Serum Levels of Tumor Necrosis Factor-Alpha and

Interleukin-6 in Ocular Cicatricial Pemphigoid

Soon Jin Lee, Zhengzhi Li, Brian Sherman, and C. Stephen Foster

Purpose. These studies examined regulation of the cytokines interleukin-6 and tumor necrosis
factor-alpha in ocular cicatricial pemphigoid (OCP), a systemic autoimmune disease.
Methods. Serum levels of interleukin-6 and tumor necrosis factor-alpha in sera collected from
35 patients with OCP, 29 normal persons and 17 patients with ocular inflammatory diseases
were determined using an enzyme-linked immunosorbent assay.
Results. Levels of interleukin-6 were significantly decreased in sera of patients with OCP (me-
dian, 28.9; range, 7.5 to 136.7 pg/ml, P < 0.001) compared with sera from normal subjects
(median, 65.2; range, 21.1 to 303.9 pg/ml). Sera from patients with non-OCP, extraocular
inflammatory diseases and uveitis, showed no such decrease. In contrast, tumor necrosis fac-
tor-alpha levels were significantly elevated in OCP patients (median, 22.5; range, 8.3 to 44.4
pg/ml, P < 0.001), whereas no such increase was observed in sera from patients with extraocu-
lar inflammatory disease or uveitis, compared to normal sera controls (median, 17.4; range, 5
to 27.2 pg/ml).
Conclusions. These results suggest that elevated serum tumor necrosis factor-alpha levels and
decreased serum interleukin-6 levels can be added to the increasing list of systemic immuno-
logic correlates of active OCP, again emphasizing that OCP is a systemic disease whose primary
manifestation is ocular. Invest Ophthalmol Vis Sci. 1993;34:3522-3525.

1 he term cytokine refers to a class of proteins, pro- lin secretion by B lymphocytes and serves as a second
duced by a variety of cells, that play critical roles in signal in murine T cell activation.2-3 Increased serum
many biologic processes including inflammation and levels of IL-6 have been reported during renal trans-
immunity.1 Cytokines play important roles in many plant rejection episodes, and IL-6 has been associated
physiological responses and are involved in the patho- with certain autoimmune diseases.4
physiology of a wide range of diseases. Because of Tumor necrosis factor-alpha (TNF-a) also is pro-
their regulatory nature, their potential as therapeutic duced by many different cell types, including macro-
agents has been explored. The clinical application of phages, T cells, and NK cells; it has a variety of physio-
pure recombinant cytokines or their inhibitors in logical properties, including stimulation of fibroblast
cancer, infectious disease, and autoimmune disease growth and neutrophil function.5 TNF-a enhances
has recently met with some success. both IL-2 receptor and human leukocyte antigen-II
Interleukin-6 (IL-6) has a wide variety of activities expression on T cells and is an autocrine T cell growth
on numerous cell types. IL-6 is expressed by tumor factor. Elevated TNF-a gene expression has been ob-
cells, mitogen-stimulated lymphocytes and cytokine- served in graft versus host disease, and TNF-a has
stimulated fibroblasts.2 IL-6 enhances immunoglobu- been associated with rheumatoid arthritis.6
As a first step in understanding the abnormal regu-
lation of cytokine networks in autoimmune eye dis-
eases, we examined IL-6 and TNF-a levels in the sera
From the Immunology and Uveitis Unit, Rhodes Molecular Immunology
Laboratory, Department of. Ophthalmology, Massachusetts Eye and Ear Infirmary, of patients with ocular cicatricial pemphigoid (OCP), a
Harvard Medical School, Boston, Massachusetts. systemic autoimmune disease, as well as in sera from
This work was supported by the Mr. & Mrs. Richard Rhoads Fund.
Submitted for publication: October 28, 1992; accepted June 15, 1993.
patients with other ocular inflammatory eye diseases,
Proprietary interest category: N. and in sera from normal subjects. We found decreased
Reprint requests: C. Stephen Foster, Department of Ophthalmology, Massachusetts
Eye & Ear Infirmary, Harvard Medical School, 243 Charles Street, Boston, MA levels of IL-6 and increased levels of TNF-a in OCP
02114. sera compared to sera from normal groups. These re-

Investigative Ophthalmology & Visual Science, December 1993, Vol. 34, No. 13
3522 Copyright © Association for Research in Vision and Ophthalmology

Downloaded from iovs.arvojournals.org on 05/05/2020

 Serum TNF-a and IL-6 Levels in Cicatricial Pemphigoid 3523

suits support the notion that OCP is an autoimmune cluded seven diagnosed with scleritis, one with sar-
disease associated with an abnormal cytokine network. coidosis-associated chronic conjunctivitis, and one
with episcleritis.
First, we studied TNF-a in the sera of normal sub-
MATERIALS AND METHODS jects grouped by either age or sex. No significant dif-
Sample Collection ferences were observed between subjects older or
younger than 50 years (P = 0.429) or between male
Before initiating treatment, informed consent about and female subjects (P = 0.154; Fig. 1). For this rea-
the nature and possible consequences of this study was son, samples from patients or control subjects were
obtained from all participants. Sera were collected not divided into groups before comparison of serum
from patients with OCP (n = 35), uveitis (n = 8), other TNF-a levels.
extraocular inflammatory diseases (n = 9), and normal TNF-a levels were significantly increased in the
(control) individuals (n = 29) and stored at -70°C sera of patients with OCP (median, 22.5; range, 8.3 to
until assayed. The diagnosis of OCP was confirmed by 44.4 pg/ml, P < 0.001) compared to serum levels in
immunohistochemical demonstration of immunoreac- normal subjects (median, 17.4; range, 5 to 27.2 pg/ml;
tants at the epithelial basement membrane zone. The Table 1). Statistically significant increases in serum
protocols used in this study followed the tenets of the TNF-a levels were not observed in patients with uveitis
Declaration of Helsinki and were approved by the (median, 22.5; range, 8.9 to 30.1 pg/ml; P = 0.257) or
Massachusetts Eye and Ear Infirmary Human Experi- extraocular inflammatory diseases (median, 19.5;
mentation Committee. range, 14.2 to 22.5 pg/ml; P = 0.057) compared to
normal subjects (Table 1).
Determination of Tumor Necrosis Factor-
Serum IL-6 levels also did not significantly differ
Alpha and Interleukin-6 Levels
in normal subjects grouped by either age or sex (Fig.
Serum TNF-a and IL-6 levels were measured by a two- 2). Serum IL-6 levels were significantly lower in OCP
site sandwich enzyme-linked immunosorbent assay, patients (median, 28.9; range, 7.5 to 136.7 pg/ml; P <
using a commercial kit (T-cell Sciences, Inc., Cam- 0.001) compared to normal subjects (median, 65.2;
bridge MA). Briefly, 96-well plates coated with either range, 21.1 to 303.9 pg/ml; Table 1). There was no
an anti-TNF-a or an anti-IL-6 monoclonal antibody significant difference in the mean serum IL-6 levels of
were incubated with sera samples from each group. patients with extraocular inflammatory diseases (me-
After washing the plates to remove unreacted sera dian, 45.8; range, 0 to 450 pg/ml; P = 0.476) or with
components, enzyme-conjugated monoclonal antibod-
ies directed against either TNF-a or IL-6 were added
and incubated. Unbound enzyme-conjugated antibod- 50 ~
o
ies were removed by washing and the substrate, O- -
o
phenylenediamine, was added. The reaction was termi-
40 -
nated and absorbance at 490 nm was measured. All
samples were tested in duplicate and mean values were
determined. The values for TNF-a and IL-6 were cal-
30 -
culated using a standard curve determined using puri- •
fied TNF-a and IL-6 provided with these kits.
Statistical Method 20 -
. _l_
—W~ o
The Wilcoxon test was used to compare the mean -
serum levels of TNF-a and IL-6 between the different
-€- 0

o 8
v 10 -
• 51
groups of patients and control subjects. The level of o
a
significance chosen was P < 0.05. a : . 8
i i i i I i I i
RESULTS >60yr <50yr Female Male
p=0.429 p-0.154

Patients included in this study varied in age from 20 to FIGURE l. TNF-a levels in the sera of normal subjects. Sera
60 years. No significant differences were observed be- from normal subjects were obtained as described in Meth-
tween normal subjects grouped on the basis of age or ods. Normal subjects were divided into groups on the basis
sex, so normal subjects were not divided into groups of either age or gender and their serum TNF-a levels were
for comparison with patients with ocular inflamma- compared. P values are presented for the significance of
tion. Extraocular inflammatory disease patients in- differences between the means of each group.

Downloaded from iovs.arvojournals.org on 05/05/2020

 3524 Investigative Ophthalmology & Visual Science, December 1993, Vol. 34, No. 13

uveitis (median, 79.6; range, 0 to 435.6 pg/ml; P = 400

0.200) compared to normal subjects (Table 1). Be-

cause the difference in age of the normal group for
IL-6 had borderline significance, we reanalyzed the
300 -
data with age-matched group for comparison of serum
IL-6 levels from patients with OCP, extraocular in-
flammatory disease, and uveitis and those from normal
subjects. The new P values were OCP = 0.01 (n = 9), 200
uveitis = 0.163 (n = 8) and extraocular inflammatory
diseases = 0.199 (n = 7), all similar to the values ob-
tained without age matching.
100 -

DISCUSSION
Elevated serum levels of IL-6 and TNF-a have been
previously reported in autoimmune diseases such as
rheumatoid arthritis, suggesting abnormal regulation
of these two cytokines in the peripheral immune sys-
tem.4-6 Our studies examined the serum levels of these FIGURE 2. IL-6 levels in the sera of normal subjects. Sera
from normal subjects were obtained as described in Meth-
two cytokines in patients with OCP, an extraocular
ods. Normal subjects were divided into groups on the basis
inflammatory problem that we and others have con- of either age or gender and their serum IL-6 levels were
tended represents local expression of a systemic au- compared. P values are presented for the significance of
toimmune disease.7'8 Serum TNF-a levels were sub- differences between the means of each group.
stantially elevated in OCP patients, suggesting a po-
tential role for this cytokine in the increased activation
of the peripheral immune system, as judged by ele- autoimmune diseases, such as rheumatoid arthritis.5
vated serum levels of soluble IL-2 receptor and solu- Increased IL-6 expression has been implicated in
ble CD8 (unpublished observation). TNF-a has been polyclonal B cell activation, which is observed in au-
shown to potentiate both CD8 T cell function in vivo toimmune diseases such as rheumatoid arthritis.3 How-
and in vitro and the release of soluble CD8 from syno- ever, IL-6 levels in the sera of our OCP patients were
vial fluid mononuclear cells in vitro.9 The observed decreased, suggesting that a different mechanism may
increase of serum-soluble CD8 levels in OCP patients be involved in the pathogenesis of OCP. Although
may result from elevated levels of TNF-a, which has these studies comprised a relatively small number of
been shown to trigger CD8 T cell activation.9 The ele- subjects and a large variation in mean values was
vated serum TNF-a levels observed in OCP patients in noted, the findings obtained from this study warrant
this study suggest an association between expression further investigation with a larger study population.
of this cytokine and OCP, a finding reported for other Reduced IL-6 serum levels were reported in cyclo-
sporin-induced nephrotoxicity in human renal allo-
graft recipients.10 Decreased serum levels of IL-6 ob-
TABLE l.Comparison of Serum TNF-a served in this study may have resulted from abnormal
and IL-6 Levels function of cells secreting IL-6 or from secondary ef-
fects of other cytokines involved in the cytokine regula-
Median Value No. of tory network. More studies are needed to fully estab-
(pg/ml) Samples P Value* lish the potential relationship between serum levels of
TNF-a IL-6 and TNF-a and their relationship to conjunctival
OCP 22.5 35 <0.001 inflammation.
Uveitis 22.5 8 0.257
Extraocular 19.5 9 0.057 Key Words
Normal 17.4 29
IL-6 tumor necrosis factor, interleukin-6, cicatricial pemphigoid,
OCP 28.9 26 <0.001 ocular immunology, cytokines
Uveitis 79.6 8 0.2
Extraocular 45.8 9 0.476
Normal 65.2 26 — Acknowledgments
* P values are relative to normal controls as calculated by the Wil- The authors thank Dr. George T. Cicila for critically review-
coxon test. ing the manuscript.

Downloaded from iovs.arvojournals.org on 05/05/2020

 Serum TNF-a and IL-6 Levels in Cicatricial Pemphigoid 3525

References sis factor in synovial exudates. Ann Rheum Dis.

1. Taniguchi T. Regulation of cytokine gene expression. 1988;47:768-772.
Ann Rev Immunol. 1988;6:439-464. 7. Waltman SR, and Yarian D. Circulating autoantibod-
2. Snick J. Interleukin-6: an overview. Ann Rev Immunol. ies in ocular pemphigoid. Am J Ophthalmol.
1990;8:253-278. 1974;77:891-894.
3. Hirano T, Kishimoto T. Interleukin-6: possible impli- 8. Foster CS. Cicatricial pemphigoid. Trans Am Ophthal-
cations in human diseases. Res Clin Lab. 1989; 19:1- mol Soc. 1986;84:527-663.
10. 9. SymonsJA, Wood NC, DiGiovine FS, Duff GW. Solu-
4. Hirano T, Matsuda T, Turner M, et al. Excessive pro- ble CD8 in patients with rheumatoid diseases. Clin
duction of IL-6/B cell stimulatory factor-2 in rheuma- Exp Immunol. 1990;80:354-359.
toid arthritis. Eur JImmunol. 1988; 18:1797-1801. 10. Yoshimura N, Oka T, Kahan BD. Sequential determi-
5. Beutler B, Cerami A. Cachectin and tumor necrosis nations of serum interleukin-6 levels as an immuno-
factor as two sides of the same biological coin. Nature. diagnostic tool to differentiate rejection from nephro-
1986;320:584-588. toxicity in renal allograft recipients. Transplantation
6. DiGiovine FS, Nuki G, and Duff GW. Tumour necro- 1991;51:172-176.

Downloaded from iovs.arvojournals.org on 05/05/2020