eyedrops 1% were used in the study. ACV was dissolved in balanced salt solution (BSS, ICN Biomedicals Inc. Aurora, Ohio, USA) to 50 mg/ml; ACV solution and TFU eyedrops were further diluted in BSS, down to 10 mg/ml for the first and 0.08 mg/ml for the latter.

AM preparation

The AMs were trephined into 7 mm diameter samples, with a corneal trephine (Operaid, OPHTEC, Groningen, The Netherlands). The samples were washed for 5 min in BSS, laid onto a sterile gauze to remove the excess of liquid and then placed in individual wells of 24-well microplates containing 0.5 ml of the antiviral solution or BSS (control). After 60 min incubation at 37°C on a benchtop shaker, the fluid was aspirated. The AM samples were washed three times × 5 min in BSS to remove the non-adsorbed drug and the excess of liquid removed by draining.

Cell cultures

Vero cells, a continuous cell line from monkey kidney, were purchased from Istituto Zooprofilattico Sperimentale della Lombardia e dell'Emilia-Romagna, Brescia, Italy. Vero cell monolayers appear by direct inspection of the culture plates through an inverted tissue culture microscope (LEICA DMIL, LEICA, Wetzlar, Germany) as polygonal-shaped cells firmly attached to the bottom of the culture vessel. Cells were grown in minimal essential medium (MEM)–10% fetal calf serum (ICN Biomedicals Inc, Aurora, Ohio, USA).

Cell cultures infection and assessment of viral replication

Vero cells were used 24 h after seeding in 24-well plates, when they reached semiconfluence as assessed by inverted microscope examination. The growth medium was removed and the cell monolayers washed. Cells were inoculated with 200 μl viral suspension. After a 90 min incubation at 37°C in 5% CO2, the inoculum was removed, and 1 ml of the maintenance medium (minimal essential medium–2% fetal calf serum) was added. One AM sample (either antiviral or BSS treated) was placed, epithelial side up, over the cell culture in each well. Plates were incubated at 37°C in 5% CO2 up to 5 days. Cell monolayers were inspected daily to evaluate, in comparison with monolayers of non-infected cells and infected cells without addition of AM, the development of the typical cytopathic effect (CPE): rounding up (large round cells starting in cluster or small foci, which then extend to the entire monolayer) and detachment of cultured cells.

Antiviral activity CPE extent

Each monolayer was observed at low magnification, 40×, with the inverted microscope; 10 diameters per well, that is the experimental point, was considered for the presence of clusters of enlarged round refractile cells in individual wells. The presence and extension of CPE were recorded as a percentage of each monolayer showing the morphological changes. Viral DNA assessment was carried out by quantitative polymerase chain reaction by means of HSV-2 Q-PCR Alert (Nanogen Advanced Diagnostics, Buttigliera Alta (TO), Italy) following the manufacturer s instructions. Briefly, at the end of the incubation period, the cultures were harvested, pooled according to treatment, washed and resuspended in equal volumes of lysing solution. The oligonucleotide mixture was added to the reaction mixture, together with the specific fluorescent probe, and 20 µl/well were transferred to the amplification microplate. Five microliters of lysed samples was added, the plate sealed and transferred to the real-time thermal cycler. Positive (105 to 102 HSV DNA copies) and negative (sterile distilled water) controls were carried out in parallel. Viral DNA copies were evaluated in comparison with the positive controls.

Results were expressed as means (SD) and evaluated by one-way analysis of variance and the Bonferroni multiple comparison test.


The mean (SD) weight of the 10 AM samples was 0.23 (0.06) g (range, 0.16–0.30 g), indicating a certain variation among the samples. A preliminary experiment showed that AM soaked in 50 mg/ml ACV was capable to inhibit significantly the development of HSV-1 in cell cultures, as judged by the CPE extent compared with the controls (25% (7.1) of the monolayer vs 75% (7.1), p<0.02). The microscopic examination of various microscopic fields allows a comprehensive evaluation of each monolayer, thus making possible to quantitate the extension of the typical morphological alterations in each experimental point. The preliminary observation could not rule out a possible intrinsic antiviral effect of the AM; therefore, antiviral-soaked AM was compared with BSS-soaked AM and control-infected cells (table 1). Whereas ACV-AM did significantly inhibit viral replication, the addition of non-treated AM had no significant effect on viral replication. By considering the viral genomes on pooled cell cultures, the figures were consistent with the results shown in table 1: 6.2×105 HSV DNA copies for ACV-AM–treated cultures compared with 20.0×106 and 25.2×106 for AM-treated cultures and controls, respectively.

Table 1

Antiviral activity of AM treated 60 min with 50 mg/ml ACV

Table 2 shows the antiviral activity of AM treated with different concentrations of either ACV or TFU. Antiviral activity decreased by lowering the concentration of both antivirals. The inhibition of viral growth was significant in comparison with control or non-treated AM, with the exception of AM treated with 0.08 mg/ml TFU. It has to be added that AM soaked in 10 mg and 2 mg/ml TFU produced an aspecific toxic effect, evidenced microscopically by the rapid degeneration, within 1 day, of the entire monolayer.

Table 2

Antiviral activity of AM treated 60 min with different concentrations of ACV or TFU


HSV and varicella-zoster virus are potentially sight threatening. The tissue damage is sometimes directly dependent on the infectious agent replication but is also because of the inflammatory response, originally aimed to limit the infective process (harmful to the conjunctival and corneal epithelium).

AMT has been reported to be an effective surgical procedure for the reconstruction of the ocular surface and to reduce corneal inflammation.14 22–24

The use of AMT both in the early treatment of infectious corneal ulcers and in the management of acute ulcerative and necrotising stromal herpes keratitis has been recently suggested.14

Anti-infective properties of AM are probably related to its anti-inflammatory effects, thus counteracting the strong inflammatory response elicited by micro-organisms and not against the agent itself; therefore, an effective and adequate antimicrobial therapy is mandatory to ensure control of infection.

Heilingenhaus et al25 26 demonstrated how AMT significantly modifies the course of necrotising stromal keratitis induced by HSV-1. The effect was associated with suppression of inflammation, rapid epithelialisation and reduction of stromal necrosis.

In addition, in a mouse model of HSV stromal keratitis, AMT was effective in promoting corneal wound healing and in reducing inflammation, probably related to the reduced expression and activity of matrix and metalloproteinases and increased expression of tissue inhibitors of metalloproteinases.25 27

In a murine experimental model, the improvement of herpetic ulcerative keratitis after AMT is suggested to be because of a reduced expression and activity of matrix metalloproteases 8 and 9, increased or sustained expression of tissue inhibitors of metalloproteases 1 and 2.27 Concerns have been raised on the possibility that AMT in the management of infectious keratitis could be a risk factor for superinfections or viral reactivations, thus leading to a severe corneal damage and a disturbance of the normal healing process. However, besides this well-known anti-inflammatory effect of AM, an intrinsic antiviral property should be taken in consideration.

Paradowska et al28 reported that human placenta contains endogenous tumour necrosis factors and interferons, possible mediators of the non-specific antiviral immunity.

Moreover, AM may express cistatin E, a novel human cysteine proteinase effective in inhibition of viral replication.29–31 Chronic infection of the cornea by HSV remains an important cause of unilateral blindness. The development of non-toxic topical antiviral agents has been an important step forward in management of herpes keratitis.

AMT could be taken into consideration as adjunctive therapy for the anti-inflammatory and supposed direct antiviral properties of AM.26

The possibility to use AM to deliver drugs has been first suggested by Kim et al14 and Heilinghaus et al26 and then by Mencucci et al19 and Gicquel et al.13 As expected, AM can absorb ACV or TFU; antiviral uptake occurs quite rapidly. The antiviral effect of treated AM was evidenced by the development of the typical CPE and quantitation of viral genomes in tissue culture cells compared with control-infected cells.

The reduction compared with control cultures was significant with both antivirals, and it was dose dependent, as it was with antibiotic-treated AM. The incorporation of TFU into cellular DNA explains the aspecific toxic effect on tissue culture cells by AM treated with higher concentrations (10 and 2 mg/ml) of the drug. Its lack of selectivity is well known; on the other hand, AM treated with higher concentrations might be used in vivo: 10 mg/ml correspond to the concentration in the eye drop preparation. The difference could be explained by the fact that in vivo dilution and dispersion of the instilled drug occur and also that AM could achieve drug concentration. These observations confirm and extend our previous work on antibiotic-treated AM. As in the previous study, we could not find a significant effect by non-treated AM: the viral yield was only slightly reduced compared with the controls. Our in vitro model might not be suitable to detect the intrinsic antiviral effect. On the other hand, the lack of intrinsic anti-infective activity in our in vitro model does not rule out such an activity in vivo, where the anti-inflammatory and the healing properties of AM might be implemented.

The aim of our study was not to prove the superiority of AM treated with antivirals to the antiviral alone but the possibility to combine the anti-inflammatory properties of AM with the capability of delivering drugs. Moreover, the question about the patient's compliance: the intensive topical treatment to ensure adequate drug concentrations could be not so strictly necessary, thus improving the patient's comfort and also reducing the burden on the nursing staff; moreover, drug-soaked AM should prove superior when there are difficulties in frequent instillations such as with children, elderly or handicapped people.

In conclusion, our study, confirming and extending our previous observations on antibiotic-treated AM, suggests the potential application of antiviral-treated AM in the management of herpetic ocular infections as adjunctive therapy for its anti-inflammatory, supposed direct antiviral properties and as a drug delivery system.


View Abstract


  • See Editorial, p 1

  • Linked articles 184259.

  • Competing interests None.

  • Patient consent Obtained.

  • Provenance and peer review Not commissioned; externally peer reviewed.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Linked Articles