J Nutr Sci Vitaminoi, 48, 325-331,

2002

Review

Characterization

and

Bioavailability from Edible

of Vitamin Algae

B12-Compounds

Fumio WATANABE1, *, Shigeo TAKENAKA2, Hiromi KITTAKA-KATSURA3, Shuhei EBARA4and Emi MIYAMOTO1 1Department of Health Science, KochiWomen's University, Kochi780-8515, Japan 2 Department of Veterinary Science, and 4 Department of AppliedBiologicalChemistry, Osaka Prefecture University, Sakai, Osaka 599-8531, Japan 3 Department of Health Science, Hiroshima Prefectural Women's University, Hiroshima 734-8558, Japan (Received May 22, 2002)

Summary Substantial amounts of vitamin B12 were found in some edible algae (green and purple layers) and algal health food (chlorella and spirulina tablets) using the Lactobacillus delbrueckiisubsp. lactis ATCC7830 microbiological assay method. Corrinoid compounds were purified and characterized from these algae to clarify the chemical proper ties and bioavailability of the algal vitamin B12. True vitamin B12 is the predominate cobamide of green and purple layers and chlorella tablets. Feeding the purple layer to vita min B12-deficientrats significantly improved the vitamin B12status. The results suggest that algal vitamin B12is a bioavailable source for mammals. Pseudovitamin B12 (an inactive cor rinoid) predominated in the spirulina tablets, which are not suitable for use as a vitamin B12 source, especially for vegetarians. Key Words algal health food, bioavailability, cobalamin, edible algae, vitamin B12

B12; Lactobacillus delbrueckii subsp. lactis (formerly Lactobacillus leichmannii) ATCC7830 has been used widely (5). The radioisotope dilution assay (RIDA) method using radioactively labeled B12 and hog intrin sic factor (IF), the most specific B12-binding protein, has also been used to determine the B12 content in foods (9-11). Recently, a chemiluminescence (acridinium ester)-labeled B12derivative has been devised for use in stead of radioactive B12, Several researchers (12, 13) have attempted to assay B12in foods using a fully auto mated chemiluminescence B12 analyzer (Chiron Diagnostics, East Walpole, MA, USA) with the acri dinium ester-labeled B12 derivative and IF, and have demonstrated that, except for foods containing substan tial amounts of inactive corrinoids, the observed corre lation coefficient between the microbiological and chemiluminescence methods in foods is excellent (12, 13). Various types of edible algae are available in the world. Edible algae are known to be rich in vitamins, minerals and dietary fiber (14). The B12in various edi ble algae was assayed by IF-based competitive inhibition methods and/or the microbiological method using L. delbrueckii ATCC7830 (14-16) (Table 1). Although dried green (Enteromorpha sp.) and purple (Porphyra sp.) layers (non), which appear to be most widely eaten among the edible algae, contained substantial amounts of B12,other edible algae contained none or only traces of B12.The B12 content determined using the microbio logical method was about 7-fold greater in the spirulina used for human food supplement than the values deter mined using RIDA.

Vitamin B12(B12or CN-B12)belongs to the corrinoids, which are compounds having a corrin nucleus in com mon, and is synthesized only in certain bacteria (1). There are some naturally occurring B12-compounds with different upper ligands (L) (Fig. 1). Especially no table are McB12and AdoB12, which function as coen zymes of methionine synthase (EC 2.1.1.13) involved in methionine biosynthesis (2) and methylmalonyl-CoA mutase (EC 5.4.99.2) involved in the amino acid and odd-chain fatty acid metabolisms (3) in mammals, re spectively. The corrinoids carrying a base other than 5,6-dimethylbenzimidazole in the lower ligand (cobalt coordinated nucleotide) are also found in nature (4). Usual dietary sources of B12 are animal food products (meat, milk, eggs and shellfish), but not plant food prod ucts (5). Several papers have reported that some plant foods (edible algae) contain large amounts of B12,which appears to be in the form of inactive corrinoids, there fore they may not be a bioavailable source for mammals (6-8). It is still unclear whether or not 12-compounds found in edible algae are true B12or inactive corrinoids because there is little available information on the chemical properties of the algal B12-compounds. This review summarizes current studies (mainly the works of the authors of this paper) on the characteriza tion and bioavailability of B12-compounds from various edible algae. 1.

B12 Content in Edible Algae Historically, the B12 content of foods has been deter mined by bioassay using microorganisms that require * E-mail: [email protected] 325

326

WATANABEF et al .

Fig.

1.

Structural

of B12-compounds.

formula

of B12 and partial

The partial

structures

structures of B12-com

pounds show only those portions of the molecule that differ from B12. 1, AdoB12, 2, McB12; 3, OH-B12; 4, SO3 B12; 5, CN-B12 7,

pseudovitamin

cobamide; Table

1.

or B12; 6, benzimidazolyl B12;

9, p-cresolyl Content

8,

cobamide;

5-hydroxybenzimidazolyl

cobamide.

of B12 in edible algae.

2. Purification and Characterization of B12 Compounds from Edible Algae To determine whether or not the dried purple and green layers and/or microaglae used for human food supplements contain true B12 or inactive corrinoids , some B12-compounds were purified and characterized . 2-1. Driedpurple and green layers (norin) A B12-compound was purified from an edible purple layer (Porphyra yezoensis) and partially characterized (17). The silica gel 60 TLC and C18 reversed-phase HPLC patterns of the purified pink-colored compound were identical to those of authentic B12 (Table 2) , but not to those of the corrinoids inactive for humans , The purple layer contained five types of biologically active B12-compounds (CN-B12,OH-B12,SO3-B12,AdoB12 and McB12),in which B12 coenzymes (AdoB12 and McBl2) comprised about 60% of the total B12(16). A B12-compound was also purified from a dried green layer (Entromopha prolifera) and partially characterized (18). The silica gel 60 TLC and C18 reversed-phase HPLC patterns of the purified pink-colored compound were identical to those of authentic B12,but not those of the inactive corrinoids for humans. Most B12-com pounds found in the dried green layer are the non-coen zyme forms of B12(about 75%, mainly OH-B12). 2-2. Microalgae usedfor human health foods Spirulina (Spirulina sp., a blue-green alga) tablets also contain large amounts of B12 (7). The B12concentration of spirulina tablets was determined by both Lactobacillus microbiological and IF-chemiluminescence methods

Vitamin B12-Compounds Table

2.

TLC

Rf values

and

and

HPLC.

(benzimidazolyl sheets v/v) of

and

and as

the

developed

solvents purified

5C18RS). 40•Ž,

times

I and

B12-compound

(2ƒÊL)

of

5-hydroxybenzimidazolyl with II,

and were

by

the

purified

in of

the

the

the

purple

and

from

dark

at

room

eluted absorbance

were with at

a 20%

In

(v/v)

361nm.

case

with

were

flow

cyanocobamides

on

silica

concentrated

was

gel

60

column

1%

TLC

(7:1:2

solutions

HPLC

containing

rate

upon

of

(28%)/water

of HPLC,

solution

solvent

spotted

a reversed-phase

methanol

The

cyanocobamides and

2-propanol/NH40H

the

analyzed

and

layer,

B12)

and

temperature.

CN-B12, purple

pseudovitamin

(10:7:10v/v)

cyanocobamides

327 layer,

purified

cyanocobamides,

isocratically

measuring

from

B12-compound

1-butanol/2-propanol/water

respectively,

corrinoids

monitored

of the

solutions

B12-compound The

and

retention

Concentrated

from Edible Algae

(2ƒÊL)

(wakosil-II

(v/v)

acetic

acid

at

1mL/min.

(Adapted, with permission, from Ref. 17)

Table

3.

B12 concentration

of spirulina

tablets.

cal

and

ues

(200.9-211.6ƒÊg/100g

IF-chemiluminescence

the

chemiluminescence

ues

(201.3-285.7g/100g

the

microbiological

fied

from

ized B12

as

HPLC,

All values a Determined with

permission,

obtained by the from

represent microbiological Ref.

mean•}SEM assay

(n=4). .

1H-NMR

tablets

true

contain

The

duces

(12,19). The values determined using the microbiolog ical method were about 9-fold greater than the values determined using the chemiluminescence method (Table 3). Although Herbert and Drivas (6) reported that most types of B12found in the spirulina tablets are biologically inactive corrinoids, probably cobinamide like compounds, there is no detailed information on the chemical properties of the spirulina corrinoids. Recently, two corrinoids were purified from spirulina tablets and partially characterized (19). The major (83%) and minor (17%) compounds were identified as pseudo-B12 and B12,respectively, as judged from data of TLC, C18reversed-phase HPLC (Table 4), 1H-NMR spec troscopy and ultraviolet-visible spectroscopy (data not shown). These results indicate that pseudo-B12, which is inactive for humans, is the predominant corrinoid in the spirulina tablets. Some varieties of green algae, Chlorellasp., have been reported to have the ability to take up and accumulate exogenous B12 (20). Indeed, B12-enriched ChloreIlacells have been prepared and used as food for rotifer (a food for fry in fish farming); rotifer growth significantly in creases with the feeding of Chlorella cells (21). Chlorella (Chlorella sp.) tablets have already been introduced as a human health food. The B12 concentration of chlorella tablets was assayed by both Lactobacillus microbiologi

been

the

as

by puri

character identified

as

reversed-phase

ultraviolet-visible that

chlorella

Pleurochrysis

of as

food

carterae

a human

health

food

et algal

human

pro

around

Takenaka

lyophilized

a

is a

and

scales

coccoliths.

safety

carterae,

phototroph

calcium-rich P.

used

C18

(calcified

called

consumption

Lyophilized

val

B12.

structures

clarified

human

by the

was

was

and

val

determined

indicate

alga,

are

to

partially

TLC,

results

calcareous

which

24)

of

marine CaCO3

cells),

and

data

coccolithophorid

unicellular

weight)

spectroscopy These

(Adapted,

19)

the

spectroscopy.

similar

B12-compound

from

The

determined

A B12-compound

tablets

purified

judged

were dry

chlorella

The

(22).

weight)

method

method.

the

(22).

methods dry

the al.

(23,

cells

for

supplement.

cells

have

(mainly

already

calcium

sup

plement).P. carterae B12, of

could

take

up

of which

was

converted

most B12

(25).

human

The

health

100g

dry

pound

weight

(26).

versed-phase

HPLC

compound those

3.

of

algal the

were

identical corrinoids

Bioavailability

to

cells

(26).

gel

cells 60

forms for

of B12

per

A

B12-com partially

TLC

and

purified

those

of authentic for

used

and

of the

inactive

of

exogenous coenzyme and

algal

silica

patterns

of the

into

125.4•}1.2ƒÊg

from The

accumulate

lyophilized

contained

purified

characterized

not

carterae

food

cell

was

P

and

C18

re

pink-colored B12,

but

humans.

B12-Compounds

from

Edible

Algae Dagnelie and

et

spirulina) status

Although

an

children

B12/d) baseline

(8)

and

tological

in

al.

reported

fermented of

of

the

consuming

values

that of

effect foods

B12-deficient

increase

indicated

the plant

mean

algae

on

the

children plasma

only the

of

plant

B12

corpuscular

was

B12

(noni hema

(Table

5).

concentrations

foods absorbed, volume

(0.1-2.7ƒÊg elevated (MCV)

de

328

WATANABEF et al.

Table 4. Rf values and retention times (min) of the purified spirulina B12-compounds , authentic B12,and cyanocobamides on TLCand HPLC.

a Solvent I:1 -butanol/2-propanol/water (10:7:10v/v) , Solvent II: 2-propanol/NH4OH (28%)/water (7:1:2v/v). b Isocratic:20% (v/v) methanol solution containing 1% (v/v) a cetic acid, Gradient: a linear gradient of methanol (5-70% , v/v) in 1% (v/v) acetic acid solution. (Adapted, with permission , from Ref. 19) Table bi

5.

B12 intake

and blood

values

before

and after changing

the diet by supplying

a B

otic children.

12 source

to B12-deficient

macro

a F , fish; (125ƒÊg/d); b Reference

B,

barley-malt and ranges:

W,

syrup; whole-meal B 12, 136-552

K,

Kombu

sourdough pmol/L;

algae;

M,

milk

products;

N,

noni

(Adapted,

with

algae;

Sp,

spirulina

algae;

Su,

B12

supplement

bread. MCV,

73-82fL.

teriorated further. Rauma et al. (27) reported that veg ans consuming nori and/or chlorella had serum B12 concentrations twice as high as those not consuming these seaweeds (Table 6). In a longitudinal study, how ever, six of nine vegans showed slow, but consistent de terioration of B12status over a 2-y observation period. It is entirely obscure how algal B12 sources can raise serum B12levels, but fail to improve hematological signs of B12deficiency. It has been speculated that edible algae contain substantial amounts of the corrinoids inactive for humans. Recent studies indicate that dried layers (non) and chlorella (Chlorella sp.) contain substantial amounts of true B12 (17, 18, 22), but the other edible algae (kombu, duce or arame) shown in Tables 5 and 6 contain none or only traces of B12, especially, most of

the

permission,

corrinoids

found

pseudo-B12 shows

that

(28).

the

of

competitive

tain

and and

true

clarify

but

not

bioavailability

layer P.

yezonensis,

deficient

rats

methylmalonic

be is

rats,

inactive

acid

the

intes

consider by and

both

IF-based

(Phorphyra

been

in the

of feeding

excreted

(16).

(17

lyophilized the When

substantial

(71.7•}20.2ƒÊmol/d)

yezo

reported

corrinoids

investigated which

in

detectable

layers have

effects

is

(19).

of B12

the were

sp.) Pseudo-B12

absorbed

method

purple tenera)

(19).

IF involved

can

method

Phorphyra

B12-deficient

the

and

(Spirulina

humans

pseudo-B12

lyophilized

B12,

the

spirulina

microbiological inhibition

Raw

8)

for to

B12

Moreover,

Lactobacillus

nensis

in

affinity

absorption

ably

Ref.

is inactive

moderate

tinal

of

from

to con , 29). To purple

layer

to

B12

9-wk-old amounts in

urine

.

Vitamin

B12-Compounds

from Edible Algae

329

Table 6. Serum B12concentrations and use of seaweeds in long-term adherents of a strict uncooked vegan diet ("living food diet").

a Arame -Kelbamare-Spirulina . (Adapted, with permission, from Ref. 27)

were

fed

a

diet

(10ƒÊg/kg excretion

and

nificantly

of

The

the

dried

layer

et

al.

(13)

human

raw

noni

dried

one

dried

nori

total

level

crease be

drying

to

65%

converted

properties

of the

The

dried

amounts

of

weight)

relative

the

that

excessive

sult

in

the

source

that of B12 papers

of the

among

of

of

the

daily, given

acid

40g

of

27%

of

is derived

which

the

from

levels

suggesting

in

that

B12-compounds

B12

by

on

(nori)

edible

intake

have

nori

(about

other

the

320g 40g

the

the

chemical

B12-compounds.

iodine

intake

the

About

information

layer

harmful

raw

non,

unidentified

dietary

indicate

dried

is no

purple

to

to

period.

harmful

There

the

change

given

weight)

of

the

from

not

were

test

in

into

process.

Several

the

in

by

although

were

B12-compounds, up

in

absorbed

did

volunteers

during

the

evaluated

B12

that

of dehydrated

determined

unidentified

that

(equivalent

when

daily

B12

may

basis

sig

from

rats.

volunteers

increased

unde levels

B12

been

excretion

tenera)

the

also

indicate

acid

female

on

of

demonstrated

(Porphyra

excretion

became

is significantly

in

layer acid

AdoB12)

has

results

methylmalonic

when

deficiency)

availability

tenera)

purple

methylmalonic

(especially

is bioavailable

Yamada urinary

B12

B12

These

dried

urinary

(P,

(30). layers

with

d,

layer

from

rats

purple

lent

index hepatic

purple B12

sults

20

increased.

other

the

for

(an

tectable

rats,

supplemented

diet)

algae

dried of

dried edible described

contained

lesser

6mg/100g (15),

layers dietary

layers

of

is unlikely iodine.

(nori)

dry

suggesting

are

to

These an

re re

excel

algae. that

spirulina

B12

may

not be bioavailable in mammals (6-8). Our studies have demonstrated that most B12-compound found in spir ulina tablets is pseudo-B12 inactive for humans (19). IF involved in the intestinal absorption of B12 strictly rec ognizes the structure of the B12molecule (30). Pseudo B12 has been reported to show moderate affinity to IF among various B12-compounds (28). Intestinal absorp tion and ileal content of pseudo-B12 24h after oral ad ministration of radioactive-labeled pseudo-B12 to rabbits have been shown to be about 13-21% of those of au thentic B12 (30). Intravenous injection of transcobal amin II-pseudo-B12 complex has revealed that plasma clearance and tissue distribution of pseudo-B12 are very similar to those of authentic B12,but urinary corrinoid excretion was slightly greater in pseudo-B12-injected rabbits than in authentic B12-injected rabbits (30). Although the adenosyl coenzyme form of pseudo-B12 has a Km1,000-fold higher for AdoB12-dependent mammalian methylmalonyl-CoA mutase than AdoB12 (31), pseudo-B12 can be fully active for human McB12 dependent methionine synthase under the experimen tal conditions used by Kolhouse et al. (32). These obser vations suggest that pseudo-B12 does not have the abil ity to act as a B12-antagonist in mammals. Herbert (33) reported that an extract of Spirulina contains two B12 compounds that can block the B12-metabolism. Van den Berg et al. (34) demonstrated that a spirulina-supple mented diet does not induce severe B12 deficiency in rats, implying that the feeding of spirulina may not in terfere with the B12metabolism. To clarify the B12-com

330

WATANABEFetal.

Table

7.

Hepatic

B12 concentrations

of B12-deficient

rats fed CN-B12 and P. carterae-supplemented

3)

diets.

4)

5) The mean values with different superscript letters within a column are significantly different (n=4 rats/group), p