Previous Article | Next Article ![]()
Journal of Virology, September 2005, p. 11501-11506, Vol. 79, No. 17
0022-538X/05/$08.00+0 doi:10.1128/JVI.79.17.11501-11506.2005
Copyright © 2005, American Society for Microbiology. All Rights Reserved.
Department of Microbiology and Immunology, SUNY Upstate Medical University, Syracuse, New York 13210
Received 18 April 2005/ Accepted 25 May 2005
|
|
|---|
100-fold over 6 days, virions were abundant, and lesions developed that contained VZV antigens and resembled varicella and zoster lesions. The SOC system for VZV replication has applications for testing virus mutants and antiviral drugs. |
|
|---|
Alternatives to studying VZV in skin implanted in SCID-hu mice are organotypic raft cultures and skin organ culture (SOC). There are few reports of organ cultures being used to study infectious diseases (8, 21, 22). Instead, organotypic raft cultures, composed of fibroblasts embedded in a collagen matrix that feeds overlying keratinocytes, are frequently employed to study human papillomaviruses and less often for herpesviruses (6, 12, 25). Although both SOC and organotypic raft cultures are composed of dermis and epidermis, hair follicles and sebaceous glands are only found in SOC. There is mounting evidence that hair follicles are the site of VZV transfer to the skin from infected T cells, which express skin-homing markers that direct them to endothelial cells in the dermis (10). Thus, SOC appears highly suited for studying VZV in full-thickness skin where all the major cell types are present and are correctly positioned and differentiated.
Given the importance of studying VZV in human skin, we set out to facilitate this in several ways: increase the sampling frequency and number of skin tissues tested, shorten experiments from months to days, and reduce the number of SCID mice used. To achieve these goals, we developed an ex vivo system to study VZV in human fetal skin organ culture, which maintains skin tissues on NetWell supports as opposed to xenografts in SCID mice. Our hypothesis that SOC would be an effective system for investigating VZV skin tropism was supported by the results. We found that VZV replicated in all layers of the skin, causing typical lesions and producing abundant virions.
The first step in development of the SOC model for VZV replication was to determine the proper conditions for tissue sterility, integrity, and differentiation. Human fetal skin tissue (18 to 23 weeks gestational age) was purchased from Advance Biosciences Resources (Alameda, CA), in accordance with all local, state, and federal guidelines. The skin was disinfected by swabbing with Betadyne (10% povidone iodine), dipped in 70% ethanol, and rinsed in tissue culture medium [TCM: minimal essential medium with Earle's salts and L-glutamine, supplemented with 10% heat-inactivated fetal bovine serum, penicillin/streptomycin (5,000 IU/ml), amphotericin B (250 µg/ml), and nonessential amino acids (Mediatech, Washington, D.C.)]. TCM contains 1.4 mM Ca2+, which is needed for preservation of skin integrity since it quickly dedifferentiates in 0.15 mM calcium (26).
Skin tissue was cut into 1.0-cm2 pieces, placed individually in wells of a 24-well plate, and submerged in TCM supplemented with nystatin (10,000 U/ml) and ciprofloxacin (10 µg/ml) (Sigma, St. Louis MO) for 24 h in a tissue culture incubator (humidified 5% CO2 at 37°C). For long-term cultivation at 37°C, the tissues were placed individually in 500-µm mesh NetWell inserts (Corning, Corning, NY) that rested above 1.0 ml of TCM containing nystatin and ciprofloxacin in each well of a 12-well plate, thereby situating the tissues at the air-liquid interface. The skin explant was nourished through the dermis by contact with the TCM, which was refreshed every 2 days.
To observe the architecture of the skin over time, tissues were harvested after 0, 2, 4, and 6 days, then prepared for histological analysis. After 6 days, the skin appeared grossly normal, with only slight contraction from its original size, and hematoxylin and eosin-stained sections indicated that the cellular anatomy was preserved (Fig. 1A to D). Specifically, the dermis was composed of fibroblasts in a collagen matrix, with prominent hair follicles. The basement membrane separated the dermis from the epidermis, where stratified layers of basal cells and flattened keratinocytes were observed. The presence of keratin sheets on the top surface indicated that differentiation was occurring in the epidermis. The histological appearance of tissues from SOC was nearly identical to skin implants from SCID-hu mice (for review see 14) with the exception that SOC did not produce a thick layer of subcutaneous fat.
![]() View larger version (145K): [in a new window] |
FIG. 1. Histological analysis of VZV-infected skin organ cultures. Skin tissues were mock-infected (A to D) or infected with VZV (E to H), and then processed for histological analysis on days 0, 2, 4, and 6 after inoculation. Sections were stained with hematoxylin and eosin and examined by light microscopy and digital photomicrography. From the top down, the tissue compartments are sheets and whorls of keratin (pink), epidermis (blue), and dermis (pink), and hair follicles are oval structures in the dermis. Mock-infected tissues were unchanged over 6 days, while lesions formed in the VZV-infected tissue (white arrow). The black arrow indicates destruction of hair follicles. These sections are representative of more than 20 tissues examined. Magnification, 20x.
|
Tissues were incubated at 37°C for 2 h to allow the virus to adhere, and then placed individually on NetWells for cultivation as above. On days 0, 2, 4, and 6, VZV-infected skin tissues were harvested and processed for histological analysis to observe the location of VZV replication and any pathological changes. Light microscopy of fixed sections stained with hematoxylin and eosin revealed that mock-infected and VZV-infected tissues were indistinguishable on days 0, 2, and 4 (Fig. 1, A to C and E to G). By 6 days after inoculation, no lesions were visible by gross examination, although VZV-infected explants showed histopathology typical of varicella and zoster lesions, such as thickening of the epidermis, sloughing of keratinocytes (white arrow), and the presence of balloon cells (Fig. 1H). Destruction of hair follicles, as evidenced by loss of hematoxylin staining, was widespread (black arrow, Fig. 1H).
To confirm that lesions and tissue damage were caused by VZV infection, immunohistochemistry was performed. VZV proteins were detected using a polyclonal human immune serum (GK serum, kindly provided by Ann Arvin, Stanford University); a secondary antibody conjugated to alkaline phosphatase (goat anti-human alkaline phosphatase, Jackson ImmunoResearch Laboratories, Inc., Westgrove PA), and the signal was developed with Vector Red (Vector Laboratories, Burlingame CA). Tissues were counterstained with hematoxylin for contrast. After 6 days postinoculation, mock-infected skin reacted slightly with the antibodies in the outer keratinocytes, giving a faint red color (Fig. 2A). A strong red signal in the epidermis of VZV-infected skin indicated where viral antigens were present in a lesion (Fig. 2B). This lesion had been observed in adjacent sections that were stained with hematoxylin and eosin (Fig. 1H).
![]() View larger version (164K): [in a new window] |
FIG. 2. Immunohistochemistry of VZV-infected skin organ cultures. Tissues were mock-infected (A) or VZV-infected (B to F) and processed for immunohistochemistry on day 6 after inoculation. Virus antigens were detected with a high-titer human serum and visualized with the chromogen Vector Red, then counterstained with hematoxylin. Using a 20x objective, VZV-positive cells and debris (black arrows) were found in lesions (B), the epidermis, and hair follicles (D). Details within the white boxes are shown under higher magnification (C, E, and F, 40x).
|
To measure the extent and kinetics of VZV replication in the SOC system, we compared two modes of virus inoculation: intradermal injection and scarification. The rationale was that VZV infects skin from within, via dissemination of infected T cells (1), and externally, via breaks in the epidermis. Tissues were inoculated by injection as described above, or by scarification using a 27-gauge needle dragged across the skin 10 times while flooded in VZV-infected MeWo cells. Since the inoculum was freshly prepared to obtain the highest possible VZV titer, the virus concentration was not calculated before use. Instead, tissues were harvested 2 h after inoculation (time zero) and processed for titration by mincing tissue and placing them in 1 ml of TCM that contained 2.5-mm glass beads. After vortexing for 3 min., an infectious focus assay was performed to assess the actual amount of input virus received by the tissues following the method described (27). The tissues were incubated at 37°C on NetWells, and then four tissues were removed for virus quantitation by infectious focus assay after 2, 4, and 6 days.
Both inoculation methods resulted in a 100-fold increase in virus between days 2 and 6, and the doubling time was 17 h in tissues injected intradermally and 23 h in scarified tissues (brackets, Fig. 3). This is similar to the 16 to 20 h measured in cultured human foreskin fibroblasts (J. Moffat, unpublished observations). Scarification introduced approximately 10 PFU per tissue, whereas intradermal injection introduced over 100 PFU. Another difference was that VZV titer increased exponentially following scarification, but there was a decrease in titer in the first 2 days after intradermal injection. There is a possibility that some of the increase in titer was from VZV replication in the MeWo cells used for the inoculum, so the values stated may slightly overestimate the extent of virus growth in the skin tissue.
![]() View larger version (14K): [in a new window] |
FIG. 3. VZV replication in SOC. Tissues were inoculated by intradermal injection (circles) or scarification (triangles). The PFU/tissue was determined by standard plaque assay; each symbol represents one tissue. The doubling times were measured between 2 and 6 days (brackets).
|
![]() View larger version (156K): [in a new window] |
FIG.4. Electron microscopy of VZV in SOC. Tissues were infected for 6 days and then examined by electron microscopy. (A) Stages of VZV assembly were observed in an infected keratinocyte in the nucleus (N) and the cytoplasm (C), and the areas within boxes 1 to 3 are shown at higher magnification in the corresponding panels. Capsids formed in the nucleus and a DNA-containing capsid were adjacent to the inner nuclear membrane (1, white arrowhead). Enveloped virions were found in cytoplasmic vesicles (2 and 3, black arrowheads). Extracellular, intact virions are found in the spaces between cells (2 and 3, white arrowheads). Viral envelopes were also found in the intercellular spaces. (B) An uninfected dermal fibroblast (left) was coated with VZV particles and an extracellular (Ex) mass of VZV virions and viral debris was adjacent. The area within the box is shown at high magnification (right) and white arrowheads indicate three complete virions. The magnifications are labeled and the lengths of the black bars indicate relative size.
|
An additional question was whether VZV replication and assembly would differ in the various skin cell types. Thus, we examined infected dermal fibroblasts, notable for their lack of keratin, by transmission electron microscopy. VZV assembly in fibroblasts proceeded similarly to that observed in keratinocytes with the exception that very large accumulations of virions were seen outside cells (Fig. 4B), which may indicate differences in VZV egress in these cell types. This phenomenon, of cell membranes crowded with particles near masses of virions, was observed throughout the dermis. Of the extracellular virions, defective particles and viral debris were abundant, which is typical of VZV morphology, although many complete virions were also seen (Fig. 4B, white arrowheads). After several different tissues were examined, it was apparent that the predominant skin cell types, keratinocytes and fibroblasts, were susceptible to VZV infection and the assembly patterns were generally similar.
The SOC model presented here demonstrates the convenience and relevance of studying VZV replication in human skin ex vivo. The virus grew rapidly in the tissues, forming typical lesions and clusters of extracellular virions and viral debris. We were hopeful that these virions could be collected from the tissue, however, large amounts of infectious cell-free virus were not obtained using mechanical or enzymatic methods. VZV was difficult to extract due to the architecture of the skin, where cells are tightly joined and embedded in a collagen matrix. The inability to isolate concentrated, infectious VZV virions has been a barrier to using common and powerful techniques such as plaque purification and high-multiplicity infections, and so we continue to explore methods for isolating them in usable quantities.
The success of the SOC system could lead to several new research directions. These include VZV pathogenesis studies that are designed to distinguish the viral genes required in vitro from in vivo, and studying VZV assembly in differentiated skin, specifically the roles of tegument proteins and glycoproteins in this process. The SOC model could also provide a screening tool for antiviral agents because results can be obtained so rapidly, making it a useful first step before confirming results in mice with the SCID-hu model. Lastly, SOC could be applied to other skin-tropic viruses such as papillomaviruses and the human poxvirus molluscum contagiosum virus, which lack good tissue culture and animal models (4).
Transmission electron microscopy was performed with expert technical assistance from Maureen Barcza, Department of Pathology.
|
|
|---|
This article has been cited by other articles:
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Copyright © 2009 by the American Society for Microbiology. For an alternate route to Journals.ASM.org, visit: http://intl-journals.asm.org | More Info»