| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
* Grup de Recerca en Remugants, Departament de Ciència Animal i dels Aliments, Universitat Autònoma de Barcelona, 08193 Bellaterra, Spain
Sheep and Goat Research Department, Animal Production Research Institute, 4 Nadi El-Said, 12311 Dokki, Giza, Egypt
1 Corresponding author: gerardo.caja{at}uab.es
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSACKNOWLEDGEMENTSREFERENCES |
|---|
Key Words: ultrasonography • milking frequency • cisternal milk • dairy sheep
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSACKNOWLEDGEMENTSREFERENCES |
|---|
Ultrasonography allows the noninvasive exploration of the internal structure of the mammary gland and has been used for measuring cisternal size in ewes (Bruckmaier and Blum, 1992; Ruberte et al., 1994; Nudda et al., 2000), goats (Bruckmaier and Blum, 1992; Salama et al., 2004), and cows (Bruckmaier and Blum, 1992; Ayadi et al., 2003). Several studies have shown that ultrasonography could be useful for studying the udder changes that occur to accommodate milk during different milking intervals and after milk letdown in cows (Ayadi et al., 2003, Caja et al., 2004) and goats (Salama et al., 2004). However, as far as we know, ultrasonography has not been used to evaluate cisternal size changes when different milking intervals are performed in dairy ewes.
The main objective was to study the effect of milking intervals on cisternal size and milk partitioning in the udder of 2 dairy ewe breeds (Manchega and Lacaune) differing in lactational performance. Composition of milk obtained from each udder compartment was evaluated at different milking intervals. The experiment allowed validation of the udder scanning methodology in ewes of different cistern sizes and at different intervals after milking.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSACKNOWLEDGEMENTSREFERENCES |
|---|
Animals and Management Conditions
Twenty-four adult lactating dairy ewes of 2 breeds (Manchega, n = 12, 70.2 ± 2.4 kg of BW; Lacaune, n = 12, 75.5 ± 2.2 kg of BW) with symmetrical and healthy udders (SCC, 64 ± 48 x 103 cells/mL; <5 cfu/mL) from the experimental farm of SGCE (Servei de Granges i Camps Experimentals) of the Universitat Autònoma de Barcelona in Bellaterra (Spain) were used in a short-term experiment performed at wk 11 and 16 of lactation.
Ewes had similar stage of lactation (70 ± 3 DIM) and milk yield by breed (Manchega, 1.11 ± 0.09 L/d; Lacaune, 2.32 ± 0.11 L/d) in the week before (wk 10) the start of the experiment. Ewes were allocated at wk 10 and 15 of lactation into 6 groups of 4 ewes each (Manchega, n = 2; Lacaune, n = 2), and were housed indoors, fed a dehydrated forage-based diet, and managed as described previously by Castillo et al. (2008).
Ewes were milked in a double-12 stall parallel milking parlor (Westfalia Surge Ibérica, Granollers, Spain), set to 42 kPa, 120 pulses/min, and 50% pulsation ratio, and equipped with recording jars and low milk pipeline. Before the experiment, animals were milked twice daily (0800 and 1800 h), and the milking included machine milking (without udder preparation or teat cleaning), machine stripping, and teat dipping in an iodine solution (P3-Ioshield, Ecolab Hispano-Portuguesa, Barcelona, Spain).
Experimental Procedure
The experiment consisted of a crossover with 6 milking intervals (4, 8, 12, 16, 20, and 24 h) at random, replicated at wk 11 and 16 of lactation. Milking interval schedule during each experimental week was applied successively (total length 84 h) without any period of washing-out between milking intervals. Carryover effects between milking intervals were reduced by randomizing the milking order of ewe groups and by removing alveolar milk at milk fraction collection (Castillo et al., 2008). The experiment started (0 h) after complete udder emptying by machine milking with the help of an i.v. injection of oxytocin (2 IU/ewe; Veterin Lobulor, Laboratorios Andreu, Barcelona, Spain) to remove residual milk. Thereafter, 1 of the 6 milking interval treatments was randomly applied, and the same udder emptying procedure was repeated for each milking interval treatment.
Scanning Methodology
To prevent undesired milk letdown during scanning and to enable recording of cisternal and alveolar milk fractions separately, each ewe was injected i.v. with 0.8 mg/ewe of atosiban, an oxytocin receptor blocking agent (Tractocile 7.5 mg/mL, Ferring, Madrid, Spain) while in its pen before being taken to the milking parlor. Cisternal size was evaluated for each half udder by measuring cisternal area by ultrasonography. Ultrasonography was conducted using a real-time B-mode ultrasonograph (Ultra Scan 900, Ami Medical Alliance Inc., Montreal, Canada) equipped with a 5-MHz sectorial probe (2 dB power, 80° scanning angle, 0.5 mm axial and 1.5 mm lateral resolution). Exploration depth was 14 cm. The probe was placed directly against the upper part of the medium suspensory ligament, caudally to the udder, and between the inguinal lymph nodes (Ruberte et al., 1994; Rovai, 2001) using the teat as scan axis. Contact gel (Geleco, Laboratorios Carreras, Barcelona, Spain) was applied to udder skin to exclude air between probe and udder. Scans were done for each udder half and transferred to a personal computer for image analysis. Cisternal area was calculated in triplicate for each scan using image treatment software (MIP4 Advanced System, Microm España, Barcelona, Spain). Area in pixels was converted to centimeters squared using a conversion rate of 1,024 pixels/cm2.
As the calculated half-life of atosiban is approximately 18 min (Wellnitz et al., 1999), a single dose was sufficient to prevent milk letdown during the time that ewes were being moved to the milking parlor (approximately 4 min), having scans performed on both half udders (approximately 5 min), and having cisternal milk evacuated by machine milking (<2 min). Approximately 20 min after the atosiban injection, ewes were injected i.v. with 2 IU of oxytocin and machine-milked to obtain letdown alveolar milk.
Sampling and Analyses
Milk volume was recorded and sampled for each milking and udder fraction. Samples of approximately 100 mL were preserved with potassium dichromate (0.3 g/L) at 4°C until analysis of milk composition. Un-homogenized milk samples were analyzed with a near-infrared spectrometer (InfraAlyzer-450, Bran+Luebbe, Nordersted, Germany) for content of total solids, fat, CP (N x 6.38), true protein, casein, and lactose. Samples for SCC were preserved with an antimicrobial tablet (Bronopol, Broad Spectrum Micro-tabs II, D&F Control Systems Inc., San Ramon, CA) and kept at 4°C until analysis. The SCC was determined in the DHI Laboratory of Catalonia (Allic, Cabrils, Barcelona, Spain) using an automatic cell counter (Fossomatic 5000, Foss Electric, Hillerød, Denmark) specifically calibrated for sheep milk.
Statistical Analysis
Results were analyzed by the PROC MIXED procedure for repeated measurements of SAS (version 8.2, SAS Institute Inc., Cary, NC). The mixed model used included the fixed effects of milking interval (4, 8, 12, 16, 20, and 24 h), breed (Manchega and Lacaune), milk fraction (cisternal and alveolar), group (1 to 6), period (wk 11 and 16), and order of treatment application (first to sixth); the random effect of animal nested within group and breed; the interactions breed by milking interval, milk fraction by milking interval, period by milking interval, breed by milk fraction, and period by milk fraction; and residual error. Differences between least squares means were determined with the PDIFF test of SAS and significance was declared at P < 0.05. Pearson correlation coefficients between measurements were calculated. Significance was declared as P < 0.05 unless otherwise indicated.
|
RESULTS AND DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSACKNOWLEDGEMENTSREFERENCES |
|---|
Volumes of cisternal and alveolar milk varied (P < 0.001) according to milking interval (Figure 1
). Cisternal milk increased (P < 0.05) linearly until 24 h of udder filling (Figure 1a) in both breeds. Alveolar milk increased linearly (P < 0.05) until 16 h in Manchega and 20 h in Lacaune dairy ewes (Figure 1b). Thus, the alveolar compartment saturated earlier than the cisternal compartment in both breeds. These results agree with the pattern of milk accumulation for extended milking intervals in dairy ewes (McKusick et al., 2002) and in dairy goats (Salama et al., 2004) whose alveolar milk reached plateaus 20 and 16 h after milking, respectively. These authors reported that cisternal milk increased continuously until 24 h, in both ewes and goats.
|
Cisternal milk (mL):
 |
Alveolar milk (mL):
 |
The analysis of the regression lines showed that the dose of oxytocin administered was not enough to evacuate completely the alveolar compartment. The volume of milk left in the compartment, 66% greater in Lacaune than Manchega, agreed with the expected size of the alveoli population in each breed. Lollivier and Marnet (2005) showed differences between cows in the dose necessary to induce milk ejection (0.25 to 4 IU), but the doses used by these researchers were markedly lower (0.005 to 0.008 UI/kg of BW for a 500-kg cow) than in our case. Thus, dose of oxytocin used (2 IU/ewe) was considered to be supraphysiological for both breeds (0.029 IU/kg of BW for a 70-kg ewe) and did not limit the contractive response of the alveoli. Differences in udder morphology between the 2 breeds, Lacaune having larger and baggier udders than Manchega (Rovai, 2001), may explain the differences observed in the volume of milk left when measuring the alveolar milk fraction.
As expected, cisternal milk percentages were different (P < 0.01) between breeds. The contribution of cisternal milk to total milk in the udder represented 53 and 68% in Manchega and Lacaune ewes, respectively. Values were greater in Lacaune (47 to 77%) than in Manchega (33 to 65%) throughout the milking intervals (4 to 24 h). Percentages of cisternal milk obtained in Manchega ewes were similar to McKusick et al. (2002) in East Friesian crossbred dairy ewes (32 to 57%) for a similar milking interval range (4 to 24 h). On the other hand, percentage of cisternal milk in Lacaune ewes was high and close to that observed in Sarda ewes (82%) for a 24-h milking interval (Nudda et al., 2000). Moreover, the Sarda value may have been overestimated because the authors used adrenaline instead of atosiban to avoid spontaneous milk letdown.
With these differences in mind, dairy sheep breeds could be classified according to their udder cistern features: the large-cisterned ewes (e.g., Awassi, Lacaune, and Sarda), usually selected for high milk yield, present a great capacity of milk storage in the cisternal compartment, and the medium-cisterned ewes (e.g., East Friesian crossbred and Manchega), which have been selected as dual-purpose or local breeds, show a medium capacity for storing milk inside the cisterns (Labussière, 1988). Small-cisterned ewe breeds usually correspond to unselected or nondairy sheep.
Percentages of cisternal milk obtained for dairy ewes (Figure 1a
, dotted lines) were greater than values reported in Ripollesa meat sheep (24% at 4 h; Caja et al., 1999), but close to those in Murciano-Granadina goats (75% to 24 h; Salama et al., 2004). Consequently, efficient udder stimulation in dairy sheep does not seem to be a key to maximizing machine-milking fraction and minimizing stripping fraction during milking.
Percentages of cisternal milk increased (P < 0.05) with time after milking and reached the maximum at 24 h, although values did not differ from 16 to 20 h in Manchega and from 12 to 24 h in Lacaune ewes (Figure 1a). A similar evolution pattern of cisternal milk percentage with increasing milking interval was found by McKusick et al. (2002) in East Friesian crossbred ewes, which had the maximum percentage of cisternal milk at 24 h but did not show differences between 12 and 20 h of udder filling. Experiments in dairy goats (Salama et al., 2004) showed that the ability to tolerate extended milking intervals was related to the size of cisternal compartment. Thus, animals that store a larger proportion of milk in the gland cistern were able to tolerate extended milking intervals better than those that stored a relatively larger proportion in the alveolar compartment. In our experiment, greater percentages of cisternal milk were detected in Lacaune than in Manchega (Figure 1a), so smaller milk losses were expected in Lacaune than in Manchega ewes when once daily milking is performed. Nudda et al. (2002) reported moderate milk yield losses in Awassi (18%) and Sarda (24%) ewes when a 24-h milking interval was applied for only 4 d in mid lactation. Salama et al. (2003) reported an 18% milk yield loss in dairy goats milked once daily throughout lactation. Studies on the long-term effect of once-daily milking have been inadequate in dairy ewes, so further research is necessary to support the hypothesis of applying long-term reduced milking frequencies (e.g., once-daily milking) in large-cisterned dairy ewes.
Transfer of milk from the alveoli to the cistern seemed to start very soon after milking (Figure 1), without requiring that the alveolar compartment reach a saturation stage, which is consistent with early observations in dairy ewes (McKusick et al., 2002) and dairy goats (Salama et al., 2004). In our results, milk transfer from alveolar tissue to cistern according to time after milking appeared to occur in different ways by breed. From the 4-h milking interval, Manchega and Lacaune ewes differed (P < 0.05) in cisternal (57 vs. 161 mL, respectively; SEM, 53 mL) and alveolar (98 vs. 182 mL, respectively; SEM, 38 mL) milk compartments, and the differences were maintained until 24 h (Figure 1a). The cisternal:alveolar ratio at the 24-h milking interval was 2.0:1 and 3.4:1 for Manchega and Lacaune ewes, respectively. McKusick et al. (2002) reported values in East Friesian crossbred ewes that were greater for alveolar milk at the 4-h milking interval, supporting a large and baggy udder, although cisternal:alveolar ratios at 24 h were lower (1.4:1) than in our results. These results confirm the importance of the cisternal compartment in accommodating milk storage when the milking interval is extended in dairy ewes, allowing significant increments of total milk yield until 24-h of udder filling.
Cisternal Scans
During ultrasonography, milk in the gland cistern was evident in the scans (Figure 2) as a dark area (an-echogenic) and the glandular parenchyma as a gray-white area (echogenic), in agreement with other works carried out in ewes, goats, and cows (Bruckmaier and Blum, 1992; Ruberte et al., 1994). No differences were detected between left and right half udder cisternal areas in the dairy ewes (23.18 ± 0.74 cm2), probably because experimental ewes were selected based on udder symmetry, and machine milking may have contributed to the udder symmetry maintenance. Therefore, left and right udder half values were averaged and discussed jointly.
|
). Cistern area value for Manchega ewes was 52% lower (P < 0.001) than for Lacaune ewes (mean of all the milking intervals; Manchega: 15.01 ± 1.00 cm2 and Lacaune: 31.36 ± 1.00 cm2). Other values of cisternal area in sheep are those obtained in East Friesian (Bruckmaier and Blum, 1992; Bruckmaier et al., 1997; 19 to 40 cm2), Sarda (Nudda et al., 2000; 19 cm2), Lacaune (Bruckmaier et al., 1997; 33 cm2), and Ripollesa (Caja et al., 1999; 5.6 cm2). Differences in sheep breed, stage of lactation, and hours of udder filling could explain the large variability among the studies cited. Our results confirm the importance of the cisternal compartment in dairy ewes (Manchega: 4 to 21 cm2; Lacaune: 16 to 38 cm2, for 4- and 24-h intervals, respectively), especially in Lacaune ewes, which presented cisternal areas greater than those of Murciano-Granadina dairy goats (9 to 28 cm2 for 8- and 24-h interval; Salama et al., 2004).
|
). Cisternal areas showed curvilinear increases with milking interval, reaching a plateau at 20 h. These plateaus could be controversial because the cisternal milk volume increased linearly up to 24-h interval in both breeds. We hypothesize that either the frequency probe used (5 MHz) or the exploration depth (14 cm) was not optimum for visualizing the full extent of the cisternal-cross section in the scan at extended milking intervals.
Relationship Between Cisternal Area and Milk Fractions
Correlations between cisternal area, and cisternal, alveolar and total milk were calculated for all the milking intervals tested (Table 1). Manchega and Lacaune ewes presented the greatest correlations between cisternal area and cisternal milk at the 8-h interval (Manchega, r = 0.70, P < 0.001; Lacaune, r = 0.56, P < 0.05), which were similar to those reported by Rovai (2001) in the same breeds (Manchega, r = 0.89; and Lacaune, r = 0.57) and to those reported in dairy goats (r = 0.72 at 8 h; Salama et al., 2004), but lower than correlations reported in meat sheep (r = 0.90 at 4 h; Caja et al., 1999).
|
Correlation coefficients between cisternal area and total milk yield varied according to breed and were less in Lacaune than in Manchega ewes (Table 1). Values in Manchega ranged between 0.48 and 0.76, indicating that udder cistern size was related to the amount of milk produced, and that ultrasonography could be used to identify high-yielding animals in this breed. In Lacaune ewes, correlation coefficients were only significant at the 8-h interval (r = 0.68; P < 0.01), showing the methodological problem indicated above with regard to probe depth. In both breeds, volumes of milk fractions (cisternal and alveolar) correlated with total milk yield in all the milking intervals (Manchega, r = 0.41 to 0.93; Lacaune, r = 0.50 to 0.90; Table 1).
Composition of Milk Fractions
Percentage of fat in cisternal and alveolar fractions varied inversely at different milking intervals (Figure 4a). Fat content of milk fractions was affected by breed (P < 0.001), and was always greater in Manchega than in Lacaune ewes. Fat percentage in cisternal milk dramatically decreased (P < 0.05) as milking interval increased (Manchega, –30%; Lacaune, –36%; between 4- and 24-h milking intervals). Conversely, milk fat content in alveolar milk increased (P < 0.05) as milking interval increased, but reached a plateau at 12 h of udder filling in Manchega (8.9% on average) and at 20 h in Lacaune (8.7% on average). The data confirm that transfer of milk fat from the alveoli to the cistern slowed with the increase in milking interval, producing a backup of milk fat in the alveolar compartment, as described previously in ewes (McKusick et al., 2002). Fat content in alveolar milk was greater (P < 0.05) than the percentage of fat in cisternal milk for most milking intervals studied (Manchega, 8.55 to 9.16% vs. 8.24 to 5.74%; Lacaune, 7.26 to 8.93% vs. 7.54 to 4.84%, respectively), except for the 4-h interval. The large size of fat globules in conjunction with the active expulsion required for their drainage to the cistern favored their accumulation in the alveolar compartment. Thus, milk fat concentration was usually lower in cisternal than in alveolar milk in dairy ewes (Labussière, 1969; Rovai, 2001).
|
). Milk protein content was greater (P < 0.001) in Manchega than in Lacaune ewes in both cisternal (6.10 ± 0.09 vs. 5.57 ± 0.09%, respectively) and alveolar milk (6.47 ± 0.08 vs. 5.82 ± 0.08%, respectively) fractions. As the milking interval increased, the protein content of the cisternal milk fraction was not modified (Manchega, 6.10 ± 0.10%; Lacaune, 5.57 ± 0.09%), but protein content did increase in alveolar milk, remaining constant after the 12-h interval (Manchega, 6.56 ± 0.11%; Lacaune, 5.83 ± 0.11%). No significant differences were observed between milk protein percentage of cisternal and alveolar milk (Manchega: 6.10 vs. 6.47%, SEM, 0.09; Lacaune: 5.57 vs. 5.82%, SEM, 0.09; P > 0.05; respectively), which may be because milk protein is found primarily in the form of small casein micelles (Cowie and Tindal, 1971) within the aqueous fraction of milk. Thus, proteins pass freely from the alveolar to the cistern compartment without being as dependent on milk ejection reflex as is fat content. Nevertheless, at some milking intervals, we found a small increment of protein content in the alveolar fraction compared with the cisternal fraction. McKusick et al. (2002) observed differences of milk protein percentages between cisternal and alveolar fractions in East Friesian crossbred ewes, but only at the 24-h interval. With respect to milk lactose, milking interval and breed did not affect lactose content either in alveolar or in cisternal milk, except at the 24-h interval, when a significant difference between cisternal and alveolar milk (4.20 vs. 4.79%, P < 0.001; respectively) was observed only in Manchega ewes, indicating that Manchega ewes should suffer greater milk yield losses than Lacaune ewes at the 24-h milking interval. Decreased lactose percentage associated with longer milking intervals may be due to leaky tight junctions between mammary epithelial cells. When mammary tight junctions were disrupted, the small lactose molecules passed freely from milk to blood (Stelwagen et al., 1994) and the influx of interstitial fluid into the milk occurred (Stelwagen et al., 1997). Both events probably favored important changes in milk osmosis, provoking the decrease of lactose content observed in cisternal milk of Manchega ewes. Lactose content in alveolar milk did not vary according to milking interval and breed. Content of total solids showed a similar trend to milk fat content (data not shown).
Intramammary infections were not observed during the experiment. Both breeds showed low SCC in alveolar (131 x 103 cells/mL, P = 0.74) and cisternal milk fractions (110 x 103 cells/mL, P = 0.46). Moreover, milking interval had no effect (P > 0.10) on SCC in either breed. Regardless of milk fractions, SCC content was similar between cisternal and alveolar milk when 4-, 8-, and 12-h intervals were performed. Yet alveolar milk had a greater SCC (P < 0.05) at 20 and 24 h (Figure 5). Similar results were obtained in milk fractions of East Friesian crossbred ewes (McKusick et al., 2002). Taking into account that the impairment of tight junctions starts from 20 h of udder filling in dairy ewes (Castillo et al., 2008), tight junction disruption might have facilitated a paracellular influx of somatic cells into milk (Stelwagen and Lacy-Hulbert, 1996), provoking the differences observed between cisternal and alveolar milk fractions from 20 h after milking.
|
|
CONCLUSIONS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSACKNOWLEDGEMENTSREFERENCES |
|---|
|
ACKNOWLEDGEMENTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSACKNOWLEDGEMENTSREFERENCES |
|---|
Received for publication February 13, 2008. Accepted for publication May 23, 2008.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSACKNOWLEDGEMENTSREFERENCES |
|---|
- and β-adrenergic agonist and oxytocin administration. J. Dairy Res. 59:151–159.[Medline]This article has been cited by other articles:
|
|
 |
|
  V. Castillo, X. Such, G. Caja, R. Casals, A. A. K. Salama, and E. Albanell Long- and short-term effects of omitting two weekend milkings on the lactational performance and mammary tight junction permeability of dairy ewes J Dairy Sci, August 1, 2009; 92(8): 3684 - 3695. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
) and Lacaune (
) ewes at different milking intervals. Percentages of cisternal milk of total milk in the udder of Manchega (–
) dairy ewes. Values are least squares means. Significant differences between alveolar and cisternal fractions within a time category are indicated by * (P < 0.05). Vertical bars represent SEM; a–eMeans within the same breed with different letters differ at P < 0.05.