A MANUAL FOR THE PROCESSING OF AGAR FROM GRACILARIA
by
Gertrudes A. Santos, Ph.D.1
1 Seaweed Processing Consultant, 1717 Ala Wai #1109, Honolulu, Hawaii 96815
The name agar originated from the Malay word “Agar-agar”, the local name in the Dutch East Indies for Eucheuma muricatum (spinosum) (Tseng, 1944) which was exported to China for more than a century. For the sake of simplicity agar-agar was shortened to just agar and is now accepted universally whether in the food and other industries or as culture media. The introduction of agar in bacteriology was achieved by a German housewife, Frau Hesse, who prepared the bouillon from agar-agar for her husband's bacterial cultures. The agar-agar came from Frau Hesse's mother who lived in America which was given to her by some friends who lived in Java. Dr. Walther Hesse was so excited about the efficiency of the new medium which his wife prepared, so he relayed his findings immediately to Dr. Robert Koch who was working at the time on the tubercle bacilli. In 1882, Koch reported on the tubercle bacilli and mentioned the new culture medium (Hitchens and Leikind, 1939). Hitchens and Leikind even suggested agar should be called Frau Hesse's medium in honor of the woman who discovered it. According to Tseng (1944) the agar-agar of Frau Hesse from Java could have been carrageenan from Eucheuma.
Tseng (1944) defines agar as the dried amorphous, gelatin-like, non-nitrogenous extract from Gelidium and other red algae, a linear galactan sulfate, insoluble in cold but soluble in hot water, a 1 to 2 percent solution of which upon cooking solidifies to a firm gel at 35° to 50° and melting at 90° to 100°.
The USP XVIII defines and describes agar as the dried hydrophilic colloidal substance extracted from Gelidium cartilagenium (Linne) Gaillon (Fam. Gelidiaceae), Gracilaria confervoides (Linne) Greville (Fam. Sphaerococcaceae) and related red algae. Unground agar usually occurs in bundles consisting of thin, membranous, agglutinated strips or in cut, flaked or granulated forms. It may be weak yellowish orange, yellowish gray to pale yellow, or colorless. It is tough when damp, brittle when dry. It is colorless or has a slight odor and has a mucilaginous taste. Powdered agar is a white to a yellowish-white or pale yellow, insoluble in cold water, but soluble in boiling water. When boiled with 65 times its weight of water for 10 minutes, with constant stirring, and adjusted to a concentration of 1.5 percent, by weight, with hot water, agar forms a clear liquid which congeals at 32° to 39° to form a firm resilient gel, which does not melt below 85°. Armisen and Galatas (1987) reported a wider range of 34° to 43° for the gelling temperarature. Actually according to our observation, the gelling temperature as well as the melting temperature of a 1,5 percent concentration of agar vary according to the seaweed source, the method of preparation and the purity of the sample. The gelling temperature of the agar sols ranges from 30° to 50° and the melting temperature from 82° to 92°.
The chemical nature of agar varies according to the seaweed source, the environment where the seaweeds grow and on the method of preparation of the agar. Meer (1980) recognized two types of agar, the Gelidium and Gracilaria agars. Bacteriological agar is prepared mostly from Gelidium and Pterocladia. Gelidium cartilagenium collected along the West Coast of North America (Mexico) (Durrant & Sanford, 1970) is the common source of bacteriological agar in the U.S.A.
Araki (1965) reported that agar of Gelidium amansii is a mixture of two different polysaccharides, one a neutral agarose which consists of alternating 1,3-linked 3-D-galactopyranose and a 1,4-linked 3,6-anhydro-α-L-galactopyranose and the other a charged agaropectin. Agaropectin contains galactopyranose residues with sulfate and other charged groups present in varying degrees in the molecule. Hirase (1957) reported the presence of pyruvic acid in the agar of Gelidium amansii as a ketal attached to the C4 and C6 of the l,3-linked-$-D-galactopyranose residues (Fig, 1). Araki (1965) also proved that the D-galactose residues are 6-0-methylated to certain degrees.
Duckworth and Yaphe (1971) fractionated Difco Bacto agar using DEAE-Sephadex A-50 and showed that agar consists of a complex mixture of polysaccharides having the backbone structure of alternating 1,3-linked β-D-galactopyranose and 1,4-linked 3,6-anhydro-α-L-galactopyranose charged in varying degrees by sulfate and pyruvate and a galactan sulfate. Agarose is then the mixture of agar molecules having the lowest charge content and the greatest gelling ability. The gel strength of agar decreases with an increase in sulfate and a decrease in 3,6-anhydro-α-L-galactose concentration (Yaphe, 1984).
Gelation occurs when a chain of macromolecules forms a network capable of entrapping the dispersing medium. Such a gel has that characteristic of having a composition approaching a pure liquid but may resemble a solid. It is an elastic colloid which actually retains the shape of the containing vessel even when removed from it. Gelation can be characterized (Whitney, 1977) by the time of gelation, the gelation temperature and the minimum concentration of the dispersed phase required for gelation. Agar gels (Rees, 1969) may contain as much as 99.9% water. Such gels exhibit strong syneresis (“weeping”) and behave like free water which can easily be separated by freezing and thawing. The stiffness of agarose gels may be due to the aggregation of the double helices forming a network phase which may contain as much as 100 parts of water for each part of agarose. Such a structural network would have relatively large voids through which large molecules and particles could diffuse. The aggregate in agarose gels may actually contain 10 to 104 double helices rather than what is shown in Fig. 2 (Arnott, et al, 1974). Formation of such a gel network is the property which makes agarose very useful in immunology, biotechnology and genetic engineering.
Fig. 1. a) Agarose, b) 6-galactan sulfate, and c) pyruvated agarose
Fig. 2. A schematic representation of the agarose gel network
Guiseley (1970) studied the relationship between methoxyl content and the gelling temperature of agarose using 50 agarose samples. Majority of the samples were prepared by Blethen's method (1966) from Chilean agar which could have been manufactured from Gracilaria lemanaeformis, the predominant Gracilaria species in Chile. The gelling temperature increases with an increase in methoxyl content of the agarose. Optical rotation studies on the solgel transition of agarose (Rees, 1972; Rees, et al, 1970; Rees and Scott, 1971; Arnott, et al, 1974) showed that agarose has a specific rotation of-44° at 589 nm with a left-handed double-helical conformation in contrast to the right-handed carrageenan. The temperature dependence of the optical rotation was determined for several agarose derivatives (Arnott, et al, 1974). The cooling curve for agarose which is almost free from any substituent is centered at almost 25° and has a transition width of 4°. Upon reheating the reverse transition occurred at 80° with very marked hysteresis but with no change in width (4°). The presence of a few 6-0-methyl substituents closed the hysteresis loop slightly, the cooking tansition moving to 30°. More 6-0-methyl groups displaced the cooking transition at a higher temperature and a substantial broadening of the hysteresis look (Fig. 3).
 | Fig. 3. Changes in optical rotation with the solgel transition for agarose samples. 1. Agarose, 2. Agarose with a few 6-0-methyl-D-galactose residues, and 3. Agarose with more 6-0-methyl-D-galactose |
Santelices and Doty (1989) reported that close to 5 000 tons of agar are processed annually from 25 000 to 30 000 tons of agar Gracilaria and the volume of the farmed production is not known but perhaps about 15 000 tons of dried seaweeds. In the manufacture of agar from Gracilaria not only one species is utilized so it is but appropriate to have an idea of the properties and chemical nature of the agars of the different species which have already been investigated.
The chemical studies on the majority of the Gracilaria agars were done by the Yaphe group. Polysaccharides from Gracilaria debilis, G. compressa, G. foliifera, G. domingensis, G. damaecornis and G. ferox were evaluated as sources of agar (Duckworth, et al, 1971). The agars obtained were different from one another as shown by the chemical and enzymatic hydrolysis and fractionation on DEAE-Sephadex A-50. Only G. debilis gave an agar of high gel strength. Agars containing 4,6-0-(l-carboxyethylidene)-D-galactose are always found in regions of the molecule that are low in sulfate. Replacement of 3,6-anhydro-L-galactose sulfate causes kinks in the helix thus forming an agar of lower gel strength. If the sulfate groups are in C6 of the L-galactose molecules, alkali treatment converts L-galactose 6-sulfate into 3,6-anhydro-L-galactose which causes an increase in gel strength of the agar. The acid hydrolyzates of G. foliifera, G. damaecornis, G. domingensis and G. ferox agars contain 6-0-methyl-D-galactose and 4-0-methyl-L-galactose. The amount of 3,6-anhydro-L-galactose increased (except that of G. domingensis) after alkali treatment and no significant change in the 6-0-methyl-D-galactose content. The pyruvic acid values which vary in the agars of the six species are not affected by alkali treatment. Low pyruvic acid concentration favors higher 6-0-methyl-D-galactose values. Before alkali treatment only G. debilis agar (sulfate, 3.4%) has high gel strength. The sulfate content of G. compressa agar decreased after alkali treatment but it did not gel, may be because of the high pyruvic acid value. The sulfate groups of G. ferox, G. damaecornis and G. demingensis were alkali-stable indicating that such groups are not located at the C6 of the L-galactose residues. The agars of the above six species discussed have gel strengths not comparable to the agars of Gelidium cartilagenium and G. sesquipedale which are known commercial sources of agar and so such Gracilaria agars can only be of application in the food industry. Agars of Gracilaria cf. verrucosa and Pterocladia capillacea from the Mediterranean were compared (Friendlander, et al, 1981). Fractionation on DEAE-Sephadex A-50 gave three fractions, namely, neutral agarose, sulfated agarose and a galactan sulfate. Alkali treatment increased the neutral agarose content for both agars. Agar from P. capillacea had higher agarose content and lower sulfate than the agar of G. cf. verrucosa and the former agar after alkali treatment had a structure approaching that of theoretical agarose.
Agars from Gracilaria verrucosa, G. tenuistipitata, G. blodgettii and G. eucheumioides from China were studied (Ji Minghou, et al, 1985) by fractionation on DEAE-Sephadex A-50 and 1 3 CNMR spectroscopy. The latter analysis showed L-galactose 6-sulfate as a minor constituent of the agars of G. verrucosa, G. tenuistipitata and G. blodgettii but not sensitive enough to detect the other sugars contributing to the charge density of the agarose molecules. 6-0-methyl-D-galactose was found as a minor sugar of the agarose from G. verrucosa and G. tenuistipitata and 2-0-methyl-3,6-anhydro-L-galactose as the major component of that of G. eucheumioides. The yield and quality of agar obtained from Gracilaria spp. collected from Taiwan and Micronesia were studied (Nelson, et al, 1983). G. edulis from Taiwan gave the highest agar yield while G. lichenoides from Micronesia gave the highest gel strength. Young algal tissues of G. tikvahiae (Craigie and Wen, 1984) synthesize agarose polymers with low methoxyl and high L-galactose-6-sulfate while the methylated agars are formed more in the older tissues and at higher temperature. Genetic improvement in clones of G. tikvahiae produced a mutant (MP-40) which gives an agar of high gel strength (over 1 000 g/cm2 , conc. 1%) will be of value as raw material in the agar industry. G. lemanaeformis and G. verrucosa the two most important seaweeds of Chile (Kim, 1970) needed only 3-5% NaOH, heating at 95° for 60-90 minutes to yield an agar of high gel strength (600 g/cm2 ) which is in great demand in the food industry especially in the U.S.A. and commands a price of $12 500/ton in 1986 (Santelices and Doty, 1989). Enzymatic hydrolysis showed the presence of 6-0-methyl-D-galactose in the hydrolytic product. Another methoxylated agar is obtained from G. secundata (Brasch, et al, 1983) of New Zealand which gives a gel strength of 495 g/cm2 . Seven species of the genus Gracilaria from the Philippines have been studied (Hurtado-Ponce & Umezaki, 1988; Santos & Doty, 1978), G. arcuata, G. coronopifolia, G. edulis, G. eucheumioides, G. salicornia, G. verrucosa and G. sp. G. arcuata, G. salicornia and G. eucheumioides agars were investigated for possible raw material in the manufacture of agar in the country (Santos and Doty, 1978). The gel strength of the agar G. arcuata and G. salicornia were relatively high but although the G. eucheumioides agar has low gel strength it has a good consistency and mouth-feel that it could be quite applicable for fruit jellies and other jellied desserts.
The genus Gracilaria is widely distributed and can be found in temperate as well as in tropical countries. It is interesting to note that Gracilaria species in temperate countries yield agars of higher gel strength than the agars of the tropical species. The plausible explanation is that the Gracilaria species in cold countries grow very much slower giving the sugar molecules time to polymerize and form bigger molecules than the tropical species. Some tropical species yield agars of high gel strength after being subjected to the proper treatment. The Gracilaria species of Thailand were surveyed and the agar analyzed (Edward, et al, 1982; Tarn and Edwards, 1982; Edwards and Tam, 1984). In 1986, 1987 and 1988 another team made a renewed survey and screening of the Thai Gracilaria species. The Gracilaria samples were not treated with alkali before extraction but was done on the agar at room temperature (Chinadit and Chandrkrachang, 1986). The gel strength obtained for the agars of Polycavernosa changii, P. fastigiata and P. fisherii were 714, 1 100 and 947 g/cm2 , respectively and the sulfate values were 0.07, 0.08 and 0.94%. Such agars can be considered of bacterial grade (Chandrkrachang, 1989).
The agar from Gracilaria cylindrica (now identified as Polycavernosa chang ii) has high gel strength, low gelling temperature and the melting temperature approaching those of Gelidium agars (Doty, et al, 1983) so the agar was fractionated using the method of Blethen (1966) to obtain agarose (Santos and Doty, 1983). The separation of agarose directly from the seaweed gave a better yield than when agar was used for fractionation. The gel strength of the agarose obtained range from 747–950 g/cm2 of a 1% gel concentration and sulfate content of from 0.17 to 0.42% comparable to the sulfate values of commercial agarose. The study showed that there are some tropical Gracilaria species which could be possible raw material for the manufacture of agar not only for the food industry but also for the preparation of bacteriological grade agar and even agarose.
