Localization of Tumor Necrosis Factor Receptor ...

Localization of Tumor Necrosis Factor Receptor Messenger RNA in Normal and Herpes Simplex VirusInfected Mouse Eyes Emmett T. Cunningham, Jr.,*f Anna Stalder,% Pietro Paolo Sanna,% Stephanie S. Liu* Floyd E. Bloom,X Edward L. Howes, /r.,f Iain L. Campbell,X and Todd P. Margolis*-\

Purpose. To investigate the distribution of p75 and p55 tumor necrosis factor receptor (TNFR) mRNA in normal mouse eyes and in mouse eyes acutely infected with McKrae strain herpes simplex virus (HSV). Methods. In situ hybridization with antisense 3oS-labeled riboprobes for p55 and p75 TNFR subtypes was used in uninfected and HSV-infected mouse eyes. Controls included the use of sense riboprobes and corneas inoculated with vehicle alone. Results. In uninfected and infected mouse eyes, in situ hybridization produced an autoradiographic signal for mRNA, encoding both p75 and p55 over the corneal endothelium, iris, ciliary body, choroid, and arachnoid layers of the optic nerve sheath. In addition, the signal was observed over scattered cells at the vitreoretinal interface. Signal for p75, but not p55, was observed over cells in the retinal ganglion cell layer. Acute HSV infection was accompanied by an intense leukocytic infiltrate in the conjunctiva, the corneal subepithelium and stroma, the anterior and posterior chambers, the iris root and ciliary body, and the vitreous cavity. In this setting, increased p75 and p55 mRNA signal was correlated closely with the number and location of receptor-bearing white blood cells. Signal over control sections hybridized with sense p75 and p55 TNFR cRNA probes was comparable to background. Signal over control eyes inoculated with sterile vehicle showed slight increased signal in the immediate vicinity of the traumatic keratitis, but otherwise it was comparable to that observed in uninfected animals. Conclusions. The observed distribution of p75 and p55 TNFR mRNA in normal and acutely infected mouse eyes, and particularly over the heavily vascularized uveal tract and over cells at the vitreoretinal interface, supports a role for TNF as a mediator of intraocular inflammation, perhaps as a key regulator of the blood-ocular barrier. Invest Ophthalmol VisSci. 1997;38:9-15.

A umor necrosis factor (TNF) exists in two forms: TNF-a and TNF-/?, also known as lymphotoxin. Both forms are secreted by activated white blood cells and share largely overlapping biologic activities, including potent proinflammatory and mitogenic effects in virtually all tissues studied to date.1"3 From * The Francis I. Proctor Foundation; the ^Department of Ophthalmology, School of Medicine, University of California, San Francisco; and the %Department of Neuropharmacology, Scripps Research Institute, La folia, California. Supported by a Career Development Award from Research to Prevent Blindness (TPM) and by National Institutes of Health grants EY10008 (TPM), MH47680 (FEB), and MH50426 (ILC). Submitted for publication February 7, 1996; revised August 29, 1996; accepted September5, 1996. Proprietary interest category: N. Reprint requests: Todd P. Margolis, The Francis I. Proctor Foundation, University of California Medical Center, 95 Kirkham Street, San Francisco, CA 94143-0944.

Investigative Ophthalmology & Visual Science, January 1997, Vol. 38, No. 1 Copyright © Association for Research in Vision and Ophthalmology

Intravitreal injection of TNF in laboratory animals produces a pronounced panuveitis, with disruption of the blood-ocular barriers, elevation of protein levels in the aqueous, and an influx of polymorphonuclear cells, monocytes, macrophages, and lymphocytes into aqueous and vitreous cavities.4"8 This observation is made more relevant by the observation that TNF appears to be produced by normal ocular tissues9"12 and has been implicated in the pathogenesis of experimental10'1213"15 and clinical forms of uveitis,16"18 corneal neovascularization,19"21 proliferative vitreoretinopathy,22"24 and central nervous system demyelination of the type found in optic neuritis.20"28 Both TNF-of and TNF-/? have been shown to interact with two distinct membrane-bound receptor subtypes

10

Investigative Ophthalmology & Visual Science, January 1997, Vol. 38, No. 1

termed p75 and p55 in a number of tissues.29 31 Sippy and associates recently provided evidence for expression of p55 mRNA by cultured human retinal pigment epithelial (RPE) cells, although p55 expression was not examined in RPE cells in situ.32 Moreover, no evidence has been provided for TNF-receptor expression by other ocular cell types. In this study, we applied a sensitive and specific in situ hybridization protocol33 to localize messenger RNA (mRNA) encoding the two TNF receptor (TNFR) subtypes in normal mouse eyes. We also examined eyes acutely infected with McKrae strain herpes simplex virus (HSV) because TNF has potent inhibitory effects on herpes viral replication,34'35 including replication in corneal epithelial cells and fibroblasts36; local TNF expression is correlated with decreased corneal scarring caused by HSV37; and the expression of mRNA for TNF and its receptor has been shown to increase dramatically in response to acute viral infection in other tissues.38'39

with a sterile 27-gauge syringe, and 20 to 30 fi\ of McKrae strain HSV (109 PFU/ml) were applied topically. The mice were allowed to recover from the anesthetic and were maintained for 3 days after inoculation, a time previously shown to correspond with the onset of a prominent ocular inflammatory response to corneal infection with McKrae strain HSV.40

MATERIALS AND METHODS

In Situ Hybridization Histochemistry Bluescript SK transcription vectors (Stratagene, La Jolla, CA) containing the full-length cDNA for the p55 and p75 murine TNFR subtypes were generously provided by Dr. Raymond Goodwin (Immunex Corporation, Seattle, WA).29 Detailed protocols for probe synthesis, prehybridization, hybridization, posthybridization, and autoradiographic localization have been published.33 In brief, high-specific activity sense and antisense riboprobes were synthesized by the incorporation of 35S-labeled cytidine triphosphate and uridine triphosphate with specific activities greater than 800

These studies were carried out in accordance with the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research and with the National Institutes of Heal til Guide for the Care and Use of Laboratory Animals. All experiments were performed on 8- to 10week-old female BALB/c mice (Simonsen, Gilroy, CA). Ocular Herpes Simplex Virus Infection Mice were anesthetized with CO2. A few drops of 0.5% proparacaine (Alcon, Fort Worth, TX) were then applied to the corneas, the epithelium was lightly scored

Tissue Preparation Deeply anesthetized normal and infected mice were perfused transcardially with 10 ml of 0.15 M NaCl, followed immediately by 20 ml of 10% neutral buffered formalin. The eyes were removed and allowed to postfix in the final perfusate at 4°C for 2 to 3 weeks, after which they were transferred to the same perfusate with 10% sucrose added as a cryoprotectant for 12 to 24 hours. Four l-in-4 series of 10 //m-thick cryostat sections through the entire eye were collected onto uncoated, positively charged slides (Fischer Scientific, Pittsburgh, PA).

FIGURE l. (top) Photomicrographs showing the distribution of mRNA encoding the p75 tumor necrosis factor receptor in uninfected (A,B) and McKrae herpes simplex virus (HSV)infected (C,D) mouse eye. Darkfield images demonstrate autoradiographic signal over the corneal endothelium, iris, ciliary body, choroid, and retinal ganglion cell layer in an uninfected animal (A,B). Darkfield (C) and brightfield (D) images of the same section demonstrate autoradiographic signal over the bulbar conjunctiva, corneal limbus, iris, ciliary body, choroid, and retinal ganglion cell layer of an animal infected with McKrae strain HSV, a distribution comparable to that observed in vininfected mice. In addition, infection with McKrae strain HSV was accompanied by an intense white blood cell infiltrate, seen here in the ciliary body, iris, and anterior vitreous (C,D). All photomicrographs are reproduced at the same magnification. Scale bar = 100 //m. FIGURE 2. (bottom) Photomicrographs showing the distribution of mRNA encoding the p55 tumor necrosis factor receptor (TNFR) in uninfected (A,B) and McKrae herpes simplex virus (HSV)-infected (C) mouse eye. Darkfield images demonstrate autoradiographic signal over the corneal endothelium, iris, ciliary body, and choroid. Darkfield image (C) demonstrates autoradiographic signal over the bulbar conjunctiva, cornea, anterior chamber, iris, ciliary body, and anterior choroid of an eye acutely infected with McKrae strain HSV. As with p75, this increase signal was accompanied by an intense influx of white blood cells, although increased signal over intrinsic cells, particularly keratocytes, could not be excluded. Darkfield image of an adjacent control section (D) hybridized with a sense riboprobe for the p55 TNFR demonstrates signal comparable to background. Scale bars = 100 //m.

TNF Receptor mRNA in Normal and HSV-Infected Mouse Eye

11

11


Ci/mmol with the Riboprobe II System (Promega, Madison, WI). Full-length templates used to generate antisense probes were obtained by linearization of the p55 and p75 cDNAs with Bgl II and Eco RI, respectively, and by the use of T3 RNA polymerase for transcription. Full-length p55 and p75 sense probes used as controls were linearized with Bam HI and Sac I, respectively, and transcribed with T7 RNA polymerase. Mounted sections were postfixed further in 10% neutral buffered formalin for 30 minutes, after which they were treated with proteinase K, acetylated, and dehydrated. The 35S-labeled cRNA probes were diluted in hybridization buffer at 107 cpm/ml, applied to the sections, and allowed to hybridize at 55°C to 60°C for 12 to 24 hours. Sections were then treated with RNase A to reduce background and were washed through progressively lower concentrations of saline-sodium citrate to reduce the salt content and to increase the stringency of hybridization. Finally, sections were dehydrated, exposed to Beta Max film (Amersham, Arlington Heights, IL) for 4 to 7 days, dipped in Kodak nuclear emulsion NTB3 (Eastman Kodak, Rochester, NY), dried, exposed for 2 to 4 weeks, developed, and counterstained with hematoxylin-eosin.

RESULTS Uninfected Eyes Twenty eyes from 12 animals were hybridized with antisense probes for mRNA encoding the p55 and p75 TNFR subtypes. In each case, autoradiographic signal was observed over the corneal endothelium, iris, ciliary body, choroid, and arachnoid layers of the optic nerve sheath (Figs. 1A, IB, 2A, 2B, 3). In addition, signal was observed over scattered cells at the vitreoretinal interface, particularly near the optic nerve (Fig. 4). Whether these cells were vascular or hyalocytic was impossible to discern with certainty. Autoradiographic signal for p75 (Fig. IB), but not p55 (Fig. 2B), was found over the retinal ganglion cell layer. Herpes Simplex Virus-Infected Eyes Ten eyes fromfiveanimals infected with McKrae strain HSV were hybridized with antisense probes for mRNA encoding the p55 and p75 TNFR subtypes. In each case, signal was observed in all regions described above for uninfected eyes. In addition, acute HSV infections were accompanied by intense leukocytic infiltrate in the bulbar conjunctiva, the corneal subepithelium and stroma, die anterior and posterior chambers, the vitreous cavity, and over the iris root and ciliary body. This increased autoradiographic signal for the TNFR subtypes was associated closely with the distribution of infiltrating white blood cells, although the possibility that some of this signal represented upregu-

FIGURE 3. High-power darkfield {top) and brightfield (bottom) photomicrographs of the same section demonstrate autoradiographic signal for mRNA encoding the p55 tumor necrosis factor receptor over the choroid in uninfected mouse eye. Some autoradiographic signal may be present in the closely apposed retinal pigment epithelium. Scale bar = 100 /un. lated TNFR mRNA expression by intrinsic cells, particularly keratocytes, could not be excluded (Figs. 1C, ID, 2C, 2D). Controls Tissue from four uninfected and six infected eyes from the same number of animals was hybridized with sense probes for mRNA encoding the p55 and p75 TNFR subtypes. In each case, autoradiographic signal was comparable to background (Fig. 2, p75 data not shown). In addition, ocular tissue from four animals whose corneas had been scored and treated with sterile vehicle was hybridized with antisense probes for the mRNA encoding the p55 and p75 receptor subtypes. Although the trauma associated with such sham inoculations typically produced mild keratitis with increased autoradiographic signal for the two TNFR subtypes observed over the few infiltrating white blood cells, signal was otherwise comparable to that observed in uninfected animals (data not shown). Recendy, we

TNF Receptor mRNA in Normal and HSV-Infected Mouse Eye

13

leukocytic infiltrate,40 and, in these cases, signal for both TNFR subtypes was found over white blood cells in the conjunctiva, the corneal subepithelium and stroma, the anterior and posterior chambers, the vitreous cavity, and the iris root and ciliary body. Together, these results support a role for TNF as an important mediator of intraocular inflammation. The cell type(s) in the uveal tract synthesizing mRNA for the p55 and p75 TNFR stibtypes was difficult to determine precisely because of the relatively high energy of sr'S-riboprobes and because of the compromised cellular morphology due the heavy proteinase K treatment required by the in situ hybridization protocol.33 However, both TNFR subtypes are known to be expressed on vascular endothelium,'12'43 and the uveal tract is heavily vascularized. This suggests that much of the TNFR signal observed over the uveal tract was, in fact, associated with iris, ciliary body, and choroidal blood vessels, a notion supported by the potent blood-ocular barrier-permeabilizing effects produced by systemic and intravitreal TNF,4"8 Expression of TNFR mRNAs by other cellular components of the uveal tract, particularly resident macrophages, is possible, especially given the presence of signal for the two TNFR mRNAs observed over scattered cells at the vitreoretinal interface, perhaps representing hyalocytes. The role of p75 TNFR over the retinal ganglion cell layer is unclear because direct effects of TNF on retinal ganglion cell function are not known to exist. Interestingly, we recently demonstrated differences in FIGURE 4. Darkfield {lop) and briglufieM {bottom) photomi- the distribution of the two TNFR subtypes in die cencrographs of the same section demonstrate autoradiotral nervous system, where the p75 subtype is observed graphic signal for mRNA encoding the p55 tumor necrosis more commonly over neurons than is the more abunfactor receptor over the choroid, the arachnoid layer of the dant p55 subtype.41 We were unable to demonstrate optic nerve, and clusters of cells at the vitreoretinal interface {open arrows) in uninfected mouse eye. Some autoradio- p55 mRNA signal over RPE cells in situ, a finding at odds with a recent study by Sippy and colleagues42 graphic signal may be present in the closely apposed retinal pigment epithelium. Scale bar = 100 //m. using cultured human RPE cells. Although the reason for this difference is unclear, it may represent differential mRNA expression between human and murine RPE cells, altered mRNA expression in vitro, or possiused a sensitive and specific mRNA protection assay41 to demonstrate TNF receptor subtypes in normal bly increased sensitivity of the polymerase chain reacmouse eyes (unpublished observation, 1994) supporttion-based assay used by Sippy and colleagues.32 ing our in situ results. Acute ocular HSV infection resulted in an intense inflammatory response, characterized by a leukocytic infiltrate into the conjunctiva, the corneal subepithelDISCUSSION ium and stroma, the anterior and posterior chambers, the vitreous cavity, and the iris root and ciliary body In situ hybridization histochemistry was used to localthat was maximal at 3 days after inoculation.40 Our ize mRNA encoding the p75 and p55 TNF receptor in situ hybridization studies demonstrated that these subtypes in uninfected and HSV-infected mouse eyes. In both cases, autoradiographic signal was observed infiltrating leukocytes expressed high levels of both over the corneal endothelium, iris, ciliary body, chop55 and p75 TNFR subtypes. Although upregulation roid, and arachnoid layers of the optic nerve sheath. of endogenous TNFR expression cannot be ruled out, In addition, signal was seen over scattered cells at the particularly over corneal keratocytes that have been vitreoretinal interface. Signal for p75, but not p55, was shown to be responsive to TNF,44 these results suggest observed over cells in the retinal ganglion cell layer. that a large portion of the overall increase in TNFR Acute HSV infection was accompanied by an intense content in eyes acutely infected with HSV may be ac-

14


counted for by an influx of receptor-bearing white blood cells. Similar findings have been reported for the expression of mRNA encoding TNF 39 and its receptor 38 in brain after acute infection with the lymphocytic choriomeningitis virus. Key Words choroid, iris, receptor, tumor necrosis factor, uveitis References 1. Beutler B. TNF, immunity and inflammatory disease: Lessons of the past decade. JInvest Med. 1995;43:227235. 2. Tracey K, Cerami A. Tumor necrosis factor: A pleiotropic cytokine and therapeutic target. Ann Rev Med. 1994; 45:491-503. 3. Fiers W. Tumor necrosis factor: Characterization at the molecular, cellular and in vivo level. FEBS Lett. 1991;285:199-212. 4. Kulkarni PS, Srinivasan BD. Cachectin: A novel polypeptide induces uveitis in the rabbit eye. Exp Eye Res. 1988; 46:631-633. 5. Rosenbaum JT, Howes ELJr, Rubin RM, Samples JR. Ocular inflammatory effects of intravitreally injected tumor necrosis factor. AmJPathol. 1988; 133:47-53. 6. Fleisher LN, Ferrell JB, McGahan MC. Ocular inflammatory effects of intravitreally injected tumor necrosis factor-alpha and endotoxin. Inflammation. 1990; 14:325-335. 7. Fleisher LN, Ferrell JB, Smith MG, McGahan MC. Lipid mediators of tumor necrosis factor-alpha-induced uveitis. Invest Ophthalmol Vis Sci. 1991; 32:23932399. 8. De Vos AF, Hoekzema R, Kijlstra A. Cytokines and uveitis, a review. Cmr Eye Res. 1992;ll:581-597. 9. Tamihara H, Yoshida M, Yoshimura N. Tumor necrosis factor-a gene is expressed in stimulated retinal pigment epithelial cells in culture. Biochem Biophys Res Commun. 1992; 187:1029-1034. 10. de Kozak Y, Goureau O, BellotJ, Naud MC, FaureJP, Hicks D. Tumor necrosis factor synthesis by resident retinal cells from uveitis-susceptible and resistant rat strains: Synthesis and regulation of nitric oxide from Lewis rat retinal Muller cells. In: Nussenblatt RB, Whitcup SM, Caspi RR, Gery I, eds. Advances in Ocular Immunology. New York: Elsevier; 1994:3-6. 11. Chandrasekharam NN, Detrick B, Hooks JJ. Human retinal pigment epithelial cells secrete cytokines in response to inflammatory mediates. In: Nussenblatt RB, Whitcup SM, Caspi RR, Gery I, eds. Advances in Ocular Immunology. New York: Elsevier; 1994:17-20. 12. Planck SR, Huang X-N, Robertson JE, Rosenbaum JT. Cytokine mRNA levels in rat ocular tissues after systemic endotoxin treatment. Invest Ophthalmol Vis Sci. 1994; 35:924-930. 13. Rosenbaum JT, Boney RS. Failure to inhibit endotoxin-induced uveitis with antibodies that neutralize tumor necrosis factor. Reg Immunol. 1993; 5:299-303. 14. Nakamura S, Yamakawa T, Sugita M, et al. The role of tumor necrosis factor-alpha in the induction of ex-

perimental autoimmune uveoretinitis in mice. Invest Ophthalmol Vis Sci. 1994;35:3884-3889. 15. Ferguson TA, Herndon JM, Dube P. The immune response and the eye: A role for TNFa in anterior chamber-associated immune deviation. Invest Ophthalmol Vis Sci. 1994;35:2643-2651. 16. Franks WA, Limb GA, Stanford MR, et al. Cytokines in human intraocular inflammation. Cuw Eye Res. 1992;11:187-191. 17. Nolle B, Gordes RS, Wiechens B. Small amounts of TNF-a in vitreous specimens of patients with intermediate uveitis. In: Nussenblatt RB, Whitcup SM, Caspi RR, Gery I, eds. Advances in Ocular Immunology. New

18.

19.

20. 21. 22.

23. 24.

25. 26.

27.

28.

29.

30.

York: Elsevier; 1994:361-364. Rahi AH, Al-Kaff A, Rahi SL, RahiJS. Cytokine profile in uveitis: A point prevalence study. In: Nussenblatt RB, Whitcup SM, Caspi RR, Gery I, eds. Advances in Ocular Immunology. New York: Elsevier; 1994:369-372. Leibovich JS, Poiverini PJ, Shepard HM, Wiseman DM, Shively V, Nuseir N. Macrophage-induced angiogenesis is mediated by tumour necrosis factor-alpha. Nature. 1987; 329:630-632. Elner VM, Strieter RM, Pavilack MA, et al. Human corneal interleukin-8: IL-1 and TNF-induced gene expression and secretion. AmJ Pathol. 1991; 139:977-988. Strieter RM, Kunkel SL, Elner VM, et al. Interleukin8: A corneal factor that induces neovascularization. AmJ Pathol. 1992; 141:1279-1284. Burke JM. Stimulation of DNA synthesis in human and bovine RPE by peptide growth factors: The response to TNF-alpha and EGF is dependent upon culture density. Curr Eye Res. 1989;8:1279-1286. Limb GA, Little BC, Meager A, et al. Cytokines in proliferative vitreoretinopathy. Eye. 1994;5:686-693. Limb GA, Earley O, Jones SE, LeRoy F, Chignell AH, Dumonde DC. Expression of mRNA coding for TNFa, IL-1/? and IL-6 by cells inflitrating retinal membranes. Graefe'sArch Clin Exp Ophthalmol. 1994;232:646-651. Hofman FM, Hinton DR, Johnson K, Merril JE. Tumor necrosis factor identified in multiple sclerosis brain./ Exp Med. 1989; 170:607-612. Selmaj K, Raine CS, Cannella B, Brosnan CF. Identification of lymphotoxin and tumor necrosis factor in multiple sclerosis lesions. J Clin Invest. 1991;87:949954. Sharief MK, Hentges R. Association between tumor necrosis factor-alpha and disease progression in patients with multiple sclerosis. N EnglJ Med. 1991; 325: 467-472. Powell MB, Mitchell D, Lederman J, et al. Lymphotoxin and tumor necrosis factor-alpha production by myelin basic protein-specific T cell clones correlates with encephalitogenicity. Int Immunol. 1990; 2:540544. Goodwin RG, Anderson D, Jerzy R, et al. Molecular cloning and expression of the type 1 and type 2 murine receptors for tumor necrosis factor. Mol Cell Biol. 1991;ll:3020-3026. Loetscher H, Steinmetz M, Lessiauer W. Tumor necrosis factor: Receptors and inhibitors. Cancer Cells. 1991;3:221-226.

TNF Receptor mRNA in Normal and HSV-Infected Mouse Eye 31. Tartaglia LA, Rothe M, Hu Y-F, Goeddel DV. Tumor necrosis factor's cytotoxic activity is signaled by the p55 TNF receptor. Cell. 1993;73:213-216. 32. Sippy BD, Hofman FM, He S, et al. SV40-immortalized and primary cultured human retinal pigment epithelial cells share similar pattarns of cytokine-receptor expression and cytokine responsiveness. Curr Eye Res. 1995; 14:495-503. 33. Cunningham ET Jr, De Souza EB. Localization of type IIL-1 receptor rriRNA in brain and endocrine tissues using in situ hybridization histochemistry. In: De Souza EB, ed. Methods in Neuroscience. Neurobiology of

Cytokines, Part A. New York: CRC Press; 1994:112-127. 34. Feduchi E, Alonso MA, Carrasco L. Human gamma interferon and tumor necrosis factor exert a synergistic blockage on the replication of herpes simplex virus. / Virol. 1989; 63:398-409. 35. Rossol-Voth R, Rossol S, Schutt KH, Corridori S, de Cian W, Falke D. In vivo protective effect of tumour necrosis factor alpha against experimental infection with herpes simplex virus type 1./ Gen Virol. 1991; 72: 143-147. 36. Shun-Hua C, Oakes JE, Lausch RN. Synergistic antiherpes effect of TNF-a and IFN-g in human corneal epithelial cells with that in cornealfibroblasts.Antiviral Res. 1994; 25:201-213. 37. Gashi H, Wechsler SL, Kaiwar R, Nesburn AB, Hofman FM. Local expression of tumor necrosis factor alpha and interleukin-2 correlates with protection against corneal acarring after ocular challenge of vaccinated mice with herpes simplex virus type \. J Virol. 1995; 69:334-340.

15

38. Stalder A, Campbell IL. Cytokine receptor gene expression in lymphocytic choriomeningitis. Soc Neurosci Abstrs. 1993; 326. 39. Campbell IL, Hobbs MV, Kemper P, Oldstone MBA. Cerebral expression of multiple cytokine genes in mice with lymphocytic choriomeningitis. / Immunol. 1994; 152:716-723. 40. Shimeld C, Whiteland JL, Nicholls SM, et al. Immune cell infiltration and persistence in the mouse trigeminal ganglion after infection of the cornea with herpes simplex virus type 1. / Neuroimmunol. 1995; 61:7-16. 41. Stalder A, Campbell IL. Simultaneous detection of multiple cytokine receptor mRNAs by RNase protection assay in LPS-induced endotoxiemia. Lymphokine Cytokine Res. 1994; 13:107-109. 42. Mackay F, Loetscher H, Stueber D, Gehr G, Lesslauer W. Tumor necrosis factor alpha (TNF alpha)-induced cell adhesion to human endothelial cells is under dominant control of one TNF receptor type, TNFR55. JExp Med. 1993; 177:1277-1286. 43. Slowik MR, De Luca LG, Fiers W, Pober JS. Tumor necrosis factor activates human endothelium cells through the p55 tumor necrosis factor receptor but the p75 receptor contributes to activation at low tumor necrosis factor concentrations. Am J Pathol. 1993; 143:1724-1730. 44. Tran MT, Tellaetxe-Isusi M, Elner V, Stricter RM, Lausch RN, Oakes JE. Proinflammatory cytokines induce RANTES and MCP-1 synthesis in human corneal keratocytes but not in corneal epithelial cells: /5-chemokine synthesis in corneal cells. Invest Ophthalmol Vis Sci. 1996; 37:987-996.