A Biochemical Study of Fasting, Subfeeding, and Recovery Processes in Yellow‐Legged Gulls

TL;DR: The gulls quickly recovered body mass during the refeeding period, but while some plasma substances quickly reached their initial values, others showed many changes before the end of the experiment, which could reflect a process of metabolic restabilization.

Abstract: An investigation of the effects of fasting, subfeeding, and refeeding on plasma biochemistry was carried out on 22 captive yellow‐legged gulls Larus cachinnans Pallas. These birds showed the same fasting endurance model described in other species, but with an important decrease in glucose plasma concentration and very great differences between individuals when reaching the deterioration limit, suggesting a moderate physiological adaptation to long periods of fasting. A different model was proposed in subfed gulls in relation to fasted gulls, based on lipid and protein use, which could be reflected by changes in nitrogen wastes and triglyceride levels in this experiment. Thus, the subfed gulls might use protein directly from the diet as an energy source, thereby reducing the use of fat stores. The gulls quickly recovered body mass during the refeeding period, but while some plasma substances quickly reached their initial values, others showed many changes before the end of the experiment, which co...

Summary

Introduction

• Many studies have been developed on the physiological response of birds when enduring food restriction.
• The second phase is a period of steady body mass and metabolism.
• In summary, this study has three objectives: (1) to know whether a seabird, the yellow-legged gull Larus cachinnans Pallas, uses the above-cited classic model of resource allocation during starvation, as shown in other bird species (thus, the authors will analyze differences with respect to the described pattern; Fig. 1).
• Moreover, (2) this species was chosen in order to compare changes during periods of moderate food restriction with periods of absolute fasting.

Experimental Procedure

• These birds were transported to the wildlife recovery center La Cañada de los Pájaros (Huelva, Spain) and were housed in individual cages ( m).
• The study was performed with the per-4 # 4 # 4 mission of the appropriate authorities, avoiding any damage to the birds.
• For 2 wk, sardines (Sardina pilchardus Walbaum) were provided ad lib.
• Variable total body-mass loss was defined as the proportion of body-mass loss regarding weight at the beginning of the experiment.
• Nevertheless, after 8 d, three gulls from the fasting group died on the same day, without symptoms, and their lesser total bodymass loss the day before (15%) was finally established as the final limit before placing the gulls in the recovery period, refeeding them with food ad lib.

Blood Extraction and Weighing Procedures

• Blood samples were taken from the humeral vein (2.5 mL) every 2 d throughout the experiment, always before feeding, at the middle of the day (1100–1500 hours) to avoid any variation in blood chemicals caused by the circadian rhythm (Ferrer This content downloaded on Fri, 15 Feb 2013 07:10:25 AM.
• All use subject to JSTOR Terms and Conditions 1993).
• Winged infusion sets (Valu-Set, Becton Dickinson, Sandy, Utah) were used to prevent damage to the veins, applying them on alternate wings each time.
• Blood sampling was done immediately after capture.
• The gulls were weighed after blood collection with a dynamometer (Pesola; accuracy 5 g).

Data Analysis

• Mean values of parameters were tested for differences between groups on the same day or in the same mass-loss rank by the Mann-Whitney U-test for independent samples.
• Within-group variations were tested with Wilcoxon matched pairs signedranks test.
• These nonparametric tests were used as a precaution since, as a result of small sample sizes in some analyses, normal distribution could not be ascertained for all parameters.
• The experiment effects were examined with repeated-measures ANOVA, where the treatment (fasting or subfeeding) was used as a factor (between-subject effect) and the samples obtained from the same bird throughout the experiment were used as repeated measures (within-subject effect).
• Moreover, repeatedmeasures ANOVA was used to analyze changes in mass or biochemical parameters in each group separately.

Initial and Final Values in the Deterioration Period

• That day, there were no differences among the three groups or between sexes (MannWhitney: in all parameters).
• The four birds from theP 1 0.05 control group did not show significant variations in total bodymass loss (repeated-measures ANOVA: , )F p 1.07 P p 0.4110, 30 and plasma biochemical traits (always ) throughout theP 1 0.05 experiment and are not used in the rest of the statistical analyses.
• Urea, uric acid, cholesterol, glucose, and alkaline phosphatase changed in both groups.
• There were no significant differences between these two groups the last day of the deterioration period regarding all the parameters, but inorganic phosphorus, calcium, and magnesium showed a tendency toward higher values in the fasting group (MannWhitney: ).P ! 0.12.

Weight and Biochemical Changes with Respect to the Classic Model

• In order to explain the changes in body mass throughout the deterioration period, the authors analyzed daily body-mass loss and total body-mass loss during the fasting phases according to the classic model (Fig. 2).
• For both variables, data of the first four sampling days from the beginning and, separately, data of the last four sampling days to reach the final limit of the deterioration period were analyzed ( in each group) in order to equil-n p 9 ibrate the sample size between the groups.
• A descent in daily body-mass loss between the second and the fourth day (proposed phase 1) were not significant in either group (Wilcoxon: , in both groups).
• All use subject to JSTOR Terms and Conditions gulls than in the restricted gulls (Mann-Whitney: ,Z p 1.99 ; see Fig. 2).
• P p 0.047 Concerning the biochemical parameters, the authors focused on uric acid and triglycerides as representatives of nitrogen residuals and fat use, respectively, synchronizing newly recorded data with respect to the first and the last day of the deterioration period (in Fig. 3, backward from last day).

Weight Changes in the Recovery Period

• Changes in total body-mass loss during the recovery period were used to explain the return of their gulls to the initial body weight (Fig. 4).
• The values at last day of the deterioration period were significantly higher than the values at the first sampling day of the recovery period (Wilcoxon; fasting group: Z p , ; restricted group: , ).

Changes in Biochemical Variables throughout the Experiment

• The first one was the individual differences in the number of days to attain the deterioration limit (commented on above), which prevents changes from being analyzed with respect to a chronological order.
• With the aim of avoiding these problems, the changes were analyzed using a general linear model (Table 2), which allowed testing of the linear relationship of each biochemical trait with the proportion of total body-mass loss, but not with time.
• All use subject to JSTOR Terms and Conditions 15%–25% ranks as a consequence of the sampling interval.
• The effects of the treatment in this period were significant in triglycerides (cited above), creatinine, and amylase concentrations (see Fig. 5).

Fasting Model

• One of the objectives of this study was to test whether the changes during fasting in yellow-legged gulls could be adjusted to the classic model with three periods (see Fig. 1) used by different authors (e.g., Cherel and Le Maho 1985; Boismenu et al.
• Nevertheless, the next two phases were similar to the classic model, with a clear increase in daily bodymass loss the last day (phase 3).
• Changes in triglyceride concentrations in this group were similar to those changes of free fatty acids and b-hydroxybutirate observed in the classic model (Groscolas 1986; Cherel et al.
• F values from variance analysis (general linear model).
• Their results suggest that adult yellow-legged gulls show a moderate adaptation to prolonged fasting, with a biochemical pattern very similar to the classic model but with great differences between individuals.

Strategy of Subfed Gulls

• The second objective of this study was to compare the changes during periods of moderate food restriction with respect to periods of absolute fasting.
• All use subject to JSTOR Terms and Conditions in addition to the product from structural or muscular protein catabolism.
• Triglycerides were also different among the treatments in the deterioration period.
• Some studies have shown a regular maintenance of total protein levels during fasting (Jeffrey et al.

Changes in Other Biochemical Traits throughout the Experiment

• Other substances that were not used to explain the deterioration models could clarify the significance of the exposed results.
• Nevertheless, in this work a regular decline in glycemia occurred in both groups (see Fig. 5).
• In other gull species cholesterol remained stable during fasting, pathological processes, and migration (Jeffrey et al.
• An increment in diet protein produces a rise in the synthesis of cholesterol in the liver and intestines but a reduction in plasma concentration, while a lowprotein diet causes a high plasma cholesterol level, reducing its excretion (Leveille and Sauberlich 1961; Yeh and Leveille 1972; Lewandowski et al. 1986).

Recovery Period

• The third objective of their study was to analyze the recovery process in a bird species.
• After this jump, body weight continued to be more or less stable but below the initial level of the experiment.
• In their study, many plasma parameters quickly recovered their initial values after refeeding (see Fig. 5), while other traits showed many changes before the experiment ended (see Fig. 5).
• This could mean that gulls did not reach their prefasting steady metabolism even though they quickly recovered their initial body mass.
• Production of this enzyme decreases during fasting in chickens and increases when food is restored (Kokue and Hayama 1972).

Conclusions

• Yellow-legged gulls showed a moderate physiological adaptation to extended fasting but the same model of biochemical changes in plasma that other more adapted species showed.
• Subfed birds seem to use lipids and proteins in a different way than fasted birds, probably suffering a lesser impact on their health.
• The gulls quickly recovered body mass during the refeeding period, but whereas some plasma substances quickly reached their initial values, others showed many changes before the experiment end, which could reflect a process of metabolic restabilization.
• Some differences in the results reported here with respect to a recent study on fasting in herring gulls can be explained by methodological interferences.

Figures

Figure 2. Changes in daily body-mass loss (body-weight proportion with respect to the previous sampling; top) and total body-mass loss (body-weight proportion in relation to day 0 of the experiment; bottom) throughout the deterioration period (food restriction). Individual data synchronized, starting forward from the first or backward from the last day of this period. Solid squares and continuous lines represent gulls in the fasting group, and open triangles and dotted lines represent the restricted gulls group. Proposed phases marked over graphics. ( ; , both groups.)Means SE n p 9

Figure 3. Comparisons between the fasting and restricted groups in urea (left) and triglyceride (right) plasma concentrations throughout the deterioration period (food restriction). Individual data synchronized backward from the last day of this period. Solid squares and continuous lines represent gulls in the fasting group, and open triangles and dotted lines represent the restricted gulls group. Proposed phases marked over graphics. ( ; , both groups.)Means SE n p 9

Figure 1. Ideal pattern of changes in plasma composition and body mass during fasting in bird species (classic model). Daily body-mass loss is the body-weight proportion with respect to the previous fasting day (the scale on Y-axis is only illustrative).

Figure 4. Variation in total body-mass loss (body-weight proportion with respect to the first experimental day) backward from the last day of the deterioration period (divided by the vertical line) and throughout the recovery period. Solid squares and continuous lines represent gulls in the fasting group, and open triangles and dotted lines represent the restricted gulls group. ( ; fasting group: ; re-Means SE n p 5 stricted group: .)n p 8

Figure 5. Variation on different plasma biochemical parameters ( ) in relation to total body-mass loss (body-weight proportion withmeans SE respect to the first experimental day). Proportions are divided in ranks of 5% separately in the deterioration period (food restriction) and the recovery period (food ad lib.). Ranks from 1% to 5% are represented with number 5, and so on. Because these are not the cases, they do not show intermediate ranks between the last rank of the deterioration period and the first of the recovery period. Solid squares represent gulls in the fasting group, and open triangles represent the restricted gulls group. (Deterioration period: , both groups; recovery period:n p 9 , fast group; , restricted group.)n p 5 n p 8

Table 1: Mean values ( SE) of plasma parameters the first and the last day of the deterioration period, differentiated by groups

Full text

A Biochemical Study of Fasting, Subfeeding, and Recovery Processes in Yellow‐Legged Gulls
Author(s): CarlosAlonso‐Alvarez and MiguelFerrer
Reviewed work(s):
Source:
Physiological and Biochemical Zoology,
Vol. 74, No. 5 (September/October 2001), pp.
703-713
Published by: The University of Chicago Press
Stable URL: http://www.jstor.org/stable/10.1086/322932 .
Accessed: 15/02/2013 07:10
Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at .
http://www.jstor.org/page/info/about/policies/terms.jsp
.
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of
content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms
of scholarship. For more information about JSTOR, please contact support@jstor.org.
.
The University of Chicago Press is collaborating with JSTOR to digitize, preserve and extend access to
Physiological and Biochemical Zoology.
http://www.jstor.org
This content downloaded on Fri, 15 Feb 2013 07:10:25 AM
All use subject to JSTOR Terms and Conditions

703
A Biochemical Study of Fasting, Subfeeding, and Recovery Processes
in Yellow-Legged Gulls
Carlos Alonso-Alvarez*
Miguel Ferrer
Department of Applied Biology, Estacio´n Biolo´ gica de
Don˜ana, Pabello´n del Peru´ , Consejo Superior de
Investigaciones Cientificas, Avenida de Marı´a Luisa s/n 41013
Sevilla, Spain
Accepted 5/22/01
ABSTRACT
An investigation of the effects of fasting, subfeeding, and re-
feeding on plasma biochemistry was carried out on 22 captive
yellow-legged gulls Larus cachinnans Pallas. These birds showed
the same fasting endurance model described in other species,
but with an important decrease in glucose plasma concentration
and very great differences between individuals when reaching
the deterioration limit, suggesting a moderate physiological ad-
aptation to long periods of fasting. A different model was pro-
posed in subfed gulls in relation to fasted gulls, based on lipid
and protein use, which could be reflected by changes in nitrogen
wastes and triglyceride levels in this experiment. Thus, the sub-
fed gulls might use protein directly from the diet as an energy
source, thereby reducing the use of fat stores. The gulls quickly
recovered body mass during the refeeding period, but while
some plasma substances quickly reached their initial values,
others showed many changes before the end of the experiment,
which could reflect a process of metabolic restabilization. These
results contribute to a better knowledge of fasting, subfeeding,
and refeeding processes in birds and can be added to a recent
study about fasting in gulls.
Introduction
Many studies have been developed on the physiological re-
sponse of birds when enduring food restriction. These works
suggest three different phases based on changes in body weight
and plasma biochemistry (see Fig. 1), which we will name the
“classic model” (e.g., Le Maho et al. 1981; Boismenu et al. 1992;
* Corresponding author; e-mail: alonso@ebd.csic.es.
Physiological and Biochemical Zoology 74(5):703–713. 2001. 䉷 2001 by The
University of Chicago. All rights reserved. 1522-2152/2001/7405-99138$03.00
Handrich et al. 1993b). In a first phase (phase 1), body weight
shows an important reduction in a short interval of time,
whereas the second phase (phase 2) presents a slow and stable
daily weight descent, during a more or less long period, de-
pending on the species. The final phase reveals a quick and
strong increase in body-mass loss to reach a critical level close
to death (phase 3). The fasting-adapted species relies on fat as
the primary energy source during fasting periods, spares pro-
tein, and relies primarily on protein only when fat reserves are
depleted (see, e.g., Cherel et al. 1988a). Protein is spared due
to its key role in body structure and muscle function and as
enzymes (Felig 1979; Castellini and Rea 1992). In this model,
plasma levels of residuals from protein catabolism (urea and
uric acid) decrease during phase 1, maintain a stable low con-
centration or a slow increment in phase 2, and rise suddenly
to reach the highest levels during phase 3 (use of structural
proteins as energy source), which might indicate the bird’s
death. Changes in excretion of these nitrogen residuals are in
fact parallel to those observed in daily body-mass change. How-
ever, in birds, as in mammals, by far the largest reservoir of
body fuel is in the form of fat, stored as triglycerides (Cherel
et al. 1988a; Castellini and Rea 1992). In the classic model, free
fatty acids and ketone bodies increased in blood plasma in phase
1 as a consequence of triglycerides breaking down, which re-
flects their use as an energy source; they maintain high con-
centrations in phase 2 and abruptly decrease in phase 3 (ex-
haustion of fat stores). Plasma triglycerides steadily decrease
during fasting (Fig. 1), though they were not described on the
basis of the classic model (Jenni-Eiermann and Jenni 1994).
Finally, glucose maintains its concentration, only decreasing in
phase 3 (Cherel et al. 1988b; Boismenu et al. 1992), reflecting
its importance in birds’ metabolism. In fact, this carbohydrate
is a critical fuel for the central nervous system, and its circu-
lating concentration is tightly regulated (Castellini and Rea
1992).
On the other hand, the physiological means for supporting
absolute fasting could be different from a limited food restric-
tion, a phenomenon probably more extended in species with
very diverse food resources, such as gulls (Cramp and Simmons
1983; Munilla 1997). Comparison between the effects of a re-
duced diet in relation to the effects of fasting in wild species
of birds has not been documented in bibliographies, as far as
we know. This kind of study might help develop understanding
of the usual physiological state of many individuals during
subfeeding periods.
In contrast with the studies on fasting, the refeeding period
This content downloaded on Fri, 15 Feb 2013 07:10:25 AM
All use subject to JSTOR Terms and Conditions

704 C. Alonso-Alvarez and M. Ferrer
Figure 1. Ideal pattern of changes in plasma composition and body
mass during fasting in bird species (classic model). Daily body-mass
loss is the body-weight proportion with respect to the previous fasting
day (the scale on Y-axis is only illustrative).
in wild bird species has been poorly studied. Handrich et al.
(1993a) proposed a model consisting of two phases. The first
one is the recovery of initial body weight and the restoration
of prefasting metabolic rates. The second phase is a period of
steady body mass and metabolism. The Garcı´a-Rodrı´guez et al.
(1987) study showed a slow recovery of body weight in buzzards
(Buteo buteo Linnaeus) throughout the refeeding period, while
plasma uric acid levels declined abruptly at the beginning but
slowly reached the original concentrations of fed birds, in agree-
ment with the second phase proposed by Handrich et al.
(1993a). More research on this process is needed.
In summary, this study has three objectives: (1) to know
whether a seabird, the yellow-legged gull Larus cachinnans Pal-
las, uses the above-cited classic model of resource allocation
during starvation, as shown in other bird species (thus, we will
analyze differences with respect to the described pattern; Fig.
1). Moreover, (2) this species was chosen in order to compare
changes during periods of moderate food restriction with pe-
riods of absolute fasting. Therefore, we might obtain a more
realistic approach to some bird life histories. Finally, (3) we
also studied the recovery process in the yellow-legged gull. With
these three objectives, we analyzed 12 plasma parameters, rep-
resentatives of proteins and their catabolic residuals (total pro-
tein, urea, and uric acid), fats such as triglycerides and cho-
lesterol, carbohydrates such as glucose, and some enzymes and
ions.
Material and Methods
Experimental Procedure
On February 28, 1999 (2 mo before the laying date in the
colony), we captured 22 adult yellow-legged gulls on a refuse
dump close to the city of Porrin˜o (Pontevedra, Spain). These
birds were transported to the wildlife recovery center La Can˜ada
de los Pa´jaros (Huelva, Spain) and were housed in individual
cages ( m). The study was performed with the per-4 # 4 # 4
mission of the appropriate authorities, avoiding any damage to
the birds. The gulls were divided into three groups, and the
sex ratio was balanced. Sex was determined through the PCR
amplification of CHD gene fragment sequences following Grif-
fiths et al. (1998). In this way, nine individuals formed the
fasting group (four males, five females), nine formed the re-
stricted group (four males, five females), and the last four birds
constituted the control group (two males, two females). For 2
wk, sardines (Sardina pilchardus Walbaum) were provided ad
lib. Fish is present in 32% of the pellets in the original pop-
ulation of the experimental birds (sardines included; Munilla
1997), and its biomass proportion might be even higher. After
this, from day 0 of the experiment the fasting group remained
without food, the restricted group was fed one-third of the
mean daily intake (calculated individually for each gull during
the previous 2 wk), and the control group remained with ad
lib. food. This interval was called the deterioration period. All
birds had water ad lib. during the experiment. Variable total
body-mass loss was defined as the proportion of body-mass
loss regarding weight at the beginning of the experiment. The
return to feeding (recovery period) was planned when birds
would reach phase 3 of the classic model described in other
species (i.e., Boismenu et al. 1992), but since there was no
previous information about critical levels of body-mass change
or biochemical parameters in this species, an a priori limit of
total body-mass loss was fixed at 25% to start refeeding the
birds. This limit was estimated conservatively from the pro-
portion of total body-mass loss of three ill individuals (captured
in previous years), which was calculated from the expected body
weights regarding their body size in the original population (C.
Alonso-Alvarez and M. Ferrer, unpublished observations).Nev-
ertheless, after 8 d, three gulls from the fasting group died on
the same day, without symptoms, and their lesser total body-
mass loss the day before (15%) was finally established as the
final limit before placing the gulls in the recovery period, re-
feeding them with food ad lib. Another two birds, one from
the fasting group and the other from the restricted group, were
retired after phase 3 because of a clear risk of death (inability
to walk), and they were treated with glucosaline serum and
vitamins so that they could recover their initial healthy state.
The experiment finished when all the birds reached the con-
fidence limits of total body-mass loss in relation to capture.
After this, the gulls were set free in the original capture location.
Blood Extraction and Weighing Procedures
Blood samples were taken from the humeral vein (2.5 mL)
every 2 d throughout the experiment, always before feeding, at
the middle of the day (1100–1500 hours) to avoid any variation
in blood chemicals caused by the circadian rhythm (Ferrer
This content downloaded on Fri, 15 Feb 2013 07:10:25 AM
All use subject to JSTOR Terms and Conditions

Fasting, Subfeeding, and Recovery in Yellow-Legged Gulls 705
1993). Winged infusion sets (Valu-Set, Becton Dickinson,
Sandy, Utah) were used to prevent damage to the veins, ap-
plying them on alternate wings each time. Blood sampling was
done immediately after capture. Lithium-heparin was employed
as anticoagulant, and the samples were stored between 0⬚ and
4⬚C until they were carried to the laboratory a few hours after
collection. Plasma was separated by centrifugation (550 g for
10 min) and was stored in a freezer for 1 d until the analysis.
The gulls were weighed after blood collection with a dyna-
mometer (Pesola; accuracy Ⳳ5 g). Daily body-mass loss rep-
resented the change in body weight with respect to the day
before sampling, allowing us to see small changes from day to
day.
Tested Parameters
Twelve biochemical components of blood were measured using
a spectrophotometer (Hitachi 747, Tokyo, Japan) and com-
mercial kits (Boehringer-Mannheim Biochemica, Mannheim,
Germany). The analyzed biochemical parameters were (abbre-
viations and methods indicated in parentheses): urea (UREA;
urease method), uric acid (URIC; uricase method), triglycerides
(TRIG; enzymatic method that includes amounts of free glyc-
erol), total protein (TP; biuret reaction), creatinine (CREA;
kinetic Jaffe´ reaction), inorganic phosphorus (iP; molybdenum
blue reaction), calcium (Ca; cresolphtalein complexone reac-
tion), magnesium (Mg; blue xilidil reaction), glucose (GLUC;
hexocinase method), cholesterol (CHOL; cholesterol esterase),
amylase (AMY; maltoheptaose reaction), and alkaline phos-
phatase (AP; paranitrophenyl-phosphate method).
Data Analysis
Mean values of parameters were tested for differences between
groups on the same day or in the same mass-loss rank by the
Mann-Whitney U-test for independent samples. Within-group
variations were tested with Wilcoxon matched pairs signed-
ranks test. These nonparametric tests were used as a precaution
since, as a result of small sample sizes in some analyses, normal
distribution could not be ascertained for all parameters. The
experiment effects were examined with repeated-measures
ANOVA, where the treatment (fasting or subfeeding) was used
as a factor (between-subject effect) and the samples obtained
from the same bird throughout the experiment were used as
repeated measures (within-subject effect). Moreover, repeated-
measures ANOVA was used to analyze changes in mass or
biochemical parameters in each group separately. A general
linear model of variance analysis was developed in order to
determine the influence of the treatment (group as fixed factor)
and the influence of proportion of body-mass loss (total body-
mass loss as covariable) in each biochemical parameter (de-
pendent variable) throughout the experiment, using the indi-
vidual as a random factor in order to avoid pseudoreplication.
All tests were performed with SPSS software (Norusis 1993).
Results
Initial and Final Values in the Deterioration Period
Body mass ( ; males: ; females: 733.1 Ⳳmean Ⳳ SE 898.9 Ⳳ 24.2 g
7.9 g) and plasma biochemical values were measured the first
day of the experiment in all birds. That day, there were no
differences among the three groups or between sexes (Mann-
Whitney: in all parameters). The four birds from theP
1 0.05
control group did not show significant variations in total body-
mass loss (repeated-measures ANOVA: , )F p 1.07 P p 0.41
10, 30
and plasma biochemical traits (always ) throughout theP 1 0.05
experiment and are not used in the rest of the statistical anal-
yses. There were many significant differences (Wilcoxon: P
!
) in plasma concentrations between the first day and the0.05
last day of the deterioration period in gulls suffering fasting or
food restriction (see Table 1). Urea, uric acid, cholesterol, glu-
cose, and alkaline phosphatase changed in both groups. There
were no significant differences between these two groups the
last day of the deterioration period regarding all the parameters,
but inorganic phosphorus, calcium, and magnesium showed a
tendency toward higher values in the fasting group (Mann-
Whitney: ).P
! 0.12
Weight and Biochemical Changes with Respect to the Classic
Model
In order to explain the changes in body mass throughout the
deterioration period, we analyzed daily body-mass loss and total
body-mass loss during the fasting phases according to the classic
model (Fig. 2). For both variables, data of the first four sam-
pling days from the beginning and, separately, data of the last
four sampling days to reach the final limit of the deterioration
period were analyzed ( in each group) in order to equil-n p 9
ibrate the sample size between the groups. The sample size on
some days was not equilibrated because of the highly variable
number of days to reach the fixed deterioration limit among
individuals (fasting group: 8–12 d; restricted group: 10–18 d).
Daily body-mass loss did not show significant within-subject
differences in the first four sampling days (repeated-measures
ANOVA: , ), but it did show such a dif-F p 1.41 P p 0.25
3, 48
ference in the last 4 d ( , ). The differencesF p 15.77 P ! 0.001
3, 48
between groups in the first four measurements and in the last
4 d (Fig. 2) showed a tendency to statistical significance
( , ; and , , respec-F p 3.07 P p 0.09 F p 3.47 P p 0.08
1, 16 1, 16
tively), showing lower values in the restricted group (Fig. 2).
A descent in daily body-mass loss between the second and the
fourth day (proposed phase 1) were not significant in either
group (Wilcoxon: , in both groups). DailyZ p 0.77 P p 0.44
body-mass loss measurements in the last day of the deterio-
ration period (in proposed phase 3) were higher in the fasted
This content downloaded on Fri, 15 Feb 2013 07:10:25 AM
All use subject to JSTOR Terms and Conditions

706 C. Alonso-Alvarez and M. Ferrer
Table 1: Mean values (ⳲSE) of plasma parameters the first and the last day of the
deterioration period, differentiated by groups
Biochemicals
Fasting Group
Restricted Group
First Day Last Day First Day Last Day
Urea (mg/dL) 5.78 (.22) 13.22 (1.63)* 5.33 (.33) 11.67 (1.54)**
Uric acid (mg/dL) 9.97 (1.05) 20.32 (2.93)* 10.51 (1.98) 18.17 (2.49)*
Total protein (g/dL) 3.35 (.36) 3.05 (.46) 3.17 (.19) 2.22 (.28)**
Triglycerides (mg/dL) 75.56 (10.34) 53.56 (9.90) 73.67 (4.92) 32.22 (4.42)**
Cholesterol (mg/dL) 324.6 (35.4) 201.9 (24.9)* 375.6 (19.7) 181.6 (15.5)**
Glucose (mg/dL) 365.8 (15.9) 267.8 (11.9)** 325.8 (9.49) 282.4 (9.0)*
Amylase (U/L) 815.4 (62.7) 653.9 (22.4) 927.6 (88.3) 890.1 (115.3)
Creatinine (mg/dL) .26 (.003) .22 (.003) .26 (.002) .17 (.002)**
AP (U/L) 230 (52.2) 99.56 (34.1)* 223.2 (57) 172.9 (44.1)*
Pi (mg/dL) 3.28 (.25) 4.75 (.63)* 3.18 (.16) 3.57 (.30)
Ca (mg/dL) 8.73 (.50) 8.25 (.54) 8.69 (.23) 7.17 (24)**
Mg (mg/dL) 2.23 (.006) 2.26 (.13) 2.22 (.008) 2.01 (.005)
Note. Wilcoxon matched pairs signed-ranks test; in each group. There are no differences among groupsn p 9
on the first ( ) or the last day ( ).P
1 0.1 P 1 0.05
*.P
! 0.05
** .P
! 0.01
gulls than in the restricted gulls (Mann-Whitney: ,Z p 1.99
; see Fig. 2).P p 0.047
Concerning the biochemical parameters, we focused on uric
acid and triglycerides as representatives of nitrogen residuals
and fat use, respectively, synchronizing newly recorded data
with respect to the first and the last day of the deterioration
period (in Fig. 3, backward from last day). In the fasting gulls,
uric acid increased in the first 4 d of the sampling and in the
last four (repeated-measures ANOVA: , ;F p 7. 2 1 P p 0.001
3, 24
and , , respectively). In the same group,F p 4.48 P p 0.012
3, 24
triglycerides increased in the first four measures ( ,F p 6.66
3, 24
) and decreased in the last four ( , ).P ! 0.01 F p 4.03 P ! 0.05
3, 24
Differences between groups were detected in the last 4 d of the
deterioration period (uric acid: , ; triglyc-F p 8.24 P p 0.01
1, 16
erides: , ), but they were not significant inF p 11.23 P ! 0.01
1, 16
the first 4 d (uric acid: , ; triglycerides:F p 4.29 P p 0.06
1, 16
, ). When we observed the deteriorationF p 2.25 P p 0.16
1, 16
period as a whole (see Table 2, “Group” column), we observed
significant differences in triglycerides but not in uric acid, al-
though it was close to statistical significance ( ; see alsoP p 0.07
Fig. 5).
Weight Changes in the Recovery Period
Changes in total body-mass loss during the recovery period
were used to explain the return of our gulls to the initial body
weight (Fig. 4). The values at last day of the deterioration period
were significantly higher than the values at the first sampling
day of the recovery period (Wilcoxon; fasting group: Z p
, ; restricted group: , ). Thus,2.02 P p 0.043 Z p 2.52 P p 0.012
afteronly2dofrefeeding, the birds recovered a great part of
mass they had lost, without differences between both groups
( ; fasting group: ; restricted group:mean Ⳳ SE 49.3% Ⳳ 10.5%
; Mann-Whitney: , ). There69.9% Ⳳ 6.17% Z p 1.71 P p 0.24
were not significant differences between groups throughout the
recovery period (repeated-measures ANOVA: ,F p 0.04
1, 11
). However, there was a significant decrease in totalP p 0.86
body-mass loss in the fasting group (within-subject effect:
, ) but not in the restricted groupF p 7. 9 0 P p 0.001
4, 24
( , ). Nevertheless, only a nonsignificantF p 1.77 P p 0.16
4, 39
tendency to higher values of mass loss in the restricted group
in the last day was detected (Mann-Whitney: ,Z p 1.70 P p
; Fig. 4).0.09
Changes in Biochemical Variables throughout the Experiment
In this study we confronted two problems in the interpretation
of data. The first one was the individual differences in the
number of days to attain the deterioration limit (commented
on above), which prevents changes from being analyzed with
respect to a chronological order. The second problem was the
obvious differences in total body-mass loss level between the
groups in the same day, promoted by the effect of the treatment.
This issue prevents between-group comparisons. With the aim
of avoiding these problems, the changes were analyzed using a
general linear model (Table 2), which allowed testing of the
linear relationship of each biochemical trait with the proportion
of total body-mass loss, but not with time.
In addition, we analyzed data by means of total body-mass
loss ranks (Fig. 5), which allowed comparison of the plasma
levels among the groups when the birds were in a similar body-
mass proportion. Figure 5 included in the deterioration period
This content downloaded on Fri, 15 Feb 2013 07:10:25 AM
All use subject to JSTOR Terms and Conditions

Frequently asked questions

Q1. What are the contributions mentioned in the paper "A biochemical study of fasting, subfeeding, and recovery processes in yellow-legged gulls" ?

These results contribute to a better knowledge of fasting, subfeeding, and refeeding processes in birds and can be added to a recent study about fasting in gulls. 

Q2. What future works have the authors mentioned in the paper "A biochemical study of fasting, subfeeding, and recovery processes in yellow-legged gulls" ?

Yellow-legged gulls showed a moderate physiological adaptation to extended fasting but the same model of biochemical changes in plasma that other more adapted species showed. Some differences in the results reported here with respect to a recent study on fasting in herring gulls can be explained by methodological interferences.