Synthesis of phosphatidylcholine by two distinct methyltransferases in rat colonic brush-border membranes: evidence for extrinsic and intrinsic membrane activities

Synthesis of phosphatidylcholine by two distinct methyltransferases in rat colonic brush-border membranes: evidence for extrinsic and intrinsic membrane activities

Biochimica et Biophysics Elsevier Acta 875 (1986) 493-500 493 BBA 52109 Synthesis of phosphatidylcholine by two distinct methyltransferases in ra...

799KB Sizes 0 Downloads 3 Views

Biochimica et Biophysics Elsevier

Acta 875 (1986) 493-500

493

BBA 52109

Synthesis of phosphatidylcholine by two distinct methyltransferases

in rat colonic

brush-border membranes: evidence for extrinsic and intrinsic membrane activities Pradeep Department

K. Dudeja,

Emily S. Foster and Thomas

A. Brasitus

*

of Medicine, Michael Reese Hospital and Medical Center, Pritrker School of Medicine of the lJnioersit_y of Chicago, 4 K&K, 31st Street & Lake Shore Drive, Chicago, IL 60616 (U.S.A.) (Received June 26th. 1985) (Revised manuscript received October 16th. 1985)

Key words:

Phosphatidylcholine

synthesis;

Methyltransferase;

(Rat colon membrane)

The enzymatic synthesis of phosphatidylcholine from phosphatidylethanolamine via a transmethylation pathway has not been shown to occur in the small intestine and has been assumed to be absent from the entire gut. The existence of this pathway, however, has not been investigated in the large intestine. Utilizing a recently developed method for the isolation of brush-border membranes from rat colonocytes, the present studies were designed to determine whether phospholipid methylation activity was present in the large intestine. The results demonstrate that this pathway for synthesis of phosphatidylcholine exists in rat colonic plasma membranes and involves at least two distinct methyltransferases. The predominant product of the first enzyme (methyltransferase I) is phosphatidyl-N-monomethylethanolamine; phosphatidylcholine and phosphatidyl-N-monomethylethanohunine are the principal products of the second enzyme (methyltransferase II). Methyltransferase I has an apparent K, for S-adenosyk-methionine of 100.0 PM and a pH optimum of 8.0, while methyltransferase II has an apparent K, of 0.3 PM and a pH optimum of 6.0. Additional evidence to support the presence of two distinct enzymes includes the differential effects of ATP, Triton X-100, trypsin treatment, and temperature on their activities.

Introduction Phosphatidylcholine is a major phospholipid of many plasma membranes [l]. The enzymatic synthesis of PC from PE via a methylation pathway has been shown to occur in a number of cell types [2-41. The enzyme(s) responsible for this conversion of PE to PC utilize S-adenosyl-Lmethionine as the methyl donor and appear to be membrane bound [l]. While initially reported in

* To whom correspondence should be addressed. Abbreviations: PME, phosphatidyl-N-monomethylethanolamine;PDE, phosphatidyl-N,N-dimethylethanolamine; phosphatidylcholine; PE, phosphatidylethanolamine.

0005-2760/86/$03.50

0 1986 Elsevier Science Publishers

PC,

microsomal membranes [2,5,6], it has become evident that the transmethylation pathway is also present in several plasma membranes [4,7-lo] and may play a role in a number of important membrane processes [4,5,11]. The importance of phospholipid methylation in these processes, however, has recently been questioned [9,12,13]. The number of methyltransferases involved in the conversion of PE to PC is also controversial. In 1961 Bremer and Greenberg [2] suggested that three different enzymes were present in rat liver microsomes. Based on kinetic studies, Mg2+ requirements and pH optima for the various substrates, several laboratories subsequently reported the existence of at least two methyltransferases in

B.V. (Biomedical

Division)

494

various mammalian tissues [5,7,8,11,29]. One enzyme appeared to methylate PE to PME and the other(s) to catalyze the transmethylation of PME to PC via the intermediate PDE. Other investigators [6,12,14,15], however, have suggested that a single methyltransferase in rat hepatic microsomes may catalyze the conversion of PE to PC. The transmethylation pathway for the synthesis of PC has been assumed to be absent in the gut. Bremer and Greenberg [2] were unable to identify phospholipid methylation activity in the small intestine of the rat. To date, this pathway for PC synthesis has not been looked for in the large intestine. Utilizing a recently developed method for the isolation of brush-border membranes from rat colonocytes [ 161, our laboratory has established that phosphatidylcholine is the major phospholipid of these membranes [16]. The present studies were, therefore, designed to examine whether the transmethylation pathway existed in these membranes and to determine the number of enzymes involved in the conversion of PE to PC. The results demonstrate that this pathway for the synthesis of PC is present in rat colonic brushborder membranes and involves at least two distinct methyltransferases. Materials and Methods Preparation

of colonic brush-border

membranes.

Male albino rats of the Sherman strain weighing 250-300 g were fasted for 18 h with water ad libitum before death. The colons were excised and epithelial cells, relatively devoid of goblet cells, were obtained using a technique which combined chelation of divalent cations with mild mechanical dissociation [16]. Brush-border membranes were then prepared as described by Brasitus and Keresztes [16]. The purity of membrane suspensions and the degree of contamination with intracellular organelles were assessed by marker enzymes. The specific activity ratio ((purified brush-border membrane)/(crude homogenate)) for the brush-border enzyme markers, total alkaline phosphatase (p-nitrophenylphosphatase) and cysteine-sensitive alkaline phosphatase, were approx. 15-20 in all membrane preparations. The corresponding values for succinic dehydrogenase, NADPH-cytochrome-c re-

ductase and sodium-potassium-dependent adenosine triphosphatase, marker enzymes for mitochondrial, microsomal and basolateral membranes, respectively, ranged from 0.50 to 1.50 in all membrane preparations. Brush-border membranes were suspended in 2 mM Tris/50 mM mannitol (pH 7.4) and used immediately. Subcellular fractionation of colonocyte. Isolated colonocytes were homogenized at 4°C in 2 mM Tris/SO mM mannitol (pH 7.4) using a Polytron PCU-2-110 (Brinkman Instruments, Westbury, NY) at a power setting of 4 for 4 x 15 s and then centrifuged at 1500 x g 10 min. The nuclear enriched pellet was recovered and the supernatant was centrifuged at 15000 x g for 20 min. The pelleted fraction, enriched in mitochondria, was kept for analysis, the supernatant was centrifuged at 105 000 X g for 60 min and the microsomal pellet was saved. The nuclear, mitochondrial and microsomal pellets were then suspended in appropriate buffers and assayed for phospholipid methyltransferase activity (see below). Assay of phospholipid methylation. The methylation of phospholipids was measured by incorporation of [ ‘HImethyl groups from S-adenosyl-L[melhyL3H]methionine into phospholipids [5]. The reaction mixture (500 ~1) contained appropriate buffer (see below), S-adenosyl-L-[ methyl- ‘H] methionine [200 PM, 4 r.lCi] and plasma membranes (200 pg protein). The assay was initiated by the addition of membranes, and was incubated at 37°C for 60 min, unless otherwise indicated. The reaction was terminated by the addition of 3 ml chloroform/methanol/2 M HCl (6 : 3 : 1, v/v), followed by the addition of 2 ml 0.1 M KC1 in 50% methanol. This mixture was vigorously vortexed twice and centrifuged at 200 X g for 10 min. The aqueous phase was aspirated and the chloroform phase rewashed with 2 ml of 0.1 M KC1 in 50% methanol. To identify the products of phospholipid methylation, the chloroform phase was then evaporated to dryness under nitrogen and the residue dissolved in 100 ~1 chloroform. The sample was applied on a Silica Gel G plate and the chromatogram was developed in chloroform/proprionic acid/n-propyl alcohol/water (2 : 2 : 3 : 1, v/v). The phospholipid standards were simultaneously chromatographed, and their positions were visualized using a saturated solution of

495

iodine in chloroform. The areas corresponding to the standard phospholipids (PE, PME, PDE and PC) were each scraped individually, extracted with chloroform/methanol (2 : 1, v/v), and their radioactivity was measured separately in a liquid scintillation counter as reported [5]. The chemical identity of the methylated products was further established by two-dimensional chromatography [17] and by hydrolysis of the phospholipids and identification of their free bases as described by Schneider and Vance [6]. Additionally, both labeled and unlabeled S-adenosylL-methionine were routinely purified by ion exchange before use [12]. The effect of trypsin and Triton X-100 treatment on methyltransferase activity. Membranes (200 pg protein) in 50 mM mannitol/2 mM Tris-HCl (pH 7.0) were incubated with trypsin (1 pg/50 pg membrane protein) for 30 min at 37°C in the presence or absence of 0.05% Triton X-100. At the end of the incubation period, trypsin inhibitor (2 pLg/pg trypsin) was added. In control experiments, the inhibitor was added with trypsin at the beginning of the incubation. After trypsin treatment the phospholipid methyltransferase activity was assayed as described above. Phospholipid methylation of the colonic membranes was also assayed in the presence of varying concentrations of Triton X-100 (0.05%1.0%) alone. Temperature dependence of methyltransferase activities. Arrhenius plots were determined for methyltransferase activities over the temperature range of 13-40°C as previously reported [18]. Maximum velocity conditions were used and the pH was kept constant at each temperature measured. Statistical method. Values are expressed as mean +S.E. Paired or unpaired t-tests were used for all statistical analysis. A P value of less than 0.05 was considered significant. Materials. S-Adenosyl-L-[methyl’H]methionine (5-15 Ci/ mmol) was purchased from New England Nuclear (Boston, MA). PE, PME, PDE and PC were obtained from Calbiochem-Behring (San Diego, CA). Non-radioactive S-adenosyl-rmethionine was a product of Boehringer-Mannheim (Indianapolis, IN). S-Adenosyl-L-homocysteine and other chemicals were obtained from Sigma Chemical Co. (St. Louis, MO) and/or Fisher Chemical Co. (Fairlawn, NJ), unless otherwise indicated.

Results Properties of phospholipid methyltransferase(s) To determine the optimal pH for phospholipid methylation, rat colonic brush-border membranes were incubated with 200 FM S-adenosyl-t[ methyl-‘Hlmethionine in the presence of various buffers including: 50 mM sodium acetate (pH 5.0 and 6.0) 50 mM/Tris acetate (pH 7.0, 8.0 and 9.0) and 50 mM sodium borate (pH 10.0). The amount of [ 3H]methyl radioactivity incorporated into phospholipids was then determined as described in Materials and Methods. The data presented in Table I demonstrates that phospholipid methylation using 200 PM S-adenosyl-Lmethionine was highest at pH 8.0. The incorporation of [‘HImethyl groups into phospholipids was linear up to 500 pg of membrane protein. The [ 3HImethyl phosphohpids were not formed at 0°C or following incubation of S-adenosyl-r_-[ methyl-’ Hlmethionine with heat-denatured membranes (data not shown). The possibility that the radioactivity came from [‘HI methanol released after an enzymatic transmethylation [19,20] was also considered. The addition of 0.1% methanol to the reaction mixture, however, had no effect. Furthermore, the radioactivity extracted into the chloroform phase was not volatile after heating at 80-85°C to dryness. The effects of various S-adenosyl-L-methionine concentrations on phospholipid methylation Brush-border membranes were incubated with various concentrations of S-adenosyl-L-methionine (between 0.1 and 400 PM) and the extent of formation of [ 3HImethyl phospholipids measured at pH 8.0. When the total radioactivity in the methylated phospholipids was plotted as a function of S-adenosyl-L-methionine concentrations (0.1 to 12 PM) (Fig. l), the resulting curve showed two phases with a relatively sharp inflection point at 6 PM. A double reciprocal plot [21] of the initial linear velocity for the formation of [ 3H]methylated phospholipids and low S-adenosyl-L-methionine concentrations (0.1 to 6 PM) yielded a straight line with an apparent K, of 0.31 PM and a V,,, of 51.3 pmol/mg protein per h (not shown). When [ 3H]methyl groups incorporated into phospholipids were plotted as a function of higher S-adeno-

496

J 200

400

600

Ad0Met~Mr

800

1000

d]

Fig. 1. Enzymatic methylation of colonic brush-border membrane phospholipids as a function of the concentration of S-adenosyl-t-[ merhy/-3H]methionine. Colonic brush-border membranes (200 pg protein) were incubated at 37’C in the presence of 0.1-12.0 PM S-adenosyl-L-methionine concentrations (pH 8.0). Reaction time was 60 min during which the incorporation of [ ‘HImethyl groups into phospholipids was linear. Values are expressed as pmol [ -‘HImethyl groups incorporated/mg protein and represent the mean of three separate determinations.

syl-L-methionine concentrations (40-400 PM), a rectangular hyperbola was obtained with no break in the curve. A double reciprocal plot of the linear velocity for the formation of [3H]methyl phospholipids and high S-adenosyl-L-methionine con-

TABLE

I

EFFECT OF pH ON PHOSPHOLIPID METHYLATION RAT COLONIC BRUSH-BORDER MEMBRANES Values represent

mean+

S-Adenosyl-tmethionine

PH

200 pM

5.0 6.0 7.0 8.0 9.0 10.0

1pM

SE. of three separate

determinations.

[ 3HIMethyl groups incorporated into phospholipid (pmol/mg protein

5.0 6.0 7.0 8.0 9.0 10.0

135*11 180*20 167k14 270+18 134&12 35+ 8 21+ 39* 27+ 31* 25* 3_+

4 4 2 3 3 1

per h)

IN

centrations demonstrated a straight line with an apparent K, of 100 FM and a V,,, of 307.6 pmol/mg protein per h (not shown). Similar results were obtained when [3H]methyl groups incorporated into PME, PDE or PC were plotted as a function of S-adenosyl-L-methionine concentrations (not shown). These observations suggested that at least two enzymes were involved in the methylation of phospholipids. When the pH-optima studies were repeated using 1 PM instead of 200 PM S-adenosyl-Lmethionine, as shown in Table I, the highest phospholipid methylation activity occurred at pH 6.0 not at pH 8.0, again suggesting the existence of at least two distinct methyltransferases. All subsequent experiments were, therefore, performed at pH 6.0 using 1 PM S-adenosyl-L-methionine and pH 8.0 using 200 PM S-adenosyl-L-methionine. Audubert and Vance [12], have recently suggested that prior studies erroneously assumed that radioactivity from S-adenosyl-L-[ methyl-’ H] methionine found in PME after incubation was indicative of all the PME formed, and ignored the subsequent conversion of PME to PDE and PC. They have suggested formulas which may more accurately estimate the apparent K, values for S-adenosyl+methionine in the conversion of PE to PC [12]. Using these equations, the pH optimum for the membranes incubated with low Sadenosyl-L-methionine concentrations was 6.0 and the apparent K, for S-adenosyl-L-methionine for the conversion of PE to PME was 0.36 PM; for the conversion of PME to PDE, it was 0.34 PM; and for PDE to PC, it was 0.32 PM. The pH optimum for the membranes incubated with higher concentrations of S-adenosyl-L-methionine was 8.0 and the apparent K, for S-adenosyl+methionine for the conversion of PE to PME was 110 PM; for the conversion of PME to PDE, it was 91 PM; and for PDE to PC, it was 110 PM. Separation and identification of methylated phospholipids Identification of [ 3H]methylated phospholipids formed after colonic membranes were incubated with S-adenosyl-L-[ merhyf- 3Hlmethionine was determined by thin-layer chromatography. When colonic membranes were incubated with 1 PM S-adenosyl+methionine (pH 6.0), three major

497

peaks were found with R, values corresponding to PME, PDE and PC. Values for these three products formed were 18.4 + 2.6, 5.8 f 1.1 and 16.5 + 2.8 pmol/mg protein per h, (n = 8) respectively. Incubation of the membranes with 200 PM S-adenosyl-L-methionine (pH 8.0) also yielded the same three products. Values for PME, PDE and PC were 115.2+ 6.1, 57.6 f 3.8 and 52.4+ 7.1 pmol/mg protein per h, (n = 16), respectively. Therefore, PME and PC were the major products found at 1 PM S-adenosyl-L-methionine (pH 6.0) while PME was the predominant product formed at 200 PM S-adenosyl-r_-methionine (pH 8.0). Enzymatic methylation of nonpolar lipids com. igratmg with the solvent front was also observed in these experiments. The ratio of [ ‘HImethyl groups incorporated in nonpolar lipids to phospholipids was 4.1 f 0.7 at 1 PM S-adenosyl-L-methionine (pH 6.0) and 1.5 k 0.1 at 200 PM S-adenosyl-Lmethionine (pH 8.0). Preliminary studies in our laboratory suggest that these nonpolar lipids are not methylated fatty acids [22] but methylated neutral glycosphingolipids (unpublished observations). The effect of addition of exogenous substrates on phospholipid methylation To examine the effects of exogenous phospholipids on phospholipid methylation, sonicated suspensions of 500 pg of PE, PME and DPE were added to colonic membranes incubated with 1 PM S-adenosyl-L-methione (pH 6.0) or 200 PM Sadenosyl-L-methionine (pH 8.0). In agreement with earlier studies performed in rat brain synaptosomes [8] and rat liver microsomes [ll], the addition of PE did not change the methylation activity of rat colonic brush-border membranes using 1 PM or 200 PM S-adenosyl-Lmethionine (not shown). The addition of PME and PDE, however, also failed to stimulate phospholipid methylation in rat colonic membranes at either concentration of S-adenosyl-L-methionine (not shown). While this is similar to the findings of Sastry et al. [ll] in rat liver microsomes, a number of investigators [8,9,23] have shown that these two exogenous substrates stimulate phospholipid methylation activity in other tissues. The reason for this lack of stimulation of phospholipid methylation by these exogenous substrates in rat colonic brush-

border require

membranes is presently further examination.

unclear

and

will

Time-course of phospholipid methylation The time-course of the [’ HImethyl incorporation into phospholipids was studied at different S-adenosyl-L-methionine concentrations and pH values. At 1 PM S-adenosyl-L-methionine (pH 6.0), [ 3HImethyl incorporation into total phospholipids was linear for 60 min, into PME for 120 min, into PC for 45 min, and into PDE for 45 min. At 200 PM S-adenosyl-L-methionine (pH 8.0) incorporation of [ 3HImethyl groups into total phospholipids as well as PME, PC and PDE was linear for 120 min. The effect of ATP and divalent cations on phospholipid methylation Since prior studies [24] have suggested that calcium (300 PM) and ATP (10 PM) stimulated phospholipid methylation in rat liver microsomes, the effects of these substances on phospholipid methylation in rat colonocytes were evaluated. ATP (lo-100 PM) alone or in the presence of calcium (0.2-1.0 mM) did not alter the activity of phospholipid methylation at 1 PM or 200 PM Sadenosyl-L-methionine concentrations (data not shown). Higher concentrations of ATP (5 mM and 10 mM) alone also failed to inhibit phospholipid methylation at 1 PM S-adenosyl-t-methionine. However, 5 mM and 10 mM ATP significantly inhibited phospholipid methylation activity by 45 + 5% (P -C0.05) and 75 + 3% (P < 0.01) using 200 yM S-adenosyl-L-methionine (n = 3). Addition of varying concentrations of Mg*+ or Ca’+ (l-10 mM) did not stimulate phospholipid methylation at 1 PM or 200 PM S-adenosyl-L-methionine concentrations (not shown). Inhibition of phospholipid methyltransferases activity by S-adenosyi-L-homocysteine At an S-adenosyl-r_-methionine concentration of 1 PM (pH 6.0), 2 mM S-adenosyl-L-homocysteine equally inhibited PME, PDE and PC synthesis by 98.5 + 1.0% (n = 3). The same concentration of S-adenosyl+homocysteine inhibited the synthesis of each of these reaction products by 45.0 + 6.1% at 200 PM S-adenosyl-L-methionine (pH 8.0) (n = 3).

498

The effect of Triton X-100 and trypsin treatment on methylation of phospholipids The incorporation of [‘HImethyl groups into phospholipids was assayed in the presence of varying concentrations (0.05-1.0%) of Triton X-100. At all concentrations tested, this detergent significantly inhibited phospholipid methylation at 1 PM S-adenosyl-L-methionine (Fig. 2A). Incubation of Triton X-100 with 200 PM S-adenosyl-Lmethionine however, showed a biphasic response (Fig. 2B). At 0.05% methylation was increased by approx. 20%, whereas at higher concentrations (O.l-1.0%) methylation was decreased by about 20%. Incubation of membranes with trypsin (1 pg/50 pg membrane protein) at 1 PM S-adenosyl-Lmethionine produced a 60 + 3% reduction of

[ ‘HImethyl group incorporation into total phospholipids (n = 3). PME, PDE and PC were equally affected. At 200 PM S-adenosyl-L-methionine, however, trypsin treatment resulted in only a 17 _t 2% reduction in these activities (n = 3). Trypsin plus 0.05% Triton X-100 yielded similar results to those with trypsin alone at both S-adenosyl-rmethionine concentrations (not shown). Temperature dependence of phospholipid methylation The temperature dependence of phospholipid methylation at 1 PM and 200 PM S-adenosyl-r_methionine concentrations over the temperature range of 13-40°C was examined. Representative Arrhenius plots of these activities are shown in Fig. 3. A linear plot with a single slope was obtained for the methylation activities at 1 PM

A.

40

SO

I'; a

./-.‘\

20

2 g

E 3 st

*

0.0

+

o.oa

0.10

1.0

Mton X-fOO(%J

c

t8 f

400

1

6.

Trlton

X- 1OOlXJ

Fig. 2. The effect of varying concentrations of Triton X-100 on incorporation into total phospholipids. Phospholipid methylation of the colonic brush-border membranes was assayed in the presence of varying concentrations (0.05-1.0%) of Triton X-100 using 1 /.IM S-adenosyl-L-[merhyl3H]methionine at pH 6.0 (A) or 200 PM S-adenosyl-L-[methyl‘Hlmethionine at pH 8.0 (B) for 60 min. Values are expressed as pmol [ ‘HImethyl groups incorporated/mg protein and represent the mean of three separate determinations of each concentration of detergent tested.

[ ‘HImethyl

l/Kx

103

Fig. 3. Temperature-dependence of phospholipid methyltransferase activities. Representative Arrhenius plots of rat colonic brush-border membrane methyltransferase activities are shown. Mean values of phospholipid methylation assayed with 200 PM S-adenosyl-L-[merhyl-3H]methionine at pH 8.0 are plotted in A and with 1 pM S-adenosyl-L-[merhyl-3H]methionine at pH 6.0 are plotted in B. The number of preparations tested, break-point temperatures and apparent energies of activation for the individual experiments are given in the text. Linear plots were determined by the method of least-squares.

499

TABLE

II

DIFFERENTIAL MEMBRANES

CHARACTERISTICS

OF

METHYLTRANSFERASE

Characteratics

Methyltran\ferasr

Principal product(s) ‘Apparent’ K,,, (PM) ‘Apparent’ r/,,,,, (pmol/mg per h) pH optimum ATP inhihiti~~n (10 mM) Triton X-100 effect (0.05-1.0%) Trypsin treatment

PMF. 100.0 307.6 * 20.3 8.0 75% hiphasic re\ponx 20%’ inhihition

~-adenosyl-L-methionine (pH 6.0). Arrhenius plots of methyltransferase activities at 200 PM Sadenosyl-I.-methionine (pH 8.0), however. showed a distinct breakpoint at 30.1 + l.l”C. The apparent energy of activation, d E, was 11.8 t_ 1.2 kcal/mol (n = 3) for the methylation activity measured using 1 I_LM~-adenosyl-~.-methionil~e. Using 200 PM S-adenosyl-L-methionine, values for A E above and below the breakpoint were 4.7 _t 1.2 and 28.3 & 1.4 kcal/ mol (n = 3), respectively.

Compared to microsomal, mitochondrial, nulcear or original homogenate the highest specific activity of phospholipid methylation was present in brush-border membranes measured at I ,F(.Mor 200 FM ~-adenosyl-L-methi~~nine (40.7 i_ 3.1 and 224.0 + 4.2 pmol/mg protein per h, respectively). The activity of phospholipid methylation in subcellular fractions ranged from 25550% of brushborder membranes (not shown). Discussion The present results demonstrate that at least two methyltransferases with distinct characteristics are present in rat colonic brush-border membranes. As shown in Table II. these enzymes differ with respect to their principal products, kinetic parameters and pH optima as well as the effects of ATP, Triton X-100 and trypsin treatment on their activities. Additionally, methyltransferase I demonstrated a discontinuity in the slope of its Arrhenius plot at approx. 30°C (Fig. 3, plot A),

I

I AND

II OF

RAT

COLONIC

Methyltransferase

BRUSH-BORDER

II

PME and I’( 0.31 51.3+ 1.7 6.0 0% inhibited nt all cnncentration~ 609, inhibition

i.e., in the vicinity of the lipid tl~crmotropic transition temperature of these colonic plasma membranes [16]. In contrast, methyltransferase II activity showed a single slope (Fig. 3, plot B). Since temperature-sensitive changes in enzymatic activity may be independent of the membrane lipids [25], all experiments were performed under maximum velocity conditions and constant pH to preclude artificial breaks in the slope. Prior studies in our laboratory [1X] have also demonstrated the removal of such breakpoints seen in Arrhenius plots of enzymes by delipidation and restoration of the original breakpoint temperature by relipidation. Therefore, these findings suggest that methyltransferase I functionally experienced the effects of the lipid thermotropic transition, whereas methyltransferase II activity did not. Based on operational criteria previously established in our laboratory 1181, the first enzyme would be characterized as an ‘intrinsic’ membrane activity, while the second would be an ‘extrinsic’ membrane activity. It should be emphasized, however, that these criteria focus specifically on whether the rate-limiting step of a given membrane function is subject to modulation by the lipids; there is no necessary implication concerning the location of the proteins [18]. While the trypsin data would suggest that methyltransferase I might be less accessible to degradation than methyltransferase II, other explanations are also possible [26], and the exact location of these two enzymes in the membrane is unclear at the present time. These results are consistent with earlier reports in Neurospara crassu, which provided genetic evidence for two enzymatic activities for the conver-

500

sion of PE to PC 127,181. At least two distinct methyltransferases have also been described in the mammalian membranes of red blood cells, adrenal glands, brain synaptosomes, pituitary and liver [.5,7,8,11,29]. In these studies, one enzyme was shown to methylate PE to PME and the other(s) to catalyze the conversion of PME to PC through the intermediate PDE. In the present experiments, while PME was the predominant product of methyltransferase I, both PME and PC were the major products of methyltransferase II. This pattern of methylated derivatives of PE formed by the latter enzyme is distinctly unusual, and its significance remains unclear at this time. While speculative, methyltransferase II may represent a complex of more than one enzymatic activity but further studies will be necessary to clarify this issue. In summary, these experiments demonstrate for the first time that phospholipid methylation activity is present in rat cofonic brush-border membranes and involves at least two distinct methyltransferases. Although the importance of phospholipid methylation in the regulation of several important membrane processes has been questioned [9,12,13], recent investigations have indicated that the transmethylation pathway for PC may modulate transepithelial sodium transport [10.30]. Preliminary observations in our laboratory have suggested that electroneutral sodium absorption, the predominant sodium absorptive process in the rat colon, may also be influenced by the transmethylation pathway for PC 1311. Further studies of phospholipid methylation activity in these membranes will, therefore, be of interest. Acknowledgements

The authors would like to thank Ms. Dolores Gordon for her secretarial support and Ms. Kimberli Coleman for her excellent technical assistance. This investigation was supported by PHS grant number CA36745 awarded by the National Cancer Institute, DHHS. References 1 Pelech.S.L. and Vance, D.E. (1984) Biochim. Biophys. Acta 179, 217-251 2 Bremer, J. and Greenberg, Acta 46. 205-216

D.M. (1361) Biochim.

Biophys.

3 Hiram. F.. Strittmatter. W.J. and Axelrod. J. (1979) Proc. Natl. Acad. Sci. USA 76. 368-372 4 Hirata, F. and Ax&rod. J. (1978) Nature (Lond) 275, 219-220 5 Hirata. F., Viveros, OH., Diliberto, EM. and Axeirod. J. (1978) Proc. Nat]. Acad. Sci. USA 75. 1718-1721 6 Schneider. W. and Vance, D.E. (1979) J. Biol. Chem. 254. 3886-3891 7 Strittmatter, W.J., Hirata, F. and Axelrod, J. (1979) Biothem. Biophys. Res. Commun. 88, 147-151 8 Crews, F.T., Hirata. F. and Axelrod, J. (1980) J. Neurothem. 34, 1491-1498 9 Chauhan, V.P.S., Sikka, S.C. and Kalra, V.K. (1982) Biochim. Biophys. Acta 688. 357-368 10 Sariban-Sohraby. S.. Burg. M., Wiesmann, W.P.. Chiang. P.K. and Johnson. P.J. (1984) Science 225, 745-746 11 Sastry, B.V.R., Statham, C.N., Axelrod. J. and Hirata. F. (1981) Arch. Biochem. Biophys. 211, 762-769 12 Audubert, F. and Vance, D.E. (1983) J. Biol. Chem. 258, 10695-10701 13 Moore, J.P.. Johannsson. A., Hesketch, T.A.. Smith, G.A. and Metcalfe, J.C. (1984) Biochem. J. 221, 675-684 14 Rehbinder, R. and Greenberg. D.M. (1965) Arch. Biochem. Biophys. 109, 110-I 15 15 Tanaka, Y., Doi, 0. and Akamatsu. Y. (1979) B&hem. Biophys. Res. Commun. 87. 1100-l 115 16 Brasitus. T.A. and Keresztes, R.S. (1984) Biochim. Biophys. Acta 773, 290-300 17 Katyal, S.L. and Lombardi, B. (1974) Lipids 9, 81-86 18 Bra&us, T.A., Schachter, D. and Mamouneas, T.G. (1979) Biochemistry 18. 41364144 19 Diliberto, E.J. and Axehod. J. (1976) J. Neurochem. 26, 1159-1165 20 Paik, W.K. and Kim, S. (1980) Protein Methylation pp. 119-125. John Wiley & Sons. New York 21 Lineweaver. H. and Burk, D. (1934) J. Am. Chem. Sot. 56, 6.58-666 22 Zatz, M., Dudley, P.A., Kloog, Y. and Markey, P.S. (1981) J. Biol. them. 256. 10028-10032 23 Jaiswai. R.K., Landon, E.J. and Sastry. B.V.R. (1983) Biochim. Biophys. Acta 735, 367-379 24 Alemany, S., Varela. 1.. Harper, J.F. and Mato. J.M. (1982) J. Biol. Chem. 257, 9249-9251 25 Silvius, J.R., Read, B.D. and McEhaney. R.N. (1978) Science 299, 903-905 26 Audubert, F. and Vance, D.E. (1984) Biochim. Biophys. Acta 792, 359-362 27 Scarborough, G.A. and Nyc, J.F. (1967) Biochim. Biophys. Acta 146, 111-119 28 Scarborough, G.A. and Nyc, J.F. (1967) J. Biol. Chem. 242, 238-242 29 Prasad, C. and Edwards. R.M. (1981) J. Biol. Chem. 256, 1300& 13003 30 Weismann. W.P., Johnson, J.P., Miura, G.A. and Chiang, P.K. (1984) Am. J. Physiol. 248, F43-F47 31 Dudeja, P.K.. Foster, ES. and Brasitus. T.A. (1985) Gastroenterology 88, 1370(A).