| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
ANIMAL GROWTH, PHYSIOLOGY, AND REPRODUCTION |
* Caine Veterinary Teaching Center, and
and
Research and Extension Center, University of Idaho, Caldwell 83607; and and
Dairy and Animal Science Department, Pennsylvania State University, University Park 16802
 |
Abstract |
|---|
 Top Abstract INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
. Heifers in the BRED group were inseminated on detection of estrus or at a fixed time, 72 h after insert removal concomitant with a GnRH treatment. Heifers in the NONBRED group received an injection of GnRH 48 h after insert removal. Blood samples collected on d 0 (d of AI or estrus) and 18 were used to determine steady-state levels of MX2 mRNA. Samples collected on d 0, 7, 14, and 21 were analyzed for progesterone concentration. Pregnancy was determined retrospectively by progesterone concentration on d 21 and was diagnosed at 30 ± 1 and 60 ± 3 d after AI. The fold change in levels of MX2 mRNA from d 0 to 18 was greater for heifers classified and diagnosed as pregnant on d 21 (P < 0.05) and 30 ± 1 (P < 0.05) and 60 ± 3 (P < 0.05) d after AI compared with nonpregnant (bred but not pregnant) and NONBRED heifers. Heifers that experienced pregnancy loss from 21 to 30 ± 1 (P = 0.11) or 21 to 60 ± 3 (P = 0.08) d of gestation tended to have smaller fold increases in MX2 mRNA expression than those that maintained pregnancy. The sensitivity (range 57.1 to 65.6%) and negative predictive values (range 47.9 to 57.1%) of determining reproductive status on d 18 according to the change in the level of MX2 mRNA expression in PBL were low, and the correlation between diagnosis of pregnancy by fold change in MX2 mRNA expression and other methods was small (r = 0.20 to 0.36). The current study indicates that increased expression of MX2 mRNA in PBL is related to pregnancy approximately 21, 30, and 60 d after AI in dairy heifers and that losses that occurred later in pregnancy were associated with lower fold increases in MX2 mRNA. However, using the change in MX2 mRNA expression was not a reliable method for diagnosis of pregnancy at 18 d after AI because of the low sensitivity and negative predictive value.
Key Words: gene expression • heifer • leukocyte • pregnancy
|
INTRODUCTION |
|---|
|
TopAbstractINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
(IFN-), which is a member of the type I IFN family (Binelli and Thatcher, 1999
; Demmers et al., 2001). During pregnancy, IFN- is produced by the conceptus beginning on d 13 to 15 of pregnancy and inhibits the luteolytic cascade (Helmer et al., 1987, 1989; Kazemi et al., 1988; Thatcher et al., 1989). Like other type I IFN, IFN- has antiviral, antiproliferative, and immunoregulatory activities and increases the expression of IFN-stimulated genes, including the Myxovirus resistance (MX) gene, whose functions are to limit viral replication and clear infected cells (Pontzer et al., 1997; Malmgaard, 2004; Bazer and Spencer, 2006).
Until recently, it was thought that IFN- acted locally in the endometrium, because IFN activity had not been detected in uterine venous or lymphatic drainage (Charleston and Stewart, 1993; Ott et al., 1998; Hicks et al., 2003). However, Yankey et al. (2001) demonstrated that pregnant ewes had elevated expression of MX1 messenger RNA (mRNA) in peripheral blood leukocytes (PBL) beginning on d 15 after mating compared with bred, but not pregnant, ewes, suggesting that IFN- produced a rapid systemic response. This observation was recently confirmed in lactating dairy cows, in which the increase in levels of MX1 and MX2 genes in PBL was greater in pregnant compared with bred, nonpregnant cows (Gifford et al., 2007). However, previous studies did not evaluate the reliability of using changes in MX2 mRNA in PBL to determine reproductive status.
The hypotheses of the current study were that changes in levels of MX2 mRNA in PBL of dairy heifers after insemination are correlated with reproductive status and that these changes can be used to determine reproductive status.
Therefore, the objectives of the current study were to quantify changes in expression of MX2 mRNA in PBL of dairy heifers of different reproductive statuses and to evaluate the reliability of using change in expression of MX2 mRNA for determination of nonpregnancy.
|
MATERIALS AND METHODS |
|---|
|
TopAbstractINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
Animals and Housing
Holstein heifers (n = 266), 13 ± 1 mo of age and originating from a dairy in the Treasure Valley of Idaho, were used in this study. Heifers were housed in open lots of similar size, design, and number of animals and were fed a total mixed ration to meet the nutritional requirements for Holstein heifers weighing 360 kg and gaining 0.8 kg/d (NRC, 2001).
Synchronization Protocols and AI
Heifers were assigned randomly to receive AI (BRED) or to not receive AI (NONBRED) after the completion of a synchronization protocol (Figure 1). Therefore, all heifers received an intravaginal insert (EAZI-BREED CIDR insert, Pfizer Animal Health, New York, NY) containing 1.38 g of progesterone for 7 d. Upon insert removal, all heifers received an injection of 25 mg of PGF2 (dinoprost tromethamine, Lutalyse sterile solution, Pfizer Animal Health). Heifers in the NONBRED group received 1 injection of 100 µg of GnRH (gonadorelin diacetate tetrahydrate, Cystorelin, Merial Ltd., Iselin, NJ) 48 h after insert removal and were not inseminated. During the 3 d after insert removal, signs of estrus in the BRED group were detected once daily in the morning with the aid of tail paint (All-weather Paintstick, LA-CO Industries, Chicago, IL). Those heifers not inseminated at estrus received fixed-time insemination 72 h after insert removal, concomitant with an injection of 100 µg of GnRH. One technician inseminated the majority of the heifers, and a relief technician inseminated heifers once a week.
|
Blood Sampling and PBL Isolation
Blood samples (7 mL) were collected from the coccygeal vein or artery into EDTA-containing evacuated tubes (Becton Dickinson, Franklin Lakes, NJ) on study d 0, 7, 14, and 21 for measurement of progesterone concentration (Figure 1
). After collection, blood tubes were placed in ice and transported to the laboratory within 4 ± 2 h of collection, where they were immediately centrifuged at 2,000 x g for 20 min. Plasma was harvested and frozen at –65°C and later analyzed for concentrations of progesterone by RIA (Kulick et al., 1999). The sensitivity of the assay was 0.15 ng/mL, and the inter- and intraassay CV were 9.41 and 2.69%, respectively.
On study d 0 and 18, blood samples (9 mL) were also collected as described above for harvesting of PBL (Figure 1). Samples were centrifuged at 300 x g for 20 min at 4°C. The buffy coat was removed and resuspended in 0.87% Tris-NH4CL lysis buffer at a 1:5 (vol/vol) ratio. Samples were then incubated for 5 min at 37°C and centrifuged at 300 x g for 10 min. The supernatant was removed, and the pellets were washed with 10 mL of 1x PBS and were again centrifuged at 300 x g for 10 min. The supernatant was removed, and the pellets were resuspended and lysed with 2 mL of Trizol (Invitrogen Life Technologies, Grand Island, NY) and stored at –80°C for RNA extraction. The average (±SD) time that elapsed from collection of the blood samples to stabilization of the mRNA was 5.89 ± 4.85 h.
mRNA Extraction and Quantitative, Real-Time PCR
Total cellular mRNA was extracted from PBL using Trizol according to the instructions of the manufacturer. The concentration of RNA was calculated by measuring absorbance at 260 nm, and 5 µg of total cellular RNA was used to synthesize complementary DNA using Strata Script RT (Stratagene, La Jolla, CA). Aliquots of complementary DNA were then combined with MX2 forward (5' to 3': CTTCAGAGACGCCTCAGTCG) and reverse (5' to 3': TGAAGCAGCCAGGAATAGTG) primers to make up the qrt-PCR master mix. SYBR Green (Finnzymes DyNAmo SYBR Green qPCR Kit, New England BioLabs, Ispwich, MA) was added to the qrt-PCR master mix, and aliquots were pipetted in duplicate into 96-well plates. ß-Actin was also measured in duplicate for each sample to adjust for loading errors and interplate variation (Gifford et al., 2007). Samples were then subjected to the following thermal cycling parameters: a) 95°C, 30 s; b) annealing temperature of 56.5°C for 30 s; and c) 72°C, 30 s for 40 cycles. At the completion of the 40 cycles, the samples were then subjected to a melting curve analysis from 60°C to 92°C to confirm amplicon specificity. The results are expressed as fold change adjusted for the levels of ß-actin (control housekeeping gene), as described by Kubista et al. (2006).
Classification of Synchrony of the Estrous Cycle
Heifers were classified according to progesterone concentrations on study d 0, 7, and 14 as having their estrous cycle synchronized or not. Heifers that had a progesterone concentration >1.0 ng/mL on study d 0 and those that had a progesterone concentration 
1.0 ng/mL on study d 7 or 14 were classified as not synchronized and were not used for statistical analysis. Therefore, heifers were classified as synchronized and used in the statistical analyses when their progesterone concentrations were 1.0 ng/mL on study d 0 and >1.0 ng/mL on study d 7 and 14.
Classification of Reproductive Status Based on Change in MX2 mRNA Steady-State Levels
Previous studies have demonstrated that the average fold change in expression of MX2 mRNA in PBL of bred, nonpregnant cows is <2.0 (Gifford et al., 2007). However, to minimize the risk of classifying a pregnant heifer as nonpregnant according to fold change in MX2 mRNA in the current study, heifers were classified retrospectively as nonpregnant on study d 18 (18 to 19 d after AI), when the fold change in the steady-state level of MX2 mRNA was 1.47. This value was obtained by subtracting the SEM from the mean fold change for heifers diagnosed nonpregnant at 30 ± 1 d after AI.
Pregnancy Diagnosis
It was determined retrospectively that among heifers diagnosed as pregnant by ultrasonography at 30 ± 1 d after AI, the lowest concentration of progesterone on study d 21 was > 2.9 ng/mL. Therefore, heifers in the BRED group with progesterone concentrations > 2.9 ng/mL on study d 21 were classified as pregnant. Similarly, heifers in the NONBRED group were classified as pregnant on d 21 of the estrous cycle if progesterone concentration was >2.9 ng/mL to assess reliability of diagnosing pregnancy by progesterone concentration. Pregnancy was diagnosed 30 ± 1 d after AI by scanning of the uterus with ultrasound (7.5 MHz transrectal linear probe, Sonovet 600, Alliance Medical, Bedford Hills, NY) and visualization of uttering fluid, and embryo, and a fetal heart beat. Heifers diagnosed as pregnant at d 30 ± 1 after insemination were reexamined by palpation per rectum of the uterine contents 60 ± 3 d after AI.
Pregnancy Loss
Pregnancy loss (PL) was characterized between d 21 and 30 ± 1 after AI, d 30 ± 1 and 60 ± 3 after AI, and d 21 and 60 ± 3 after AI. Heifers determined to be pregnant on study d 21 by progesterone concentration and diagnosed as nonpregnant by ultrasonography on d 30 ± 1 after AI were considered to have experienced PL. Heifers diagnosed as pregnant by ultrasonography at 30 ± 1 d after AI and diagnosed as nonpregnant by palpation per rectum of the uterine contents on d 60 ± 3 after AI were considered to have experienced PL. Finally, those heifers classified as pregnant on study d 21 according to progesterone concentrations and diagnosed as nonpregnant 60 ± 3 d after AI were considered to have experienced PL.
Calculation of Sensitivity, Specificity, Positive Predictive Value, and Negative Predictive Value
Classification of the heifers as true positive (TP), false positive (FP), true negative (TN), and false negative (FN) was based on the comparison between the classification of reproductive status determined by the fold change in expression of MX2 mRNA (study d 18 = 18 to 19 d after AI or 18 d after estrus) and the classification or diagnosis of pregnancy based on progesterone concentration (study d 21 = 21 to 22 d after AI or 21 d after estrus), ultrasonography (30 ± 1 d after AI), and palpation per rectum (60 ± 3 d after AI).
Therefore, heifers were considered to be TP when they had an increase in expression of MX2 mRNA > 1.47 and were classified and diagnosed pregnant by other means at different time points (i.e., progesterone concentration on study d 21, ultrasonography 30 ± 1 d after AI, and palpation per rectum 60 ± 3 d after AI). Heifers were classified as FP when the increase in MX2 mRNA expression was > 1.47, and heifers were classified and diagnosed nonpregnant by other means at different time points. True negatives were those heifers with change in MX2 mRNA steady-state levels 1.47 and that were diagnosed nonpregnant by other methods at different time points. Finally, those heifers that had change in steady-state levels of MX2 mRNA 1.47 but were classified and diagnosed pregnant by other means at different time points were considered FN.
For determination of TP, FP, TN, and FN compared with presumptive reproductive status on study d 21 based on progesterone concentration, all heifers were used except for one that did not have a blood sample collected on study d 21. However, for determination of TP, FP, TN, and FN compared with pregnancy status at 30 ± 1 and 60 ± 1 d after estrus or AI, all heifers in the NONBRED group were used and only heifers diagnosed pregnant from the BRED group were used. This procedure was adopted to avoid overestimation of FP, because some of the heifers in the BRED group that were diagnosed nonpregnant at 30 ± 1 and 60 ± 3 d after AI could have been pregnant on study d 18 and lost the pregnancy thereafter. Therefore, for calculation of TP, FP, TN, and FN at 30 ± 1 d after AI or estrus, 72 heifers in the BRED group diagnosed nonpregnant were not included, and for calculation of TP, FP, TN, and FN 60 ± 3 d after AI or estrus, 77 heifers in the BRED group diagnosed as nonpregnant were not included (Table 1).
|
). The equations used to calculate each of these outcomes were as follows:
 |
Data Analyses
The average fold changes in steady-state level of MX2 mRNA from study d 0 to 18 according to reproductive status (BRED pregnant, BRED nonpregnant, and NONBRED) at different times after AI or estrus (study d 21 and 30 ± 1 and 60 ± 3 d after AI or estrus) were analyzed by ANOVA using the GLM procedure (SAS Inst. Inc., Cary, NC) with a model that included reproductive status.
The level of agreement between classification of reproductive status based on the fold change in steady-state level of MX2 mRNA and classification or diagnosis of pregnancy on study d 21 and 30 ± 1 and 60 ± 3 d after AI was evaluated by the Kappa procedure using the FREQ procedure of SAS.
Changes in progesterone concentration over time were analyzed by ANOVA for repeated measures using the MIXED procedure of SAS with a model that included the effect of reproductive status (BRED pregnant, BRED nonpregnant, and NONBRED), study day, and their interaction or the occurrence of PL (maintained pregnancy vs. lost pregnancy), study day, and their interaction. The covariance structures (unstructured, compound symmetry, toeplitz, and autoregressive order 1) for the repeated measures model were tested (Littell et al., 2000).
|
RESULTS |
|---|
|
TopAbstractINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
The proportion of heifers in the BRED group classified and diagnosed as pregnant on study d 21 and 30 ± 1 and 60 ± 3 d after AI was 65.3, 57.7, and 53.9%, respectively. Three (6.5%) heifers in the NONBRED group had progesterone concentration 
2.96 ng/mL on study d 21 and were incorrectly classified as pregnant. The incidence of PL from study d 21 to 30 ± 1 d after AI, study d 21 to 60 ± 3 d after AI, and between 30 ± 1 and 60 ± 3 d of gestation was 10.1, 17.4, and 8.2%, respectively.
The average fold change in steady-state level of MX2 mRNA from study d 0 to 18 was greater for BRED heifers classified and diagnosed as pregnant on study d 21 (P < 0.05) and 30 ± 1 (P < 0.05) and 60 ± 3 (P < 0.05) d after AI than BRED heifers classified and diagnosed as nonpregnant and NONBRED heifers (Figure 2). The average fold change in steady-state levels of MX2 mRNA for heifers in the NONBRED group was not different than that of heifers in the BRED group that were diagnosed as nonpregnant on study d 21 (P = 0.21) and 30 ± 1 (P = 0.25) and 60 ± 3 (P = 0.23) d after AI.
|
). However, there was an interaction (P < 0.05) between reproductive status and study day on progesterone concentration in that heifers classified as nonpregnant tended (P = 0.08) to have greater progesterone concentration on study d 14 and had (P < 0.05) smaller concentration on study d 21 than those heifers classified as pregnant (Figure 3).
|
). However, there was no difference (P = 0.47) in change in expression of mRNA for MX2 between heifers that experienced or did not experience PL from 30 ± 1 to 60 ± 3 d of gestation (Figure 4).
|
).
|
).
The frequency distribution of heifers classified as TP, FP, TN, and FN based on change in MX2 mRNA compared with other methods of pregnancy classification or diagnosis is depicted in Table 2. Classifying heifers as pregnant on study d 18 when change in steady-state level of MX2 mRNA was > 1.47 and as nonpregnant when change in steady-state level of MX2 mRNA was 1.47 resulted in sensitivity of 57.1, 62.2, and 65.6% compared with classification or diagnosis of reproductive status on study d 21 and 30 ± 1 and 60 ± 3 d after AI, respectively (Table 1). The specificity of classifying heifers as pregnant or nonpregnant according to change in steady-state level of MX2 mRNA resulted in specificity of 63.4, 73.9, and 73.9% compared with pregnancy classification or diagnosis on study d 21 and at 30 ± 1 and 60 ± 3 d after AI or estrus, respectively (Table 1). Compared with classification or diagnosis of reproductive status on study d 21 and 30 ± 1 and 60 ± 3 d after AI or estrus, using change in steady-state level of MX2 mRNA to identify pregnant and nonpregnant heifers on study d 18 resulted in negative predictive values of 57.14, 47.89, and 52.31%, respectively, and positive predictive values of 63.37, 83.56, and 83.10%, respectively (Table 1).
|
|
DISCUSSION |
|---|
|
TopAbstractINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
). Because these procedures require skilled personnel and because of the need to reinseminate nonpregnant cows as early as possible after the previous AI, alternative methods for early pregnancy diagnosis are constantly being pursued.
Interferon- is produced by the conceptus beginning on d 13 to 15 of pregnancy to inhibit the luteolytic cascade (Helmer et al., 1987, 1989; Kazemi et al., 1988; Thatcher et al., 1989), and peak production of IFN- is observed around d 17 of gestation (Bartol et al., 1985). Aside from blocking the secretion of PGF2, IFN- modulates the immune system supposedly to protect the allogeneic conceptus from the maternal immune response to the fetal antigens (Skopets et al., 1992). In vitro studies have demonstrated that IFN-, formerly known as bovine trophoblast protein-1, inhibits the proliferation and the activity of lymphocytes (Skopets et al., 1992). However, innate immunity was increased by the presence of IFN- and pregnancy in sheep, because PBL had increased lytic activity against bovine herpes virus-1-infected cells on d 17 after infection (Tekin and Hansen, 2002). Although the latter seems to antagonize the former activity of IFN-, some activation of the innate immune system may be beneficial to pregnancy, because knock-out mice without natural killer cells had reduced litter size due to an alteration in the vascularization of the placenta (Tekin and Hansen, 2002). Furthermore, the activation of natural killer cells may have a positive effect on establishment of pregnancy and placentation by causing a limited lysis of the luminal epithelial cells (Tekin and Hansen, 2002).
Reproductive status is correlated with changes in expression of MX2 mRNA and protein in the endometrium of ewes (Charleston and Stewart, 1993; Ott et al., 1998) and cows (Hicks et al., 2003) and in the PBL of ewes (Yankey et al., 2001) and cows (Gifford et al., 2007). In the current study, dairy heifers classified as pregnant at approximately 21 d after AI according to their progesterone concentrations had steady-state levels of MX2 mRNA 2.75 times greater compared with their d-0 levels. However, for heifers that were not inseminated and heifers that were bred but were not pregnant, the average fold change ranged from 1.5 to 2.0. Furthermore, a similar correlation between reproductive status at 30 and 60 d after AI and change in expression of MX2 mRNA was also observed.
Interferon- increases expression of many proteins in the uteri of pregnant cows and ewes (Johnson et al., 2002). Charleston and Stewart (1993) demonstrated a local increase in expression of MX1 mRNA in the endometrium of unilaterally pregnant ewes and in ewes receiving intrauterine treatment of IFN-. Furthermore, reproductive status of ewes affected levels of MX1 mRNA and MX1 protein in the endometrial epithelium, stroma, and myometrium from d 13 to 19 after mating (Ott et al., 1998). More recent studies have demonstrated increased levels of MX1 mRNA and protein in peripheral blood mononuclear cells of pregnant ewes beginning on d 15 after mating compared with nonpregnant ewes (Yankey et al., 2001). The effects of IFN- on levels of MX1 mRNA have also been demonstrated in cattle. Hicks et al. (2003) showed a 15-times greater level of MX1 mRNA in the endometrium of pregnant cattle between d 12 and 15 after AI, whereas nonpregnant cows showed little change in the expression of MX1 mRNA or protein throughout the estrous cycle.
Interestingly, fold change in expression of MX2 mRNA from d 0 to 18 after insemination tended to be correlated with PL from 21 to 30 and 21 to 60 d of gestation. Heifers that maintained their pregnancies had a greater increase in MX2 mRNA than those that experienced embryonic or fetal loss. Previous studies have demonstrated that cows that had a delayed increase in progesterone concentration after insemination had embryos that were poorly developed 16 d after insemination and that produced less IFN- (Mann and Lamming, 2001). Furthermore, when cows that had poorly developed embryos were challenged with oxytocin, they had a similar increase in concentrations of PGF2 metabolite compared with cows that were not inseminated, and their concentrations of PGF2 metabolite were greater than that of cows that had an early rise in progesterone concentration after ovulation and had a well-developed embryo (Mann and Lamming, 2001). More recently, it has been demonstrated that ewes treated with exogenous progesterone from 36 h to 9 d after mating were bearing blastocysts that were 220% larger on d 9 after mating than those not supplemented (Satterfield et al., 2006). Furthermore, although ewes supplemented with exogenous progesterone from 36 h to 12 d after mating were bearing embryos that were elongated or filamentous on d 12 after mating, ewes not supplemented were bearing embryos that were still in the spherical or tubular blastocyst phase (Satterfield et al., 2006). In the same study, it was demonstrated that ewes treated with exogenous progesterone up to 12 d after mating had significantly greater concentrations of IFN- in the uterine luminal flush than ewes not treated (Satterfield et al., 2006). Similarly, greater expression of other IFN-stimulated genes, such as cathepsin L and radical S-adenosyl methionine domain containing 2, was observed only on d 12 after mating in ewes supplemented with progesterone from 36 h to 12 d after mating (Satterfield et al., 2006). Therefore, it is possible that pregnant heifers with a small increase in levels of MX2 mRNA could have been bearing underdeveloped embryos that were producing smaller amounts of IFN- and did not survive. The lack of difference in change in steady-state levels of MX2 mRNA for heifers that maintained or lost pregnancy between 30 and 60 d after AI was probably due to the small number of heifers experiencing PL during this period (n = 6).
Although the concentration of progesterone from study d 0 to 21 was not different for heifers that lost pregnancy from 30 to 60 d of gestation compared with heifers that maintained pregnancy, concentrations of progesterone on d 21 after AI were greater for heifers that did not experience PL. The role of progesterone in the establishment and maintenance of pregnancy is well-documented (Spencer and Bazer, 2002), and although the early rise in progesterone after AI is critical for the establishment and maintenance of pregnancy in cattle (Mann and Lamming, 2001), treatment of lactating dairy cows with progesterone inserts from 14 to 21 d after AI has been observed to reduce PL (El-Zarkouny and Stevenson, 2004; Chebel et al., 2006), indicating that low progesterone concentrations as late as 14 to 21 d of the estrous cycle may affect fetal survival. Therefore, it is possible that lower concentrations of progesterone around d 21 of the estrous cycle predisposed heifers to PL from 30 to 60 d after AI.
There was a large variability in fold change in steady-state levels of MX2 mRNA from study d 0 to 18 among heifers of similar reproductive status as classified by progesterone concentration on study d 21 and diagnosed at 30 ± 1 and 60 ± 3 d after AI. Because of this, when we used fold change in MX2 mRNA of 1.47 to determine failed pregnancy on study d 18 (18 to 19 d after AI), the sensitivity and negative predictive values of the test were low. Similarly, the level of agreement between classification of reproductive status, based on changes in MX2 mRNA levels and pregnancy classification or diagnosis on study d 21 and 30 ± 1 and 60 ± 3 d after AI, was relatively low and ranged from 0.20 to 0.36, as measured by the Kappa procedure. This could be related to the nature of MX proteins and their role in the innate immune response; MX proteins are large GTPases that belong to the dynamin superfamily (Haller et al., 1998; van der Bliek, 1999) and are potent inhibitors of replication of negative-stranded RNA viruses (Horisberger et al., 1983; Haller et al., 1998). They were first discovered in mice, where MX1 protein confers resistance to influenza viruses (Lindenmann, 1962; Horisberger et al., 1983) by blocking transcription of the viral genome in the nucleus (Haller and Kochs, 2002). The MX2 protein, on the other hand, is localized in the cytoplasm and blocks vesicular stomatitis virus and hantavirus replication (Jin et al., 2001).
Although all heifers were vaccinated with viral vaccine (Bovi-Shield Gold 5, Pfizer Animal Health) 6 wk before enrollment, it is possible that exposure of some heifers to viral agents around the time of enrollment resulted in upregulation of MX2 mRNA, which could have caused a steep increase in MX2 mRNA levels on study d 0. In such cases, heifers that were pregnant on study d 18 could have had an increase in MX2 mRNA expression related to pregnancy that was not large enough to result in significant fold change in steady-state level of MX2 mRNA compared with study d 0, resulting in FN. Furthermore, the relatively long and variable time from collection of the blood samples to stabilization of mRNA (3 to 24 h) could have affected its integrity and its isolation and quantification. Debey et al. (2004) demonstrated that storage of blood samples at room temperature for 20 to 24 h before stabilization of mRNA from peripheral blood mononuclear cells resulted in upregulation of genes related to stress and apoptosis and downregulation of genes related to the cell cycle, metabolism, immune function, and proliferation. However, in the current study, blood samples that were not immediately processed for stabilization of mRNA were kept in ice or in refrigerators. The variability in change in steady-state levels of MX2 mRNA within reproductive statuses may also have been influenced by the time of blood sample collection relative to AI, because some heifers had their first blood sample, on study d 0, collected 0 or 24 h after AI and their second blood sample, on study d 18, collected from 18 to 19 d after AI. Peak production of IFN- occurs at d 16 of pregnancy in sheep (Godkin et al., 1982) and d 17 of pregnancy in cattle (Bartol et al., 1985) and can be detected until d 20 and 35 in these species, respectively. Although, it has been demonstrated that in sheep (Yankey et al., 2001) and cattle (Gifford et al., 2007), the expression of MX mRNA increases substantially between 15 and 21 d after mating, pregnant cows had significantly greater increases in expression of MX2 mRNA as early as d 16 after AI compared with nonpregnant cows (Gifford et al., 2007). Furthermore, approximately 15% of heifers in the BRED group were inseminated 48 h after intravaginal insert removal. Therefore, blood samples collected on study d 0 and 18 corresponded to 24 h and 19 d after AI, respectively, in approximately 15% of heifers in the BRED group.
Fold change in expression of MX2 mRNA in PBL between d 0 and 18 was affected by pregnancy status, and the magnitude of change in the level of MX2 mRNA from estrous cycle d 0 and 18 was correlated with PL later in gestation. However, predicting reproductive status according to fold change in expression of MX2 mRNA in PBL was not accurate, because the sensitivity and negative predictive values of the test were low. Further studies are needed to determine practical methodologies and optimal sampling time for assaying levels of MX2 or other IFN-stimulated genes as indicators of pregnancy status in dairy cattle. Should this approach prove successful, it could open the possibility for utilizing a rapid resynchronization procedure that could allow reinsemination of nonpregnant cows with an approximate 21-d AI interval.
|
Footnotes |
|---|
2 We thank Frank Hurtig from Merial Ltd. for providing Cystorelin and Frederico Moreira from Pfizer Animal Health for providing Lutalyse sterile solution and EAZI-BREED CIDR inserts.
4 Present address: Veterinary Medicine Teaching and Research Center, University of California Davis, 18830 Road 112, Tulare, CA 93274.
3 Corresponding author: rchebel{at}vmtrc.ucdavis.edu
Received for publication January 5, 2007. Accepted for publication April 9, 2007.
|
LITERATURE CITED |
|---|
|
TopAbstractINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
stimulated genes. Methods Mol. Med. 122:367–380.[Medline]I 1. Vet. Immunol. Immunopathol. 34:81–96.[CrossRef][Medline]
., and P. J. Hansen. 2002. Natural killer-like cells in the sheep: Functional characterization and regulation by pregnancy-associated proteins. Exp. Biol. Med. 227:803–811.
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
