Fatty acid chain elongation by microsomal enzymes from the bovine meibomian gland

ARCHIVES OF BIOCHEMISTRY AND BIOPHYSICS Vol. 237, No. 1, February 15, pp. 177-185, 1985

Fatty Acid Chain Elongation by Microsomal Enzymes from the Bovine Meibomian Gland’ GREGORY Institute

of Biological

Chemistry Received

J. ANDERSON

AND

and Biochemistry/Biophysics Pullman, Washington July

19, 1984, and in revised

P. E. KOLATTUKUDY’ Program,

Washington

State

University,

9916.4

form

October

26, 1984

The composition of meibomian gland lipids suggested that fatty acid chain elongation might play a major role in the synthesis of such lipids. A fatty acid synthase preparation from the bovine meibomian gland catalyzed the formation of Cl6 acid and the enzyme was immunologically quite similar to that in the mammary gland. The microsomal fraction from the gland, on the other hand, catalyzed elongation of endogenous fatty acids in the presence of ATP and Mgz+ and of exogenous C1s-CoA using malonyl-CoA and NADPH as the preferred reductant. The elongated products, ranging up to Czs in chain length, were found mainly as CoA esters and products derived from them. With Cre-CoA as the exogenous primer, the elongation rate was linear with incubation time up to 20 min but the rate changed in a sigmoidal manner with increasing protein concentration. The elongation rate was maximal at a pH around 7.0. Typical Michaelis-Menten-type substrate saturation patterns were observed with both malonyl-CoA and NADPH. From linear double-reciprocal plots, the K, values for the two substrates were calculated to be 52 and 11 pM, respectively, with a Vof about 340 pmol min-’ mg protein-’ with respect to malonyl-CoA. Exogenous CoA esters of Cl6 to Czz fatty acids were elongated to give products up to Czs without exhibiting any preference for the primer. The present elongation system could account for the formation of most of the very long chains found in meibomian lipids. o 1% Academic

Press. Inc.

The outer surfaces of most animals and plants are coated with a complex mixture of nonpolar lipids which are quite unlike the internal lipids (l-3). In animals these surface lipids, which perform unique functions, are generated by sebaceous glands. One of the unusual features of the surface lipids is the presence of very long aliphatic chains. For example, compounds containing 20-30 carbon atoms are common in mammalian surface lipids (2, 3). The meibomian gland, a special sebaceous gland located in the eyelids, provides the

lipids which play a key role in the proper functioning of the eye (4, 5). A major portion of the aliphatic chains found in the meibomian gland lipids have more than 16 carbon atoms (6-9) and such compounds might be derived by elongation of the usual product generated by fatty acid synthase. However, in view of the fact that fatty acid synthase from some organisms can generate very-long-chain fatty acids (lo), the possibility that such a synthase might be present in this specialized tissue cannot be ruled out. Biosynthesis of meibomian lipids has not been previously investigated except for a study of incorporation of exogenous labeled precursors into the unique lipids by excised glands (L. Rogers, P. E. Kolattukudy, and N. Nicolaides, manuscript in

‘Scientific Paper 6886, Project 2001, College of Agriculture Research Center. This work was supported in part by Grant GM-18278 from the U. S. Public Health Service. a To whom corrspondence should be addressed. 177

0093-9861/85 Copyright All rights

$3.06

0 1985 by Academic Press. Inc. of reproduction in any form resewed.

178

ANDERSON

AND

preparation). In the present paper we demonstrate that the fatty acid synthase present in the bovine meibomian gland is immunologically quite similar, if not identical, to the synthase present in other tissues of the animal, and that a microsomal enzyme system present in the gland is responsible for the production of the very long chain fatty acids. Some of the characteristics of the elongating system of the meibomian gland are described. MATERIALS

AND

METHODS

Steer eyelids were obtained from an area slaughterhouse and frozen at -70°C shortly after removal. Dithioerythritol (DTE): NADPH, NADH, ATP, glucose g-phosphate, glucose-g-phosphate dehydrogenase, tetradecanoyl-CoA, hexadecanoyl-CoA, octadecanoyl-CoA, eicosanoyl-CoA, and docosanoyl-CoA were purchased from Sigma Chemical Company. [2“C]Malonyl-CoA and [1-“Ckctadecanoyl-CoA were purchased from New England Nuclear Corporation. [2-i4C]Malonyl-CoA was also synthesized by the thiophenyl ester exchange method (11). Before use, the product was diluted to a specific activity of 2.5-12.3 Ci/mol. Boron trifluoride in methanol was prepared by bubbling BF, gas into anhydrous methanol until the weight of the mixture increased by 14%. The concentration of acyl-CoA and NAD(P)H stock solutions was determined spectrophotometrically using reference constants supplied by P-L Biochemicals and/or by using weighed amounts. Stock solutions of octadecanoyl-CoA, eicosanoyl-CoA, and docosanoyl-CoA were sonicated for 30 s with the needle probe of a Biosonik III immediately before use. Prepwahm of m&osomea A longitudinal excision was made along the line of the eyelashes exposing the meibomian glands in partially thawed eyelids. The glands were excised and any adhering tissue was removed with a razor blade, yielding about 35 mg tissue per lid. They were homogenized in buffer containing 0.1 M sodium phosphate, pH 7.6, 0.25 M sucrose, 1 mM DTE. Organelles and cell debris were removed by centrifugation at looos for 5 min followed by recentrifugation of the supernatant for 15 min at 12,060g. The resulting supernatant was centrifuged at 105,OOOg for 90 min and the pellet was washed by resuspension in the homogenizing buffer and centrifugation for 60 min at 105,ooOg. The washed microsomes were gently resuspended in homogenization buffer (0.5 ml per original eyelid). Protein concentration was detemined by the method of Lowry et al (12).

a Abbreviation

used: DTE,

dithioerythritol.

KOLATTUKUDY Enzyme assays. Unless otherwise specified, microsomal chain elongating activity was assayed in a reaction mixture containing 0.1 M sodium phosphate, pH 7.6, 1 mrd DTE, 1 mrd NADPH, 30 @d octadecanoyl-CoA, 80 PM [2-“Clmalonyl-CoA, and 0.5 mg microsomes in a total volume of 0.5 ml. To test for incorporation of labeled primer, 30 jtM [lWJoctadecanoyl-CoA (59 Ci/mol) and 80 m malonylCoA were substituted in the above reaction mixture. Reactions were initiated by addition of the microsomes and, after gentle shaking for 20 min at 31°C. the mixture was acidified with 6 N HCI. Total lipids were obtained by extraction of the acidified reaction mixture once with ehloroform:methanol (2:l) and twice with chloroform. The chloroform extract was washed with acidic water and evaporated to dryness under reduced pressure. Control experiments with [I-“C]C,-CoA showed that 295% of the acyl-CoA was recovered by this procedure. As a further check, the lipids were extracted as above after exposure of the reaction mixture to 0.1 M NaOH for 15 min on a steam bath and subsequent acidification to release any water-soluble thioester-bound acyl chains. Since this method yielded essentially the same amount of products as that obtained without the base treatment, the direct extraction procedure was routinely used. Produet an&six The lipid products recovered from the reaction mixture were examined by thinlayer chromatography on silica gel G with either hexane:ethyl ether:formie acid (40:1&l, v/v) or CHClz:MeOH:HOAc:HzO (85:15:10:4, v/v) as the solvent system. Nonradioactive standards (eicosanoic acid, octadecanol, methyl palmitate, palmitoyl palmitate, phosphatidylethanolamine, and phosphatidylcholine) were identified under uv light after spraying of the TLC plate with a 0.1% solution of 2’,7’-dichlorofluoresceine in ethanol. For analysis by radio-gas-liquid chromatography, the total lipids were refluxed in 14% BF, in methanol for 2.5 h. After addition of water, the lipids were recovered by CHCls extraction and the fatty acid methyl esters were purified by thin-layer chromatography on silica gel G with hexane:ethyl ether:formic acid (40:1&l, v/v) as the solvent. After recovery from the gel by ether extraction, the methyl esters were subjected to gas-liquid chromatography in a coiled stainlesssteel column (198 x 0.3 cm) packed with 5% OV-1 on 80-100 mesh Gas Chrome Q with a 185 to 310°C temperature program at lO”C/min. A Perkin-Elmer 810 gas chromatograph connected to a Barber-Colman radioactivity monitor was used and the peak areas were determined by triangulation. Identification was by cochromatography with standards. Determinationof rcuhactivity. Lipid samples were assayed for “C in 0.4% Omnifluor as described earlier (15). Radioactivity on thin-layer platea was determined directly by assaying the gel in the scin-

BIOSYNTHESIS

OF

MEIBOMIAN

tillation fluid. A Berthold thin-layer radioactivity scanner was used to locate the labeled components on thin-layer chromatograms. Preparation of antiserum Fatty acid synthaae antiserum was prepared basically as described earlier (16). Injections consisted of 1.2 mg enzyme (containing 50 mM potassium phosphate, pH 7.0, 0.1 mM EDTA, 1 mM DTE in 0.5 ml) emulsified with 0.5 ml adjuvant. Blood was collected by heart puncture and the antiserum was stored at -80°C without further purification. Immuncd(@sion Double-diffusion analysis was done in Petri dishes on 1% agar containing 0.9% NaCl according to the method of Ouchterlony (17). After 18 h of diffusion the nonagglutinated protein was removed by repeated washes with 0.9% NaCl followed by distilled water, and protein was fixed with 10% acetic acid. Preparation of fatty acid synthase. Bovine mammary gland fatty acid synthase was purified to homogeneity by the method of Kumar and Dodds (18). A fatty acid synthase preparation from bovine meibomian gland was obtained by 30% ammonium sulfate precipitation of the 105,000~ supernatant followed by dialysis of the suspended precipitate against 0.1 M sodium phosphate, pH 7.0, 0.1 mM EDTA, 1.0 mM DTE. RESULTS

AND

DISCUSSION

The meibomian gland could generate very-long-chain acids either by virtue of a special chain length specificity of the fatty acid synthase, such as that observed with mycobacteria (lo), or by an elongation system similar to those found in other mammalian tissues (19). To distinguish between these two possibilities, the subcellular fractions isolated from the gland homogenate were examined for their ability to generate fatty acids from malonyl-CoA. The high-speed (105,OOOg) supernatant readily incorporated malonylCoA into fatty acids with NADPH as the preferred reductant. The products were free fatty acids and radio-gas-liquid chromatographic analysis showed that the major product was n-Cl6 (83%) with a small amount of n-Cl8 (8%). The fatty acid synthase activity was precipitated by 30% saturation with ammonium sulfate, as is usual with vertebrate fatty acid synthases. This enzyme preparation cross-reacted with rabbit antibodies prepared against fatty acid synthase from the bovine mammary gland. Since Ouch-

GLAND

LIPIDS

179

terlony double-diffusion analysis showed complete fusion of the precipitant lines (Fig. l), it is highly likely that the fatty acid synthase from the meibomian gland is identical to the synthase present in the mammary gland and probably other tissues as well (20, 21). Obviously, the verylong-chain acids produced by the gland are not generated by the synthase. Elongation of endogenous primers. The microsomal preparation from the meibomian gland incorporated [2-‘4C]malonylCoA into fatty acids in the presence of ATP and Mgz+. After treatment of the lipid products with 14% BFa in methanol, it was found that 76% of the label was present in fatty acid methyl esters, while 24% represented fatty alcohols. Radiogas-liquid chromatographic analysis of the methyl esters showed that the bulk of the label was contained in fatty acids longer than Ci6, suggesting that malonylCoA was incorporated into fatty acids mainly by chain elongation of preformed fatty acids (Fig. 2). With no added fatty acyl-CoA as primers, Cz, was the longest fatty acid detected when ATP was provided to activate the endogenous fatty acid. When either ATP or Mgz+ was omitted, only Cl6 fatty acid was generated (Fig. 2), probably representing de nova synthesis catalyzed by contaminating fatty acid synthase. Such a conclusion is supported by the observation that substitution of NADPH by NADH (22) abolished the formation of Cl6 as indicated below. Radio-gas-liquid chromatography of the fatty alcohols as acetates showed that the chain length distribution of label in this fraction was similar to that in the fatty acids (data not shown). Cofactor and primer requirements. NADPH and NADH gave similar rates of elongation, and addition of an NADPHregenerating system was not stimulatory (Table I). ATP and Mgz+, but not CoA, were required to give maximal levels of incorporation of malonyl-CoA into lipids. The lack of a requirement for added CoA has been noted before in both plant and animal systems (23-25) and is probably due to free CoA generated from malonylCoA. In the absence of exogenous ATP

180

ANDERSON

AND

KOLA’JTUKUDY

FIG. 1. Ouchterlony double-diffusion comparison of the fatty acid synthase from bovine mammary and meibomian glands. The center well contained antiserum raised against purified bovine mammary gland fatty acid synthase. The wells marked A contained the purified mammary gland enzyme (1.2 mg/ml), while the other wells contained partially purified meibomian gland enzyme (6.2 mg/ml).

addition of octadecanoyl-CoA stimulated incorporation of malonyl-CoA, suggesting that this exogenous primer could substitute for endogenous primers. When the exogenous octadecanoyl-CoA concentration was increased, the rate of malonylCoA incorporation increased until the primer concentration was 30 PM. Optimum concentrations of the primer gave about threefold stimulation and much higher concentrations were inhibitory. For example, with 100 PM Cl*-CoA malonyl-CoA incorporation was only 30% of that observed with 30 PM CIB-CoA. The maximal rates obtained with optimal concentrations of octadecanoyl-CoA were only 60% of that obtained with ATP/Mga+. These lower rates obtained with exogenous

primer(s) may be due to the detergent effect of the added acyl-CoA. Elongation of exogenous primer. Of the total radioactivity recovered in the lipids generated from [2-‘*C]malonyl-CoA in the presence of CIB-CoA primer, 20-25% was in free fatty acids whereas the major part remained in a polar fraction with -15% in a nonpolar fraction representing wax esters and sterol esters. However, when the lipids were isolated after a mild treatment (15 min) with 0.1 N NaOH, the bulk (50%) of the label was in the free fatty acids with 35% still remaining in a polar fraction, possibly representing phospholipids. Isolation of the lipids after saponification, followed by treatment of the lipids with 14% BF3 in methanol and thin-

BIOSYNTHESIS

I

6

TIME

OF

MEIBOMIAN

12

(min.)

FIG. 2. Radio-gas-liquid chromatogram of the fatty acids (as methyl esters) produced by bovine meibomian gland microsomes from [2-‘%]malonyl-CoA and endogenous primer. The upper tracing shows the products from the complete reaction mixture as described in Table I and the lower tracing shows the product formed when ATP was omitted. Numbers above the peaks indicate chain length.

GLAND

erated from these substrates showed labeled CaO, Cz, and CB acids. Since elongation was the major process responsible for incorporation of label into lipids from [2-‘4C]malonyl-CoA in the presence of C&CoA, this system was used to further characterize the elongation process. In the absence of any exogenous reduced pyridine nucleotide the rate of elongation was ~30% of that observed with added NADPH (Table II). NADH partially substituted for NADPH but the the relative effectiveness of NADH depended on the concentration. At higher concentrations NADH became nearly as effective as NADPH, but when NADH was included with NADPH in the incubation mixture no increase in rate was obtained over that with NADPH alone. Effect of pH, itition time, and protein concentration, The pH dependence of elongation of exogenous Cia-CoA showed that the maximal rate was obtained at around pH 7.0 (Fig. 4). The rate of elongation TABLE COFACTOR

layer chromatography, showed that 88% of the radioactivity was in methyl esters with 7% in fatty alcohols. These results suggest that some of the products of the meibomian microsomal elongation reaction may have been CoA esters, as suggested previously with other mammalian systems (13, 14), although partial hydrolysis of phospholipids under these conditions cannot be ruled out. Radio-gas-liquid chromatographic analysis of the fatty acids showed that almost all of the labeled products were longer than Cl8 and labeled fatty acids up to Czs were found (Fig. 3). A small amount of label was found in Cl6 acid which probably represents de nova synthesis. In any case, over 85% of the total incorporation of malonyl-CoA represented elongation of the Cl8 primer. To confirm that the exogenous primer was incorporated into elongated products [l-‘4CJoctadecanoyl-CoA and unlabeled malonyl-CoA were used as substrates and radio-gas-liquid chromatographic analysis of the products gen-

181

LIPIDS

REQUIREMENTS

ELONGATION

I FOR THE MICROSOMAL

OF ENDOGENOUS

PRIMER’ [Z-“C]Msl~hyl-COA

incorporated Cofactor

deleted

(pm01

NOlE None, + CoA (5 PM) M&l, ATP ATP + 10 PM octadecanoyl-CoA NADPH NADH Glucose B-phosphate and glucose-6-phosphate dehydrogenase NADPH, glucose 6-phosphate, and glucose-g-phosphate dehydrogenase NADPH, NADH, glucose 6phosphate, and glucose-6phosphate dehydrogenase “Complete

0.1 M sodium

mixture

mg-‘)

437 405 169 113 230 345 465

514

490

161

(total volume, 0.5 ml) conpH 7.2, 1 rn~ DTE, 1 rn~ NADH, 1 rn~ NADPH, 0.25 M sucrose, 80 CM [S-“C]malonylCoA, 6.6 rn~ ATP, 1 rn~ MgClz, 14 mM glucose g-phosphate, and 1 unit glucose-6-phosphate dehydrogenase. Reactions were initiated by addition of the microsomes containing 0.71 mg protein, and incubation was for 1 h at 31°C.

tained

reaction

min-’

phosphate,

ANDERSON

AND

26

28

IL I

1

I

cl

0

12

TIME

_-

(min.)

FIG. 3. Radio-gas-liquid chromatogram of the fatty acids (as methyl esters) produced by bovine meibomian gland microsomes from [2-“C]malonyl-CoA and octadecanoyl-CoA as primer. Numbers above the peaks indicate chain length.

was nearly linear with time of incubation until about 20 min, after which the rate declined (Fig. 4). Longer incubation time did not affect the de nova synthesis as adversely as it did the elongation process and, therefore, the proportion of label found in the de nova product increased with the incubation G6) time. Therefore, radio-gas-liquid chromatographic analysis was always used to distinguish between de nova synthesis and elongation. With increasing protein concentration, the amount of lipids produced increased but, instead of a linear relationship, a sigmoidal pattern was observed (Fig. 4). Such a relationship has been observed in other enzymes which use acylCoA as a substrate (26,27) and is probably due to the inhibitory interaction between the substrate and the low amount of enzyme. At higher protein concentrations the proportion of de novo product increased. To avoid the complication created by the formation of de nova products, three conditions which greatly reduced the formation of Cl6 acid were used. Thus

KOLATTUKUDY

with low amounts of protein (0.5 mg), a short incubation period (20 min) and a slightly higher pH (7.6), which preferentially inhibited de nouo synthesis (28), essentially only elongation was observed. These conditions were, therefore, used for all subsequent experiments. Eflect of substrate concentratim. Increasing concentrations of malonyl-CoA gave increasing rates of elongation until a typical substrate saturation was approached (Fig. 5). From the linear doublereciprocal plot, K, and V were calculated to be 52 PM and 330 pmol mine1 mg-‘, respectively. The rate of reaction also showed a typical Michaelis-Menten-type substrate saturation pattern for NADPH. From the linear double-reciprocal plot, Km and Vfor NADPH were calculated to be 11 PM and 350 pmol min-’ mg-‘, respectively. These values are within the range observed for microsomal preparations from mammalian liver and brain (23, 29-31), but should be taken with caution since the concentration of acyl-CoA cosubstrate might not be optimal. Since de nova synthesis was not totally blocked, the reaction products from the experiments shown in Fig. 5 were examined by radio-gas-liquid chromatography. No significant deviation from the pattern shown in Fig. 3 was observed.

TABLE

II

NUCLEOTIDE SPECIFICITY FOR MICROSOMAL ELONGATION OF OCTADECANOYL-COA’ [2-W]Malonyl-CoA incorporated (pm01 min-’ mg-‘)

Nucleotide NADPH NADH None

(1 (1

257 181 65

mM)

mM)

NADPH (100 PM) NADH (100 /.tM) NADPH (100 PM) + NADH (100

315 137 PM)

323

‘Assays were carried out as described under Materials and Methods, using 160 PM [2-“ClmalonylCoA.

BIOSYNTHESIS

OF

MEIBOMIAN

GLAND

183

LIPIDS

TIME (min.)

PH

PROTEIN (mg)

FIG. 4. Effect of pH, time, and protein concentration on the elongation of octadecanoyl-CoA catalyzed by bovine meibomian gland microsomes. Conditions were as described under Materials and Methods, except that 0.1 M sodium phosphate, pH 7.0, was used for the time and protein concentration studies. Elongation products (3 C& relative to total products (determined by radiogas-liquid chromatography) are indicated explicitly (middle and right) or by shading of the appropriate symbols (left). Prouct distribution was not determined for the experiments using Tris buffer.

could use CoA esters of different fatty acids. When exogenous primers of different chain lengths were used, it was found that Cl6 and longer CoA esters could effectively stimulate incorporation of [2-14C]malonyl-CoA into chain-elongated products (Table III). To determine whether the exogenous primers did, in fact, get elongated, the products were subjected to radio-gas-liquid chromatographic analysis (Table III).

Effect of the chain length of the p*mer cm elongation. Radio-gas-liquid chromatographic analysis of the products showed that the proportion of the longer chain elongation products increased with time of incubation when ClgCoA was used as the primer (data not shown). This observation, together with the finding by others (13) that the products of mammalian microsomal chain elongation are CoA esters, suggested that the elongation process

/I

Km=:,,, 0.02 1 /El

&J 0

I

o.,

1

2.Or

K,:ll,M

,

0.05

0.10

0.04

l/[Sl

a

100 f2-‘*Cl

,

200 Malonyl-CoA

(PM)

0

200 NADPH

400 (/LM)

FIG. 5. Effect of [2-‘%]malonyl-CoA and NADPH concentration on the rate of elongation of octadecanoyl-Cob by bovine meibomian gland microsomes. Conditions were as described under Materials and Methods, except that the concentration of [2-“C]malonyl-CoA was 160 pM in the experiment shown on the right.

184

ANDERSON

AND TABLE

DISTRIBUTION

KOLATTUKUDY III

OF RADIOACTIVITY IN MICROSOMAL ELONGATION VARIOUS Acn-CoA PRIMERS’

PRODUCTS

USING

[2-‘“C]Malonyi-CoA incorporated (pm01 min-’ mg-‘) Radioactivity Primer

Total products

None C,,-CoA

151 191

-

4.7 -

85.2 100

C,,-CoA C,,-CoA &,-CoA &CoA

285 269 372 271

285 249 346 236

-

-

a Assays undetectable

were carried levels.

Elongation products

out as described

Cl4

under

Cl6

6.5 5.7 12.8 Materials

The fatty acid synthase contamination was probably responsible for the formation of the small amount of Cl6 observed with all exogenous primers. Exogenous Ci4-CoA caused only a small increase in malonyl-CoA incorporation into lipids and all of the label incorporated into lipids was found in Ci6, with no evidence for true chain elongation products. This incorporation could represent normal de nauo synthesis with Cil-CoA as the primer (32) or a single elongation of the primer by microsomal enzymes. With C&-CoA, the major product was Cl8 and only small amounts of longer chains up to Czs were found (Table III). The unique dominance of the product of a single elongation with Cl6 as the primer suggests that there might be an elongating system specific for elongation of Cl6 to Cl8 in addition to the system which elongated Cl8 to longer acids. Such a possibility has been suggested by results obtained with microsomes from other mammalian tissues (33, 34). With Cl*-CoA as primer, CZo to C% fatty acids were generated and, similarly, Czo- and C&-CoA esters served as primers for the production of longer acids. Similar results were obtained by other workers using microsomes from leek epidermis (24). Although longer primers were not tested, it is clear that the present elongation system

Cl8

in products

(W total)

Cal

CL?2

c24

c26

czs

10.1 -

-

-

-

-

-

77.4 1.0 1.3 -

6.0 17.7 0.7 -

4.4 19.5 39.8 -

9.1 36.5 41.4 68.3

3.1 15.5 11.2 18.9

3.2 -

and Methods,

using

30

pM

acyl-CoA,

dash indicates

is capable of synthesizing the long-chain products of the size found in the bovine meibomian glands (8, 9). Sebaceous glands are known to have a large amount of endomembranes, and in developing sebaceous glands proliferation of such membranes precedes production of sebum (35-38). Although the biochemical functions of such membranes have not been fully elucidated, enzymes which catalyze several key steps involved in the synthesis of sebaceous gland lipids were found to be in such membranes. Thus the acyl-CoA reductase and acyl-CoA:fatty alcohol transacylase involved in wax ester synthesis were found in a particulate fraction obtained from the uropygial gland of white-crowned sparrow (26, 27) and domestic goose (J. S. Buckner and P. E. Kolattukudy, unpublished results). a-Hydroxyacyl-CoA reductase, acyloin reductase, and alkane diol:acyl-CoA transacylase involved in the biosynthesis of diol diesters in uropygial glands were also found in the microsomal fraction (39, 40). The present results show that chain elongation is catalyzed by the microsomal fractions from the meibomian gland. Thus the general picture which emerges concerning the biosynthesis of the sebaceous gland lipids is that the usual cytoplasmic enzymes, acetyl-CoA carboxylase and fatty

BIOSYNTHESIS

OF

MEIBOMIAN

acid synthase, generate fatty acids which are subsequently modified by reactions such as chain elongation, hydroxylation, reduction, and esterification catalyzed by enzymes located in the endomembranes to generate the unusual wax esters and diol diesters secreted by the glands. ACKNOWLEDGMENTS We

thank

and

Linda

Dan

providing

Rogers

Schneider the

of

bovine

for

technical

Colfax

Meat

GLAND

in

Enzymology

‘71, pp.

20.

Biochem 1,29-40. SMITH, S. (1973) Arch.

21.

W.,

Academic

SEUBERT,

22.

163. MAITRA, Chem

23.

CAREY,

24.

eyelids.

PODACK,

S. K., AND 249,111-117. E. M.,

ed.),

New

York.

E. R. (1973)

MoL

Biochem

Cell.

S.

ROGERS, 186, 152-

(1974)

L.

156,

BioL

J.

B&hem

(1975)

380, 176-189.

V. P., LESSIRE,

Arch.

NUGTEREN,

P. E., AND

PARKIN,

Vol.

Biophys.

Biophys.

KUMAR,

AND

Ada

AGRAWAL,

(1984) 25.

AND

J. M., Press,

751-758. BUCKNER, J. S., KOLATTUKUDY, L. (1978) Arch Biochem

Biophys. for

(Lowenstein,

86-97,

19.

assistance, Packing

185

LIPIDS

R., AND

Biochem.

D. H.

STUMPF,

Biophys. B&him.

(1965)

P. K.

230,580-589. Bio-phys. Actu

106,280-290. REFERENCES

26.

KOLATTUKUDY,

Arch. 1. KOLA~IJKUDY, Biochemistry

P. E. (1976) in of Natural Waxes

P. E., ed.), pp. 1-15, 2. DOWNING, D. T. (1976) chemistry of Natural

Eye 5. WOLFF,

and

Elsevier, Science

27.

Bio-

(1961)

Sot.

8. BARON,

C.,

AND

9. NICOLAIDES,

29.

AND

Exp. UK 66,

Exp. Eye Rs. 10,223-227. Exp. Eye Res. 27,289-300.

BLOUGH,

30.

11.

RUTKOSKI,

B&hem.

H.

A.

(1976)

J. Lipid

13.

BERNERT,

14.

BROPHY,

15.

J. 152,495-501. KOLA’ITUKUDY,

Arch,

17. 18.

J.

BLOCH,

K.

A., AND JAWORSKI, 91,370-373.

LOWRY, 0. H., AND RANDALL, 265-275.

Biophys.

KAITARANTA,

TARIK,

J. T., AND

Acta

32.

.I 152,495-501. KNUDSEN, J., AND GOLDBERG,

34.

(1973) 498. POLLET,

Anal.

N. J., FARR, A. L., J. BioL Chem. 193,

AND 341.

Actu

573,436-442.

P. J., AND

VANCE,

P.

E.,

ROGERS,

L.

Biophys.

191, 244-258.

BEDFORD, C. J., KOLATXJKUDY, L. (1978) Arch B&hem 151.

P. E., AND

OUCHTERLONY, 0. (1953) Scam-L 32,231-240. KUMAR, S., AND DODDS,

Biophys. Path&

38.

P. F. (1981)

MicrcbioL in

Methods

D. E. (1975)

Biochem.

I. (1980)

Biochem

GRUNNET,

95, 1808-1814. I., AND BLOCH,

(Washington,

S., BOURRE,

J.-M.,

D. C.) 182,

CHAIX,

N. (1973)

BAUMANN,

Annu.

R.,

P. (1963)

Rev.

K. 497-

G., DAUDU,

Biochimie

O.,

55, 333-

PhysioL

eds.),

40.

4, pp.

(Kolattukudy, Amsterdam.

WAGNER,

R.

MwphoL 39.

Vol.

167-187,

Oxford. BUCKNER, J. S., AND KOLAITUKUDY, in Chemistry and Biochemistry

SAWAYA,

Arch. Acta

32,

Biochim

H. (1979)

VANCE,

SCHECHTER,

Science

Waxes Elsevier,

(1978)

ROGERS, 186, 139-

N. A.,

25, 15-

NICOLAIDES, N. (1963) in Advances in Biology of Skin (Montagna, W., Ellis, R. A., and Silver

37.

AND

BAUMANN, J. Neurochem

36.

B&hem.

D. E. (1975)

A.,

SIEKEVITZ, 40.

B&him

H. (1979)

L.

254,8153-

35.

A. SPRECHER,

Chem

J.-M., (1979)

Res. Commun I.,

DAVIDSON,

.I BioL

SPRECHER,

P. J., AND

33.

L.,

(LowAca-

574.18-24.

BROPHY,

J. BioL

(1977).

J. G. (1978)

ROSEBROUGH, R. J. (1951)

J. T., AND

Biochem

K.,

H.

31.

J. I., BOSWELL, F. M., AND SMITH, Invest. OpthalmoL Vti. Sci 20,

D. O., AND 252,5735-5739.

12.

16.

85-90. BERNERT,

in Enzymology 71, pp. 263-275,

New York. GALEOS, W.

SPRECHER,

Biophys.

N.,

T. N., MACY, R. E. (1981) 522-536. PETERSON, Chem.

P. E.,

AND KUMAR, S. (1979) 8158. BERNERT, J. T., BOURRE,

Res. 17, 373-376.

10.

KOLATXJKUDY,

Biqphys. J. S. (1970) J. M. (1978)

ROGERS, L. (1978) 191,244-258. ROGERS, L., AND LARSON,

demic Press, STROM, K. A.,

291. 6. ANDREWS, 7. TIFFANY,

AND

28.

Amsterdam. (Washington,

S., AND MAURICE, D. M. Res. 1, 39-45. E. (1946) Trans. OpthdmoL

E.,

Biophys.

J. D. (1981) in Methods enstein, J. M., ed.), Vol.

Elsevier, Amsterdam. in Chemistry and Waxes (Kolattukudy,

P. E., ed.), pp. 17-48, 3. NICOLAIDES, N. (1974) D. C.) 18. 19-26. 4. MISHIMA,

Chemistry (Kolattukudy,

P.

Biodem.

C.,

AND

Pergamon, P. E. (1976) of Natural

P. E.,

ed.),

BOORD,

R.

pp. L.

147-200, (1975)

146.395-413. W.

N., AND

Biochem.

KOLA~UKUDY, L. (1978) 440.

KOLATTUKUDY, P. E. (1973) 157, 309-319.

Biophys.

P. E., RILEY,

Arch

B&hem.

R. G., AND

Biophys.

ROGERS, 189, 433-

J.