**4. Assessment of feed stability in water**

Water stability of feed is of paramount importance in the manufacture of aquaculture diets. Water stability is greatly influenced by the properties of binders, although the ingredients themselves have a direct influence on the characteristics of the binders (Dominy & Lim, 1991). Durazo-Beltran and Viana (2001) report difficulty in bonding the constituents of fish silage to generate pellets for abalone farming, probably because of the high content of hydrolyzed protein. Although the water stability of aquatic feed is a major concern of the aquaculture industry, there is no standard method to determine feed water stability. It is usually estimated by the method of dry matter weight loss, according to which a certain amount of feed, usually in the form of pellets, is placed in a water containing beaker and allowed to stay for a variable length of time with occasional shaking. Pellets are then decanted and dried. Variability can be found in literature about the amount of water in which pellets are placed, temperature, time intervals, methods in pellet drying after incubation in water (Ruscoe et al., 2005; Johnston & Johnston, 2007; Palma et al., 2008; Orire et al., 2010). A slight modification of the dry matter weight loss is suggested by the American Feed Industry Association (AFIA 1999 in Liu et al., 2008). According to the AFIA water stability is expressed as a percentage of pellets retained on a wire mesh sieve after immersion in a shaking water bath for a period on time after which the pellets are retrieved and dried. Obaldo et al. (2002) developed three methods for measuring the water stability of shrimp pelleted feeds: a static method, a horizontal method and a vertical shaking method. Several factors were taken into consideration (water and pellet agitation, direction of agitation, water temperature, salinity, leaching container, filtration medium) among which temperature and water salinity influenced the rates of dry matter retention the most. Higher water temperatures and lower levels of salinity produced more leaching and lower dry matter retention. In general, the three methods appeared to provide an accurate and precise means of determining the water stability of shrimp feeds. The employment of water shaking is particularly interesting inasmuch as it may mimic the actual indoor and outdoor culture conditions and can be usefully adjusted to variable conditions.

Water stability is expressed as a percentage of immersed diet weight/initial sample weight (Liu et al., 2008; Orire et al., 2010), or as percent loss of dry matter (%LDM) calculated as percent difference in sample weight (minus the initial diet moisture) after reweighing (Fagbenro & Jauncey, 1995; Johnston & Johnston, 2007), or as percentage of dry matter remaining (%DMR) calculated with the formula %DMR = Wo X (1-M) – Wt/Wo X (1-M) X 100 where Wo = pellet weight as-fed, Wt = weight after immersion and drying, and M = moisture content of diet as a proportion (Ruscoe et al., 2005).

Caltagirone et al. (1992) suggest a subjective parameter, to be taken into account together with the %LDM, in the evaluation of binder efficiency, that is the consistency of food, defined according to the following scale: 1) inconsistency (feed completely disintegrates on removal from the water; 2) weak consistency (feed partially disintegrates on removal from the water; 3) good consistency (feed does not disintegrate on removal from the water but disaggregates after the application of a weak pressure; 4) very good consistency (feed does not disintegrate on removal from the water and resists weak pressure). According to these evaluation parameters Caltagirone et al. (1992) evaluated six binding agents after 48 hours of water immersion: agar-agar, cellulose, CMC, guar gum, gelatine and sodium alginate. Cellulose and CMC were unsuitable as binders even at strong concentrations; sodium alginate retained a good consistency but showed a high loss of weight, while guar gum, gelatine and agar-agar appeared to be good binders in terms of pellet stability. Pearce et al. (2002b) tested prepared feed with different binders (gelatin, guar gum, sodium alginate, and corn starch), evaluating their effectiveness according to the above described methodology up to 216 hours of immersion in sea water. Guar gum and starch-based pellets became inconsistent after 24 hours of immersion; sodium alginate retained a good consistence for up to 48 hours, while gelatine remained firm and intact even after 216 hours in sea water. Again with the same evaluation procedure Mortensen et al. (2004) tested the stability of a pellet in which gelatine from fishskin was strengthened with the enzyme transglutaminase. Feed retained a good consistency for up to 7 days in sea water and the loss in dry matter, even increasing with time, reached 17.6%, with respect to 60.1% of the salmon feed used as control.

In our laboratory, we adopted a rather different methodology, according to which pellet water stability was analyzed by monitoring the diameter of the released particles in water over progressive time intervals, employing a Low Angle Laser Light Scattering Technique. (Volpe et al., 2008; Coccia et al., 2010; Fabbrocini et al., 2011). The diameter of particles

American Feed Industry Association (AFIA 1999 in Liu et al., 2008). According to the AFIA water stability is expressed as a percentage of pellets retained on a wire mesh sieve after immersion in a shaking water bath for a period on time after which the pellets are retrieved and dried. Obaldo et al. (2002) developed three methods for measuring the water stability of shrimp pelleted feeds: a static method, a horizontal method and a vertical shaking method. Several factors were taken into consideration (water and pellet agitation, direction of agitation, water temperature, salinity, leaching container, filtration medium) among which temperature and water salinity influenced the rates of dry matter retention the most. Higher water temperatures and lower levels of salinity produced more leaching and lower dry matter retention. In general, the three methods appeared to provide an accurate and precise means of determining the water stability of shrimp feeds. The employment of water shaking is particularly interesting inasmuch as it may mimic the actual indoor and outdoor culture

Water stability is expressed as a percentage of immersed diet weight/initial sample weight (Liu et al., 2008; Orire et al., 2010), or as percent loss of dry matter (%LDM) calculated as percent difference in sample weight (minus the initial diet moisture) after reweighing (Fagbenro & Jauncey, 1995; Johnston & Johnston, 2007), or as percentage of dry matter remaining (%DMR) calculated with the formula %DMR = Wo X (1-M) – Wt/Wo X (1-M) X 100 where Wo = pellet weight as-fed, Wt = weight after immersion and drying, and M =

Caltagirone et al. (1992) suggest a subjective parameter, to be taken into account together with the %LDM, in the evaluation of binder efficiency, that is the consistency of food, defined according to the following scale: 1) inconsistency (feed completely disintegrates on removal from the water; 2) weak consistency (feed partially disintegrates on removal from the water; 3) good consistency (feed does not disintegrate on removal from the water but disaggregates after the application of a weak pressure; 4) very good consistency (feed does not disintegrate on removal from the water and resists weak pressure). According to these evaluation parameters Caltagirone et al. (1992) evaluated six binding agents after 48 hours of water immersion: agar-agar, cellulose, CMC, guar gum, gelatine and sodium alginate. Cellulose and CMC were unsuitable as binders even at strong concentrations; sodium alginate retained a good consistency but showed a high loss of weight, while guar gum, gelatine and agar-agar appeared to be good binders in terms of pellet stability. Pearce et al. (2002b) tested prepared feed with different binders (gelatin, guar gum, sodium alginate, and corn starch), evaluating their effectiveness according to the above described methodology up to 216 hours of immersion in sea water. Guar gum and starch-based pellets became inconsistent after 24 hours of immersion; sodium alginate retained a good consistence for up to 48 hours, while gelatine remained firm and intact even after 216 hours in sea water. Again with the same evaluation procedure Mortensen et al. (2004) tested the stability of a pellet in which gelatine from fishskin was strengthened with the enzyme transglutaminase. Feed retained a good consistency for up to 7 days in sea water and the loss in dry matter, even increasing with time, reached

In our laboratory, we adopted a rather different methodology, according to which pellet water stability was analyzed by monitoring the diameter of the released particles in water over progressive time intervals, employing a Low Angle Laser Light Scattering Technique. (Volpe et al., 2008; Coccia et al., 2010; Fabbrocini et al., 2011). The diameter of particles

conditions and can be usefully adjusted to variable conditions.

moisture content of diet as a proportion (Ruscoe et al., 2005).

17.6%, with respect to 60.1% of the salmon feed used as control.

released by pellets are continuously monitored over time, thus providing a time-course indication about the water stability as a function of the released particle diameter inasmuch as pellets that disaggregate into small particles are less stable in water than pellets that disaggregate into particles of a larger diameter. According to this evaluation methodology the binder capacity of two algal polysaccharides (alginate and agar) and one polysaccharide from fruit (pectin) revealed that pectin showed better water stability than alginate and agar (Volpe et al., 2008). When a coating was added to the microbound pellets water stability was improved. Indeed, microbound pellets disaggregated into particles of a derived diameter almost twice as large as pellets manufactured without coating (Coccia et al., 2010). Palma et al. (2008) report that pellets without binder, but with a microcoat of lignosol behaved less well in water than pellets made with lignosol as a binder, indicating that the presence of an internal binder is necessary to provide firmness to the feed. Using the Low Angle Laser Light Scattering Technique 3% and 6% agar-based round pellets specifically designed for sea urchin feeding, proved to have good water stability with a significant increase in the diameter of the released particles recorded only after 5 days of immersion in sea water for 3% agar- and 6 days for 6% agar-based pellets (Fabbrocini et al., 2011).

Due to the ample variability in feed ingredients, percentage of binders included and manufacture technology, it is impossible to come to the conclusion that a certain binder is better than another with respect to its water stability performances. Moreover, experimental outcomes are reported in literature as relative, that is a binder behaving better than another under specific conditions, making impossible an objective evaluation of binder performances. Nonetheless, in table 2, literature data on feed water stability are reported.



1 The reported outcome refers to the maximum time interval tested . > means that the binder(s) on the left perform better that the binder(s) on the right. \* a commercial binder.

Table 2. Effect of binders on feed water stability.

### **5. Assessment of nutrient leaching**

16 Aquaculture

OUTCOME1 REFERENCE

Mortensen et al., 2004

Pearce et al., 2002b

Liu et al., 2008

Ruscoe et al.,

2005

al., 2008

Knauer et al., 1993

Volpe et al., 2008

Coccia et al.,

Fabbrocini et al., 2011

2010

Alginate>gelatin Valverde et

Gelatin + TG > commercial salmon

Gelatin > sodium alginate > guar gum,

corn starch

Carrageenan> alginate> gelatin>

Carrageenan , CMC> agar, gelatin

Agar/gelatin >

alginate>gelatin

Pectin>agar, alginate

Chitosan, pectin> agar, alginate

agar 6% > agar 3%

agar>

feed

CMS

% OF BINDER EMPLOYED

gelatin+0.13% transglutaminase

sodium alginate, corn starch

2% carrageenan, CMS, alginate,

carrageenan, CMC, agar, gelatin

hexametaphosphate or D-gluconic acis as

30% alginate

sequestrants 9 and 13% agar 6 and 15% gelatin 20% agar/gelatin

mixture

24 hours 2.5% Agar, alginate, pectin

24 hours 1% chitosan, pectin,

left perform better that the binder(s) on the right. \* a commercial binder.

agar, alginate

1 The reported outcome refers to the maximum time interval tested . > means that the binder(s) on the

3% agar, 6% agar No differences

24 hours 2% alginate plus 1% sodium

(TG)

gelatin

216h Guar gum, gelatin,

WATER STABILITY TEST

Dry matter weight loss

Dry matter weight loss

Dry matter retention

Dry matter remaining after immersion in

Variation in weight after soaking in water

Variation in weight after soaking in water

Long Angle Laser Light Scattering Technique

Long Angle Laser Light Scattering Technique

Long Angle Laser Light Scattering Technique

96h 144h

Table 2. Effect of binders on feed water stability.

water

MAX TIME INTERVAL CONSIDERED

168h Fish-skin

30 min 7% starch

180 min 3 and 5%

24 hours 10% Gelatin

Binder type may affect feed stability in water, and by leaching of attractant molecules may determine attractiveness of microbound diets. Partridge and Southgate (1999) report that leaching increased with decreasing binder concentration. O'Mahoney et al. (2011) in a study aimed at investigating the characteristics of new substances to be employed as binders for abalone farming, found no dry matter leaching after up to 4 days of water immersion of feed containing glucomannan-xanthan gum. In spite of the importance of assessing nutrient leaching there is still no standard method to determine it. Some authors report nutrient leaching and dry matter loss after water immersion as synonyms. Other authors use instead different procedures to determine nutrient leaching such as radiolabelled substances included into the feed, internal markers, detection of specific nutrients, protein or lipid. Genodepa et al. (2007) included 14C-labelled rotifers in the feed and tested the effect of zein, agar, alginate and carrageenan on nutrient leaching. They report that leaching occurred primarily within the first 30 minutes of immersion and longer immersion periods (up to 240 min) only resulted in a very modest increase in leaching rates. The leaching rate of zeinbound diets was significantly lower than that of agar-, alginate- and carrageenan-bound diets. On the contrary, Fagbenro and Jauncey (1995) report that the leaching of total protein content was very low, after 10 min immersion in water of pellets made with 3% CMC, corn starch, guar gum or wheat gluten. Accordingly, Simon (2009) found no significant difference in chromic oxide (as internal standard) and nutrient concentrations after 1 hour immersion in water of pellets containing gelatine (8%), alginate (7%), or agar (8%). Kovalenko et al. (2002) evaluated dietary water soluble vitamins riboflavin and thiamine levels before and after water immersion of feed containing 5.38% of alginate and report a substantially high loss of both vitamins after 90 min of immersion, with the greatest percent loss within 1 to 15 min interval. Falayi et al. (2006) evaluated total protein and lipid content in pelleted feed samples containing 10% wheat grain starch and cassava tuber starch as both binding agents and carbohydrate source. They found that after 1 hour of water immersion protein retention dropped to about 80% while lipid retention was about 97-95%.
