159

Biochimica et Biophysicu A cta, 1081 ( 1091 ) 159- t 66 ' 1991 Elsevier Science Publishers B,V. (Biomedical Division) (}005-2760/91/$03.50 A D O N I S 000527609100069W

BBALIP 53566

A sensitive assay of lysogangliosides using high-performance

liquid chromatography Takuro Kobayashi and lkuo Goto Department of Neurolo.t(v. Neurologwal lnstttute, ~'ctdty of 3,h'dictne. Kvushu Unit'er.~i(v,Fukuoka (Japan) tRcceivcd 21 Ma,, Iqg(D

Key words: Lysoganglioside: Ganglioside: Ganglio.xidosis; H P L C

Lysogangliosides, LGM I, LGM 2 and LGM 3, each carrying a single sphingoid base (i.e., Cts:t, Cta:0, Czo:t, C20:0), were prepared and a sensitive assay method of these lipids using HPLC was developed. The mcthod involves fluorescence derivatization of the free amino group of the molecule with o-phthalaldehyde, separation of the molecular species of each lysoganglioside using reversed-phase HPLC and assay on the basis of a known amount of one of the lysogangliosides, as the internal standard. Using this method, lysoganglioside can be accurately assayed in the range of 5 - 1 0 0 0 pmol. For assay of the lipid in Ihe tissue, crude isolation procedures including extraclion of lipids, Folch's partition and DEAE-Sepharose and AG I-X2 column chromatographies were required before the fluorescence derivatizafion. In the normal human and the bovine cerebral cortex, 0.4-2.0 p m o l / m g protein of LGM I containing C ta:l and C2o:l sphingosine residues were detected. In the frontal cortex from a patient with Sandhoff disease, an abnormal accumulation (55-78 p m o l / m g protein) of LGM 2 was noted. Among various molecular species, L G M , containing Cts:l wasi the most abundant. Introduction Glycosphingolipids including gangliosides are composed of a hydrophilie oligosaccharide moiety and a hydrophobic eeramide. The ceramide consists of sphingoid (long chain) bases and fatty acids linked to the free amino group of tbe sphingoid bases. These compounds are natural constituents of plasma membranes of mammalian cells and are particularly rich in neuronal cells. Glycosphingolipids devoid of the fatty acid moiety are called lysoglycosphingolipids in analogy to lysophospholipids. It has been considered that such lipids do not occur in the tissue because of the cyto-

Abbreviations: GM~, ll~NeuSAc-L~ctcec:GM_~, I/~NeuSAc-Ggse~Cer; GMI, 113Neu5Ac-GgOse4Cer:, LGM~ (LysoOM3). lI Neu5AcLactSph; LGMz (LysoGMa), II:lNeu5Ac-GgOse~Sph; LGM. {LysoGMt), II~Neu5Ac-GgOse4Sph; HPTLC, high-performance thin-layer chromatography; HPLC, high-performante liquid chromato~aphy; FAB-MS, fast-atom bombardment mass spectrometry; DEC, l-(3-dimethylaminopropyl)-3-ethylcarbodiimide.Sphingosineis a generic term of sphingoid (long chain) bases. The alkylchain length ana the number of double bonds in the moleculeare expressed such as CIs:~.

Correspondence: T. Kobayashi, Department of Neurology. Neurological h~st~tute, Faculty of Medicine, Kyushu University 60, 3-1-1 Maidashi. Higashi-ku, Fukuoka812. Japan,

toxicity. Recently however, there have been ,~'ports of an abnormal accumulation of the lysoglycosphingolipids in tissues from patients with glycosphingolipidoses; galactosylsphingosine in Krabbe disease [1,2], glucosylsphingosine in Gancher disease [3]. lysosulfatide in metachromatic leukodystrophy [4] and LGM 2 in G M : gangliosidosis [5.6]. In such glycosphingolipidoses, the accumulation of lysoglycosphingolipid is well explained by the deficiency of degrading enzyme of the lipid and is postulated to contribute to the pathogenesis of these diseases. Even in normal tissues or cells, the presence of lysoglycosphingolipids has been demonstrated by the development of sensitive assay methods, albeit the concentration being very low. Free sphingoid bases exist in almost all tissues and cells [7,8]. Galactosylsphingosine and lysosulfatide are found in the white matter of the central nervous system and peripheral nerves [2,9]. LGM 3 was noted in cultured cells [10]. At present, the meaning of these cytotoxic lysosphingolipids is not understood but the lipids, especially free sphingoid bases, are considered to have physiological functions such as modulation of cell growth and signal transduction systems. We report here a sensitive assay of lysocompounds of monosialogangliosides, LGM~, LGM 2 and LGM~. Using the method, we found evidence of the natural occurrence of LGM~ in the human and the bovine

160 cerebral cortex and an abnormal accumulation of LGM 2 in the brain from a patient with Sandhoff disease. Materials and Methods

Materials The sources of some of the commercial materials were as described in previous reports [7,11]. Sialidase (A. ureafaciens) was obtained from Nakarai Chemicals (Kyoto. Japan). AG l-X2 (200-400 mesh, acetate form) and AG 50w × 8 (200-400 mesh, hydrogen form) were from Bio-Rad (Richmond, U.S.A.). DEAE-Sepharose CL-6B was obtained from Pharmacia (Uppsala, Sweden) and was converted to the acetate form, according to Iwamori and Nagai [12]. The bovine brain ganglioside mixture was purchased from Sigma (St. Louis, U.S.A.) or was provided by Eisai Co. (Tokyo, Japan). GM I was prepared from the bovine brain ganglioside mixture according to Momoi et al. [13]. GMs was purified from human liver by the method of Seyfried et al. [14]. GM 2 was purchased from latron Co. (Tokyo, Japan) or was purified from the brain of a Sandhoff disease patient according to Momoi et al. [13]. The purity of these gangliosides was higher than 98%, as checked using H PTIC.

Preparation of lysogangliosides The preparation of lysogangliosides was carried out essentially according to Neuenhoffer et al. [15] and Nores et al. [16]. 100 mg of GM t was hydrolyzed in 1 M KOH in n-butanol/water (9: 1, v / v ) at 117°C for 2 h, according to Taketomi and Yamakawa 117]. The hydrolysate was neutralized by the addition of hydrochloric acid and evaporated to dryness under a stream of nitrogen gas. The dried sample was dissolved in a small volume of distilled water, dialyzed against water for 24 h and lyophilized. The sample was dissolved in chlorof o r m / m e t h a n o l / w a t e r (80:20 : 1, v / v ) and applied onto an latro beads (6RS-8060) column (1.0 × 30 cm), which had been equilibrated with the same solvent. De-N-acetyl LGM~, obtained by the alkaline hydrolysis of GM~, was isolated by a gradient elution with chlorof o r m / m e t h a n o l / w a t e r from 80 : 20 : 1 (v/v) to 30:70:10. The fractions were checked using HPTLC, with a developing solvent of chloroform/methanol/15 mM CaCI 2 (55:45:10, v/v) and visualization with a resorcinol reagent. Reacetylation of the neuraminic acid residue of de-N-acetyl LGM t was carried out with DEC and acetate buffer in phosphatidyicholine liposomes. according to Nores et al. [16] The acetylated sample was lyophilized and LGM 1 was purified using normalphase HPLC, in which the gradient elution was conductcd with chloroform/methanol/water from 80: 20 : 1 (v/v) to 30 : 70 : 10 for 90 rain at a flow rate of 1.2 ml/min. Each fraction (1.2 ml) was monitored using H PTLC as described above. Fractions containing LGM l

were collected and concentrated to a small volume by nitrogen gas. To obtain LGM~ containing the saturated sphingoid base (Cts:,, or C20:c~), LGM t was reduced with extensive use of sodium borohydride according to Schwarzmann [18]. subsequently dialyzed against water and lyophilized. In other cases. GM I was first reduced usi!ag sodium borohydride and LGM 1was purified using the procedure described above.

Tissues The brain of a patient who died with infantile Sandhoff disease (2 years of age) was provided by Dr. M. Yoshino (Department of Pediatrics, Kurume University School of Medicine). Brains of three individuals without neurological disease (34-64 years of age) were from our clinic. All these brains were kept frozen below - 2 0 ° C until use. Bovine brains were obtained from a local slaughterhouse.

HPLC equipment The HPLC system including a solvent delivery system, programmer, sample injector and fluorescence monitor was as described [7,11]. In addition, a column oven (CTO-6A, Shimadzu, Japan) and a integrator (CR4A, Shimadzu) were also used. The reversed-phase chromatography was run with a Chemcosorb column (5-ODS-H, 4 . 6 × 150 mm. Chemco, Japan) and a Nucleosil column (Chemcopak, 3/~m, 12 nm pore size, 5.2 × 250 ram, Chemco) was used for the normal-phase chromatography.

Mass spectrometry Negative-ion FAB-MS of lysogangliosides was carried out essentially as described [191.

Isolation of lysogangliosides from the tissue All the procedure was done below 30°C. Tissues (10-300 mg wet weight) were homogenized in 2 mi of distilled water using an all-glass Potter-Elvehjem homogenizer and a portion was taken for the determination of protein content, which was assayed according to Lowry et al. [20]. The homogenate was put into a 50 ml graduated tube and 8 ml of methanol, 4 ml of chloroform and 100-200 pmol of LGM I containing C20:o sphingosine (an internal standard) were added. After keeping at room temperature for 2 h with occasional shaking, the homogenate was centrifuged at 1000 x g for 5 min, the supernatant removed and the pellet was again homogenized in 12 ml of chloroform/methanol (2 : 1, v/v). The second supernatant obtained after centrifugation was combined with the previous one, to which 12 ml of chloroform was added to make the solution of chloroform/methanol (2 : 1, v/v). Then the Folch's partition was carried out after addition of 0.2 vol. of distilled water, the upper phase saved and the lower phase was washed with the theoretical upper

161 phase. The two solutions of the upper phase were combined, evaporated to a small volume by a stream of nitrogen gas. dialyzed against water and lyophilized. The lyophilized sample was dissolved in 2 ml of chlorof o r m / m e t h a n o l / w a t e r ( 3 0 : 6 0 : 8, v / v ) and applied to a column of DEAE-Sepharose (0.7 × 2 cm) which had been equilibrated with the same solvent mixture. Unadsorbed materials were collected by elation wtth an additional 13 ml of the same solvent, which was then applied to an A G l-X2 column (0.7 × 2 cm). The column was washed successively with l0 ml of the same solvent and 10 ml of m e t h a n o l / w a t e r ( 9 : 1 , v / v ) and lysogangliosides were elated from the column with 4 ml of m e t h a n o l / w a t e r / a c e t i c acid ( 9 0 : 5 : 5 , v / v ) . The solution was then evaporated to dryness.

Detection of lysogangliosides using HPLC Fluorescence derivatizalion and detection using H P L C of lysogangliosides were essentially the same as described [7,11] except that ethanthiol in o-phthalaldehyde solution was increased to 50 .al to stabilize the fluorescence, Fluorescent derivatives of authentic lysogangliosides were directly injected onto reversed-phase HPLC. In the case of lysoganglioside assay in the tissue. the following manipulations had to be made before injection onto HPLC, to eliminate several fluorescence peaks other than those of lysogangliosides. T h e !ysoganglioside fraction eluted from anion-exchange{columns was derivatized with o-phthalaldehyde ant, iaPplied to a column of A G 50W-X8 (0.7 × 0.7 cml. ~ h e pass-through fractions were evaporated to dryness, ~itissolved in 0.1 ml of methanol and passed throu~:~l a Millipore column guard filter (0.22 .am). An isocrtatic elution of fluorescent lysogangliosides was carried lout with t e t r a h y d r o f u r a n / m e t h a n o l / water ( 5 : 5 : 7 . v/vt~ as the carrier solvent at a flow rate of 1.5 m l / m i n and ~ith a column temperature at 55°C. T h e fluorescence ~;as monitored at an excitation of 335 nm and emissiot~ of 420 nm, as described [7,1 l].

HPLC. The peak ! corresponded to the lower spot on H P T L C and the peak ill to the upper spot (Fig. 1. lanes 3 and 4). The peaks ! and 11I thus isolated were then subjected for FAB-MS analysis (Fig. 2). in the peak l, pseudo molecular ion ( M - 1) was at m / : 1278, while that of the peak lit was at m / : 1306. These data and several fra?~ment ions seen in Fig, 2 were essentially the same as those reported by Neuenhofer et al. [151. Hence, the peaks I and 111 were identified to be LGM~ containing sphingosine residues of C~s.~ and C_-a:~, respectively. When the peaks ! and I11 were reduced with sodiv.'n borohydride, peaks II and IV appeared, respectively, thereby suggesting that the peak il contains Cts . sphingosine and the peak IV has C_~0:0 sphingosine. When the bovine brain G M t was first reduced with sodium borohydride, main sphingoid bases of which were C~.~, and C,_u:~, as examined by the method of Kadowaki et al. [21], and then LGM~ was prepared,

a

i

io

dlb

Results and Discussion

Preparation and detection of authentic lysogangfiosides L O M l prepared from bovine brain GMI asing fl~e procedure described undec Materials and Methods ga~e two spots on H P T L C , both reacting with rcsorcinol aJ~d ninhydrin (Fig. la). When the L G M u was derivatiz~d with o-phthalaldehyde and analyzed using rcversdlphase HPLC, two large peaks (peak I and IZI) and t~.o small peaks (peak II and I V ) w e r e observed (Fig. lb). lm order to separate these four peaks, the LGM~ dissolvt~d in a small volume of methanol was injected onto Oeversed-phase HPLC, and elated isocratically w|th m e t h a n o l / w a t e r (75:25, v / v ) containing 10 mM NalCI as the carrier solvent at a flow rate of 1.2 m l / m i n . E~ch fraction (1.2 ml) was monitored using H P T L C ~ad

m

1 2 3 4 Fi~.. I (ak HPTLC chromatogram of LGM 1 Tolal LGM n was isclat:d from bovine brain GM, and two spols were isolated using ~,:',,:rsed-phase HPLC. Technica[ details are described ;n the lext. HP'FL.(" wa,, developed with chloroform/methanol/15 mM ('aCf 2 t55:45: I0. v/v) and spots were visualized with rcsorcinol reagent. L, GMI: 2, total LGMn: 3. isolated lower spot of LGM~: 4. isolated upper spot of LGMI. (hi, HPLC chromatogram of LGMn. Total LGM I i~olated from bovine brain GM nwas dcrivalizcd with ~-phthalak!':hyde and applied ont~ HPLC as described in the text.

162

b >. E ¢0 4.a t(J

b'l O 3 LL

II IV

~o

Time (min) Fig. ] continued. there were small peaks of I and III and large peaks of II and IV on HPLC. These peaks II and IV were purified using reversed-phase HPLC as described above. In the FAB-MS, peaks II and IV gave pseudo ions at m/z 1280 and 1308, respectively. When the purified LGM 1 was coupled with palmitic acid according to Sonnino et ;,.i, :~

-.:..

............................ -"'

A

i

: ~f

[i

UULL........ .L.,a j.~. . . . . . L~ . . . . , Z . . , .J ....... ' i '-,,-'~. I~1m [llXllllmrrllH ililHllllll311Fff~TIH lira ~lllllTmffr I/T'Tl~i]ml~X~lI] ~" llTmmll roll m1111HTTIIT1

II

0

,~o.~""" ---~ .... --;-7,;, ........................

8

I I

-

0

~'~:~"

13Uh

~ T ~ l l ~ I

i rt t I t i n'l/t'l'lii ! ~l~i p t

g

Fig. 2. N e g a t i v e - i o n F A B - M S s p e c t r u m s o f L G M I. T e c h n i c a l d e t a i l s ace d , ' s c f i b e d in the lexl. A . p e a k I in Fig, 2: B. p e a k I!1 in Fig, 2.

al. [22], the product was identical to the authentic GM~ ganglioside on HPTLC, developed with two solvent systems, chloroform/methanol/10 mM CaCI 2 (50: 40: 10, v / v ) and c h l o r o f o r m / m e t h a n o l / a m m o n i a (65 : 25 : 5, v/v). LGM 3 was prepared from human liver GM 3 using the method described above. There was one large early peak and one small late peak on the HPLC. When the LGMs was reduced with sodium borohydride, there was only the second peak, In the FAB-MS analysis of the first peak, pseudo molecular ion at m/z 913 was observed. The findings of several other fragment ions (m/z 622, 470, 308, 290) and the pseudo molecular ion were the same as those reported by Nt, aenhofer et al. 115] and Nores et al. [16l. Thus, the first peak of LGM3 was concluded to contain the sphingosine of C]s:l and the second peak Cts:o. LGM2 prepared from human or bovine brain G M 2 showed four peaks on HPLC. Because of the lira!ted amount of the lipid, further separation of the lipid was not feasible. The HPLC chromatograms of the authentic lysogangliosides thus obtained are shown in Fig. 3, in the carrier solvent, the addition of ions was necessary, as is the case in the lysosulfatide assay [9]. Without ions, fluorescent lysogangliosides were eluted near the void volume, even if the water concentration was increased in the carrier solvent. Among the ions tested, NaCI (more than 10 raM) was proved to be the most satisfactory. Phosphoric acid (10-35 raM), acetic acid (10 mM) and trifluoroacetic acid (10 raM) prevented fluorescent lysogangliosides from passing through the HPLC col-

163

IIi~1

til

u u o ;3 u.

2

Jli,

i 0

!'~

3

A ......... ;" ' '. ................ ~'i,...................................................

1

=.'..'.. . . . .

...,

2

/=":.

f

4

....................aC...... $~

liO

'

IJ~,

--'~

Time

?~5

310

31~

4.10

(rain)

Fig. 3. HPLC chr0matograms of ~uthenti¢ lys0gangli0sidcs. L,;s0gangli~sides each carrying several sphingoid bases were dem'atized with o-phthalaldehyde, injected onto a reversed-phase HPLC ;.~qd elated a'izh tctrahydrofuran/methanol/water (5:5:7. ,~'/~') at a now rate of 1.5 mi/min at a column temperature of 55°C. Technical details are described in the text. A. LGM~: B. LGMz: C. LGMf 1. containing ('1.', i sphingosine: 2. containing C~ . sphingo_~ine- 3. containing C_-,, ~~phing,~ine: 4. containing C:,, ,, sphingosine. umn but made the peaks broad, As to the solvent system, t e t r a h y d r o f u r a n / m e t h a n o l / water (5 5 : 7, v / v ) was the most appropriate for the separation of each molecular species of L G M I. L G M , and LGM~. But when the H P L C was run at room temperature, the pressure of the column became too high, and the column temperature was raised to 55°C. Using the method described above, authentic lysogang'iosides could be accurately assayed in the range of 5-1000 pmol (Fig. 4). Lysoganglioside in the normal brain For the assay of lysoganglioside in the tissue, crude isolation of the lipid using Folch's partition, DEAESepharose and A G l-X2 column chromatographies was needed before the derivatization with o-phthalaldehyde. More than 98% of the lysoganglioside came to the upper-phase of the Folch's partition when an authentic lysoganglioside was used. As to the dialysis against water, most of the lysoganglio~ide remained in the dialysis tube. Because lyso-compounds of monosialogangliosides have both positive and negative charges in their molecule, they did not retain in the DEAE-Sepharose, yet they did bind to the A G I-X2 column and were eluted with acetic acid. The yield of the two chromatographic steps was each approx. 70%. When a known amount of authentic lysoganglioside was used as the starting material and lipid isolation was carried out, the final recovery ot" the lipid was approx, 50%. L G M t, L G M 2 and L G M 3 were recovered at the same ratio, irrespective of the molecular species, therefore, one could

be used as the internal standard. Usually LGMt conraining C_,,, sphingosiae has been used as the internal standard because the lipid was undetectable except in

°6t/

15

/

0.~

0

200

qo0

600

L G M I C I a : t- p m c l l a s s a y

801)

11)1)0

tube

Fig, 4, Standard curve of LGM~ containing C~:1 sphingosine

~'L(;M~C,~ ,). o-Phthalaldehydewas coupled with various amounts of LGMICt~. i, as indicated and injected onto HPLC. Technical details are described in the text. Relative fluorescence intensities of LGMIC~:~ were obtained on the bases of the peak area of LGM~ containing C'2o ~ sphingosine (It0 pmol) which had been included in the reaction mixture as the internal standard. Similar curves were obtained with other lysogangliosidesas the internal standard.

164 tissues from patients with GM~ gangliosidosis. During the isolation procedures, lysoganglioside was not newly p r o d , t e d from the parent ganglioside, as checked using GM~ as the starting material. Using the above-described method, lysoganglioside in the tissue was measured in the range of 2 0 - I 0 0 0 pmol. In the normal h u m a n or the bovine brain tissues, two peaks were observed in the H P L C (Fig. 5). The retention times of these peaks corresponded to those of L G M , with Cls:, and C=o:l sphingosine residues, respectively. When lysoganglioside was isolated from 300 g of bovine brain, a ninhydrin- and resorcinol-positive spot corresponding to that of an authentic L G M I was observed on H P T L C (Fig. 6). Because of the isolated sample was scanty (a few gg), we could not further characterize it, but from these data. there appears to exist a small a m o u n t of LGM~ in the normal h u m a n and the bovine brain. The h u m a n cerebral cortex contained 0.9-2.0 p m o l / m g protein of L G M , (n = 3). The ratio of the molecular species (C1,~: j ; C20: l) was 1 : 0 . 6 0.9. In the bovine cerebral gray matter, 0.4-1,4 p m o l / m g protein of L G M I was noted ~ n = 3 ) . The molecular species ratio was 1 : 0.3-0.8. L G M j may not be the only iysoganglioside present in normal tissues. In the usual assay of lysoganglioside, several hundred mg of brain was used as the starting material, but other lysogangliosides may possibly be detected if larger a m o u n t s of brain were used. Moreover, the method used in this study can detect only lysogangliosides containing one sialic acid in the molecule but not those with m o r e sialic acids. The natural occurrence of lysoganglioside in the tissue could be expected, because we have already d e m o n -

I

'"

U

] 2""

• ..........

A

Time (min) Fig. 5. HPLC chrom:ttogr:ml of lysogangliosides in lhe cerebral cortex ~ff a normal subject (34 years of age). From 300 mg ,,f wet tissue. lyst,gat~gliosides were isolated, derivatized with o-phthalaldehyde and analyzed using reversed-phase ttPLC. A. lysogangliosidcs from a normal human brain. LGM t containing f-'.,~j:o sphingosine ILGM t ('~.:~) as the internal standard 1100 pmol). B, authentic LGM r Number.,. over the spots arc the same as those in Fig, 3.

123 Fig. 6. HPTLC chromatogram of lys0ganglioside in the bovine brain. Lysogangliasides were isolated from 300 g of bovine brain using Folch's partition and DEAE-Sepharose and AG I-X2 column c h r o m a t o g r a p h i e s . The isolated sample was further purified using reversed-phase HPLC with the isocratic clution with methanol/water (75:25. v/v) containing l0 mM NaCI. Fractions corresponding to those of authentic LGM1 containing C1~: i sphingosine were collected. HPTLC was developed and spots were visualized as described in the legend of Fig. I. 1. GMI; 2, authentic LGM I containing Cis:l; 3. sample.

strated the existence of free sphingoid bases |7], galactosylsphingosine [2] and lysosulfatide [91 in n o r m a l tissues. At present, however, the m a n n e r in which the lysoganglioside is synthesized is unknown, It m a y be synthesized by sequential glycosylation of free sphingoid bases or there might be a m i n o r p a t h w a y of deacylation from ganglioside. T h e f o r m e r is m o r e p r o b a b l e because we found that galactosylsphingosine in the m o u s e brain is synthesized from free sphingoid bases by the action of galactosyltransferase [23.24]. The physiological role of iysoganglioside in the tissue is a n o t h e r question. Since the report of H a n n u n et al. [25], there has been much discussion concerning m o d u l a t o r y effects of the lysosphingolipid on protein kinases and on the signal transduction system (see Refs. 26 and 27 for review), but the exact role of the natural existence of lysoganglioside remains to be elucidated.

165 Lysoganglioside in the brain of a GM: gangliosido~i~" patient Lysoganglioside was analyzed in a cerebral cortex from a patient with Sandhoff disease. Fig. 7 depicts an abnormal accumulation of LGM~. The most abundant one was that containing C1~: t sphingosine. Peaks of the lipid with Cts:0, C,o. u and C20:0 were also observed, albeit in lesser amounts. When the lysoganglioside isolated from the patient brain was treated with sialidase, according to Sugano et al. [281. and analyzed using H P L C , peaks of asiato LGM,_ appeared (Fig. 8). Similar to the c h r o m a t o g r a m of lysoganglioside, the c o m p o u n d with Cts:l sphingosine was the most abundant. In the frontal cortex from the patient (3 different samples), the concentration of total L G M , wa~, 5 5 - 7 8 p m o l / m g protein. The ratio of the molecular species (C~s: ~: CIs ," C2o:t : C_,o:o) was about 1 0 0 : 6 : 8 : 7. Similar values were obtained from the sialidase-treated samples. These data are in accord with those in previous reports. Neuenhofer et al. [5] noted that the accumulated LGM_, in the brain from a patient with Tay-Sachs disease was 15 n m o l / g . Rosengren et al. [6] reported that there were 43 n m o l / g of L G M z in the brain of a Sandhoff disease patient and 12-16 n m o l / g in the brain of two patients with Tay-Sachs disease. They assayed LGM_, by measuring sialic acid in the N-acetyl L G M , fraction on TLC. In their assay system, more than 10 g of tissue were needed. Using our method described herein we could assay the lipid in a b o u t 10 mg of wet brain tissue. Moreover, the a m o u n t of each molecular species of the lipid can be identified. Rosengren et al. [6] stated that the accumulated a m o u n t of L G M z is greater in tissues from Sandhoff disease patients than those in Tay-Sachs disease patients. Unfortunately, tissues from only one

', IL"

r~

o_ ._1

3

i

i 0

"'1 110

.....

~

~" . . . . . . .

31~

410

Time

(mini Fig. 8. FtPI.(" chromatogrz~mof sialidase-treated lysogangliosides in the cerebral cortex from a patient with Sandhoff disease. LysogangIiosides i~olalcd I'rom the patient and authe~,ic L(iM z were treated with ,~lalida,,,eaccording to Sugano et al. [2Sl. After derivallzation with o-phthalaldeh)de, neutral lysosphingolipids were obtained usit~g AG I-X2 and A(i 50W-X8 columns, which were then injected onto a rcvcrsed-phax,: }'IPLCand eluted with methanol/water (9: 1. v/v). as described 171- A, sialidase-treated lysogangliosides from the cerebral cortex of the Sandhol,l, disease patient. B. sialidase-treated authentic LGM ,.

patient was available in this study. We are now collecting samples and will report data elsewhere. The metho6 described herein is applicable for the assay of lysoganglioside in tissues from patients with other ganglioside storage diseases, i.e., G M t gangliosidosis and sialidosis. Lysosphingolipids are known to be cytotoxic and may well be the pathogenetic agent for the degeneration of the tissue or cell in sphingolipidoses. T h e present method for the assay of lysoganglioside is simple, accurate and only a small sample is needed. Acknowledgments

! '"i ".,,. •

,

"!".



.

g'. ,~ ;

3

.

.

.

.

,

-..

,.L•

.

23.

Time

.,

.

.'*

.

B

. C ..

(mini

Fig. 7. HPLC chromatograms of lysogangliosides in the cerebral gray matter from a patient with Sandhoff disease and a control subject. From 10 mg of the samples from each individual, lysogangliosides were isolated, derivatized ~ith a fluorescence probe and analyzed using HPLC. as described in the text. A, control subject: B, Saadhoff disease patient; C, authentic LGM2: D, authentic LGMv The peaks of LGMiC2o:o are the internal standard (100 pmol). The numbers over the peaks are the same as those in Fig. 3.

We thank Dr, R. lsohe (Center of Advanced lnstrumenial Analysis, Faculty of Pharmaceutical Sciences, Kyushu University) for performing FAB-MS, Dr. M. Yoshino ( D e p a r t m e n t of Pediatrics, K u r u m e University School of Medicine) for providing a brain sample of a patient with Sandhoff disease and M. O h a r a for helpful comments. Thi~ work was supported ~n part by grants from the Ministry of Education, Science and Culture, Japan, and from the Ministry of Health and Welfare, Japan. References I Svennerholm, L.. Vanier. M.-T. and Mansson. J.-E. (1980)J, Lipid Res. 21.53--64.

166 2 Kohayashi, T,. Golo, I., Yamanaka. T., Suzuki, Y.. Nakano. T. and Suzuki. K. (1988) Ann, Neur01. 24, 517-522. 3 Nilsson. O. and Svennerholm, L. (1982) J. Neur~x'hem. 39, 709718. 4 Toda. K.. Kobayashi. T., ¢,,ioto. t.. Kurokawa. T. and Ogomori. K. (1989) Biochem. Bk~phys. Res, Commun. 159, 605-611, 5 Neuenhofer, S,. Conzelmann, E, Schwar-zmann. G.. Egge. H. and S,'mdhoff, K. (1986) Biol. Chem. Hoppe-Seyler 376. 241-244. 6 Rosengren, B,, Mansson, J.-E. and Svennerholm. L. (1987) J. Ncurochem, 49. 834-840, 7 Kobayashi. T.. Mitsuo. K. and Goto. 1. (1988) Eur. J. Biochern, 172~ 747-752. 8 Merrill, A.H.. Jr., Wang. E.. Mullin,s, R.E., Jamison, W.C.L.. Nimkar, S. and Liotta, D.C. (1988) Anal. Biochem, 171, 373-381. 9 Toda, K.. Kobayashi, T.. Goto. I.. Ohms, K.. Eto. Y,, Inui. K. and Okada. S. (1990) J. Neur~'hem. 55, 1585-159l. lo Hanai. N.. Nores, G.A., MacLeod, C., Torres-Mendez, C.-R. and Hakornori. S. (1988) J. Biol_ Chem. 263, 10915-10921. 11 Shinoda, H.. Kobayashi, T., Katayama, M., Goto. I. and Nagara, H, (1987)J. Neur0chem. 49, 92-99. 12 iwamori. M. and Nagai. Y. (1978) Biochim. Biophys. Acla 528, 257 267. 13 Momoi, T., Ando. S. and Nagai, Y. (1976) Biochim, Biophys. Acts 441. 488-497. 14 Seyfried. T,N.. Ando. S, and Yu. RK. (1978) J, Lipid Res, 19. 538- 543.

15 Neuenhofer. S.. Schwarzm=,nn. G., Egge. H. and Sandhoff. K. (1985) Biochemistry 24. 525-532. 16 Nores. G.A~. Hanai, N.. Levery. S.B,, Eaton. H.L.. Salyan. M.E.K. and H;:komori, S. 119881 Carhohydr. IRes. 179. 393-410. 17 TaketoHi. T. and Yamakawa. T, (10631 J. Bi~chem. 54, 444-451_ 18 Schwarzmann. G. (19781 Biochim, Bit)phys. Acta 529, 106-114. 19 lsobe, R.. Kawano. Y.. Higuchi. R and Komori. T. (1989) Anal. Biochem. 177. 206-299. 20 Lowry. O.H.. Rosebrough. N.J.. F:~rr. A.L. and Randall. R.J. (1951) J. Bit~l. Chem. 193. 265-275. 21 Kadowaki. H,. Bremer, E,G.. Evans. J,E.~ Jungalwala, F.B, and McCluer. R.H. (t983) J, Lipid Res. 1389--1397. 22 Sonnim~, S., Kirschner G., Ghidoni, R., Acquotti, D. and Tettamanti. G. (1985) J. Lipid Res. 26. 2a8-257. 23 Mitsuo, K.. Kobayashi. T., Shinnoh. N. and Goto. 1. 11989) Neurochem. Res. 14. 899-003. 24 Kobayashi. T,. Shinnoh. N, and Ooto, I. (1989) Eur. J. Biochem. 186, 493-499. 25 Hannun, Y.A.. Loomis, C.R., Merrill. A.H., Jr. and Bell, R.M. (1986) J. Biol. Chem. 261, 12604-12609. 26 Hannun, Y.A. and Bell, R.M. (1989) Science 243. 500-507, 27 Merrill. A.H.. Jr, and Stevens. V,L. (1989~ Biochim. Biophys, Acts 1010. 131-I39. 28 Sugano, K., Saito. M. and Nagai, Y. (1978) FEBS Lett. 89, 321-325.

A sensitive assay of lysogangliosides using high-performance liquid chromatography.

Lysogangliosides, LGM1, LGM2 and LGM3, each carrying a single sphingoid base (i.e., C18:1, C18:0, C20:1, C20:0), were prepared and a sensitive assay m...
602KB Sizes 0 Downloads 0 Views