It had been demonstrated that the strong and highly significant quantitative trait locus (QTL) can affect protein percentage on Bos Taurus Autosome 3 (BTA3) at the position 52 cM, near the microsatellite DIK4353, with the 95% confidence interval spanning from 25 to 57 cM in Chinese Holstein population using QTL-express, MQREML, and GRIDQTL softwares. This study herein focused on such region of fine mapping QTLs for milk production and functional traits with 16 microsatellite markers with coverage of 33 cM between the markers BMS2904 and MB099 on BTA3 in a daughter-designed Chinese Holstein population. A total of 1 298 Holstein cows and 7 sires were genotyped for 16 microsatellites with ABI 3700 DNA sequencer. The variance components QTL linkage analysis (LA) and linkage-disequilibrium (LD) analysis (LA/LD) was performed to map QTLs for 7 traits, i.e., 305-d milk yield, fat yield, protein yield, fat percentage, protein percentage, somatic cell score and persistency of milk yield. Four strong and highly significant QTLs were detected for fat yield, fat percentage, protein percentage and somatic cell score at the position 40, 30, 27 and 26 cM, respectively. Two minor QTLs for milk yield and persistency of milk yield were identified at 42 and 46 cM, respectively. These findings provided a general idea for the fine mapping of the causal mutation for milk production and functional traits on BTA3 in the future.

Our previous studies demonstrated that the region around markers BMS470 and BMS1242 on BTA6 showed a linkage to 305-d milk yield and composition traits in the Chinese Holstein population. We herein focused on such narrow region to fine map milk production QTLs with 15 SNPs across 25 Mb with each SNP in 1 Mb within most regions in a Chinese Holstein population with daughter design. 1 449 Holstein cows and 11 sires were genotyped for such SNPs by using TaqMan probe and RFLP assays. Multipoint linkage analysis across family revealed a QTL affecting milk yield between PPARGC1A C4075T and SLC34A2 T1713C. Meanwhile, within family analysis found three milk yield QTLs (two in CR T60984131G-CEP135 C501T and one in PDLIM5 A106C-OPN T3907, a fat yield QTL in UGDH T1670C-CR T60984131G region, and two protein yield QTLs in TBC1D1 G501C-UGDH T1670C and PPARGC1A C4075T-SLC34A2 T1713C, respectively. Associations between aforementioned significant SNP markers and milk production traits were further implemented. We found significant associations of PPARGC1A C4075T, SLC34A2 T1713C with milk yield (P<0.05, P<0.01, P<0.01), UGDH T1670C, and CR T60984131G with fat yield (P<0.01, P<0.01), and PPARGC1A C4075T, SLC34A2 T1713C, UGDH T1670C and OPN T3907 with protein yield (P<0.01, P<0.01, P<0.01, P<0.01). Our findings implied that QTLs affecting milk production traits on BTA6 were pleictropism or multigenic effect and PPARGC1A and OPN may be the causal mutations behind milk production QTLs on BTA6 in the Chinese Holstein population.

Current study adopted gene flow theory and selection index method to compare the breeding efficiency of three breeding plans in the Chinese Holstein cattle using ZPLAN software. Simulated conventional progeny-testing program (PT) and young sire program (YS) were compared with breeding program using genomic selection (GS) taking parameters derived from Chinese Holstein breeding system. The results showed that, GS shortened generation interval by 1.5-2.2 years, and increased the genetic progress by 30-50%, comparing to PT and YS, respectively. Economic analysis showed that GS could obtain a higher breeding efficiency, being 119 and 97% higher than that of PT and YS, respectively; and GS was also powerful in improving functional traits with a low heritability. Main factors affecting breeding efficiency in GS were further discussed, including selection intensity, accuracy and the cost of SNP genotyping. Our finding provided references for future designing and implementing GS in Chinese dairy population.