| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
* Gene Evaluation and Mapping Laboratory;
and
Germplasm and Gamete Physiology Laboratory;
and
Immunology and Disease Resistance Laboratory, Animal and Natural Resources Institute, USDA, ARS, Beltsville, MD 20705;
and
Animal and Veterinary Sciences Department, Clemson University, Clemson SC 29634;
and
¶ Department of Animal and Avian Sciences, University of Maryland, College Park 20742; and
and
# Department of Animal Science, McGill University, Ste. Anne de Bellevue, Quebec H9X 3V9 Canada
 |
Abstract |
|---|
 Top Abstract IntroductionSummaryImplicationsLiterature Cited |
|---|
Key Words: Apoptosis • Cell proliferation • Cell renewal • Lactation Efficiency • Ruminants
|
Introduction |
|---|
|
TopAbstractIntroductionSummaryImplicationsLiterature Cited |
|---|
If persistency of lactation could be increased, considerable benefits would accrue to the dairyman. Flattening the declining portion of the lactation curve promotes a more efficient lactation. By lengthening lactation, a smaller portion of a dairy animal’s life would be spent during the periparturient period with its increased health risks and associated costs. Delayed breeding could increase reproductive efficiency. Lengthening lactation and use of sexed semen for artificial insemination may serve to more efficiently match a decreased need for replacement heifers, thus reducing the need to slaughter bobby calves. Management schemes that improve maintenance of mammary cell numbers will almost certainly increase persistency. Another management goal might be elimination of a dry period between successive lactations. During this period, accelerated renewal of mammary cells, perhaps progenitor cells, may be of particular importance. Although important for dairy cows, the impact of a dry period on lactation in dairy goats has not been established.
This review will discuss changes in mammary cell number, proliferation, and death during lactation, factors that may impact these processes, and putative cellular targets for strategies to impact mammary cell number. Hypotheses pertaining to the value of a dry period between lactations will also be discussed. Although this review will focus on dairy cattle and goats, rodents will be discussed for comparative purposes and when there are limited data pertaining to ruminants.
Cell Number and Cellular Secretory Activity During Lactation
Knight and Peaker estimated mammary cell number and secretory activity during lactation of goats (Knight and Peaker, 1984b
). This was accomplished by using multiple biopsies to evaluate changes in nucleic acid concentrations and enzymatic activity. When coupled with measures of udder volume, the data were extrapolated to whole udder measures. This study demonstrated that increases in milk production during early lactation were first the result of an increase in mammary cell number followed by an increase in secretory activity per cell. After peak lactation, decreased milk yield with advancing lactation was primarily the result of declining cell number. However during late lactation, when goats were concomitantly pregnant, the secretory activity per cell also declined.
Recently, a comparable analysis of changes in mammary cell number and secretory activity during a bovine lactation was performed (Capuco et al., 2001a). This study was based on the slaughter of nonpregnant Holstein cows at four time points during lactation with quantification of total mammary DNA. Increased milk yield during early lactation appeared solely due to increased secretory activity per cell because no increase in mammary cell number was detected prior to peak lactation. The decline in milk yield with advancing lactation was solely due to decreased cell number (total mammary DNA). The secretory activity per cell, estimated as milk yield per unit of mammary DNA, increased prior to peak lactation but did not change significantly with advancing lactation. It is important to realize that these data (Capuco et al., 2001a) pertain to nonpregnant multiparous cows. When cows are concomitantly lactating and pregnant, it is likely that a decline in secretory capacity per mammary cell accompanies advanced pregnancy due to conflicting metabolic demands of gestation and lactation. Indeed, this is readily apparent during late pregnancy, when the number of mammary epithelial cells increases simultaneous with a rapid decline in milk production (Capuco et al., 1997).
In addition to evaluating the relationship between the lactation curve and the number and activity of mammary cells, we used a quantitative approach to evaluate mammary cell renewal during lactation (Capuco et al., 2001a). This was accomplished by separately evaluating rates of cell proliferation and cell death. Cell number reflects the sum of relative rates of cell proliferation and cell death. The mammary gland grows when the rate of proliferation exceeds the rate of cell death, regresses when the rate of cell death exceeds the rate of cell proliferation, and maintains constant cell number when rates of proliferation and death are equal. Regardless of the net change in cell number, a population may undergo varying degrees of cell replacement or turnover determined by the absolute rates of proliferation and cell death (Figure 1).
|
). The in situ labeling of apoptotic cells in histological sections by TUNEL detects cells with fragmented DNA, a hallmark of later stages of apoptosis. Because apoptotic cells appear to persist for about 3 h (Bursch et al., 1990), we multiplied the labeling index by eight (24/3) to obtain an estimate for a 24-h apoptotic rate (Capuco et al., 2001a).
Measures of proliferation rate and apoptotic rate were consistent with the observed decline in mammary cell number. Although the BrdU-labeling index did not vary significantly during lactation, there was a tendency for reduced rates of cell proliferation during early lactation (14 d). This suggested that the increased mammary cell proliferation evident during gestation and the dry period did not continue during early lactation. Lack of increased proliferation rate or decreased apoptotic index during early lactation precludes significant mammary growth during early lactation in cows. However, the latter conclusion is somewhat tentative because mammary growth during lactation may have occurred prior to the first sampling time, 14 d of lactation. A decline in cell number after peak lactation was predicted by the estimates of proliferation and apoptotic rate. Across lactation, 0.3% of mammary cells proliferated per day. After peak lactation, the apoptotic index averaged 0.07%. Although seemingly very low, this apoptotic index was extrapolated to an apoptotic rate of 0.56% per day—a value that exceeded the rate of cell proliferation. Using initial mammary DNA measures, the experimentally estimated rates of cell proliferation and death were used to predict mammary cell number or DNA with advancing lactation. The predicted decline in cell number closely approximated the decline in mammary DNA observed (Figure 2; Capuco et al. 2001a). In addition to the predicted curve for net mammary DNA, curves were generated for the accumulative cell proliferation and accumulative cell loss. The number of cells generated during lactation is predicted to equal the number that are present in the gland at 252 d of lactation (Figure 2, intersection of curves). If none of the cells that died by apoptosis were those that proliferated during lactation, then 100% of the mammary cells remaining after 252 d of lactation were formed during lactation. If apoptosis is random, equal death rates for preexisting cells and cells that have formed during lactation can only occur after considerable proliferation has occurred during lactation so that these two populations are equal in number. If apoptotic cell death is randomly distributed among cells of different ages, turnover of mammary cells will be less than 100%, but certainly greater than 50%.
|
). This lack of extensive cell turnover may be due to the shorter length of lactation in rodents compared with cows. Furthermore, this low degree of cell renewal may be consistent with the decline in activity per mammary secretory cell that occurs during lactation in rats, but not in nonpregnant dairy cows.
After peak lactation, there was a decline in the number of mammary epithelial cells (Capuco et al., 2001a). Expressed as a percentage of total cells, the epithelial compartment declined from 79% at 90 d to 73% at 240 d. This was consistent with previous data indicating that around the time of calving, 83% of mammary cells were epithelial, but during late lactation, the percentage had decreased to 74% (Capuco et al., 1997). Because of the error inherent in quantifying very small percentages of proliferating and apoptotic cells, we were unable to subdivide the predictive equations in Figure 2
to adequately describe changes in the relative proportions of epithelial and stromal cells. However, the data clearly model a reduction in total cells and epithelial cells of the mammary gland during lactation.
In addition to apoptotic death, another potential mechanism for declining cell numbers during lactation is the continuous loss of mammary epithelial cells in milk. In the present experiment, milk somatic cell count (SCC) averaged 52,000 cells/mL, and milk production averaged 9,920 kg for 240 d. Because epithelial cells account for less than 20% of milk SCC (Miller et al., 1991), and DNA content is 6 pg/cell, cumulative loss of mammary DNA is less than 0.62 g by 240 d of lactation. This accounts for only 1.6% of the net loss of mammary DNA (38.3 g) by 240 d of lactation. Although clearance of apoptotic cells from the mammary gland may involve passage of phagocytes into the milk, removal of intact viable mammary cells via the milk does not significantly contribute to the declining number of mammary secretory cells during lactation.
These data (Capuco et al., 2001a) provided the first quantitative demonstration that apoptotic death of mammary epithelial cells accounts for gradual regression of the bovine mammary gland during lactation and the proportional decline in milk yield. Indeed, this was the first demonstration that apoptosis can fully account for cell loss during lactation of any species.
Impact of Pregnancy on Lactation
The persistency of lactation is decreased when cows are concomitantly pregnant. A negative effect of pregnancy on milk yield has been detected as early as d 100 of gestation (Bachman et al., 1988). Because this coincided with the onset of estrogen secretion by the fetal-placental unit, estrogens were implicated as mediators of the inhibitory effects of pregnancy on lactation (Bachman et al., 1988). Others had demonstrated that estrogens administered at supraphysiological doses inhibited milk production (Folley et al., 1941; Hutton, 1958; Cowie, 1969). The time course of the inhibition and subsequent return of milk yield, along with accompanying changes in milk composition, suggested to some that high concentrations of estrogen induced an initial phase of mammary involution followed by epithelial regrowth and differentiation (Mollett et al., 1976). Thus, Mollett et al. (1976) investigated the feasibility of treating with estrogen and progesterone during lactation to determine if production of additional secretory cells could be induced. The approach was unproductive since the combined effect of estrogen and progesterone was similar to the inhibitory effect of estrogen alone. However, there was an exception. Estrogen and progesterone treatment increased milk production in cows that had been hormonally induced into lactation. This suggested that additional steroid treatment during lactation might compensate for incomplete mammary development prior to lactation.
The impact of pregnancy on apoptosis of mammary epithelial cells has been investigated in mice (Quarrie et al., 1996). When mice were remated at postpartum estrus, concomitant pregnancy accelerated the gradual mammary involution occurring during lactation due to increased levels of apoptosis. These data suggest that in addition to decreasing the secretory activity per cell, concomitant pregnancy accelerates the gradual regression of the mammary gland as lactation advances. Similarly, preliminary data for lactating dairy cows (A. V. Capuco, unpublished data) indicate that concomitant pregnancy increases the apoptotic index (TUNEL-positive cells) several times over, but is accompanied by a substantial increase in cell proliferation (BrdU labeling index). This suggests that the rate of mammary cell turnover during lactation is increased by concomitant pregnancy. Whether apoptosis or proliferation dominates might depend on the stage of gestation and the net effect of steroids and other hormones of pregnancy.
Peaker and Linzell investigated effects of estrogen on caprine lactation (Peaker and Linzell, 1974). They found that administration of estrogen at doses designed to match the secretion rate in late pregnancy caused a significant decline in milk yield. Additionally, they found that changes in milk secretion occurred during the estrous cycle. Several days preceding estrus, there was a clear change in milk composition characterized by increases in milk sodium and chloride concentrations and decreases in potassium and lactose. Milk composition returned to normal by the time estrous behavior was observed. Milk yield often declined during estrus, but was associated with a decline in feed intake. Whether increased secretion of estrogens during the normal ovarian cycle directly impact milk yield and, ultimately, persistency of lactation is not clear. However, data indicate that estrogens at concentrations observed during pregnancy can reduce milk yield.
It is clear that estrogen can inhibit milk secretion. The site of action appears to be the mammary gland, as implantation of estrogen pellets in rat mammary gland inhibited milk secretion, whereas implantation at the pituitary enhanced milk secretion (Bruce and Ramirez, 1970). However, estradiol does not appear to serve as a regulator of the IGFBP-5-regulated apoptosis described subsequently (Tonner et al., 1997; 2000).
Impact of Bovine Somatotropin, Milking Frequency, and Photoperiod on Cell Proliferation During Lactation
Small decreases in the rate of apoptosis or increases in the rate of cell proliferation will alter mammary regression during lactation. The slow rate of apoptotic cell death during lactation indicates that there are clear limitations on the capacity to globally decrease this rate in nonpregnant cows. But the opportunity to effect substantial change by decreasing apoptosis in pregnant lactating cows may exist. In all cases, a fruitful approach will likely be to increase cell proliferation during lactation so as to increase the maintenance of epithelial cell number and lactational persistency. Three potential means to alter the lactation curve have been investigated in recent years: bovine somatotropin (bST) administration, alterations in milking frequency, and photoperiodic manipulation.
The apparent lack of mammary growth during early lactation in cows (Capuco et al., 2001a) does not preclude the ability of mammary epithelial cells to proliferate in response to management conditions. For example, milking 6x daily during early lactation caused an increase in milk yield that persisted after less frequent milking (3x) was resumed (Bar-Peled et al., 1995). Although mammary growth was not assessed, this carryover effect is consistent with increased mammary growth during early lactation. Finally, net mammary growth can occur during lactation, as demonstrated by the compensatory growth response to various stimuli (Knight and Peaker, 1984a; Capuco and Akers, 1990). Suckling and milk removal during early lactation are major controlling factors of mammary cell number in rodents. Current data suggest that similar mechanisms may be operative in dairy cows and goats.
The influence of bST on the proliferative and apoptotic status of cells within the lactating bovine mammary gland was recently evaluated (Capuco et al., 2001a). Previous experiments had indicated that bST increased the persistency of lactation (Phipps et al., 1991; Van Amburgh et al., 1997; Bauman et al., 1999). In light of the demonstration that decreasing milk yield with advancing lactation is due to a decline in mammary cell number and not cellular activity, it follows that the effect of bST on persistency is due to maintenance of the mammary cell population rather than maintenance of cellular secretory rate. Indeed, bST maintained cell number in lactating caprine mammary glands (Knight et al., 1990) and data suggest that bST increases mammary cell proliferation during late gestation in sheep (Stelwagen et al., 1993) and perhaps in heifers (Stelwagen et al., 1992). The nuclear proliferation antigen Ki-67 was used as an index for the relative proliferation state of mammary tissue. The Ki-67 protein serves as a marker for cells that are engaged in cell cycle progression due to its presence during all phases of the cell cycle, except the quiescent G0 phase (Gerdes et al., 1984). This protein correlates with cellular proliferation under a variety of physiological conditions in human, rodent, and bovine tissues (Capuco et al., 2001a; Hardville and Henderson, 1966; Shayan et al., 1999). Treatment of first-lactation dairy cows with bST for 7 d increased the proportion of cells expressing the nuclear proliferation antigen by threefold (Figure 3). The effect of bST was evident when cows were given ad libitum access to feed (2.5 vs. 0.7%) or when they were restricted to 80% of ad libitum intake (0.8 vs. 0.3%) to induce a negative energy balance (Capuco et al., 2001b). Overall, cows in negative energy balance had a lower percentage of mammary cells expressing Ki-67 than did comparable cows in positive energy balance. These data strongly support the hypothesis that bST increases cellular proliferation in the bovine mammary gland and that mammary cell proliferation is blunted by reduced energy balance. Additionally, these data suggest that a proliferative response to bST would be blunted during early lactation (Capuco et al., 2001a) due to the negative energy balance cows are in during early lactation.
|
), earlier investigations had indicated that bST treatment did not cause net growth of the lactating bovine mammary gland (Capuco et al., 1989; Binelli et al., 1995). The earlier results do not infer that bST does not increase the rate of cell proliferation, but that the rate of cell proliferation did not exceed the rate of cell death. We suggest that treatment with bST increased the rate of mammary cell proliferation so as to reduce the rate of mammary regression during lactation. Additional study will be necessary to determine if extended treatment with bST increases the number of mammary cells relative to that of cohort controls. Previous slaughter studies may not have been sufficiently sensitive to detect small (e.g., 10%) differences in cell number. Alternatively, increased cell renewal may enable milk synthesis in bST-treated cows to continue at high capacity by enhanced replacement of senescent secretory cells.
Increased milking frequency (IMF) at the beginning of lactation has been shown to increase milk yield not only during IMF, but also after its cessation (Bar-Peled et al., 1995). Using a within-udder experimental design, increased milking frequency (4x) during mid-lactation was shown to increase cell proliferation when assessed 4 wk after initiation of treatment (Hillerton et al., 1990). We have recently evaluated the immediate effects of increased milking frequency initiated during early lactation on mammary growth and the long-term effects on milk yield (Hale et al., 2002). A novel approach of using four very uneven milking intervals was employed to obtain increased frequency without the added labor and other inherent costs involved in establishing a traditional pattern of increased milking frequency with fairly even milking intervals. Cows were divided into three treatment groups: 1) controls were cows milked twice daily (2x) beginning at parturition, 2) IMF1: cows milked four times daily (4x) from d 1 to 21 postpartum 3) IMF4: cows milked 2x from d 1 to 3 and 4x d 4 to 21 postpartum. The 4x cows were milked immediately before 2x cows and again (approximately 3 h later) after the 2x cows at both morning and evening milkings. All cows were milked 2x from d 21 to 305 postpartum. Milking 4x during early lactation increased milk yield not only during the treatment period, but it also elicited an increase in milk yield for the entire lactation. Milk yields were 33.5, 42.3, and 38.3 kg/d during wk 1 to 3 (P = 0.018) and 34.8, 37.4, and 37.5 kg/d during wk 4 to 40 (P = 0.058) for control, IMF1, and IMF4, respectively. Mammary biopsies from four cows per treatment were obtained on d 7 and 14 postpartum to assess mammary cell proliferation. Tritiated-thymidine incorporation by tissue slices was increased (P = 0.09) on d 7 in IMF1 cows, and arithmetic means for the percentage of cells expressing Ki-67 proliferation antigen were consistent with a proliferative response to IMF. Increased milking frequency during early lactation may increase mammary growth and thus produce a carryover effect on milk production for the majority of lactation. A carryover effect was observed with a minimal increase in labor and operating costs and was effective when increased milking was initiated on the first day of lactation or on d 4 after the routine interval for discarding milk colostrum. Whether this protocol significantly increases milk component secretion remains to be demonstrated.
Other investigations had shown that increased milking frequency enhanced mammary cell proliferation. Increased milking frequency increased milk production of goats due to a rapid increase in the activity of mammary secretory cells, often followed by proliferation of secretory tissue (Wilde et al., 1987; Knight et al., 1990). Increased milking frequency is hypothesized to increase milk production by lessening the accumulation of a feedback inhibition of milk secretion (Henderson and Peaker, 1984). Because frequent milking of one mammary gland has no effect on milk secretion by the opposing gland, it is clear that milk removal, and not systemic effects of milking, plays an important role in establishing secretion rate (Linzell and Peaker, 1971). Whether removal of inhibitors of lactation can adequately explain the impact of supplemental milkings with very short intermilking intervals is uncertain.
Lactating dairy cows exposed to long-day photoperiods (16 to 18 h of light) produce more milk than cows exposed to short-day (<12 h light) photoperiods (reviewed in Dahl et al., 2000). The endocrine mediators of this response appear elusive. Because long-day photoperiods increase systemic concentrations of prolactin, and because prolactin is known to be galactopoietic in rodents, it was hypothesized that this hormone may mediate the galactopoietic effects of long days in cattle. However, prolactin administration does not increase milk yield of lactating dairy cows (Plaut et al., 1987). Furthermore, milk production responses to long-day photoperiods are observed in the absence of photoperiod-induced changes in prolactin concentrations, such as during freezing temperatures and melatonin feeding (Dahl et al., 2000). The galactopoietic effects of long-day photoperiod are associated with increases in IGF-I secretion that precede the milk production response (Dahl et al., 1997), and IGF-I appears the most likely mediator of the galactopoietic effects of increased photoperiod (Dahl et al., 2000).
In addition to uncertainty regarding the hormonal mediator of the lactational response to increased photoperiod, the cellular mechanisms responsible are uncertain. The lactation of cows on long-day photoperiods appears more persistent than that of cows on short days. Consistent with the delay (approximately 4 wk) in onset of response, effects on persistency may be a contributing factor to the galactopoietic response to photoperiod.
Galactopoietic effects of bST, IMF, and long-day photoperiod are additive. It is reasonable to assume that effects on persistency will similarly be additive, and that the greatest persistency will be achieved by a combination of the three treatments. For the producer, there may be considerable flexibility in incorporating these treatments since IMF may be effective when restricted to the immediate postpartum period and when 4x milking is imposed with a very uneven milking intervals as previously described. This reduces problems associated with incorporating a regulated lighting regime with frequent milking. Flexibility is also increased by the recent demonstration that galactopoietic effects may be realized by restricting photoperiodic manipulation to the dry period (Miller et al., 2000).
Impact of Insulin-Like Growth Factor Survival Factors on Lactation Persistency
Insulin-like growth factors appear to be galactopoietic and to serve as participants in the galactopoietic response to exogenous bST. When the production of IGF-I is uncoupled from somatotropin regulation, such as occurs during negative energy balance, then a milk production response to bST is abrogated (Gluckman et al., 1987; McGuire et al., 1992). Infusion of IGF-I into the local arterial supply to the mammary gland of goats rapidly increased milk synthesis (Prosser et al., 1990; Prosser and Davis, 1992), whereas infusion of bST was ineffective (McDowell et al., 1987), supporting a galactopoietic action of IGF-I. Furthermore, treatment of lactating cows with bST altered the distribution pattern of IGF-I within mammary tissue from a predominantly stromal localization to a prominent epithelial localization (Glimm et al., 1988), and altered the nature of IGF-I receptor transcripts in mammary tissue (Glimm et al., 1992), suggestive of IGF-I mediation of bST action in mammary tissue. The milk yield response to local mammary infusion of IGF-I is considerably less than that obtained with systemic administration of bST; however, this is consistent with known direct effects of bST on lipid metabolism of adipose tissue (Bauman, 1999) and does not necessarily weaken the somatomedin hypothesis. Although a degree of uncertainty may remain when ascribing IGF-I as a mediator of the lactational effects of bST (Tucker, 2000), the above data and previously discussed data regarding the galactopoietic effects of photoperiod (Dahl et al., 2000) support a galactopoietic role for IGF-I.
Several levels of complexity are involved in modulating IGF-regulated functions: the local concentration of IGF, expression of IGF receptors and their downstream signaling pathways, and the types and quantities of IGF binding proteins (Peaker and Linzell, 1974; Cohick, 1998; Hadsell et al., 2002). To date, six high-affinity IGFBP (1 to 6) and nine low-affinity IGFBP, also known as IGFBP-related proteins, have been identified. Mammary epithelial cells synthesize a number of IGFBP. Depending on the specific IGFBP, the binding proteins may reduce IGF activity by competing with IGF-receptors for ligand, increase IGF-activity by serving as delivery vehicles to the target cell, or serve as a reservoir for IGF, causing their slow release and reducing IGF turnover. Furthermore, the IGFBP may have activities that are independent of their interaction with IGF, and they are subject to enzymatic modifications that may alter their various activities. The IGF system is highly complex with multiple levels of regulation, making the specific actions of IGF during mammary development and lactation difficult to resolve.
IGF-I is a mammary mitogen and survival factor. The ability of IGF-I to induce cell proliferation has been demonstrated in several in vitro and in vivo mammary model systems (Peaker and Linzell, 1974; Baumrucker and Erondu, 2000; Hadsell et al., 2002). Recently, administration of bST was shown to increase the percentage of mammary epithelial cells expressing Ki-67, a nuclear antigen marker for cell proliferation, by approximately threefold (Capuco et al., 2001a). It was proposed that this apparent proliferation response to bST was mediated by IGF-I. However, bST may directly influence mammary function without invoking mediation by IGF-I (Glimm et al., 1990; Hauser et al., 1990; Lincoln et al., 1995).
The concept that IGF-I regulates mammary gland apoptosis has evolved in recent years. In vitro studies with MCF-7 (human mammary adenocarcinoma cell line) cells demonstrated that IGF-I could serve as an inhibitor of mammary cell apoptosis (Geier et al., 1992). Subsequent studies in rats demonstrated that mammary apoptosis was correlated with enhanced expression of IGFBP-5, which serves as a negative regulator of IGF-I action by virtue of its ability to bind IGF-I and reduce its cell survival promoting activity (Tonner et al., 1995). Mammary-specific expression of IGF-I in transgenic mice demonstrated that IGF-I promoted cell survival and delayed mammary gland involution (Hadsell et al., 1996; Neuenschwander et al., 1996). Prolactin appeared to be a key factor in regulation of IGFBP-5 (Tonner et al., 1997), so that the outcome of prolactin insufficiency is increased mammary apoptosis. Recent investigations utilizing bovine mammary explants demonstrated the same relationship between level of apoptosis and expression of IGFBP-5 (Accorsi et al., 2002). Similar to the situation in rodents, apoptosis and IGFBP-5 expression were facilitated by the absence of the lactogenic hormones: prolactin, IGF-I and, in this case, somatotropin. Consistent with these findings, bST increases lactational persistency and maintains mammary cell number as lactation advances in ruminants by increasing cell renewal or survival (Knight et al., 1990; Capuco et al., 2001a).
Impact of Mastitis on Death of Mammary Epithelial Cells
A significant negative correlation exists between somatic cell count in the milk and milk yield (Raubertas and Shook, 1982). Infection is the primary reason for increased SCC. During experimentally induced Staphylococcus aureus mastitis, extensive tissue damage is evident in regions where neutrophils appear to traverse the epithelium (Harmon and Heald, 1982). The tissue damage and the decline in milk production are associated with the immunological defense mechanisms. Processes related to the bacterial infection as well as normal neutrophil function induce cell death and detrimentally affect milk secretion (Capuco et al., 1986; Long et al., 2001).
Two forms of cell death predominate: death by necrosis and by apoptosis. The relationship between mastitis and mammary cell apoptosis was evaluated in vivo after injection of Escherichia coli into mammary glands of lactating Holstein cows (Long et al., 2001). The apoptotic index was significantly increased in mastitic tissue compared to uninfected control. Infection elicited increases in expression of proapoptotic genes (Bax and interleukin-1ß converting enzyme), whereas expression of the antiapoptotic gene (Bcl-2) was decreased. Induction of matrix metalloproteinase-9, stromelysin-1, and urokinase-type plasminogen activator were also increased, consistent with degradation of the extracellular matrix and cell loss during mastitis. The Ki-67 labeling index suggested that mastitis also increased cell proliferation, perhaps as a tissue repair mechanism after mastitis. This response may be consistent with an increase in local (milk) concentrations of IGF-I in the mammary glands of cows infected with E. coli (Shuster et al., 1995). The signaling events involved in mammary cell apoptosis and proliferations induced by E. coli infection are not fully understood. The ligands, receptors, and genes participating in transmitting the death and proliferating signals have not been clarified and require further investigation.
Infection with Gram-positive bacteria similarly appears to induce apoptosis. Streptococcus agalactiae-induced bovine mastitis increased the expression of an apoptosis marker, RTPM-2 (Sheffield, 1997), and S. aureus-induced apoptosis of MAC-T cells, a bovine mammary epithelial cell line (Bayles et al., 1998; Wesson et al., 1998).
A mammary explant model has been used to dissociate tissue damage resulting from neutrophil diapadesis from effects of neutrophil function and bacterial toxins (Capuco et al., 1986). Mammary tissue from lactating Holsteins was cultured in the presence of intact or lysed neutrophils or neutrophils that were allowed to phagocytose opsonized zymosan. Phagocytosing neutrophils inflicted the greatest damage on epithelial cells. Cytological damage observed included cell sloughing from the basement membrane, nuclear condensation, cell debris in luminal areas, and epithelial vacuolation. During routine immune surveillance, neutrophils migrate through the epithelial layer and into alveolar and ductal lumena, where they phagocytose fat globules, casein micelles, and bacteria when they are present. Thus, phagocytosis is the consequence of routine immune surveillance and has the potential to damage the mammary epithelium. Such effects may account for the negative correlation between SCC and milk yield, even in the absence of infection.
Using an in vitro co-culture of MAC-T cells and bovine neutrophils, the efficacy of antioxidants as protectants against neutrophil-induced epithelial damage was investigated (Boulanger et al., 2002). Data implicated hydroxyl radicals released by activated neutrophils as agents that could damage the secretory epithelium, and that antioxidants such as melatonin may be useful for protecting mammary tissue during mastitis.
Significance of a Dry Period and the Concept of Regenerative Involution
In contrast to other species, normal management of dairy cows and goats results in an overlap of lactation and pregnancy, such that these animals are typically pregnant when milking is terminated during late lactation. Thus, when milk stasis occurs, the mammogenic and lactogenic stimulation of pregnancy opposes stimuli for mammary involution. Processes of mammary growth and involution both occur during this nonlactating or "dry period" between successive lactations. Milk production efficiency can be increased by development and utilization of schemes, which increase persistency of lactation and which minimize the duration of the dry period.
In cows, a dry period of at least 40 d has been recommended to maximize milk production in the following lactation (Swanson, 1965; Coppock et al., 1974; Sorensen and Enevoldsen, 1991). This may be to permit restoration of body reserves prior to the next lactation or to permit necessary growth and differentiation events within the mammary gland during this period. Although the data are not definitive, they strongly suggest that a dry period is necessary for reasons that center on the mammary gland rather than on the nutritional status of the animal (Swanson, 1965; Smith et al., 1966; Smith et al., 1967). Interestingly, a dry period between successive lactations was also found to be necessary for optimal lactation in rats (Paape and Tucker, 1969).
Aspects of mammary growth during the dry period have been investigated (Capuco et al., 1997). Multiparous Holstein cows were dried off 60 d prior to expected parturition or were milked twice daily during this prepartum period. Cows were slaughtered at 7, 25, 40, and 53 d into the dry period (53, 35, 20, and 7 d prepartum), and total mammary DNA and thymidine incorporation into mammary tissue slices was determined. There was no significant loss of mammary cells (DNA) during the dry period, and total number of mammary cells increased with advancing stages of the dry period. Total DNA did not differ between mammary glands of dry and lactating cows; however, increased DNA synthesis in dry cows indicated that replacement of mammary cells increased during the dry period. Autoradiographic localization of incorporated 3H-thymidine indicated that the replicating cells were primarily (>90%) epithelial. Thus, in cows, the dry period may be important for replacing senescent cells prior to the next lactation. Furthermore, although cows appeared to enter the next lactation with the same number of mammary cells regardless of whether they had a dry period, a greater percentage of those cells were epithelial in cows that had a dry period.
It is important to realize that events occurring during the dry period differ significantly from mammary involution as frequently studied in rodent models. In the absence of concomitant pregnancy mammary involution is rapid and extensive. In contrast, the mammogenic effects of pregnancy ameliorate the involution process and promote mammary cell turnover. When mice are pregnant at the time of weaning, mammary apoptosis is reduced and cell proliferation enhanced relative to their nonpregnant counterparts (Capuco et al., 2002). Consequently, we propose the term "regenerative involution" to more fully describe the processes of cell renewal and tissue remodeling that occur during involution with concomitant pregnancy.
We have hypothesized that the dry period may be critical for replacing progenitor cells that are responsible for expanding and maintaining the number of mammary secretory cells (Capuco and Akers, 1999). Indeed the mammary glands of rats that were not permitted a dry period had fewer cells at midlactation than glands of rats that were permitted a dry period of optimal length, although cell number did not differ at onset of the lactation (Paape and Tucker, 1969). If replacement of senescent cells is a critical event during the dry period, one would hypothesize that without a dry period of sufficient length to replace senescent cells, persistency of the ensuing lactation will be decreased.
In contrast to cows, a dry period may not be necessary for optimal milk production in dairy goats. Two investigations have addressed the importance of a dry period in goats. In the first (Knight and Wilde, 1988), lactating goats were induced into ovulation and mated out of season. The goats entered the next lactation without a dry period, and milk production was found to be 12% less than in the previous lactation. However, seasonal effects on lactation may have confounded the results. Subsequently, Fowler et al. (1991) investigated the necessity for a dry period using a within-animal, half-udder design. One gland was milked during the prepartum period, whereas the other was dried off 24 wk (170 d) prior to parturition. In this experiment, magnetic resonance imaging was used to monitor parenchymal volume as an index of mammary growth and involution. There was no difference in milk production between glands. Indeed, at no stage of lactation was milk yield of glands that had experienced a dry period numerically greater than that of continuously milked glands, even though those glands were larger than continuously milked glands during the first few weeks after parturition. These data suggest that a dry period is not necessary for optimizing milk production in the next lactation in goats. Issues of half-udder design may confound interpretation (Capuco and Akers, 1999); however, similar studies in cattle demonstrate a beneficial effect of a dry period. Because significant mammary growth occurs during early lactation of goats, and milking induces the release of somatotropin in goats, but not cows, the mammary gland of goats may demonstrate a greater capacity to continue cell-renewal processes, such as those detected during a bovine dry period, into early stages of lactation, thus largely negating the importance of a dry period in goats.
Bachman and colleagues have evaluated the feasibility of accelerating mammary involution during the dry period by administration of estradiol-17ß at dry off (Athie et al., 1996; 1997). By accelerating the early stages of involution, it was hoped that dry periods <40 d could be employed without a loss of milk production in the subsequent lactation. The most recent report indicated that in the absence of treatment, no milk deficit was incurred by utilizing a dry period of 34 vs. 59 d (Bachman et al., 1988), as well as the balance of other management and health costs. Management decisions regarding the length of a dry period are based on the balance of forfeited milk during the dry period and enhanced milk production in the ensuing lactation. The quantitative impact of dry period length on milk production should be reevaluated. The dairy cow population today differs significantly from that used in the classical investigations upon which current management decisions are based.
Identification of Putative Progenitor Cells
Evidence for the existence of mammary stem cells is available from a variety of sources. Numerous transplantation experiments have shown that isolated segments from any portion of the developing or even lactating gland are capable of regenerating a complete mammary ductal and alveolar network (DeOme et al., 1959; Hogg et al., 1983; Smith and Medina, 1988). Perhaps most convincingly, Kordon and Smith showed that an entire mammary gland could be regenerated with the progeny of a single cell following transplantation into cleared mammary fat pads (Kordon and Smith, 1998). Additional evidence for the existence of mammary stem cells may be derived from observations that entire mammary lobules are often comprised of cells showing identical X-inactivation patterns, and from cancer studies where mammary tumors comprised of a variety of cell types are frequently found to be of clonal origin (Tsai et al., 1996).
No genetic marker has yet been found to identify mammary stem cells in situ. However, histological analyses have indicated that a pale staining cell population present during all stages of mammary development and differentiation in mice and rats may serve as mammary stem cells (Chepko and Smith, 1997). Such "pale cells" have been described in mammary tissue from all species so far examined, including humans (Ferguson, 1985), mice (Smith and Medina, 1988), rats (Chepko and Smith, 1997), goats (Li et al., 1999), and cattle (Ellis et al., 2000; Ellis and Capuco, 2002).
To test the hypothesis that lightly staining mammary epithelial cells represent a stem cell population, we evaluated the proliferative capacity of these cells in mammary glands of prepubertal heifers (Ellis and Capuco, 2002). Prepubertal heifers were used because of the active proliferative state of the mammary gland during this stage of mammary development (Hardville and Henderson, 1966). Heifers were injected with BrdU to label cells in S-phase of the cell cycle, and the labeling index of mammary epithelium, obtained at slaughter 2 h post injection, was evaluated relative to histological appearance of the cells. We observed light, dark, and intermediate staining cells in histological sections (Figure 4A, B). Light cells comprised 10% of the total parenchymal cell population, but accounted for the majority of epithelial cell proliferation. Light and intermediate cells together accounted for more than 90% of the proliferating cells. The proportion of light cells was relatively constant across the stages of development that were evaluated (2, 5, and 8 mo). However, the proportion of light plus intermediate staining cells correlated with the tissue’s proliferative rate (Figure 5). These data strongly support the hypothesis that lightly staining mammary epithelial cells function as the primary proliferative population and provide solid justification for further studies into the light staining cell phenotype. We hypothesize (Figure 6) that the lightly staining cells provide the basal population of mammary epithelial progenitor or stem cells and that the intermediately staining cells represent a daughter population that serves to amplify the progenitor population and may have a more restricted developmental potential. The dark cells represent more differentiated mammary epithelial cells.
|
|
|
). However, their identification is often difficult because of the presence of fat globules in epithelial cells that tend to obscure their cytoplasmic staining character. Future studies will employ a variety of approaches to identify differences in gene expression that are related to the mammary epithelial cell’s staining characteristics. Development of markers for these cells would facilitate kinetic analyses of the relationship between the different staining phenotypes and permit identification of factors that regulate the proliferation of these cells. Such knowledge should provide a key for regulating mammary development and persistency of lactation. |
Summary |
|---|
|
TopAbstractIntroductionSummaryImplicationsLiterature Cited |
|---|
). With incomplete milking, regions of the mammary gland likely undergo reductions in blood flow and oxygen tension, which can initiate apoptosis. Also, with incomplete milking, mammary tight junctions are compromised and resulting changes in cytoskeletal organization lead to decreased milk synthesis and secretion (Stelwagen et al., 1997). If not remedied, these changes will ultimately induce apoptosis (Kulms et al., 2002).
|
A reevaluation of the optimal dry period length is in order. During the dry period there is extensive cell renewal, which may reflect a critical need to replace senescent mammary epithelial cells. Because dairy cows are typically pregnant at the time of dry-off, there is extensive remodeling of the mammary gland during this preparturient, nonlactating period. A descriptive term for the nature of the opposing events of involution and mammogenesis is "regenerative involution." This contrasts with the often-studied involution in rodents that occurs after forced weaning, during which the mammary gland involutes to a state that resembles that of a virgin animal. Schemes to increase the rapidity of regenerative involution will permit shorter dry periods. However, the plasticity of the mammary gland may provide opportunities to utilize dry periods of short duration.
|
Implications |
|---|
|
TopAbstractIntroductionSummaryImplicationsLiterature Cited |
|---|
1 Correspondence: Bldg. 200, Room 14 (phone: 301-504-8672; fax: 301-504-8414, E-mail: acapuco{at}anri.barc.usda.gov).
Received for publication August 8, 2002. Accepted for publication January 16, 2003.
|
Literature Cited |
|---|
|
TopAbstractIntroductionSummaryImplicationsLiterature Cited |
|---|
