Keratinocytes: a source of the transmitter L-glutamate inthe epidermis

Matthias Fischer1,2, Dagobert Glanz3, Maximilian Urbatzka2, Thomas Brzoska4 and Christoph Abels4

1Department of Dermatology and Venerology, HELIOS-Klinikum Aue, Aue, Germany;2Department of Dermatology and Venerology, Martin-Luther-University Halle-Wittenberg, Halle (Saale), Germany;3Institute of Physiological Chemistry, Martin-Luther-University Halle-Wittenberg, Halle (Saale), Germany;4Dr. August Wolff GmbH & Co. Arzneimittel, Bielefeld, Germany

Correspondence: Matthias Fischer, MD, Department of Dermatology and Venerology, HELIOS-Klinikum Aue, Gartenstraße 6, D-08280 Aue,

Germany, Tel.: +0049 (0) 3771 581416, Fax: +0049 (0) 3771 581646, e-mail: matthias.fischer@helios-kliniken.de

Accepted for publication 5 March 2009

Abstract: Various glutamate receptors have been described in

both keratinocytes and melanocytes. l-Glutamate is the

physiological agonist of the glutamate receptor family. The source

of this transmitter had not yet been identified. In normal human

epidermal keratinocytes (NHEK) and HaCaT-keratinocytes, cell

supernatants were sampled in various stages of cell density and

the l-glutamate content photometrically determined. The

following examination time-points were defined: non-confluent

(ca. 33%), subconfluent (ca. 70%) and confluent (90–100%). The

l-glutamate concentration originally in the culture medium was

14.7 mg ⁄ l (0.1 mm ⁄ l). The l-glutamate concentration in the cell

supernatant increased in NHEK with increasing cell density:

non-confluent 39.9 ± 4 mg ⁄ l, subconfluent 60.6 ± 15.8 mg ⁄ l,confluent 100.7 ± 33.2 mg ⁄ l. A linear increase of l-glutamate

concentration was also found for HaCaT cells. The investigations

show that keratinocytes are capable of producing and releasing

l-glutamate. Thus they are a source of l-glutamate which acts as a

transmitter on epidermal glutamate receptors.

Key words: differentiation – glutamate – keratinocyte –

NMDA-receptor

Please cite this paper as: Keratinocytes: a source of the transmitter l-glutamate in the epidermis. Experimental Dermatology 2009; 18: 1064–1066.

Background

Glutamate receptors play a central role in signal transmission

in the nervous system. A distinction is made between metab-

otropic (mGluR1–8) and ionotropic receptors; the ionotrop-

ic glutamate receptors are subdivided further, according to

their in vitro agonists, into amino-3-hydroxy-5-methyl-4-

isoxazolpropionate (AMPA), kainate and N-methyl-d-aspar-

tate receptors (NMDA) (1). Whereas it was previously

assumed that glutamate receptors exist predominantly in the

nervous system, increased evidence has been uncovered in

recent years that they also exist in non-neuronal cells such as

keratinocytes and melanocytes (2–10). While melanocytes

appear to express all subtypes of glutamate receptors, so far

only NMDA receptors have been reliably demonstrated in

human keratinocytes. In rats, however, AMPA receptors and

metabotropic glutamate receptors (mGluR1) have been

found, to a lesser extent, in the basal layer with immunohis-

tochemical methods (6). The particular functions of the

glutamate receptors are still unclear. Only the NMDA

receptors are known to play a role in the calcium-induced

differentiation of keratinocytes (2,3).

Question addressed

In the central nervous system, l-glutamate is produced by

glia cells and Schwann cells (11) or released at the synaptic

gap by excitatory neurons; at this location it binds to the

various subtypes of glutamate receptors (1). In contrast to

the nervous system, the origin of the transmitter l-gluta-

mate within the epidermis has not yet been determined.

For this reason, keratinocytes have been investigated as a

possible source of l-glutamate in this study.

Experimental design

The investigations were carried out with normal human

epidermal keratinocytes (NHEK; PromoCell GmbH,

Heidelberg, Germany) and HaCaT keratinocytes (supplied

magnanimously by Professor Fusenig, German Cancer

Research Institute, Heidelberg, Germany). Both cell lines

were used, because they differ in calcium dependant differ-

entiation (12), which could be influenced by NMDA-recep-

tors. The cells were cultured in complete keratinocyte

growth medium containing 0.15 mm calcium, 0.125 ng ⁄ ml

DOI:10.1111/j.1600-0625.2009.00886.x

www.blackwellpublishing.com/EXDLetter to the Editor

1064 ª 2009 John Wiley & Sons A/S, Experimental Dermatology, 18, 1064–1066

human recombinant EGF, 5.0 lg ⁄ ml insulin, 0.33 lg ⁄ ml

hydrocortisone, 10.0 lg ⁄ ml human transferrin, 0.39 lg ⁄ ml

epinephrine and 0.4% bovine pituitary extract. The cells

were kept at 37�C in a humidified atmosphere containing

5% CO2.

To determine the concentration of l-glutamate in the cell

supernatant, the HaCaT keratinocytes or NHEK were cul-

tured in 35 mm culture dishes. When the culture reached

certain stages of confluence, the cell supernatant was

removed; the concentration of l-glutamate was then deter-

mined using photometric methods. The following measure-

ment points were defined: non-confluent (approx. 33 % of

the area covered by cells), subconfluent (approx. 70 % of the

area covered with cells), and confluent (90–100 % of the

area covered by cells). The glutamate concentration was

measured using a commercially available kit (l-Glutaminsa-

ure Farb-Test; R-Biopharm, Darmstadt, Germany) in accor-

dance with the manufacturer’s instructions. During the test,

l-glutamate was deaminized oxidatively by nicotinamide

adenine dinucleotide (NAD) in the presence of the enzyme

glutamate dehydrogenase to produce 2-oxoglutarate. The

NADH produced in this way converts iodonitrotetrazolium

chloride in the presence of diaphorase to a formazan, which

was then measured at 492 nm.

Results

The l-glutamate concentration of the nutrient medium at

the factory was 14.7 mg ⁄ l (0.1 mm ⁄ l). At the non-confluent

stage, the HaCaT cells already displayed a slightly elevated

l-glutamate concentration in the cell supernatant of

19.6 ± 3.2 mg ⁄ l. This concentration rose in linear fashion as

the cell density increased; at the confluent stage it reached a

maximal concentration of 124.4 ± 13.6 mg ⁄ l (Fig. 1).

In NHEK as well the l-glutamate concentration rose in

the cell supernatant as the cell density increased: non-

confluent 39.9 ± 4 mg ⁄ l (no. experiments: n = 8), subcon-

fluent 60.6 ± 15.8 mg ⁄ l (no. experiments: n = 8), confluent

100.7 ± 33.2 mg ⁄ l (no. experiments: n = 3).

Conclusions

The investigations carried out showed that keratinocytes

are able to produce and release l-glutamate. They are thus

a source of l-glutamate, which can act as a transmitter on

epidermal glutamate receptors. This observation is congru-

ent with reports published by Hoogduijn et al. (7), who

also noted an elevated l-glutamate concentration in the cell

supernatant of HaCaT cell cultures. Because of the fact that

both autocrine and paracrine mechanisms are possible in

keratinocytes in principal (13), keratinocytes seem to stim-

ulate their glutamate receptors via a release of l-glutamate.

l-Glutamate, a non-essential amino acid, is produced in

the cell via a reductive aminiation of 2-oxoglutarate. This

process takes place in the mitochondria of all cells. Apart

from this, l-glutamate can be synthesized via transamina-

tion of 2-oxoglutarate. The enzymes required for this syn-

thesis have basically been demonstrated in keratinocytes

(14). The glutamine in the nutrient medium, which can be

converted to glutamate by the glutaminase in the keratino-

cytes, is an additional source of l-glutamate (15).

The distribution of l-glutamate in the epidermis is now

a subject of discussion. Whereas Fuziwara et al. (5)

described mainly expression in the upper epidermis in their

studies with hairless mice, Nordlind et al. (16) demon-

strated l-glutamate in the entire epidermis via immunohis-

tochemical methods. Our data, which demonstrate the

release of l-glutamate in both non-confluent cells with a

tendency to proliferation and confluent cells beginning to

cornify, point to a continuous process and thus to uniform

distribution within the epidermis. Accordingly, the biologi-

cal function of the l-glutamate released by the keratino-

cytes would not depend on the absolute concentration of

l-glutamate but rather on the distribution of the NMDA

receptors (and possible also other of glutamate receptors as

well) in the epidermis. Comparable models – in which the

function of a transmitter in keratinocytes does not depend

on the distribution of the transmitter but rather on the

expression of various receptor subtypes – have also been

described for purine and acetylcholine receptors (17,18).

Moreover, the demonstration of an amino acid transport

system that inactivates l-glutamate would be congruent

with the assumption of uniform distribution of l-glutamate

in the epidermis. The amino acid transport systems EAAC1

and GLT-1 have been found in the basal and suprabasal

epidermal layers in rats (6). In light of the fact that NMDA

receptors influence differentiation and are present in the

entire vital epidermis (particularly in the granular layer),

the distribution pattern of the amino acid transporter can

HaCaT-cells

124.4

62.9

19.6

020406080

100120140160

non-confluent subconfluent confluent

L-G

lu m

g/l

Figure 1. Increase in the L-glutamate concentration in HaCaT cells as a

function of cell density. Chequered areas: Concentration of L-glutamate

in culture medium. (n = 4 experiments each). Microscopic magnification

20x.

Letter to the Editor

ª 2009 John Wiley & Sons A/S, Experimental Dermatology, 18, 1064–1066 1065

be interpreted as a protective mechanism safeguarding

against premature differentiation induced by NMDA

receptors.

The investigation confirms that keratinocytes are a

source of l-glutamate in the epidermis. Additional studies

will have to be performed to determine whether this

mechanism plays an important role in pathophysiological

processes in the skin.

References

1 Kew J N, Kemp J A. Ionotropic and metabotropic glutamate receptor structureand pharmacology. Psychopharmacology (Berl) 2005: 179: 4–29.

2 Fischer M, Glanz D, William T, Klapperstuck T, Wohlrab J, Marsch W Ch.N-methyl-d-aspartate-receptors influence the intracellular calcium concentra-tion of keratinocytes. Exp Dermatol 2004: 13: 512–519.

3 Fischer M, William T, Helmbold P, Wohlrab J, Marsch W Ch. Expression of epi-dermal N-methyl-d-aspartate receptors (NMDAR1) depends on formation ofthe granular layer-analysis in diseases with parakeratotic cornification. ArchDermatol Res 2004: 296: 157–162.

4 Fischer M, Fiedler E, Seidel C et al. Keratinocytes express N-methyl-d-aspar-tate-receptors of the NMDAR2D-type. Arch Dermatol Res 2006: 297: 316–318.

5 Fuziwara S, Inoue K, Denda M. NMDA-type glutamate receptor is associatedwith cutaneous barrier homeostasis. J Invest Dermatol 2003: 120: 1023–1029.

6 Genever P G, Maxfield S J, Kennovin G D et al. Evidence for a novel gluta-mate-mediated signaling pathway in keratinocytes. J Invest Dermatol 1999:112: 337–342.

7 Hoogduijn M J, Hitchcock I S, Smit N P, Gillbro J M, Schallreuter K U, GeneverP G. Glutamate receptors on human melanocytes regulate the expression ofMiTF. Pigment Cell Res 2006: 19: 58–67.

8 Morhenn V B, Waleh N S, Mansbridge J N et al. Evidence for an NMDA recep-tor subunit in human keratinocytes and rat cardiocytes. Eur J Pharmacol 1994:268: 409–414.

9 Morhenn V B, Murakami M, O’Grady T, Nordberg J, Gallo R L. Characteriza-tion of the expression and function of N-methyl-d-aspartate receptor in kerati-nocytes. Exp Dermatol 2004: 13: 505–511.

10 Nahm W K, Philpot B D, Adams M M et al. Significance of N-methyl-d-aspar-tate (NMDA) receptor-mediated signalling in human keratinocytes. J Cell Phys-iol 2004: 200: 309–317.

11 Wu S Z, Jiang S, Sims T J, Barger S W. Schwann cells exhibit excitotoxicity con-sistent with release of NMDA receptor agonists. Neurosci Res 2005: 79: 638–643.

12 Micallef L, Belaubre F, Pinon A et al. Effects of extracellular calcium on thegrowth-differentiation switch in immortalized keratinocyte HaCaT cellscompared with normal human keratinocytes. Exp Dermatol 2009: 18: 143–151.

13 Wood J M, Schallreuter K U. A plaidoyer for cutaneous enzymology: our viewof some important unanswered questions on the contributions of selected keyenzymes to epidermal homeostasis. Exp Dermatol 2008: 17: 569–578.

14 Koss-Harnes D, Jahnsen F L, Wiche G, Søyland E, Brandtzaeg P, Gedde-Dahl TJr. Plectin abnormality in epidermolysis bullosa simplex Ogna: non-responsive-ness of basal keratinocytes to some anti-rat plectin antibodies. Exp Dermatol1997: 6: 41–48.

15 Benavente C A, Jacobson E L. Niacin restriction upregulates NADPH oxidaseand reactive oxygen species (ROS) in human keratinocytes. Free Radic BiolMed 2008: 44: 527–537.

16 Nordlind K, Johansson O, Liden S, Hokfelt T. Glutamate- and aspartate-like im-munoreactivities in human normal and inflamed skin. Virchows Arch B CellPathol Incl Mol Pathol. 1993: 64: 75–82.

17 Greig A V, Linge C, Cambrey A, Burnstock G. Purinergic receptors are part ofa signaling system for keratinocyte proliferation, differentiation, and apoptosisin human fetal epidermis. J Invest Dermatol 2003: 121: 1145–1149.

18 Inoue K, Hosoi J, Denda M. Extracellular ATP has stimulatory effects on theexpression and release of IL-6 via purinergic receptors in normal human epi-dermal keratinocytes. J Invest Dermatol 2007: 127: 362–371.

Letter to the Editor

1066 ª 2009 John Wiley & Sons A/S, Experimental Dermatology, 18, 1064–1066