The effect of essential fatty acid supplementation on keratinocyte replication

PROSTAGLANDINSLEUKOTRIENES AND ESSENTIALFATTYACIDS Prostaglandins Leukotrienes and Essential Fatty Acids (1995) 52, 349-355 © Pearson Professional Ltd 1995

The Effect of Essential Fatty Acid Supplementation on Keratinocyte Replication W. L. Garner, Y. Oyatsu, C. Zuccaro, J. L. Rodriquez, D. J. Smith and C. L. Marcelo Section of Plastic and Reconstructive Surgery, University of Michigan, 1500 E Medical Center Drive, Ann Arbor, MI 48109, USA (Reprint requests to WLG) ABSTRACT. Epidermal cell growth in culture, using the low calcium, low serum technique described by Boyce,

is thought to induce rapid expansion by inducing an essential fatty acid (EFA) deficiency state. To determine the mechanisms whereby EFA deficiency induces increased epidermal cell growth, keratinocytes were passaged into medium without or with the addition of EFAs, 18:2(n-6), 20:4(n-6). The resulting populations were assayed for replication rate, differentiation, and plating efficiency. Supplemental EFAs significantly decrease keratinocyte culture expansion. This is evidenced by an increase in generation time, a decrease in thymidine incorporation, and a decrease in modeled replication rate. EFA supplementation also increased the expression of cornified cell envelopes. Serum-free medium induces EFA deficient keratinocytes that demonstrate increased replication and decreased differentiation. Restoration of EFAs reverses these changes. It may be possible to manipulate keratinocyte physiology using fatty acid modifications.

INTRODUCTION

cally in patients with EFA deficiency (13). The addition of EFAs to this culture system, recreating the FA pattern found in normal epidermis, resulted in a decrease in the growth rate of keratinocyte cultures. Therefore, the success of the serum-free culture technique appears to depend, at least in part, on the induction of a cellular EFA deficiency state (14). While the efficacy of the Boyce technique for expanding cell cultures is well established, the mechanisms whereby this medium alters keratinocyte growth rates are not defined. Optimal growth in culture is the sum of several different processes, including adherence to plastic, replication rate, differentiation, and rate of cell death. Many of these activities are altered in other cell types by changes in FA composition. This study was designed to determine the specific effects of FA supplementation on keratinocyte cellular functions which might result in increased culture expansion.

Fatty acids (FAs) are a class of lipids which provide cellular membrane structure, energy source, and are precursors for eicosanoids which have many additional functions. FAs differ in length, number, and position of unsaturated double bonds. Alterations in FA composition of cells have been documented to affect cell functions including growth rate, adhesion, differentiation, and enzyme function (1-11). Human cells cannot synthesize the polyunsaturated FAs 18:2(n-6) and 20:4(n-6). These must be obtained from dietary sources and are, therefore, considered essential. In vitro, the essential fatty acids (EFAs) are supplied by serum in the medium. Cell growth in serum-free medium results in a progressive decrease in intracellular EFA content, as polyunsatured lipids are replaced by the saturated and monounsaturated FAs which human cells can synthesize. The expansion of epidermal cell cultures for research or clinical use was greatly simplified by the introduction of a low calcium, serum-free culture technique by Boyce and Ham (12). The lack of serum in this medium results in an in vitro EFA deficient state with an associated epidermal cell hyperproliferation, as is also seen clini-

METHODS Preparation of cell populations 13 primary keratinocyte populations were established as described elsewhere (15). Partial thickness skin samples were obtained from patients undergoing elective aesthetic or reconstructive post-burn procedures which included

Date received 12 September 1994 Date accepted 24 October 1994 349

350 ProstaglandinsLeukotrienesand Essential Fatty Acids the removal of skin (IRB 87-224). This population is primarily adult and is comprised of both male and female patients of all racial groups. Skin samples were soaked in solution A (30 mM HEPES, 10 mM glucose, 3 mM KC1, 130 mM NaC1, 1.0 mM Na2HPO4, pH 7.4) containing 2.5 [tg/ml gentamicin to decrease bacteria counts. The skin was cut into narrow strips and incubated with 0.15% trypsin at room temperature overnight. The epidermis was separated from the dermis, leaving the basal epidermal cells (keratinocytes) on the dermal remnant. Trypsinization of the skin sample was stopped by the addition of 10% chelated (calcium-free) serum and the keratinocytes were removed by mechanical scraping. Particulate matter was removed by filtration through nylon mesh and the cell suspension plated in MCDB 153 (KGM, Clonetics) with 0.218].tg/ml hydrocortisone, 5 ng/ml EGF, 5 gg/ml insulin, 6 mg% bovine pituitary extract, and 0.15 mM CaC12 (complete medium). The cells were initially plated in medium containing 2% calcium-free fetal calf serum (FCS) at a density of 20 x 106 cells per T-75 flask in 5% CO2 at 37°C. After cell attachment, the cell culture were grown in serum-free MCDB 153 in T-75 flasks. Cells were passed when the cultures were 85% confluent using 2 ml of 0.03% trypsin mad 0.01% EDTA at 37°C. Trypsinization was stopped by the addition of a three-fold excess of trypsin inhibitor (Sigma).

FA supplementation FAs, 16:0, 18:2(n-6), 20:4(n-6) were obtained from NuChek (Elysian, MN). Lipids were dissolved in 95% ethyl alcohol and added to FA-free bovine serum albumin (BSA) (0.6 rag%) as carrier (14). Fresh FA solutions were added to complete medium and the cells fed thrice weekly. FA supplementation was begun at first passage (P1). The media used were either control, complete medium plus the carrier protein, FA-free BSA (0.6 mg/ 100 ml) (Sigma, fraction 5), or EFA medium, control medium plus 5 ~tM 16:0, 10 gM 18:2, and 5 gM 20:4 FAs using BSA as a carrier protein. In a small number of experiments a non-essential FA control medium was used, which contained 5 gM 16:0 and 15 gM 18:1 FAs.

Cell replication Cell replication rates were obtained by comparing the average daily increase in number of plated cells. Keratinocyte strains were grown in each media for at least 5 doublings prior to determination of growth rates. Then, 50 000-75 000 cells were plated in triplicate 22m m cultures dish wells (Costar) and daily cell counts obtained by trypsinization with 1 ml of 0.03% trypsin/ 0.01% EDTA using a Coulter counter. To eliminate media-induced trypsinization or cell adherence effects on the initial plating efficiency, cell counts were compared between sequential days rather than to the original cell number plated. The increase in cell number, re-

ported as generation time (GT), was calculated using the formula (16); 0.693 Generation time =

2.3 Log (cell # day B - cell # day A) time B (days) - time A (days)

Consecutive daily cell counts allowed the determination of growth rates by linear regression in 10 of 13 experiments. The slopes of these growth curves were then compared as described below.

Density Maximal density was defined as the highest cell count obtained for each cell line. Cell counts were the average of triplicate 4-cm 2 wells.

Thymidine incorporation Cell replication was quantified by [3H]-thymidine ([3H]Tdr) incorporation using 1 gCi/ml [3H]Tdr in a 4-h pulse (17). After rinsing twice with 2 ml of cold phosphate-buffered saline (PBS), all macromolecules (DNA, RNA, protein) were precipitated with 6% trichloroacetic acid (TCA). The 6% TCA supernatant fraction, containing free [3H]Tdr and its phosphorylated nucleotides, gave a measure of [3H]Tdr transport into the cell. The pellet was then hydrolyzed with 3% perchloric acid (PCA) at 100°C for 30 min. The 3% PCA supematant provided a measure of the rate of DNA synthesis. In addition, total gg of DNA in the 3% PCA supernatant was determined using the Burton DNA assay. The 1 M NaOH dissolved pellet was assayed for protein using the Lowry technique. The cell proliferation rate was defined as cpm [3H]Tdr/~tg DNA. The total protein per plate was used as an additional measure of cell culture growth. The 6% TCA supematant (acid-soluble fraction) confirmed consistent uptake of the label into the cytoplasm of the cells grown in the various media.

Cornified envelopes Cornified envelopes were determined by a modification of the method of Reiss and Sartorelli (18). Near confluent cultures were washed with solution A and trypsinized as above. An aliquot of the resuspended cell solution was counted with a hemacytometer. The remaining cells were treated with a solution A containing 1% (w/v) sodium dodecyl sulfate (SDS) and 10 mM dithiothreitol. The cell suspension was incubated for 30 min at room temperature and the cornified cell envelopes counted using a hemocytometer.

Attachment The rate of keratinocyte attachment (plating efficiency) was determined by plating 100 000 cells into 35-mm

The Effect of Essential Fatty Acid Supplementationon Keratinocyte Replication 351 culture dishes. Non-adherent cells were obtained b y aspirating m e d i u m and a PBS wash from triplicate wells 2, 5, 8, and 11 h after plating. The remaining (adhering) cells were trypsinized and counted. Comparisons were made of percentage of adhering cells to total cell count.

Statistical analysis

THE EFFECT OF PASSAGE ON KERATI NOCYTE DOUBLI NG TI ME 7

I • DOU L,NG T,ME

]

_x I--4.

Both the absolute rate of cell replication and the response o f cultures to the F A supplemented media varied with the individual cell culture. Therefore, many of the data are compared as percent control, Comparisons between control and E F A media were made using Student's t-test. Significance was at p ~< 0.05. Regressions of growth rates were compared using analysis o f covariance ( A N O V A ) for a mixed model and a maximal likelihood approach ( B M D P Statistical Software, Inc). Chi square was used to document differences among media, passage, and individual cell population.

tD z ~ 3, o 2.

0~ ~ - - ' r - ' ~ - r - ~ - r - - ~ 1

2

3

4

5

6

PASSAGE Fig. 1 The expansion of epidermal cell cultures quantified as doubling time (days). Control epidermal cell cultures (n = 4) replicate quickly in low calcium, serum-free (complete) medium. After growth through 4 to 5 passages, replication decreases and the cultures become scenescent. Figure courtesy of W Dunham.

RESULTS Epidermal keratinocytes expanded rapidly in the control complete m e d i u m doubling cell number approximately every 37 h. This rapid growth continued for approximately 3 weeks, during which time the cultures underwent 15 to 25 doublings over five to six passages (Fig. 1) (12). By the sixth passage, keratinocytes grown in the control complete m e d i u m became senescent, replication decreased, and the cells appeared to differentiate or die. Both the absolute rate of cell replication and the specific

passage at which the cultures became senescent varied with the individual cell culture. Supplementation of m e d i u m with the E F A s (18:2 and 20:4) resulted in a significant change in the expansion of the keratinocyte cultures. These effects were seen in the replication of mature keratinocyte cultures grown in control and E F A media (Fig. 2). Control cells increased in number, being 476% of control at day 7. E F A cultures initially increased in number to 129% of control at day

KERATINOCYTE REPLICATION RATE AFTER SUPPLEMENTATION WITH EFA 600 --0-

O

CONTROL

T

~500.

z) 400. (.)

o~ aoo,

i200 . 1004 _J ILl U

0 0

I

I

3

7

10

DAY AFTERPLATING Fig. 2 Cell counts from keratinocyte cultures (n = 4) grown in complete medium without and with the addition of 10 gM 18:2, 5 ~M 20:4, and 5 ~M 16:0. Cell numbers are expressed as % control of unsupplemented ceils at the day of plating. The EFA cells demonstrate significantly decreased replication rates at days 3 and 7.

352

Prostaglandins Leukotrienes and Essential Fatty Acids

Table The effect of EFA supplementation (see Methods) on keratinocyte replication rates. 126 GTs were compared in 13 cultures during four passages

T h y m i d i n e I n c o r p o r a t i on z< a

1000 _

=~ 900-~

Control EFA

GT (days)

+ SD

Range

Difference from control (ANOVA)

_~

1.56 2.98

0.84 2.42

0.42-4.68 0.34-8.28

p = 0.002

<

8007004

ot~

600-

o u

500-

z

400-

n¢

U.l

3. However, as the cultures matured, there was a marked decrease in cell number, a decrease to 20% of control at day 7. In contrast, cells which were supplemented with non-essential FAs responded with an increased rate of expansion, reaching 707% of control by day 7 (data not shown). The above results, documenting differences in culture expansion, may have resulted from EFA-induced changes in cell replication, plating efficiency, and/or differentiation. Cell replication was determined by comparing the daily change in cell number of cultures grown without mad after supplementation with EFAs. The effects of EFAs were investigated using the GT, comparing increase in cell number between sequential days. The average GT of 13 cultures studied during four passages (126 comparisons) was 1.56 days (range 0.42-4.68 days). The addition of EFA significantly increases GT to 2.98 days (range 0.34-8.28 days) (p = 0.002) (Table). The difference in cell expansion rates was further shown by a significant decrease in Tdr incorporation by cell populations supplemented with EFAs (Fig. 3). Control cells incorporate 704 cpm [3H]Tdr/pg DNA. EFA supplemented cells incorporated significantly

a

300200-

~- 100-_-0 ~ CONTROL

EFA MEDIUM TYPE

Fig. 3 Thymidine incorporation is expressed as cpm [3H]Tdr incorporation/gg DNA for control and EFA-supplemented ceils. Tdr incorporation is significantly decreased in the EFA cells (p = 0.045).

less Tdr than control cells; 515 cpm [3H]Tdr/pg DNA (p = 0.045).

With serial passage, the effects of EFA medium increased. This can be seen in the marked increase in generation time for EFA, but not for control cell populations (Fig. 4) (p = 0.017). The sequential cell counts for a subset of eight patients were used to generate regression curves to model the effects of FAs on cell replication. The relative effects of individual cell line growth rates and the effects of passage and medium were analyzed using analysis of covariance for a mixed model using a maximal likelihood approach. The averaged slopes, increased cell

THE EFFECT OF SERIAL PASSAGE ON KERATINOCYTE GENERATION TIME -@-

CONTROL

6, 5, uJ

I--4, Z O uJ

z LIJ (.9

2¸

I

I

I

I

2

3

4

S

PASSAGE

Fig. 4 The changes in cell replication are plotted as GT (see Methods). The effects of serial passage can been seen for control and EFA-supplemented cells. GT is consistent through five passages for control cells. There is a significant increase in generation tim e during passage 5 for the EFA ceils (p = 0.017).

The Effect of Essential Fatty Acid Supplementation on Keratinocyte Replication 353 number per day, were 89 917/day for control cells and 62 107May for E F A cells. These slopes demonstrated a significant difference in growth rates between the media (p = 0.041). The growth rates of individual cultures and effects of passage were also both highly significant predictors of replication rate (p < 0.001). Changes in the F A composition of cell membranes might decrease the expansion of E F A cultures by influ-

encing other membrane activities, such as differentiation or contact inhibition. FA-deficient cultures were characterized by small cuboidal cells, with relatively little cytoplasm (Fig. 5a). The cells grew in monolayers. E F A cells demonstrated more cytoplasm and a bizarre appearance. The cells formed stratified cultures which appeared to be undergoing differentiation (Fig. 5b). The difference in appearance was reflected in a significant

(A)

(B)

Fig. 5 Photomicrographs of keratinocyte growth in each medium (200x). (A) Cells grown without EFA supplementation. Most cells are small and round, with less cytoplasm. These are some larger cells present (small arrow), usually associated with a more differentiated phenotype, (B) Cells cultured with EFA (see text). Cells are more heterogeneous and many have the larger cell appearance (small arrow). Some cell colonies have begun stratification (large arrow).

354

Prostaglandins Leukotrienes and Essential Fatty Acids

MAXI MAL DENSITY 1000

900800

trypsinization and adhere to plastic were compared. Neither the tolerance to trypsinization nor the adherence of unattached cells was affected by supplementation with essential FAs (data not shown).

7oo

x m "' "-'1 .z ",uJ u

600i 500 i 400~ 300. 200.

DISCUSSION

// /

100. 0

CONTROL

EFA MEDIUM

Fig. 6 The maximal cell densityfor each cultureper 22-mm2 culture dish is shown. Supplementationwith EFA significantlydecreasedthe maximal densityachievedby each cell line.

difference in the maximal confluence noted during serial cell passage. EFA culture did not reach the same density (Fig. 6). EFA supplementation significantly doubled the rate of cornified envelope formation compared to control cells, 4.5% to 9% of total cell number (Fig. 7) (p = 0.025). The rate of cornified envelope formation varied with individual culture, passage number, and confluence. It ranged from 0.2% to 14.6% in control cells and from 0.5% to 21.6% in EFA cells. In all comparisons, the EFA cells demonstrated a greater percentage of cells as cornified envelopes. The techniques used to expand keratinocyte cultures depend on the tolerance of the cells to typsinization and the subsequent ability of the ceils to adhere to culture dishes. Because a change in membrane FA composition might alter these properties, the ability of keratinocytes grown in control and EFA-supplemented media to tolerate

CORNI FI ED CELL ENVELOPE PRODUCTION

B1 ,,-rl,

>1 okidl

ua _lOZ e~

0

8-

a: 6 t-_j " ¢J

2 o

EFA

CONTROL

MEDIUMTYPE Fig. 7 The percentage of ceils which demonstrated comified envelopes. There is an increase in cornified envelope production after EFA supplementation (p = 0.02).

The ability to culture epidermal keratinocytes was advanced and simplified by the system described by Boyce and Ham (12). Until recently, it was assumed that this system worked by providing a specific set of nutrients which optimized keratinocyte growth or the medium lacked a serum-derived inhibitor of cell growth. The observation that a FA deficiency state was induced by the low serum conditions of this system is recent (14, 15). However, the specific mechanisms whereby FA deficiency changes the rate of population expansion is unclear. This paper reports two mechanisms for the increased culture expansion seen with this medium. EFAsupplemented medium affects keratinocyte replication rate and the subsequent rate of cell envelope formation. The finding that EFAs alter keratinocyte replication is not unexpected. Previous studies of other cell types have found significant changes in growth rates in response to various lipids (1-11). However, these studies differ from the present one because they investigated the acute effects of FA administration, rather than the effects of long-term exposure. Polyunsaturated FAs markedly inhibit the growth of normal and transformed fibroblasts, saturated and monosaturated FAs less so (3). Growth of the leukemic cell line is suppressed in a dose-dependent fashion by various polyunsaturated FAs (2). In contrast, human breast epithelial cell cultures and cell lines derived from breast neoplasms demonstrate increased growth rates in response to linoleic acid (19, 20). Several of these studies use specific inhibitors of eicosanoid synthesis which suggest leukotriene involvement in growth regulation. Evans et al have suggested that prostaglandin E 2 modulates keratinocyte differentiation induced by calcium (21). The present study supports this result since the EFA-deficient cells demonstrate fewer cornified envelopes. Further studies will be necessary to determine the relative importance of these effects on culture expansion. The lack of effects of EFA on keratinocyte adherence are relevant since other studies have suggested other effects of EFAs. Dietary supplementation with linoleic acid increased mammary tumor spread, but not proliferation (22), suggesting differences in cell adherence. Polyunsaturated FAs decrease the adhesiveness of BHK cells (5), possibly changing plating eficiency. The effects of EFAs in keratinocyte cultures differ since EFAs decreased replication and did not change plating efficiency. Equimolar concentrations of non-essential FAs increased rather than decreased replication rates. These results have not been fully characterized, but are likely

The Effect of Essential Fatty Acid Supplementation on Keratinocyte Replication to b e a n u t r i t i o n a l effect. T h e s e d i f f e r e n c e s in r e s p o n s e to d i f f e r e n t t y p e s o f F A s s u p p o r t the c o n c l u s i o n that the

9.

E F A e f f e c t s are n o t d u e to a n o n - s p e c i f i c t o x i c i t y o f exogenous FAs. D i f f e r e n c e s in k e r a t i n o c y t e e x p a n s i o n in culture are the r e s u l t o f s e v e r a l p h e n o m e n a , i n c l u d i n g d i f f e r e n c e s in

10.

rates o f cell r e p l i c a t i o n , d i f f e r e n t i a t i o n , cell death, cell a d h e r e n c e , a n d the ability to t o l e r a t e t r y p s i n i z a t i o n . This study d e m o n s t r a t e s that E F A s alter the g r o w t h o f kerat i n o c y t e c u l t u r e s b y d e c r e a s i n g t h e r e p l i c a t i o n rate a n d i n c r e a s i n g o n e i n d i c a t o r o f cell d i f f e r e n t i a t i o n . F u r t h e r

11. 12.

studies will b e n e c e s s a r y to c h a r a c t e r i z e the s p e c i f i c cellular m e c h a n i s m s w h i c h i n d u c e t h e s e c h a n g e s .

Acknowledgement

13.

14.

Supported by the Plastic Surgery Education Foundation and NIH Grant AR 26009. 15.

References 1. Bull A W, Nigro N D, Marnett L J. Structural requirements for stimulation of colonic cell proliferation by oxidized fatty acids. Cancer Res 1988; 46: 1771-1776. 2. Chow S C, Sisfontes L, Bjorkem I, Jondal M. Supression of growth in a leukemic I cell line by n-3 and n-6 polyunsaturated fatty acids. Lipids 1989; 24: no. 8. 3. Davidson B C, Girao L A F, Giangregorio A, Murphy J. Polyunsaturated fatty acids modulate fibroblast growth in culture. Anticancer Res 1991; 11: 267-272. 4. Guffy M M, North J A, Burns C P. Effect of cellular fatty acid alteration on adriamycin sensitivity in cultured L1210 murine leukemia cells. Cancer Research 1984; 44: 1863-1866. 5. Hoover R L, Lynch R D, Karnovsky M J. Decrease in adhesion of cells cultured in polyunsaturated fatty acids. Cell 1977; 12: 295-300. 6. Hori T, Yamanaka Y, Hayakawa M, Shibamoto S, Oku N, Ito F. Growth inhibition of human fibroblasts by epidermal growth factor in the presence of arachidonic acid. Biochem Biophys Res Commun 1990; 169: 959-965. 7. Ikushima S, Fujiwara F, Todo S, Imashuku S. Effects of polyunsaturated fatty acids on vincristine-resistance in human neuroblastoma cells. Anticancer Res 1991; 11: 1215-1220. 8. Kragballe K, Voorhees J J. Modulation of epidermal cell division and growth by oxygenation products of

16.

17.

18.

19.

20.

21

22.

arachidonic acid. J Allergy Clin Immunol 1984; 74, no. 3, part 2. Miller C C, Ziboh V A. Induction of epidermal hyperproliferation by topical n-3 polyunsaturated fatty acids on guinea pig skin linked to decreased levels of 3hydroxyoctadecadienoic acid (13-hode). J Invest Dermatol 1990; 94: no.3. Soyland E, Nenseter M S, Braathen L R, Drevon C A. Very long chain n-3 and n-6 polyunsaturated fatty acids inhibit proliferation of human T-lymphocytes in vitro. Eur J Clin Invest 1993; 22: 112-121. Howard B V, Kritchevsky D. Free fatty acids in cultured cells. Lipids 1969; 5: 49-55. Boyce S T, Ham R G. Calcium-regulated differentiation of normal human epidermal keratinocytes in chemically defined clonal culture and serum-free serial culture. J Invest Dermatol 1983; 81: 33s-40s. Marcelo C L, Duell E A, Rhodes L M, Dunham W R. In vitro model of essential fatty acid deficiency. J Invest Dermatol 1992; 99: 703-708. Marcelo C L, Rhodes L M, Dunham W R. Normalization of EFA-deficient keratinocytes requires palmitic acid in the medium. J Invest Dermatol 1994; in press. Garner W, Zuccaro C, Marcelo C, Rodriguez J L, Smith D J. The effects of burn blister fluid on keratinocyte replication and differentiation. J Burn Care Rehab 1993; 14: 127-131. Sheeler P, Bianchi D E. Cell biology: structure, biochemistry, and function. New York: Wiley, 1983: 56-57. Marcelo C L, Kim Y G, Kaine J L, Voorhees J J. Stratification, specialization, and proliferation of primary keratinocyte cultures: Evidence of a functioning in vitro epidermal cell system. J Cell Biol 1978; 79: 356-370. Reiss M, Sartorelli A C. Regulation of growth and differentiation of human keratinocytes by type b transforming growth factor and epidermal growth factor. Cancer Res 1987; 47: 6705-6709. Rose D R, Connolly J M. Effects of fatty acids and inhibitors of synthesis on the growth of human breast cancer cell line in culture. Cancer Res 1990; 50: 7139-7144. Balakrishnan A, Cramer S, Bandyopadhyay G K et al. Differential proliferative response to linoleate in cultures of epithelial cells from normal human breast and fibroadenomas. Cancer Res 1989; 49: 857-862. Evans C B, Pillai S, Goldyne M E. Prostaglandins leukotrienes and essential fatty acids. Prostaglandins Leukot Essent Fatty Acids 1993; 49: 777-781. Hubbard N E, Erickson K L. Effect of dietary linoleic acid level on lodgement, proliferation and survival of mammary tumor metastases. Cancer Lett 1989; 44: 118-119.

355