Human Keratinocyte Responses to Woodsmoke Chemicals

Abstract

Air pollution consists of complex mixtures of chemicals with serious deleterious health effects from acute and chronic exposure. To help understand the mechanisms by which adverse effects occur, the present work examines the responses of cultured human epidermal keratinocytes to specific chemicals commonly found in woodsmoke. Our earlier findings with liquid smoke flavoring (aqueous extract of charred wood) revealed that such extracts stimulated the expression of genes associated with oxidative stress and proinflammatory response, activated the aryl hydrocarbon receptor, thereby inducing cytochrome P4501A1 activity, and induced cross-linked envelope formation, a lethal event ordinarily occurring during terminal differentiation. The present results showed that furfural produced transcriptional responses resembling those of liquid smoke, cyclohexanedione activated the aryl hydrocarbon receptor, and several chemicals induced envelope formation. Of these, syringol permeabilized the cells to the egress of lactate dehydrogenase at a concentration close to that yielding envelope formation, while furfural induced envelope formation without permeabilization detectable in this way. Furfural (but not syringol) stimulated the incorporation of amines into cell proteins in extracts in the absence of transglutaminase activity. Nevertheless, both chemicals substantially increased the amount of cellular protein incorporated into envelopes and greatly altered the envelope protein profile. Moreover, the proportion of keratin in the envelopes was dramatically increased. These findings are consistent with the chemically induced protein cross-linking in the cells. Elucidating mechanisms by which this phenomenon occurs may help understand how smoke chemicals interact with proteins to elicit cellular responses, interpret bioassays of complex pollutant mixtures, and suggest additional sensitive ways to monitor exposures.

Figure 1

Concentration dependence of envelope formation for the three representative constituent smoke chemicals. Cultures in 24-well plates were treated for 1 day at 37 °C with guaiacol, syringol, or furfural at the indicated concentrations in serum-free medium, and then nonenvelope proteins were dissolved by the addition of SDS to 2% and DTT to 20 mM for an hour or more. The envelope structures were rinsed several times in 0.1% SDS, and protein levels were quantitated.

Figure 2

Percentage of total cell protein incorporated into SDS/DTT-insoluble envelopes. Cultures were incubated in 2 mL of serum-free medium for a day with the addition of NaCl (93 mg), ionophore X537A (200 μg), syringol (5 mg), furfural (17 mg), or no addition (none). Total and SDS-/DTT-insoluble protein fractions were quantitated in triplicate cultures. The values for furfural were significantly higher than the others (* p < 0.001).

Figure 3

Incorporation of BPA into cellular proteins, as detected by immunoblotting. Cell extracts were incubated with the amounts (mg/mL) of either furfural (F, panel A) or syringol (S, panel B) indicated, submitted to immunoblotting with a BPA antibody. The presence (+) or absence (−) of the added calcium ion, as shown, shows the activity of endogenous TGM1 (absent in the absence of calcium).

Figure 4

Differences in protein incorporation into envelopes elicited by the treatment with NaCl, syringol, or furfural. (A) Venn diagram shows the numbers of proteins that differed in amount in two-way comparisons of envelope profiles from the different treatments. Volcano plots show results of two-way comparisons of envelope protein profiles from these cultures. The x-axis shows the log2 ratio for the protein levels of furfural/syringol treatments (B), NaCl/syringol treatments (C), and furfural/NaCl treatments (D), and the y-axis shows −log10 of the unadjusted p-values. Protein levels that differ significantly at adjusted p < 0.05 are shown in blue, and the remaining proteins are shown in red.

Figure 5

Dependence of keratin protein incorporation into envelopes on chemical treatment revealed by label-free quantitation. As shown, the % of envelope proteins (Env) that consist of keratins is low (2%) for envelopes stimulated by NaCl treatment but much higher for cultures treated with syringol or furfural. The % of keratins in the SDS/DTT soluble cell protein is similar in each case but slightly lower in cultures treated with the air pollution chemicals due to their high incorporation in envelopes. Bars with different labels (a–d) were statistically different in % keratin (p < 10^–5).

Figure 6

Stimulation of pro-oxidant and proinflammatory transcription by furfural and 3-methycatechol. Treating cultures with chemicals at 10% of their EC50 values for envelope formation induced HMOX1, TXNRD, CLCl8, and PTGS2 significantly above background levels (*, p < 0.05). 3-Methyl catechol significantly induced GCLM at that low concentration.

Figure 7

Activation of the aryl hydrocarbon receptor (A) and 7-ethoxyresorufin O-deethylase expression (B) by 1,4-cyclohexanedione. Cyclohexanedione and menadione were active in rat and human hepatoma lines carrying a dioxin response element (A), but only cyclohexanedione was active in inducing CYP1A1 in human keratinocytes (B). (A) Values of each bar were significantly different from background (0), p < 0.02. (B) Value for 1 mg/mL was significantly different from 0.1 mg/mL (p < 0.002), which was not significantly different from background (0 mg/mL).