Vol. 175, No. 3, 1991 March 29, 1991
BIOCHEMICAL
INTERACTIONS
BETWEEN
ON MSH
ULTRAVIOLET
BINDING
HUMAN
AND BIOPHYSICAL RESEARCH COMMUNICATIONS Pages 839-845
IN BOTH
LIGHT MOUSE
SQUAMOUS
AND
MELANOMA
CARCINOMA
INTERLEUKIN-
1
AND
CELLS
Nicholas Birchall 1, Seth J. Orlow 2, ThomasKupper 3, and John Pawelek 4* 1 Dept. of Molecular Medicine, University of Auckland, Auckland, NZ 2 Dept. of Dermatology, New York University Medical Center, New York, NY 3 Dept. of Dermatology, Washington University School of Medicine, St. Louis, MO 4 Dept. of Dermatology, Yale University School ofMedicine, New Haven, CT 06510 Received
January
17,
1991
InteractionsbetweenB-melanotropin(MSH), interleukin l-a (IL-l), and ultraviolet light (UV) were examined in Cloudman S91 mouse melanoma and RHEK human squamous carcinoma cell lines. The following points were established: 1) both cell linesproduced IL- 1 and their production was stimulatedby exposureof the cells to UV; 2) both cell lines possessed high affinity binding sitesfor MSH, and their ability to bind MSH was modulatedby IL- 1; 3) IL- I exhibited both stimulatory and inhibitory effects on MSH binding to Cloudmancells: and 4) the stimulatory effect of IL- 1 on MSH binding to melanomacells was reflected in enhanced cellular responsivenessto MSH regarding tyrosinase activity (E.C. 1.14.18.1) and melanin content. The findings raise the possibility that interactions between keratinocytes and melanocytes may be regulated by IL-l and MSH, and suggesta possible mechanism for stimulation of cutaneousmelanogenesisby solar radiation: enhancementof MSH receptor activity by induction of IL- 1. B 1991 Academic Press', Inc.
A number of studies have linked the biological actions of interleukin-1 to those of MSH.
In mammals, at least three IL-l
responses -- fever,
inflammation,
and
immunostimulation-- are counteractedby MSH (l-5). IL-l-induced fever is accompaniedby increasedconcentrationsof MSH in the brain (6-7). The effects of MSH in modulating IL- 1 responsesare highly target cell specific, suggestingthat MSH receptorsare involved (8). UV light stimulates IL-l production in cultured human keratinocytes (9) and it also increases circulating MSH levels in both humansand horses(10-l 1). IL-l is produced by melanoma cells in culture (12-13). Since UV light is a potent stimulator of cutaneousmelanogenesis, theseobservationsprompted us to investigatewhether the combinedactionsof IL- 1 and MSH may be involved in UV-induced melanogenesis. In humans, cutaneousmelanin provides a defense from the harmful effects of UV (premature aging of skin, skin cancers) (14-16). As a protective feed-back mechanism, exposureof skin to solar radiation or selectedwave lengthsof
W
provides a positive signalto
*To whom requestsfor reprints shouldbe addressed. 0006-291X/91 839
$1.50
Copyright 0 1991 by Academic Press, Inc. All rights of reproduction in arzy form reserved.
Vol.
175, No. 3, 1991
BIOCHEMICAL
AND BIOPHYSICAL
RESEARCH COMMUNICATIONS
the pigmentary system. This resultsin an increasein the numberof active melanocytes,aswell as increasesin both the rate of melanin synthesiswithin individual melanocytes, as well as transfer of melanin from melanocytesto their surroundingkeratinocytes ( 17). Little is known about the biochemicalregulation of theseevents,but recently it hasbeendemonstratedthat UV increasesMSH receptor activity on the outer surfaceof Cloudmanmelanomacells, and that UV and MSH act synergistically to increasemelanin content in the skins of mice and guinea pigs (18). Results describedbelow suggestthat UV-induced melanogenesiscould involve the production of IL- 1 which in turn regulatesthe MSH receptorsystem. Materials andMethods The Cloudman S91 mousemelanomaline, “PS- I-HGPRT-- 1” was used throughout the study (19). This line is amelanotic unlessMSH or other agentswhich raise cyclic AMP levels are addedto the culture medium. MSH-mediated melanogenesis in this line is enhanced 2-3 fold by low levels of UVB light, however WB alone hasno effect on melanogenesis (20). The melanomacells were cultured asmonolayersin Ham’sFlO culture mediumsupplemented with horse serum(5 10%) as previously described(21). The “RHEK” human squamouscell carcinoma line was a gift of Dr. J. Rhim and was cultured in Dulbecco’s modified Eagle’s mediumsupplementedwith fetal bovine serum(10%). Synthesis, isolation, and binding of biologically active 1251-D-MSHto cells were by modifications (22-23) of the methodsof Lambert et. al. (24). Binding assayswere carried out at 10°Cto prevent internalization of the peptide. Binding data are expressedasspecific counts bound to cells. Specific binding was determinedby subtractingthe cpm bound in the presence of a 1000 fold excessof non-radioactive l3-MSH from the cpm obtainedwith carrier free 125Il3-MSH alone. Specific binding rangedbetween80-95% for Cloudmancells and 70-85% for RHEK cells. Cloudman cells were preparedfor binding assaysby harvesting in Ca/Mg-free Joklik’s medium containing ethylenediamine tetraacetic acid (EDTA, 1mM), pelleting by centrifugation (700 g, 5 minutes),and resuspendingin “MSH binding buffer” (NaCl, 140mM; KCl, 5 mM; Na2HP04, 10 mM; KH2PO4, 1 mM; glucose, 0.1%) at a concentration of 106 cells per ml. Cell numbers were determined with a Coulter Counter. RHEK cells were prepared similarly except they were harvested in CaMg-free phosphate buffered saline containing EDTA (5 mM). Failure to useCa-free harvestingsolutionsinhibited MSH binding. Bioassaysfor IL- 1 were precisely asdescribedby Kupper et al. (9). Briefly, a cloned mouse T helper cell, DlO.G4.1, (provided by Dr. C. Janeway, Jr., Yale University) proliferates in culture only if its clonotypic antibody 3D3, aswell as IL- 1 are presentin the culture medium. If the cells are cultured in the presenceof 3D3 antibody, the amount of IL- 1 in an unknown sample(conditioned culture media, cell sonicates)can be determined by its ability to promote 3H-thymidine incorporation, comparedto that of known concentrationsof authenticIL- 1. Recombinanthuman IL- 1 alpha (rhIL- la) was provided by Dr. R. Newton, Hoffman LaRoche, Nuttley, N.J., a rabbit polyclonal antibody that neutralized IL- la was provided by Immunex Corp., Inc., and 3D3 antibody was provided by Dr. C. Janeway, Jr., Yale University. To assayfor tyrosinaseactivity, cells were lysed by exposureto Triton Xl00 (0.5%) in sodium phosphate (10 mM, pH 6.8). Tyrosinase activity was measured in lysed cell prepamtionsby a modification (19) of the methodof Pomerantz(25). UV irradiation was carried out with a bank of four FS20 Westinghouse bulbs, as describedby Kupper et al. (9). 840
Vol.
175,
No.
3, 1991
BIOCHEMICAL
AND
BIOPHYSICAL
RESEARCH
COMMUNICATIONS
Results and Discussion It was previously
demonstrated that human keratinocytes
exposed to UV radiation
respond with increased synthesis of IL- 1 mRNA and biologically active IL- 1 (9). In addition, it has been shown that human melanoma cells produce IL- 1 in culture (12- 13). Here we show that cultured Cloudman melanoma cells also produce IL- 1 and in addition show increases in biologically active IL- 1 in response to UV radiation (Table 1). IL-l produced by the melanoma cells, like that of the keratinocytes (9), was largely confined to the cells and not released to the culture medium.
A monoclonal antibody to IL-la virtually
eliminated all IL- 1 biological
activity in the melanoma sonicates, indicating that the melanoma cells produced IL- la. The UVB dosages used to achieve responses in Cloudman melanoma cells ( 15 mJ/cm2) were far lower than those used for human keratinocytes
(100-300 mJ/cm2).
The UV dosages did not
affect viability, as assesed by trypan blue exclusion for melanoma cells, and propidium iodide exclusion for squamous carcinoma cells (not shown).
We noted that the same dosage of UV radiation which stimulated IL-l production by the melanoma
cells also stimulated
cell surface MSH
responsiveness to MSH in these cells (18,20). Table
1. Effect of Cloudman
receptor activity
and cellular
We therefore investigated whether IL- 1 itself
melanoma cell sonicates and conditioned
Additions to Culture Medium
culture media
3H Thymidine Incorporation (q&2 x I 04cells)
None (3D3 Antibody)
5684 + 1023
rhlL- la (lo-l2M)
131914 + 1979
Melanoma Conditioned Media
26317 + 3158
Melanoma Conditioned Media 24 hrs after WB (I5 mVcm2)
22778 + 3189
Melanoma Conditioned Media 72 hrs after WB
29590 + 2663
Melanoma Sonicates
85766 + 686 1
Melanoma Sonicates 24 hrs after WB
110115~9910
Melanoma Sonicates 72 hrs after WB
175283 + 33304
Melanoma Sonicate 72 hm after WB + anti IL- la
8003 + 400
Procedures are described by Kupper et al. (9) (see Methods). Results represent averages + SD for triplicate samples in a representative experiment. The experiments were repeated several times with similar results.
841
Vol.
175, No. 3, 1991
BIOCHEMICAL
AND BIOPHYSICAL
RESEARCH COMMUNICATIONS
------___.
10000-~/ g8 v) 0 Y
20000
a-----------------
i\
/IjT~
o-w/
’
’
’
’
’
’
’
’
’
’
’
’
’
’
’
1
- 2OOOOr c
1 Ok+
/
10-19
10.”
10-15
10-13
IL-l(M)
Figure 1.
Effects of IL-I on the bindingl2%3-MSH to Cloudman melanoma cells in culture. Cells were incubated for 24 hours with human recombinant IL-alpha at the concentrations indicated and MSH binding activity was assayed as described in Materials and Methods. Three distinct responses to IL- 1 were observed: stimulation of MSH binding (1A); inhibition of MSH binding (1B); and a biphasic combination of stimulation and inhibition (1C). The results represent the average of specific cpm bound LSD for triplicate determinations from representative experiments. Possible explanations for the variable effects of IL1 are discussed in the text.
could stimulate MSH receptor activity. The results,obtainedin many experimentsover a two year period, were complex (Figure 1). The most prevalent responsewas a dose-dependent stimulation of MSH binding by IL-l (Figure 1A). However, in some experiments IL-l inhibited MSH binding (Figure lB), and occasionallythere was a dose-relatedcombination of stimulation and inhibition within the sameexperiment (Figure 1C). The results within individual experimentswere highly statistically significant, with p values usually lessthan .OOl and never more than -005. The different responses ocurred betweenexperimentsevenwith the samelots of rhIL-la, 1251-B-MSH, and horse serum. Culture conditions (temperature, humidity, feedingschedule,cellular seedingdensity, etc.) were carefully controlled. Stimulation of MSH binding by IL- 1, the predominant response,was accompaniedby increasedcellular responsiveness to MSH in termsof tyrosinaseactivity (Table 2) and melanin content (not shown). 842
Vol.
175,
No.
BIOCHEMICAL
3, 1991
AND
BIOPHYSICAL
RESEARCH
Table 2. Enhancement by interleukin- 1 of MSH-induced Additions to Culture Medium
tyrosinase activity
Tyrosinase Activity (3HOH productioncpm/l06 cells/60 min)
none
220 + 56
IL- 1 (lo- 14M)
207 2 46
L-l (lo-13M)
174528
MSH ( 2 x lo-7M)
1235 + 153
MSH+IL-I
(lo-14M)
3922 + 320
MSH + IL- 1 (lo-13M)
3477 + 523
Tyrosinase activity was assayed by the production of 3HOH from described in Methods. Cells were from the same experiment shown represent averages of triplicate samples+ SD. The experiments were similar results. IL- 1 also enhanced MSH-mediated melanin production cell pellets (not shown).
The effects of IL-l occurred
less than one molecule
binding
whether
they were stimulatory indicating
and UV radiation
cells possessed high-affinity
was stimulated
3’5’- 3H-L-tyrosine as in Figure IA. Results performed 3 times with as assessed visually in
or inhibitory,
that IL- 1 was effective at
cell.
the effects of IL-l
We found that RHEK
that MSH
binding,
as low as lo- I6- lo-lqM,
per melanoma
We also examined cells.
on MSH
at IL- 1 concentrations
COMMUNICATIONS
by the addition
on MSH
binding
of IL-1 (lo-IOM)
binding
to RHEK
sites for I251-R-MSH, to the culture
medium
300002 $ LD 0
25000-
;
20000
E n c
15000-
-
2
m
~oooo-
2 f
5000-
zi 0 0.0
0.4
0.6
1251-MSH Figure
2.
1.2
1.6
2.0
(nM)
Effects of IL-l on the binding of r251-MSH to human squamous carcinoma cells. RHEK cells were incubated for 24 hours with human recombinant IL- 1 alpha (lo- 10~) and MSH binding activity was assayed as described in Materials and Methods. Results represent averages of specific cpm bound + SD for triplicate determinations from a representative experiment. Experiments were repeated four times with similar results. Control ( Q----o); IL-l (1. 843
and
Vol. 175, No. 3, 1991
BIOCHEMICAL
AND BIOPHYSICAL
RESEARCH COMMUNICATIONS
(Figure 2). Theseexperimentswere repeatedseveraltimes, and we never observedinhibition of MSH binding to RHEK cellsby IL- 1. We have demonstratedthat melanomacells, derived from melanocytes,and squamous carcinoma cells, derived from keratinocytes, possesshigh affinity receptors for MSH which are regulatedby IL- 1. UV light stimulatesIL- 1 production by melanomacells (Table 1) and keratinocytes (9), and it stimulatesboth cell surface and internal MSH receptor activity in melanomacells (18, 20). Evidence from studiesof mice and guinea pigs indicates that UV light acts synergistically with MSH to promote melanogenesis,apparently by stimulating the MSH receptor system(18). Our resultsraise the possibility that UV exerts its effects by first stimulating IL- 1production which in turn regulatesMSH receptors. We observed effects of IL-I at concentrationsas low as IO-19M, in our experiments the equivalent of one IL- 1 moleculeper 300 melanomacells. Such a situation could exist only if the IL-l signal were amplified in somemanner by its target cells, perhaps through the production of more IL- 1or of other cytokines (26). Although the predominant effect of IL-l
on MSH binding to melanoma cells was
stimulatory, we also observedinhibitory effects, and were unable to control theseresponses, even though the culture and assay conditions were seemingly identical. Some possible explanations for the two responsesare asfollows: 1) There was more than one population
of
IL- 1 receptorson Cloudmanmelanomacellsand the expressionof thesereceptors determined different cellular responses.In several attempts, we failed to demonstrateIL-l receptors on melanoma cells, using 1251-ILla as a probe, whereaswe were able to demonstratesuch receptors on KHEK cells (not shown). However, since the biological responsesof the melanoma cells to IL- 1 occurred at such low IL- 1 concentrations, it is possible that IL-l receptors, though functional, were below levels of detection with 12%-IL-l.
2) The signal
amplification mechanismsfor IL-l involved more than one pathway and the phenotypic responseof the cells dependedon which pathways were activated. 3) IL- 1 induced MSH or MSH-like peptide production by the melanomacells, resulting in MSH receptor occupancy by the ligand which in turn prevented binding of 1251-g-MSH. In this case, IL-l could have stimulated MSH receptor activity, but it would have appearedas inhibition becauseof prior receptor occupancy. 4) IL- 1 causedinternalization of MSH receptors,thus reducing external, while increasing internal binding sites for MSH. We have recently demonstratedinternal binding sitesfor MSH which are regulatedby UV radiation (20, 27). To our knowledge, the demonstrationof MSH binding siteson humankeratinocytes is the first such report (Figure 2). We do not understandwhat role they may play in regulating the keratinocyte phenotype, but the finding that MSH binding to kemtinocytes was stimulated by IL-l suggeststhat MSH receptors on keratinocytes could play a role in UV-induced melanogenesis.Together, our resultsindicate that there are multiple interactions betweenthe MSH and IL- 1 regulatory pathways, and that thesepathways may be involved in W-mediated 844
Vol.
175,
No.
3, 1991
melanogenesis.
BIOCHEMICAL
AND
BIOPHYSICAL
RESEARCH
COMMUNICATIONS
The findings are consistent with previously established relationships between
MSH and IL- 1. References 1. :: 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27.
Lipton J.M., and Glyn J.R. (1980) Peptides 1, 15-18 Murphy, M.T., Richards D.B., and Lipton J.M. (1983) Science 221, 192-193 Shih S.T., Khorram O., Lipton J.M., and McCann S.M. (1986) Am. J. Physiol. 250, R803-806 Robertson B.A., Gahring L.C., and Daynes R.A. (1986) Inflammation 10,371-385 Cannon J.G., Tatro J.B., Reichlin S., and Dinarello C.A. (1986) J. Immunol. 137, 2232-2236 Holdeman,M., Khorram O., Samson W.K., and Lipton J.M. ( 1985) Am. J. Physiol. 248, R125-R129 Samson W.K., Lipton J.M., Zimmer J.A., and Glyn J.R. (1981) Peptides 2,419-423 Daynes R.A., Robertson B.A., Cho B-H., Burnham D.K., and Newton R. (1987) J. Immunol. 139,103-109 Kupper T.S., Chua A.O., Flood P., McGuire J., and Gubler U. (1987) J. Clin. Invest. 80,430-436 Holzmann H., Altmeyer P.,and Schultz-Amling W. (1982) Act. Dermatol. 8, 119-123 Holzmann H., Altmeyer P., Stohr L., and Chilf G.N. (1983) Hautatzt 34,294-297 Benicelli J.L., Elias J., Kern J., and Guerry D. IV. (1989) Cancer Res. 49,930-935 Kock A., Schwatz T. , Urbanski A., Peng Z., Vetterlein M., Micksche M., Ansel J-C., Kung H.F., and Luger T.A. (1989) J. Natl. Cancer Inst. 8 I ,36-42 Kaidbey K.H., Agin P.P., Sayre R.M., and Kligman A.M. (1979) J. Am. Acad. Dermatol. 1,249-260 Laundee J., Henschke C.I., and Mohammed N. (1985) Cancer 55, 1823-1828 Kromberg J.G., Castle D., Zwane E.M., and Jenkins T. (1989) Clin. Genet. 36,4352 Pathak M.A. (1985) Ann. N.Y. Acad. Sci. 453,328-339 Bolognia J., Murray M., and Pawelek I. (1989) Dermatol. 92, 651-656 Pawelek J.M. (1979) J. Cell. Physiol. 98, 619-625 Chakraborty A., Orlow S.J., Bolognia J., and Pawelek J. (1991) J. Cell Physiol., in press. Pawelek J.M. (1978) Melanoma cells. Ip Cell Culture, A volume of Methods in Enzymology. Academic Press, New York. Vol. LVIII, 564-570 Pawelek J., M&me J., and Osber M. (1988) Melanotropin mechanisms of action: Melanogenesis. b The Melanotropic Peptides. CRC Press, Boca Raton, FL 47-58 McLane J.A., and Pawelek J.M. (1988) Biochem. 27,3743-3747 Lambert D.T., Stachelek C., Varga J.M., and Lemer A.B. (1982) J. Biol. Chem. 257, 8211-8215 Pomerantz S., and Chuang L. (1970) Endo. 87,302-310 Zhang Y., Lin J-X, Yip Y.K., and Vilcek J. (1988) Proc. Natl. Acad. Sci. USA 85, 6802-6805 Orlow S.J., Hotchkiss S., and Pawelek J.M. (1990) J. Cell. Physiol. 142, 129-136
845