FASEB J.
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
 QUICK SEARCH:   [advanced]


     


FJ EXPRESS SUMMARY ARTICLE
The
Full-length version of this article is also available, published online September 17, 2001 as doi:10.1096/fj.01-0172fje.
Published as doi: 10.1096/fj.01-0172fje.
This Article
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
15/13/2533
01-0172fjev1    most recent
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by LI, D.
Right arrow Articles by BLUMENBERG, M.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by LI, D.
Right arrow Articles by BLUMENBERG, M.
(The FASEB Journal. 2001;15:2533-2535.)
© 2001 FASEB

Rays and arrays: the transcriptional program in the response of human epidermal keratinocytes to UVB illumination 1

DELING LI*, THOMAS G. TURI{dagger}, ALYSSA SCHUCK*, IRWIN M. FREEDBERG*,{ddagger}, GREGORY KHITROV* and MIROSLAV BLUMENBERG*,§,{dagger}{dagger}2

* R. O. Perelman Department of Dermatology, Departments of
{ddagger} Cell Biology and
§ Biochemistry, and
{dagger}{dagger} The Kaplan Cancer Research Center, New York University School of Medicine, New York, New York 10016, USA; and
{dagger} Central Research Division, Pfizer Inc., Groton, Connecticut 06340, USA

2Correspondence: Department of Dermatology NYU School of Medicine, 550 First Ave., New York, NY 10016, USA. E-mail: blumem01{at}med.nyu.edu

SPECIFIC AIMS

Hypothesizing that unlike other cell types in our body, epidermal keratinocytes have to protect from UV light (UV) not only themselves but also the underlying organism, we used DNA arrays to profile the changes in gene expression during the 24 h after illumination with UVB. We confirmed the transcriptional changes in keratinocyte cultures using Northern blots and, using a new organ culture system, show that these changes occur in irradiated human skin as well.

PRINCIPAL FINDINGS

1. Three waves of transcriptional changes
Of the 6800 genes on the Affymetrix chips, approximately half are expressed in keratinocytes; among these, at least 198 were regulated by UVB light. Using a clustering algorithm, we grouped the seven time points in the first 24 h after illumination. The changes in gene expression can be grouped conveniently into three waves: the most similar changes at the 0.5, 1, and 2 h points form the early, those at the 4 and 8 h form the intermediate, and those at 16 to 24 h form the late wave.

2. Regulated genes: proteins involved in DNA protection and repair
Many of the regulated genes were known to be targets of AP1, NF-{kappa}B and p53, the transcription factors activated by UV. The regulated genes can be grouped by their function into several distinct, clearly demarcated functional clusters.

One of the most serious effects of UV is the mutagenesis caused by the damage to DNA. UV induced expression of ERCC4, the endonuclease that incises 5' from thymidine dimers; although present in keratinocytes, other repair enzymes are not further induced by UV. UV does induce the expression of several histones and enzymes that produce building blocks for DNA synthesis, namely, spermidine/spermine N1-acetyltransferase and nucleoside-diphosphate kinase. It appears that most of the DNA repair proteins are present in sufficient amounts in keratinocytes even before UV treatment, but the cells, sensing the need for extensive repair, produce more of the dNTPs and histones to build and protect the newly repaired DNA. The machinery for DNA repair is already in place, but the nuts and bolts need to be supplemented.

3. Early changes in keratinocyte physiology
UV categorically belongs to the extracellular influences that activate the immediate early genes, which presumably repair and protect the cells from the harmful effects of UV. Among these are several transcription factors such as junB, junD, c-fos, ETR101, EGR1, and HRY. UV causes strong and persistent suppression of c-Myc expression; epidermis-targeted overexpression of c-Myc causes extensive premalignant epidermal proliferation, which could be harmful in the context of UV-induced DNA damage. This is consistent with the role of c-Myc in deregulating cell growth, promoting genomic instability, sensitizing to apoptosis, and inhibiting expression of DNA damage-induced growth arrest proteins. UV suppresses many transcription factors, including ets-2, BTEB2, SRF, TGIF, HLH-1R21, AREB6, AP2, and others. It is unclear what role these transcription factors play in healthy, unirradiated keratinocytes; apparently, however, their functions may be unnecessary or harmful in irradiated ones.

Control of mRNA stability is an important alternative to transcriptional regulation by UV. The molecular mechanisms involved have not yet been defined. We note that UV regulates an unexpectedly high number of RNA processing proteins.

Most intracellular signaling processes involve protein phosphorylation, and therefore it is not surprising that kinases and phosphatases are well represented among the regulated genes. UV induces three RING3 family proteins while suppressing kinases A-Raf, casein kinases CKI-{alpha} and CKII-{alpha}, and ERK3. UV also induces CL100, a dual specificity phosphatase that attenuates the JNK pathway while suppressing phosphatases PP2A-C{alpha} and PP1{gamma}. UV also regulates expression of a dozen of the small GTP binding proteins and associated factors.

Several cell surface receptors and their associated proteins are suppressed by UV, which is concordant with the hypothesis that the illuminated cells try to shut out additional extracellular signals until the UV-caused problems are dealt with.

4. Secreted signaling proteins, chemokines, cytokines, and growth factors
In the intermediate wave, the most prominently induced genes are the secreted proteins: chemokines, cytokines, and growth factors—in particular, five members of the interleukin 8 (IL-8) family (IL-8, Gro-{alpha}, Gro-ß, MDNCF, and MIP2-ß). They serve to alert the surrounding tissue that damage has occurred. The IL-8 family chemokines are chemotactic and activating for neutrophils, basophils, and macrophages and presumably play a role in inviting inflammatory cells into UV-damaged tissue. They also activate melanocytes, which may initiate tanning in response to UV. Keratinocytes serve as specific sentinels for UV; induction of these secreted proteins alerts the organism that UV has been detected. This response seems to be a specific function of UV illuminated keratinocytes.

Several UV-induced genes, including IL-8, have been shown to be inducible by interferon {gamma} (IFN-{gamma}). It is unknown whether their induction is a direct effect of UV or an indirect, autocrine effect of a UV-induced, secreted molecule. These results point to a hitherto undescribed nexus between UV and IFN-{gamma} signaling.

5. Cornified envelope proteins and structural proteins
The genes most strongly induced by UV are keratinocyte differentiation markers, components of the cornified envelope. These include calgranulin, elafin, involucrin, and small proline-rich proteins. They are induced 16 and 24 h after illumination. Other differentiation markers, keratins and filaggrin, are not induced. Apparently, one of the epidermal responses to UV is enhancement of stratum corneum production, i.e., augmentation of the cornified, dead protective layer of skin.

In contrast, the desmosomal proteins are suppressed. Because the function of desmosomes is to keep cells firmly attached to one another, the reduction in desmosomal proteins may facilitate the movement of keratinocytes, assembly of the cornified envelope proteins, and formation of the stratum corneum.

Actin binding proteins (ABPs) are prominent among the cytoskeletal proteins regulated by UV. Actin interacts with ~60 different ABPs to affect the shape of the plasma membrane, allow cellular motility, and maintain cell shape and polarity. In the early wave, UV induces troponin and suppresses ß-spectrin expression, which increases actin cytoskeleton pliability. ABP genes induced later include MacMARCKS, myosin light chain, Arp 2/3, tropomyosin, and thymosin. These proteins enhance actin polymerization, cross-link actin filaments to other proteins, or both, thus stabilizing and strengthening the microfilament cytoskeleton. The overall picture of the regulation of the cytoskeletal proteins by UV shows an initial depolymerization, loosening of the actin cytoskeleton, which is followed by a repolymerization of the filaments, and reconstitution of the cytoskeletal network.

6. Energy procurement
Among the most strongly induced genes immediately after UV illumination are several mitochondrial proteins—cytochrome c-1, cytochrome c oxidase subunit VIIb, and cytochrome b light chain. It appears that upon illumination, cells require additional energy and induce mitochondrial proteins to address this need. Another indication that UV illumination leads to a need for additional energy is that UV induces several of the energy-producing enzymes, such as {alpha}-enolase, whereas many energy-requiring processes—such as transport, gluconeogenesis, and lipogenesis—are shut down. Strong inhibition of lipid neogenesis seems inconsistent with induction of the cornified envelope proteins because the lipid and proteinaceous components both contribute to the formation of stratum corneum. On the other hand, it may be important for saving the energy of the UV-damaged cell.

The changes in mitochondrial proteins are important in the epidermal response to UV and probably have several functions. These include a burst of respiration proteins to provide additional energy needed to cope with the UV caused damage and an increase in enzymes that remove reactive oxygen species, thus detoxifying the cells. Given the mitochondrial role in controlling apoptosis, irradiated keratinocytes may prepare their mitochondria to initiate apoptosis.

7. Genes not transcriptionally regulated
Perhaps as interesting are the genes we found not to be regulated by UV. For example, p53, a DNA damage-sensitive regulator of the cell cycle, is not among the regulated genes; UV primarily effects its stabilization through phosphorylation, leaving mRNA levels unchanged. Most cell cycle proteins, notably cyclin D1, and regulators of apoptosis are not regulated by UV. It is perhaps surprising that more dramatic changes in cell cycle and apoptosis proteins were not seen. We suspect these may occur later, when the extent of the damage and the possibility of its repair are fully assessed.

Similarly, ECM proteins, such as elastin, and the metalloproteases that degrade them were not induced in the time period we studied. Transcriptional and post-transcriptional mechanisms can increase their expression in response to UV, but the effects may not be fully manifested in the first 24 h.

8. The changes in mRNA levels reflect changes in protein levels
We found a very good correlation between the array data on the one hand and the protein and mRNA levels on the other, as measured using Northern and Western blots. For example, c-Myc protein and mRNA levels are reduced in parallel. Similarly, Western blots show induction of involucrin, cyclooxygenase-2, Gro-ß, etc., as expected from the array data.

CONCLUSIONS AND SIGNIFICANCE

UV is a major environmental danger to our aging population. The use of arrays to analyze the global changes of gene expression in epidermal keratinocytes in response to UVB allowed us to observe how the skin responds to this damaging agent. We show that the effects of UV on gene regulation, though highly specific, are catholic in their breadth, affecting transcription factors and other signaling proteins, cytoskeletal, cell surface, mitochondrial, and RNA binding proteins, secreted factors, differentiation markers, etc. Some elements of the response were to be expected because some of the responding proteins are already known stress-induced cellular repair genes. Others were unexpected and provide fresh insights in the protective role of epidermis. Our data also make a significant contribution to the expression profiling field, because the appropriate cell type was treated with the appropriate environmental agent.

Perhaps the most significant new finding from our work is the identification of many of UV-induced genes that protect the organism, such as the cornified envelope proteins and secreted signaling polypeptides—in particular, the chemokines of the IL-8 family—as well as the immediate-early genes that presumably protect the keratinocytes themselves.

The findings presented here integrate a large set of disparate findings into a congruent and comprehensive picture. For example, some of the cornified envelope proteins were known to be UV induced, but others were not; here we show that it is a general phenomenon. Both IFN and UV can activate STAT proteins, although by different mechanisms; here we show that IFN and UV induce a common set of genes. IL-8 was known to be UV induced; here we show that five members of the IL-8 family are. Mitochondrial role in apoptosis in response to the genotoxic stress is well established; here we add energy production into the picture.

Among the unanticipated consequences of UV is our discovery that the DNA repair enzymes are not induced but the enzymes that produce building blocks for DNA synthesis are; machinery is already in place, but the nuts and bolts need to be supplemented. Additional results unexpected a priori indicate that 1) UV damaged cells boost their energy supply, both by augmenting mitochondrial protein production and by shutting down energy-costly processes, 2) UV-caused cytoskeletal changes make cells first more pliable, then more rigid, and 3) UV and IFN signaling overlap to some extent.

The most dramatic changes immediately after UV are those inducing new signaling molecules, kinases, phosphatases, proteases, RNA processing enzymes, and transcription factors while shutting others down. When exposed to UV, the cell changes its physiology; it stops doing many of the things it has been doing and turns its attention in response to the stimulus. Having ‘taken care’ of its immediate needs, the cell alerts the surrounding tissue. At this time, the cell produces chemokines, cytokines, and growth factors that activate melanocytes, stimulating tanning as well as the inflammatory response, causing erythema. Still later, the cell induces expression of terminal differentiation markers, components of the cornified envelope. This may have several benefits. Enhanced cornification would provide additional protection to skin; it would eventually enhance the proliferation of keratinocytes in order to restore epidermal homeostasis. In addition, terminal differentiation has many biochemical parallels to apoptosis: cells that differentiate cannot proliferate. This may be a way to eliminate those keratinocytes that, having been damaged by UV, may have received an oncogenic mutation. The illuminated cells sense the need for an additional burst of energy and induce expression of mitochondrial components, glycolysis, and lipid degradation. The illuminated cells also change their cytoskeleton in a characteristic way, first becoming more pliable and adaptable and later more rigid.

In summary, human epidermal keratinocytes in response to UV commence DNA repair, change their assortment of transcription factors and other signal transducing proteins, alert the surrounding tissue to the damage, procure more energy, modify their cytoskeleton, and enhance the protective cornified layer of skin (Fig. 1 ).



View larger version (52K):
[in this window]
[in a new window]
 
Figure 1. Transcriptional effects of UV light on keratinocytes.

FOOTNOTES

1 To read the full text of this article, go to http://www.fasebj.org/cgi/doi/10.1096/fj.01-0172fje; to cite this article, use FASEB J. (September 17, 2001) 10.1096/fj.01-0172fje





This Article
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
15/13/2533
01-0172fjev1    most recent
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by LI, D.
Right arrow Articles by BLUMENBERG, M.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by LI, D.
Right arrow Articles by BLUMENBERG, M.


HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS