Influences of dietary guanidinoacetic acid supplementation on performance and proteins involved in milk fat and protein synthesis in dairy cows

Abstract

Guanidinoacetic acid (GAA) increases milk production in dairy cows by promoting milk fatty acid (FA) and protein synthesis; however, its mechanism is unclear. It was hypothesized that provision of rumen-protected GAA (RPGAA) would increase milk production, and milk fat and protein synthesis, while revealing the underlying mechanisms. Forty-four multiparous Holstein dairy cows (33.5 ± 1.31 kg/d of milk yield, 667 ± 11.8 kg of body weight, and 38.5 ± 2.14 d in milk [DIM]) were blocked by daily milk yield and DIM, and allocated to different treatments in a randomized-block design, namely, the control (without RPGAA), low-RPGAA (0.18 g/kg DM of GAA), medium-RPGAA (0.36 g/kg DM of GAA), and high-RPGAA (0.54 g/kg DM of GAA) groups. The experiment lasted for 95 d, including a 10-d covariate period, a 15-d adaptation period, and a 70-d sampling period. Although the DM intake and BW were not impacted (P > 0.05), the yields of actual milk, 4% fat-corrected milk, and energy-corrected milk increased linearly and quadratically (P < 0.05) with RPGAA supplementation; milk fat and protein percentage also showed quadratic increases (P < 0.05). The de novo and mixed production of FA in milk increased linearly and quadratically (P < 0.05), whereas that of preformed FA was increased quadratically (P < 0.01). Administration of RPGAA linearly increased the daily secretion yields of Arg, Cys, His, Ser, Thr, Trp, and Tyr (P < 0.05), and quadratically increased those of Ile, Lys, Met, Phe, Asp, and Glu in milk (P < 0.05). Furthermore, the apparent digestibilities of DM, OM, CP, and EE increased quadratically (P < 0.05), and those of NDF and ADF increased linearly (P < 0.05), following GAA supplementation. Serum concentrations of glucose, total protein, albumin, Arg, and creatine increased linearly (P < 0.05); concentrations of insulin-like growth factor 1, estradiol, and prolactin increased quadratically (P < 0.05), and that of blood urea nitrogen decreased quadratically (P = 0.018). Furthermore, addition of 0.36 g/kg DM of GAA from RPGAA promoted the expression of proteins concerned with mammary gland proliferation, FA synthesis, and milk protein synthesis, which were done in tissue biopsies. These results indicate that RPGAA provision enhanced milk production, FA synthesis, and milk protein synthesis by promoting the expression of proteins involved in mammary gland development, FA synthesis, and milk protein synthesis.

1. Introduction

Creatine is a vital energy source in animal body tissues and is derived from both endogenous synthesis and exogenous sources. Endogenous creatine is synthesized de novo via the methylation of guanidinoacetic acid (GAA), which is formed from the amino acids (AA), arginine (Arg), and glycine (Gly) (Ostojic, 2015a; Bonilla et al., 2021). Exogenous creatine is primarily derived from the diet, but its availability is limited to products of animal origin (Kaviani et al., 2020). Moreover, the uptake of exogenous creatine by cells and its absorption rate are further restricted due to the saturation of creatine transporters (Ostojic, 2017).

Guanidinoacetic acid has been increasingly recognized as an alternative to creatine, primarily due to its superior stability and lower cost (Speer et al., 2020). Recent studies have demonstrated that GAA supplementation in cattle can significantly increase blood creatine content (Ardalan et al., 2020; Liu et al., 2021a), enhance average daily gain (Liu et al., 2021b), improve nutrient digestion (Liu et al., 2021a,b), elevate rumen volatile fatty acid (VFA) content (Liu et al., 2021b), and upregulate the hepatic expression of genes involved in protein synthesis (Liu et al., 2021a; Li et al., 2020a). Notably, GAA supplementation has been shown to increase plasma Arg levels, indicating an Arg-sparing effect due to reduced utilization for endogenous creatine synthesis (Ardalan et al., 2020). This sparing effect allows the spared Arg to be utilized for other critical functions, such as stimulating growth hormone (GH) secretion and enhancing protein synthesis (Ostojic, 2015b; Li et al., 2023).

These findings indicate that GAA supplementation enhances the energy supply to tissues and cells, thereby improving the growth performance of ruminants (Liu et al., 2021a,b). Given that energy supply is a fundamental factor for milk synthesis and composition in the mammary glands of dairy cows (NASEM, 2021), recent studies have shown that dietary GAA supplementation can improve lactation performance, enhance dietary nutrient digestion, and optimize rumen fermentation in dairy cows (Liu et al., 2023). However, the underlying mechanisms of these effects remain unclear. Additionally, the impact of GAA on mammary gland development has not been extensively explored. It is hypothesized that GAA may promote mammary gland development by enhancing the proliferation of mammary epithelial cells, thereby improving milk production and composition.

Similar to AA, GAA is susceptible to degradation by ruminal microorganisms. Therefore, the effectiveness of GAA in ruminants is largely determined by the amount of GAA that successfully crosses the rumen and reaches the small intestine. Studies have shown that infusing GAA directly into the abomasum results in higher plasma creatine concentrations compared to infusing it into the rumen (Speer et al., 2020). Ardalan et al. (2020) further demonstrated that supplying 10 to 30 g of GAA per day to the abomasum of dairy heifers leads to a dose-dependent increase in plasma creatine concentration. The ruminal degradability of GAA has been estimated at 53%, based on plasma creatine concentration as the response criterion (Speer et al., 2020). Given these findings, rumen-protected GAA (RPGAA) has been developed to bypass ruminal degradation and be effectively released in the small intestine, thereby significantly increasing serum creatine content compared to unprotected GAA.

Accordingly, it is hypothesized that providing RPGAA could enhance lactation performance and the synthesis of milk fatty acid (FA) and protein by promoting mammary gland development and the expression of associated protein. To test these hypotheses, the current study aimed to investigate the effects of RPGAA supplementation on milk performance, nutrient digestibility, the expression of proteins involved in milk FA and protein synthesis, and mammary gland cell proliferation in lactating dairy cows.

Recent advancements in the field have further highlighted the potential benefits of GAA supplementation in dairy cows. For instance, Li et al. (2023) reported that dietary GAA can improve the efficiency of nutrient utilization and enhance the overall health status of dairy cows. These studies collectively underscore the importance of exploring GAA as a novel feed additive to optimize dairy production.

2. Materials and methods

2.1. Animal ethics statement

The test scheme was authorized by the Animal Care and Use Committee of Shanxi Agricultural University (IACUC Issue No. SXAUEAW-2023C.FU.00301403).

2.2. Preparation and properties of RPGAA

The RPGAA (53.0% GAA, on a DM basis, 95.8% DM) used in this experiment was produced through a multi-step process. Initially, a mixture was prepared by combining 541 g/kg of GAA (purity 98%), 89 g/kg of silicon dioxide, 170 g/kg of C16:0 (heated to 80 °C), and 80 g/kg of C18:0 (also heated to 80 °C). These ingredients were thoroughly blended to ensure uniform distribution. The resulting mixture was then pelletized to form cylindrical particles (1.25 cm diameter). These cylinders were subsequently introduced into a shot blasting machine and rotated for 15 min to shape them into spheres of the same diameter. Finally, the spheres were transferred to a rotary coating drum, where they were coated with 40 g/kg of preheated C18:0 (at 80 °C) and 80 g/kg of calcium stearate to create a protective layer that prevents ruminal degradation of GAA. The ruminal and intestinal disappearance rates of RPGAA were measured in 4 cows (32.8 ± 1.29 kg/d of milk yield, 662 ± 10.7 kg of body weight [BW], and 36.6 ± 2.10 days in milk [DIM]) with ruminal and duodenal cannulas. Briefly, 6 replicates of 5 g RPGAA were placed in nylon bags (12 cm × 8 cm, pore size 37 μm) and incubated in the rumen of each cow for 8 h. Subsequently, 3 of the 6 replicates from the rumen were transferred to the duodenum of each cow and collected from the feces. Nylon bags collected from the rumen and feces were washed with cool water for 3 min using a washer, dried in a freeze dryer, the DM content was determined, and samples were pooled for GAA analysis.

2.3. Animals, design, and experimental diets

Forty-four multiparous Holstein dairy cows (33.5 ± 1.31 kg/d of milk yield, 667 ± 11.8 kg of BW, and 38.5 ± 2.14 DIM) were blocked by daily milk yield and DIM, and allocated to one of 4 groups in a randomized-block design. The number of cows per treatment group was determined based on a power analysis conducted prior to the experiment. The power analysis was performed to ensure that the study had sufficient statistical power (which was set at 0.80) to detect significant differences in milk yield and milk composition parameters, assuming a medium effect size (Cohen's d = 0.5) and a significance level of 0.05. This calculation indicated that a minimum of 10 cows per group would be required to achieve the desired power. Therefore, 11 cows were allocated to each of the 4 treatment groups to account for potential dropouts or data loss during the experiment. The 4 groups included: control (without RPGAA), low RPGAA (LRPGAA; 0.18 g/kg DM of GAA), medium RPGAA (MRPGAA; 0.36 g/kg DM of GAA), and high RPGAA (HRPGAA; 0.54 g/kg DM of GAA). The GAA dose was decided based on a previous study (Liu et al., 2023), which found that supplementing 12 g/d of GAA in dairy cow diets promoted milk fat production. Considering the estimated ruminal degradation rate of GAA (53%) and the ruminal disappearance rate of RPGAA, the dose of GAA in the MRPGAA group was set as 0.36 g/kg DM. The RPGAA additive was mixed into the total mixed ration (TMR; Table 1). The experimental TMR was designed based on the nutrient requirements of 680 kg dairy cows (35 kg/d milk, 35 g/kg milk fat and 35 g/kg milk protein) as specified by NASEM (2021). All the experimental cows were housed in a naturally ventilated, head-to-head, 2-row, free-stall barn equipped with a Calan Gate Feeding System to monitor feed intake. The experiment lasted for 95 d, including a 10-d covariate period, a 15-d adaptation period, and a 70-d sampling period. All cows were milked daily at 06:10, 14:10, and 21:10, fed the same TMR ad libitum, and had free access to drinking water.

Table 1.

Ingredients and chemical composition of the basal diet (g/kg, dry-matter basis).

Item	Content
Ingredients
Whole corn silage	255.0
Alfalfa hay	113.0
Oat hay	132.0
Corn grain, ground	255.0
Wheat bran	61.1
Soybean meal	90.1
Rapeseed meal	23.8
Cottonseed cake	52.0
Calcium carbonate	5.0
Salt	5.0
Dicalcium phosphate	3.0
Mineral and vitamin premix1	5.0
Total	100.0
Chemical composition
OM	953.2
CP	164.3
EE	32.7
NDF	321.4
ADF	198.1
Non-fiber carbohydrate2	435.3
Calcium	7.2
Phosphorus	4.5
Methionine	2.3
NEL3, MJ/kg	6.52

¹One kilogram of premix contained: 18,000 mg Fe; 1640 mg Cu; 8100 mg Mn; 7000 mg Zn; 116 mg I; 56 mg Se; 17 mg Co; 826,000 IU vitamin A; 320,000 IU vitamin D; and 10,800 IU vitamin E.

²Non-fiber carbohydrate = 100 – CP – NDF – EE – Ash.

³Net energy for lactation (NE_L) was calculated as the sum of the NE_L, values of individual feed ingredients multiplied by their respective inclusion percentages (on a DM, basis) as described in NASEM (2021).

2.4. Experimental data collection and sampling

The BW of the cows were measured at 17:00 on day 1 of the covariate period and on days 1 and 70 of the sampling period. The TMR provided to the cows and the residual were measured and recorded daily for each animal to calculate DM intake (DMI). Both TMR and the residual were sampled individually every 5 d during the experiment and kept at −20 °C until analyses. The milk production of each cow was determined and recorded daily, and the mean value was recorded weekly for time course analysis. Milk samples were collected weekly from each milking session of each cow within a day and kept at 4 °C with 2-bromo-2-nitropropane-1,3-diol for subsequent analyses.

From days 1 to 10 of the covariate period and days 50 to 67 of the sampling period, each cow was dosed at 06:30 and 18:30 with 5 g of chromic oxide at each time as a digestion marker. From days 6 to 10 of the covariate period and days 58 to 67 of the sampling period, approximately 250 g of fresh feces was sampled at 07:00, 13:00, 19:00, and 01:00 from the rectum of each cow and stored at −20 °C until analyses. From days 68 to 69, the TMR, residual, and fecal samples were pooled by cow, dried at 55 °C for 72 h, and ground to pass a 1 mm screen using a cutter mill.

On day 10 of the covariate period and day 70 of the sampling period, blood samples were collected from each cow via the coccygeal vessel into 10 mL evacuated tubes. The samples were placed on ice and transported immediately to the laboratory, where they were centrifuged at 2000 × g and 4 °C for 12 min to separate the serum, and the serum was stored at −20 °C until analyses.

On day 10 of the covariate period and day 70 of the sampling period, approximately 1 g of mammary gland secretory tissue biopsies was collected from the midpoint section of the rear quarter of each cow in the control and MRPGAA groups between 16:00 and 20:00 by surgical biopsy. The procedure followed the method previously described by Farr et al. (1996) with several modifications and additional details:

Each cow was restrained in a head-to-head stanchion to ensure stability and safety during the procedure. The surgical area was thoroughly cleaned with 70% ethanol to minimize the risk of infection. Local anesthesia was administered using a 2% lidocaine solution to numb the biopsy site and ensure the cow's comfort. A small incision (approximately 1 cm) was made in the skin overlying the midpoint section of the rear quarter using sterile surgical instruments. Approximately 1 g of mammary gland secretory tissue was carefully excised using a surgical biopsy punch. The incision was then closed with a single stitch using absorbable sutures to promote healing and prevent infection. The biopsy site was treated with a topical antibiotic ointment. Biopsy samples were immediately placed in liquid nitrogen and stored at −80 °C until further analysis.

Postoperative care included monitoring each cow for any signs of discomfort or adverse reactions immediately following the procedure and for 24 h thereafter. The cows were observed daily for signs of infection, inflammation, or other complications at the biopsy site for the duration of the study. Any cow showing signs of infection or discomfort was treated with veterinary-approved antibiotics and pain management protocols.

2.5. Chemical analyses

The contents of DM (method 934.01), EE (method 973.18), nitrogen (method 976.05), and crude ash (method 942.05) in the TMR and fecal samples were analyzed according to AOAC (2000). The content of OM was estimated as the difference between DM and crude ash content. The content of NDF was measured using heat-stable alpha-amylase and sodium sulfite, as described by Van Soest et al. (1991), and expressed including residual ash; ADF content was analyzed according to AOAC (2000). The fat, true protein, and lactose percentages in milk were measured using a Milko Scan FT-120 unit (Foss Electric), as described by AOAC (2000). The chromium concentration of the fecal samples was determined via atomic absorption spectrophotometry (LB-AA-6810; Qingdao Lubo Jianye Environmental Protection Technology Co., Ltd.) according to the method of Williams et al. (1962).

Biochemical parameters including glucose (#A154-2-1), triglyceride (#A110-2-1), albumin (#A028-1-1), total protein (#A045-3-1), blood urea nitrogen (BUN; #C013-2-1), and nitric oxide (NO; #A012-1-2) in the serum were measured using an Automatic Biochemical Analyzer (HTSH-3000, Qingdao Hantang Biotechnology Co., Ltd., Qingdao, Shandong, China). The intra-assay coefficient of variation (CV) for these parameters ranged from 1.2% to 3.5%, and the inter-assay CV ranged from 2.5% to 5.0%. Insulin-like growth factor 1 (IGF-1; #BL-E21687M), estradiol (E2; #BL-E28820M) and prolactin (#BL-E28845M) were analyzed using ELISA kits purchased from Beijing Biolide Biotechnology Co., Ltd. (Beijing, China). The intra-assay CV for IGF-1, E2, and prolactin were 4.2%, 3.8%, and 4.5%, respectively, while the inter-assay CV were 6.1%, 5.7%, and 6.3%, respectively. Serum creatinine levels were analyzed using reversed-phase high-performance liquid chromatography (Waters Alliance e2695 HPLC System, Waters Corporation, Milford, MA, USA) according to Speer et al. (2022). Serum Arg concentration was evaluated using ultra-performance liquid chromatography (Waters Acquity UPLC H-Class System, Waters Corporation) as described by Ardalan et al. (2020).

2.6. Determination of milk FA

An aliquot of milk was used to obtain a fat cake through centrifugation. Milk fat was extracted and esterified, and the FA were transmethylated as previously reported (Chouinard et al., 1999). Subsequently, fatty acid methyl esters (FAME) were used to analyze milk FA composition using gas chromatography (Agilent 7890 A, Agilent Technologies, Santa Clara, CA, USA). A capillary column (100 m × 0.25 mm × 0.25 μm; CP SIL 88; Agilent Technologies) with an autosampler, flame ionization detector, and split injection was used; FAME were identified by the standards, namely, FAME Mix C4–C24 Unsatures (#18919, Sigma–Aldrich, Saint Louis, MO, USA), methyl trans-11 C18:1 (#46905, Sigma–Aldrich), and methyl cis-9, trans-11 CLA (#1255, Matreya, State College, PA, USA). The average percentage of FA in the total milk fat in each sample was corrected for other milk fat components by multiplying by 0.98885 as described by Rico and Harvatine (2013).

2.7. Determination of milk amino acids

Total AA content was determined as previously reported (Jariyasopit et al., 2021). Briefly, approximately 100 mg of milk sample was transferred to a hydrolysis bottle and mixed with 5 mL of a hydrolytic solution (0.1% phenol, 5% thioglycolic acid, and 6 mol/L HCl). The hydrolytic bottles were sealed and then hydrolyzed in a hot air oven at 110 °C for 18 h. Next, 1 mL of hydrolysate was taken to centrifuge at 10,000 × g for 10 min, and then 100 μL of the supernatant was taken to neutralize until the pH was between 1.0 and 2.0 with 1 mol/L sodium carbonate. Subsequently, 25 μL of the neutralizing solution was transferred into 2 mL gas chromatography glass bottles, and 50 μL of norleucine solution (200 nmol/mL) was added as the internal standard and dried up to 60 °C for 1 h by using a vacuum centrifuge. Thereafter, 50 μL of dichloromethane was added and dried for 30 min, then 50 μL of acetonitrile and 50 μL of N-tert-butyldimethylsilyl-N-methyltrifluoroacetamide (MTBSTFA) + 1% tert-butyldimethyl- chlorosilane (TBDMSCL, #375934; Sigma–Aldrich) were added. The bottle was sealed and incubated in a hot air oven at 100 °C for 4 h. Exactly 1.0 mL of samples were injected into a gas chromatography-mass spectrometer (Agilent Technologies) with a split ratio of 10:1. The chromatography column (30 m × 0.25 mm × 0.25 μm) was HP-5MS. The carrier gas was helium and the flow rate was 1.4 mL/min. The temperature was gradually increased from 130 to 325 °C (at 6 °C/min till 190 °C, followed by 0.5 °C/s till 230 °C, where it was held for 5 min, and finally increased to 325 °C, where it was maintained for 6 min). The temperatures of the injector, transfer lines, electronic ionization ionic source, and quadrupole were maintained at 280, 325, 240, and 180 °C, respectively. Data were obtained using MassHunter software (B0.4).

2.8. Western blotting

Following the manufacturer's specifications, protein concentration was determined by using a bicinchoninic acid (BCA) protein assay kit (#23225; Thermo Fisher Scientific, Rockford, IL, USA). Equal quantities of protein sample (20 μg) were isolated on 12% SDS-PAGE, and the isolated protein was transferred to a nitrocellulose (NC) membrane (#1704271; Bio-Rad, Hercules, CA, USA) using a Mini Trans-Blot Electrophoretic Transfer Cell (Bio-Rad) at 4 °C, 100 V for 60 to 90 min. The NC membrane was blocked with 6 % (wt/vol) bovine serum albumin (BSA) in TBST for 2 h at 25 °C. Thereafter, the NC membrane was incubated with the primary antibodies for 12 h at 4 °C and washed 5 times with TBST for 5 min to remove excess antibodies. Thereafter, the membrane was incubated at 25 °C for 2 h with secondary antibodies. The β-actin levels were used for normalization.

The following antibodies were used. Mouse anti-cyclin A1 (#13295) was purchased from Proteintech Group (Wuhan, Hubei, China); mouse anti-proliferating cell nuclear antigen (PCNA; #2586), rabbit anti-protein kinase B (Akt; #9272), and rabbit anti-β-actin (#4970) were purchased from Cell Signaling Technology (Danvers, MA, USA). Rabbit anti-sterol regulatory element-binding protein 1 (SREBP1; #NB100-2215) was purchased from Novus Biologicals (Centennial, CO, USA). Rabbit anti-B-cell lymphoma-2 (BCL2; #20351R), mouse anti-BCL2 associated X protein (BAX; #0127M), rabbit anti-cysteine-aspartic acid protease-3 (Caspase-3; #0081R), Caspase-9 (#0049R), phosphorylated (p)-Akt_Ser473 (#0876R), rapamycin target protein (mTOR; #1992R), phosphorylated-mTOR_Ser2448 (#3495R), adenosine monophosphate activated protein kinase (AMPK; #1115R), phosphorylated-AMPK (#5551R), peroxisome proliferator activated receptor gamma (PPARγ; #4590R), acetyl-coenzyme A carboxylase-α (ACACA; #11912R), phosphorylated-ACACA (#12954R), fatty acid synthase (FASN; #60347R), stearoyl-CoA desaturase 1 (SCD1; #3787R), Janus kinase 2 (JAK2; #23003R), phosphorylated-JAK2 (#2485R), signal transduction and transcriptional activator 5 (STAT5; #1142R), phosphorylated-STAT5 (#5703R), αs1-casein (#10033R), β-casein (#0466R), and κ-casein (#10031R) were purchased from Bioss Biotechnology Co., Ltd. (Beijing, China). Goat anti-rabbit secondary antibody (#E-AB-1003) was purchased from Elabscience Biotechnology (Wuhan, Hubei China), and goat anti-mouse secondary antibody (#RS0001) was purchased from ImmunoWay Biotechnology (Plano, TX, USA).

To validate these antibodies for use in bovine tissues, the following steps were performed: each primary antibody was tested for specificity using Western blotting with bovine mammary gland tissue lysates. A single band corresponding to the expected molecular weight of the target protein was observed for each antibody, confirming its specificity. Positive controls included tissues known to express high levels of the target protein, while negative controls included tissues known to lack expression. The expected bands were observed in positive controls, while no bands were detected in negative controls. The optimal dilution for each antibody was determined through a series of dilution experiments to ensure a clear and specific signal without background noise. To ensure that the antibodies did not cross-react with non-target proteins, tissues from other species (e.g., mouse and human) were tested and confirmed that they specifically recognized the bovine target proteins without cross-reactivity. Each antibody was tested in at least 3 independent Western blot experiments to ensure reproducibility. The results were consistent across experiments, confirming the reliability of the antibodies. These validation steps ensured that the antibodies used in this study were suitable for detecting their respective target proteins in bovine tissues.

2.9. Statistical analysis

The net energy for lactation (NE_L) of the dairy cow diets was estimated by multiplying the NE content of the dietary ingredients by their content (NASEM, 2021). Energy corrected milk (ECM) and 4.0% fat corrected milk (FCM) metrics were calculated as previously described (NRC, 2001). Feed efficiency was calculated as milk/DMI and ECM/DMI for each cow. Data were analyzed by using SAS 9.0 (PROC MIXED; SAS, 2002) with a randomized block design.

Data on DMI, milk production, feed efficiency, milk FA production, and AA secretion yield were analyzed using the following statistical model:

$$Y_{iklm}=\mu +\beta V_k+B_i+W_j+C_{k(i)}+T_l+W{T}_{jl}+E_{ijkl}\text{,}$$

Here, Y_iklm is the dependent variable, μ is the overall mean, β is the regression coefficient, V_k is the covariate measure, B_i is the random effects of block i, W_j is the fixed effects of week j, C_k(i) is the random effects of cow k within block i, T_l is the fixed effects of GAA treatment l, WT_jl is the interaction between week and GAA treatment, and E_ijkl is the residual error.

Data on dietary nutrient digestibility and serum indices were analyzed using the following statistical model:

$$Y_{iklm}=\mu +B_i+C_{k(i)}+T_l+E_{ikl}\text{,}$$

The spatial covariance structure SP was used to estimate the covariances, and the subject of the repeated measurements was defined as cow (block × day × RPGAA addition). Mean separations using the probability of difference test (PDIFF in SAS 9.0) were considered only for significant effects (P < 0.05). Linear and quadratic orthogonal contrasts were performed using the CONTRAST statement in SAS. Western blot data were analyzed using the SigmaPlot statistical analysis package (version 12.5; Systat Software, CA, USA). Statistical differences between the means were evaluated using the Student's t-test. The influences of the factors were considered significant at P < 0.05 unless the trends were considered at P < 0.10.

3. Results

3.1. Dietary DM intake and performance

For the RPGAA prepared, the ruminal and intestinal disappearance rates of RPGAA were 19.3% and 81.3%, respectively. The amount of metabolizable GAA was 348 g/kg RPGAA. Although the DMI and BW were not affected (Table 2), the yields of actual milk, 4.0% FCM and ECM increased linearly and quadratically (P < 0.05) with increasing RPGAA supplementation, and were greater for the RPGAA-added groups than that for the control (P < 0.05). Both the yield and percentage of milk fat and milk protein were quadratically increased (P < 0.05) and were higher for MRPGAA than for the control (P < 0.05). However, neither the yields nor the percentages of milk lactose were affected by RPGAA supplementation. Feed efficiency, defined as the ratio of milk yield to DMI or ECM yield to DMI, also increased linearly and quadratically (P < 0.05) with elevating RPGAA dosage; thus, it was higher for HRPGAA and MRPGAA than for the control (P < 0.05).

Table 2.

Effects of rumen-protected guanidinoacetic acid (RPGAA) addition on the DM intake (DMI), lactation performance, and feed efficiency in dairy cows.

Item	Treatments1				SEM	P-value
	Control	LRPGAA	MRPGAA	HRPGAA		Treatment	Linear	Quadratic
DMI, kg/d	22.2	23.0	23.4	23.1	0.18	0.373	0.359	0.153
Body weight, kg
Day 1 of covariate period	666	668	665	667	4.5	0.283	0.498	0.357
Day 1 of sampling period	670	672	669	672	4.2	0.302	0.541	0.316
Day 70 of sampling period	682	685	683	685	3.9	0.296	0.458	0.307
Milk production, kg/d
Actual	33.0b	35.3a	36.4a	35.6a	0.36	0.016	0.019	0.027
4.0% FCM2	30.5b	33.1a	34.8a	33.3a	0.38	0.013	0.022	0.025
ECM3	33.7b	36.2a	38.8a	37.0a	0.41	0.011	0.009	0.013
Fat	1.15b	1.26ab	1.35a	1.27ab	0.016	0.023	0.058	0.024
True protein	1.04b	1.14ab	1.23a	1.16ab	0.015	0.039	0.062	0.021
Lactose	1.75	1.85	1.90	1.91	0.028	0.173	0.154	0.268
Milk composition,%
Fat	3.49b	3.58ab	3.70a	3.58ab	0.026	0.030	0.079	0.036
True protein	3.16b	3.23ab	3.36a	3.27ab	0.020	0.033	0.081	0.038
Lactose	5.32	5.25	5.20	5.26	0.018	0.154	0.561	0.317
Feed efficiency4, kg/kg
Milk yield/DMI	1.48b	1.53ab	1.56a	1.54a	0.005	0.007	0.008	0.005
ECM yield/DMI	1.52b	1.57ab	1.66a	1.60a	0.007	0.008	0.007	0.006

Means with different superscripts in each row differ significantly (P < 0.05).

¹Control, low-RPGAA (LRPGAA), medium-RPGAA (MRPGAA), and high-RPGAA (HRPGAA) groups were supplemented with 0, 0.18, 0.36, and 0.54 g/kg DM of guanidinoacetic acid (GAA) derived from RPGAA, respectively.

²The 4.0% fat corrected milk (FCM) was calculated according to the NRC (2001), where 4.0% FCM = 0.4 × milk (kg/d) + 15 × fat (kg/d).

³The energy corrected milk (ECM) was calculated according to the NRC (2001), where ECM = [0.327 × milk (kg/d)] + [12.95 × fat (kg/d)] + [7.65 × protein (kg/d)].

⁴Estimated as milk yields (milk and ECM, yields) divided by DMI, for each dairy cow.

3.2. Milk FA and AA secretion

The provision of RPGAA quadratically increased (P = 0.036; Table 3) milk saturated fatty acid (SFA) concentration, and was higher for MRPGAA than the control (P = 0.039). The de novo-synthesized FA increased linearly (P = 0.006) and were higher in the MRPGAA and HRPGAA than in the control group (P = 0.004). However, concentrations of polyunsaturated fatty acids (PUFA) and preformed FA linearly decreased (P < 0.05), which was lower for MRPGAA and HRPGAA than for the control. With increasing RPGAA doses, milk FA concentrations of C8:0 and C10:0 increased quadratically (P < 0.05), that of C4:0, C12:0 and C14:0 increased linearly (P < 0.05), and that of C18:0, C18:3, and C22:6 decreased linearly (P < 0.05). Concentrations of the other FA was not affected (P > 0.05).

Table 3.

Effects of rumen-protected guanidinoacetic acid (RPGAA) addition on milk fatty acid content (g/100 g FA) in the milk of lactating dairy cows.

Item	Treatments1				SEM	P-value
	Control	LRPGAA	MRPGAA	HRPGAA		Treatment	Linear	Quadratic
C4:0	2.69b	3.03ab	3.18a	3.16a	0.082	0.020	0.049	0.206
C6:0	2.03	2.04	2.03	2.11	0.069	0.126	0.064	0.369
C8:0	0.07b	0.10a	0.11a	0.10a	0.006	0.011	0.083	0.012
C10:0	3.08b	3.20ab	3.42a	3.47a	0.098	0.018	0.066	0.020
C11:0	0.08	0.07	0.07	0.07	0.005	0.673	0.322	0.611
C12:0	4.13b	4.46ab	4.76a	4.88a	0.014	0.014	0.013	0.062
C13:0	0.07	0.07	0.06	0.06	0.003	0.065	0.087	0.106
C14:0	12.12b	13.23ab	14.55a	14.46a	0.184	0.034	0.049	0.071
C14:1	1.38	1.33	1.29	1.27	0.026	0.115	0.074	0.313
C15:0	0.92	1.02	1.03	1.03	0.019	0.106	0.080	0.365
C16:0	30.75	31.49	31.96	31.50	0.508	0.205	0.373	0.114
C16:1	1.92	1.92	1.80	1.88	0.034	0.274	0.179	0.154
C17:0	0.48	0.44	0.44	0.44	0.007	0.237	0.262	0.415
C18:0	12.71a	11.14ab	11.02ab	10.39b	0.220	0.032	0.045	0.517
C18:1n9c	16.33	15.86	14.73	15.75	0.185	0.085	0.230	0.125
C18:2n6c	7.86	7.64	7.03	7.23	0.034	0.112	0.093	0.131
C18:3	0.32a	0.27ab	0.24b	0.23b	0.011	0.045	0.019	0.530
C20:0	0.11	0.11	0.10	0.11	0.006	0.090	0.282	0.099
C20:3	0.11	0.09	0.09	0.08	0.004	0.138	0.055	0.313
C20:4	0.21	0.20	0.19	0.21	0.014	0.538	0.234	0.562
C22:6	2.61a	2.27a	1.91b	1.58c	0.058	0.005	0.025	0.186
Sum of fatty acids
SFA	69.25b	70.40ab	72.73a	71.77ab	1.233	0.039	0.098	0.036
MUFA	19.63	19.12	17.82	18.90	0.315	0.269	0.117	0.281
PUFA	11.11a	10.48ab	9.45b	9.33b	0.198	0.002	0.012	0.093
Source of fatty acids
De novo2	26.61b	28.57ab	30.51a	30.61a	3.392	0.004	0.006	0.516
Mixed3	32.67	33.42	33.77	33.37	2.672	0.105	0.107	0.317
Preformed4	40.75a	38.03ab	35.74b	36.01b	2.136	0.006	0.045	0.073

Means with different superscripts in each row differ significantly (P < 0.05).

¹Control, low-RPGAA (LRPGAA), medium-RPGAA (MRPGAA), and high-RPGAA (HRPGAA) groups were supplemented with 0, 0.18, 0.36, and 0.54 g/kg DM of guanidinoacetic acid (GAA) derived from RPGAA, respectively.

²De novo-synthesized fatty acids are those that originated via mammary de novo synthesis (<16 carbons).

³Mixed-source fatty acids are the sum of C16:0 and C16:1.

⁴Preformed fatty acids are those extracted from blood (>16 carbons).

Due to the increase in milk fat production with RPGAA supplementation, milk SFA yields quadratically increased (P = 0.026; Table 4), the yields of both de novo-synthesized and mixed FA linearly and quadratically increased (P < 0.05); they were highest for MRPGAA, lowest for the control, and intermediate for LRPGAA and HRPGAA. However, the production of preformed FA increased quadratically (P = 0.004), which was lower for HRPGAA than for MRPGAA and LRPGAA (P = 0.008). With increasing RPGAA doses, the daily milk FA production of C4:0, C8:0, C10:0, C14:0, C14:1, C16:0, C18:0, and C18:2n6c increased quadratically (P < 0.05), that of C6:0, C12:0, and C15:0 increased linearly (P < 0.05), and that of C22:6 decreased linearly (P = 0.031). The daily production of the other FA was not affected (P > 0.05).

Table 4.

Effects of rumen-protected guanidinoacetic acid (RPGAA) addition on milk fatty acid yield (g/d) in the milk of lactating dairy cows.

Item	Treatments1				SEM	P-value
	Control	LRPGAA	MRPGAA	HRPGAA		Treatment	Linear	Quadratic
C4:0	29.4c	36.4b	40.8a	38.2b	0.49	0.025	0.061	0.007
C6:0	22.2c	24.5b	26.0a	25.5a	0.20	0.032	0.042	0.086
C8:0	0.8c	1.2b	1.4a	1.2b	0.02	0.014	0.104	0.015
C10:0	33.7c	38.4b	43.9a	42.0b	0.51	0.023	0.083	0.025
C11:0	0.9	0.9	0.9	0.9	0.02	0.841	0.402	0.764
C12:0	45.2c	53.6b	61.0a	59.1a	0.88	0.018	0.016	0.077
C13:0	0.8	0.9	0.8	0.7	0.02	0.081	0.109	0.132
C14:0	132.5c	158.8b	186.6a	175.0b	2.30	0.043	0.074	0.014
C14:1	15.1c	16.0b	16.5a	15.4b	0.07	0.019	0.093	0.016
C15:0	10.1b	12.2a	13.2a	12.5a	0.24	0.008	0.012	0.081
C16:0	336.3c	378.0b	409.9a	381.2b	3.26	0.006	0.091	0.017
C16:1	21.0	23.1	23.1	22.7	0.48	0.343	0.224	0.193
C17:0	5.2	5.3	5.6	5.3	0.09	0.421	0.453	0.268
C18:0	139.0a	133.7b	141.3a	125.7c	0.90	0.015	0.081	0.018
C18:1n9c	178.6	190.4	188.9	190.6	2.32	0.106	0.412	0.281
C18:2n6c	86.0b	91.7a	90.1a	87.5b	0.43	0.014	0.079	0.039
C18:3	3.5	3.3	3.1	2.8	0.14	0.431	0.144	0.913
C20:0	1.2	1.3	1.3	1.3	0.02	0.113	0.353	0.124
C20:3	1.2	1.1	1.1	1.0	0.03	0.172	0.069	0.391
C20:4	2.3	2.4	2.4	2.5	0.05	0.673	0.292	0.702
C22:6	28.5a	27.3a	24.5b	19.1c	0.48	0.006	0.031	0.232
Sum of fatty acids
SFA	757.3c	845.2b	932.7a	868.6b	5.69	0.033	0.084	0.026
MUFA	214.7	229.5	228.5	228.7	3.84	0.076	0.246	0.093
PUFA	121.5	125.8	121.2	112.9	2.57	0.306	0.213	0.425
Source of fatty acids
De novo2	291.0c	343.0b	391.2a	370.4b	4.24	0.005	0.008	0.017
Mixed3	357.3c	401.2b	433.1a	403.9b	3.34	0.006	0.009	0.019
Preformed4	445.6ab	456.5a	458.3a	435.8b	2.67	0.008	0.231	0.004

Means with different superscripts in each row differ significantly (P < 0.05).

¹Control, low-RPGAA (LRPGAA), medium-RPGAA (MRPGAA), and high-RPGAA (HRPGAA) groups were supplemented with 0, 0.18, 0.36, and 0.54 g/kg DM of guanidinoacetic acid (GAA) derived from RPGAA, respectively.

²De novo-synthesized fatty acids are those that originated via mammary de novo synthesis (<16 carbons).

³Mixed-source fatty acids are the sum of C16:0 and C16:1.

⁴Preformed fatty acids are those extracted from blood (>16 carbons).

The provision of RPGAA increased (P < 0.05; Table 5) concentrations of His, Thr, and Arg linearly and those of Met and Glu quadratically (P < 0.05), but reduced (P = 0.014) that of Tyr linearly. Concentrations of other selected AA was not affected (P > 0.05). Due to the increase in milk protein production, supplementation with RPGAA had no influence on the daily secretion yields of Ala, Leu, Pro, Val, or Gly (P > 0.05, Table 6). However, it increased (P < 0.05) the daily secretion yields of His, Thr, Trp, Arg, Cys, and Ser linearly and those of Ile, Lys, Met, Phe, Asp, Cys, and Glu quadratically (P < 0.05), but reduced (P = 0.017) that of Tyr linearly.

Table 5.

Effects of rumen-protected guanidinoacetic acid (RPGAA) addition on amino acid content (g/100 g protein) in the milk of dairy cows.

Item	Treatments1				SEM	P-value
	Control	LRPGAA	MRPGAA	HRPGAA		Treatment	Linear	Quadratic
His	7.63c	7.93b	8.86a	8.32a	0.186	0.017	0.010	0.045
Ile	3.31	3.31	2.90	2.92	0.056	0.119	0.151	0.412
Leu	6.83	6.35	6.02	6.41	0.118	0.294	0.452	0.132
Lys	7.33	7.16	7.02	7.06	0.130	0.221	0.158	0.430
Met	0.97c	1.31b	1.48a	1.31b	0.043	0.012	0.073	0.004
Phe	2.19	2.22	2.26	2.22	0.049	0.410	0.466	0.507
Thr	3.67b	3.92ab	4.37a	4.38a	0.100	0.008	0.017	0.062
Trp	1.74	1.77	1.76	1.73	0.024	0.228	0.434	0.569
Val	5.77	5.46	5.05	5.03	0.098	0.086	0.079	0.474
Ala	10.83	10.21	9.54	10.04	0.058	0.103	0.090	0.153
Arg	4.84b	5.16a	5.15a	5.23a	0.060	0.032	0.044	0.354
Asp	6.38	6.54	6.65	6.59	0.077	0.418	0.573	0.606
Cys	0.61	0.60	0.61	0.63	0.013	0.327	0.413	0.514
Glu	17.94b	19.26ab	20.37a	19.79a	0.329	0.022	0.068	0.029
Gly	2.07	1.84	1.85	1.89	0.064	0.618	0.507	0.737
Pro	10.27	9.77	9.16	9.46	0.068	0.218	0.296	0.097
Ser	3.65	3.64	3.62	3.68	0.030	0.318	0.325	0.470
Tyr	3.96a	3.57ab	3.32b	3.29b	0.065	0.034	0.014	0.597

Means with different superscripts in each row differ significantly (P < 0.05).

¹Control, low-RPGAA (LRPGAA), medium-RPGAA (MRPGAA), and high-RPGAA (HRPGAA) groups were supplemented with 0, 0.18, 0.36, and 0.54 g/kg DM of guanidinoacetic acid (GAA) derived from RPGAA, respectively.

Table 6.

Effects of rumen-protected guanidinoacetic acid (RPGAA) addition on amino-acid yield (g/d) in the milk of dairy cows.

Item	Treatments1				SEM	P-value
	Control	LRPGAA	MRPGAA	HRPGAA		Treatment	Linear	Quadratic
His	79.7c	90.3b	108a	96.5a	1.55	0.021	0.013	0.056
Ile	34.6b	37.7a	35.3ab	33.9b	0.47	0.024	0.064	0.015
Leu	71.4	72.3	73.4	74.4	0.98	0.743	0.315	0.790
Lys	76.6c	81.5b	85.6a	81.9b	1.08	0.026	0.072	0.037
Met	10.1c	14.9b	18.1a	15.2b	0.36	0.015	0.091	0.005
Phe	22.9c	25.3b	27.6a	25.8b	0.41	0.013	0.083	0.009
Thr	38.3b	44.6b	53.3a	50.8a	0.83	0.010	0.021	0.078
Trp	18.2b	20.1a	21.5a	20.1a	0.20	0.035	0.042	0.086
Val	60.3	62.2	61.6	58.4	0.57	0.083	0.224	0.093
Ala	113.2	116.2	116.3	116.5	0.48	0.129	0.113	0.191
Arg	50.6b	58.7ab	62.8a	60.7a	0.50	0.013	0.018	0.067
Asp	66.7c	74.4b	81.1a	76.4b	0.64	0.022	0.091	0.008
Cys	6.39b	6.86b	7.40a	7.32a	0.11	0.034	0.016	0.143
Glu	187.4c	219.2b	248.3a	229.6b	2.74	0.027	0.085	0.036
Gly	21.6	20.9	22.5	21.9	0.53	0.772	0.634	0.921
Pro	107.3	111.2	111.6	109.7	0.57	0.273	0.370	0.121
Ser	38.1b	41.4ab	44.1a	42.7a	0.25	0.023	0.019	0.088
Tyr	41.4a	40.6ab	40.5ab	38.2b	0.46	0.042	0.017	0.371

Means with different superscripts in each row differ significantly (P < 0.05).

¹Control, low-RPGAA (LRPGAA), medium-RPGAA (MRPGAA), and high-RPGAA (HRPGAA) groups were supplemented with 0, 0.18, 0.36, and 0.54 g/kg DM of guanidinoacetic acid (GAA) derived from RPGAA, respectively.

3.3. Nutrient digestibility and blood metabolites

The digestibilities of dietary DM, OM, and CP increased quadratically (P < 0.05) with RPGAA supplementation (Table 7), being highest for MRPGAA, lowest for the control, and intermediate for LRPGAA and HRPGAA. Dietary EE digestibility was quadratically increased (P = 0.034) with RPGAA supplementation and was higher (P = 0.028) for MRPGAA than that of in the control. The digestibilities of NDF and ADF increased linearly (P < 0.05) with RPGAA provision with the highest values observed in the MRPGAA and HRPGAA groups, the lowest in the control group, and intermediate values in the LRPGAA group.

Table 7.

Effects of rumen-protected guanidinoacetic acid (RPGAA) addition on nutrient apparent digestibility and blood metabolite levels in dairy cows.

Item	Treatments1				SEM	P-value
	Control	LRPGAA	MRPGAA	HRPGAA		Treatment	Linear	Quadratic
Nutrient digestibility, kg/kg
DM	0.700c	0.715b	0.723a	0.714b	0.001	0.029	0.423	0.042
OM	0.708c	0.736b	0.759a	0.739b	0.002	0.037	0.156	0.035
CP	0.647c	0.667b	0.694a	0.663b	0.003	0.042	0.317	0.024
EE	0.833b	0.852ab	0.871a	0.853ab	0.005	0.028	0.142	0.034
NDF	0.584c	0.599b	0.619a	0.616a	0.003	0.024	0.037	0.095
ADF	0.479c	0.514b	0.532a	0.530a	0.004	0.039	0.045	0.086
Non-fiber carbohydrate	0.856c	0.879b	0.903a	0.857c	0.003	0.031	0.210	0.042
Blood metabolites
Glucose, mmol/L	4.18b	4.34b	4.78a	4.53a	0.037	0.029	0.017	0.208
Total protein, g/L	72.3b	76.7a	79.2a	77.4a	0.57	0.043	0.013	0.307
Albumin, g/L	32.1c	35.6a	37.0a	36.0a	0.37	0.038	0.019	0.401
BUN, mmol/L	6.91a	6.74ab	6.63b	6.93a	0.050	0.041	0.871	0.018
Triglycerides, mmol/L	2.19	2.29	2.42	2.31	0.030	0.058	0.059	0.076
Creatine, μmol/L	40.0c	44.1b	47.0a	46.0a	0.41	0.032	0.043	0.101
Arginine, μmol/L	165c	176b	189a	191a	5.4	0.045	0.039	0.208
Estradiol, pg/mL	47.8c	50.6b	60.6a	50.0b	0.55	0.018	0.079	0.019
Prolactin, mIU/L	563c	609b	692a	647b	7.7	0.024	0.088	0.023
IGF-1, ng/mL	193c	225b	236a	203b	1.9	0.033	0.102	0.027

BUN = blood urea nitrogen; IGF-1 = insulin-like growth factor 1.

Means with different superscripts in each row differ significantly (P < 0.05).

¹Control, low-RPGAA (LRPGAA), medium-RPGAA (MRPGAA), and high-RPGAA (HRPGAA) groups were supplemented with 0, 0.18, 0.36, and 0.54 g/kg DM of guanidinoacetic acid (GAA) derived from RPGAA, respe ctively.

Following the addition of RPGAA, the blood contents of glucose, albumin, and total protein increased linearly (P < 0.05). Blood glucose levels were higher (P = 0.029) in the MRPGAA and HRPGAA groups than those in the control and LRPGAA groups. Both total protein and albumin levels were greater (P < 0.05) in the RPGAA-added groups than that of in the control group. Conversely, the concentration of BUN quadratically decreased (P = 0.018) with the addition of RPGAA and was lower for MRPGAA than for the control and HRPGAA groups (P = 0.041). No significant difference was observed in blood triglyceride levels (P > 0.05). Blood arginine and creatine contents increased linearly (P < 0.05) with RPGAA provision; the levels were highest for MRPGAA and HRPGAA, lowest for the control, and intermediate for LRPGAA. Blood contents of E2, prolactin, and IGF-1 increased quadratically (P < 0.05) with the supply of RPGAA; they were highest for MRPGAA, lowest for the control, and intermediate for LRPGAA and HRPGAA.

3.4. Expression of cell proliferation-related proteins

The provision of 0.36 g/kg DM of GAA from RPGAA increased the protein levels of PCNA (P < 0.05), cyclin A1 (P < 0.05), BCL2 (P < 0.01), and BCL2/BAX (P < 0.01). Nevertheless, the levels of BAX, Caspase-3, and Caspase-9 were reduced upon treatment with GAA compared to the control (P < 0.01; Fig. 1A and B). Moreover, the protein expression ratios of p-Akt/Akt and p-mTOR/mTOR were also increased with the supplementation of 0.36 g/kg DM of GAA (P < 0.01; Fig. 1A and B).

Fig. 1.

The expression levels of proliferation-related proteins in the bovine mammary gland. (A) Western blot analysis of Akt, p-Akt, mTOR, p-mTOR, PCNA, cyclin A1, BCL2, BAX, BCL2/BAX, Caspase-3, and Caspase-9 in bovine mammary gland tissues treated with 0 g/kg (control) and 0.36 g/kg DM of GAA derived from RPGAA (MRPGAA), respectively; β-actin was used as the loading control. (B) Mean ± SEM of the immunopositive bands of PCNA, cyclin A1, BCL2, BAX, BCL2/BAX, Caspase-3, Caspase-9, p-Akt/Akt, and p-mTOR/mTOR. Akt = protein kinase B; mTOR = rapamycin target protein; PCNA = proliferating cell nuclear antigen; BCL2 = B-cell lymphoma 2; BAX = BCL2-associated X; Caspase = cysteine–aspartic acid protease; GAA = guanidinoacetic acid; RPGAA = rumen-protected guanidinoacetic acid; p = phosphorylated. ∗P < 0.05 and ∗∗P < 0.01 versus the control group.

3.5. Expression of proteins concerned with milk FA synthesis

Supplementation of 0.36 g/kg DM of GAA from RPGAA upregulated PPARγ (P < 0.05), SREBP1 (P < 0.01), FASN (P < 0.01), and SCD1 (P < 0.01) compared to their levels in the control (Fig. 2A and B), and promoted the phosphorylation of ACACA, as revealed by the higher protein expression ratio of p-ACACA/ACACA than the control (P < 0.01). Additionally, the expression ratio of p-AMPK/AMPK was lower (P < 0.05) in the RPGAA-added groups than the control, suggesting that the phosphorylation of AMPK was inhibited by 0.36 g/kg DM of GAA from RPGAA.

Fig. 2.

Changes in the expression levels of proteins concerned with fatty acid synthesis. (A) Western blot analysis of AMPK, p-AMPK, PPARγ, SREBP1, ACACA, p-ACACA, FASN, and SCD1 in bovine mammary gland tissues treated with 0 g/kg (control) and 0.36 g/kg DM of GAA derived from RPGAA (MRPGAA); β-actin was used as the loading control. (B) Mean ± SEM of immunopositive bands of PPARγ, SREBP1, p-ACACA/ACACA, FASN, SCD1, and p-AMPK/AMPK. AMPK = adenosine monophosphate activated protein kinase; PPARγ = peroxisome proliferator-activated receptor gamma; SREBP1 = sterol regulatory element-binding protein 1; ACACA = acetyl-coenzyme A carboxylase-α; FASN = fatty acid synthase; SCD1 = stearoyl-CoA desaturase 1; GAA = guanidinoacetic acid; RPGAA = rumen-protected guanidinoacetic acid; p = phosphorylated. ∗P < 0.05 and ∗∗P < 0.01 versus the control group.

3.6. Expression of proteins associated with milk protein synthesis

The supplementation of 0.36 g/kg DM of GAA from RPGAA increased the expression levels of κ-casein, β-casein, and αs1-casein proteins compared to their levels in the control (P < 0.01; Fig. 3A and B). Additionally, the expression ratios of p-JAK2/JAK2 (P < 0.01), and p-STAT5/STAT5 (P < 0.05) were higher than the control, when 0.36 g/kg DM of GAA from RPGAA was added, suggesting that the phosphorylation of JAK2 and STAT5 was promoted by RPGAA addition.

Fig. 3.

Changes in the expression levels of proteins involved in milk protein synthesis. (A) Western blot analysis of JAK2, p-JAK2, STAT5, p-STAT5, αs1-casein, β-casein, and κ-casein in bovine mammary gland tissues treated with 0 g/kg (control) and 0.36 g/kg DM of GAA derived from RPGAA (MRPGAA); β-actin was used as the loading control. (B) Mean ± SEM of immunopositive bands of κ-casein, αs1-casein, β-casein, p-JAK2/JAK2, and p-STAT5/STAT5. JAK2 = Janus kinase 2; STAT5 = signal transduction and transcriptional activator 5; GAA = guanidinoacetic acid; RPGAA = rumen-protected guanidinoacetic acid; p = phosphorylated. ∗P < 0.05 and ∗∗P < 0.01 versus the control group.

4. Discussion

The ruminal degradation rate of RPGAA (19.3%) was lower than that of GAA (estimated at 53%) (Speer et al., 2020). Thus, supplementary RPGAA can reach the small intestine more effectively and produce more creatine than GAA. In addition, most RPGAA is released in the small intestine and only some amount is released in the rumen. In this study, the supplementary RPGAA could provide 2.54, 5.08, and 7.51 g/d of GAA for small intestinal absorption in the LRPGAA, MRPGAA, and HRPGAA groups, respectively; meanwhile, the corresponding levels were 0.75, 1.49, and 2.21 g/d of GAA for ruminal bacteria in the LRPGAA, MRPGAA, and HRPGAA groups, respectively.

4.1. Dietary DM intake and milk performance

In this study, no remarkable difference in DMI was observed among treatments, which is in keeping with other studies on dairy cows (Liu et al., 2023), Angus bulls (Liu et al., 2021b), and growing cattle (Speer et al., 2022). The linearly increased production of milk, 4.0% FCM, ECM, milk fat, and milk protein, and the increase in feed efficiency after RPGAA supplementation can be attributed to the enhancement in nutrients digestibility and blood creatine concentrations, suggesting that RPGAA addition enhanced the lactogenic, lipogenic, and proteogenic capacities of the mammary gland, thereby elevating the production of fat and protein in dairy cow milk. Similarly, the addition of 0, 6, 12, and 18 g/d of GAA increased milk production, milk fat percentage, and feed efficiency linearly in Chinese Holstein dairy cows (Liu et al., 2023). The quadratic elevation in the percentage and production of fat and milk protein suggests that the biosynthesis of fat and casein in the mammary glands was enhanced by the RPGAA supply.

The linear increase in milk production and feed efficiency observed in this study can be primarily attributed to the improved nutrient digestibility and increased blood creatine concentrations following RPGAA supplementation, as creatine plays a crucial role in energy transfer within cells, facilitating the rapid supply of energy required for milk synthesis and secretion in the mammary gland (Kaviani et al., 2020). The elevated levels of creatine in the bloodstream likely enhanced the energy availability for lactating cows, thereby contributing to the improved milk production and feed efficiency. Additionally, the increased concentrations of serum Arginine (Arg) observed in this study are noteworthy. Arginine is a precursor to GAA and is known to promote milk casein synthesis in bovine mammary epithelial cells (BMEC) through the activation of signaling pathways converging on mTOR (Wang et al., 2014) and JAK2-STAT5 (Zhang et al., 2020). Therefore, the enhanced milk yield and biosynthesis of milk fat and protein with RPGAA supplementation can be attributed to both the increased energy supply due to creatine synthesis and the activation of the mTOR and JAK2-STAT5 signaling pathways by RPGAA.

4.2. Milk fat acid and AA secretion

Given that DMI remained unchanged with RPGAA addition, the increases in milk yield, milk fat/protein production, and milk fat/protein content could be due to the enhanced utilization efficiency of nutrients. This enhancement likely resulted from the improved nutrient delivery to the mammary gland rather than from increased storage or metabolism in other tissues (Ding et al., 2022), enhanced mammary epithelial cells (MEC) proliferation (Ge et al., 2022), or increased FA (de novo or mixed) synthesis (Ding et al., 2022; Zhao and Agellon, 2017). Fatty acids in milk are formed via 2 processes: de novo synthesis in the mammary gland and the incorporation of FA from the diet, microbial activity, body fat mobilization, or liver synthesis, which are then exported into the blood (Zhao and Agellon, 2017). In this study, the linear increase in FA content synthesized de novo or from mixed sources suggests that the supplementation of RPGAA enhanced the de novo synthesis of FA. De novo FA synthesis results from acetate and butyrate absorption in the rumen. Although RPGAA was used in the study, the release of GAA in the rumen (0.75, 1.49, and 2.21 g/d in the LRPGAA, MRPGAA, and HRPGAA groups, respectively) might promote the activity of rumen cellulolytic bacteria, thus increasing the production of acetate and butyrate. However, these results should be verified in future studies. Additionally, the effect of RPGAA on FA can be partly explained by an improvement in the expression of proteins involved in mammary lipogenesis.

Milk protein production is not only the result of AA absorption by the mammary, which in turn depends on the intestinal absorption and transport of AA by the mammary (Ding et al., 2019; Weston et al., 2023). Milk proteins are also produced because the mammary glands can synthesize proteins. In this study, although no impact on the daily secretion yields of Ala, Leu, Pro, Val, or Gly was observed with RPGAA addition, RPGAA supplementation linearly increased the daily secretion yields of His, Thr, Trp, Arg, Cys, and Ser and quadratically increased the daily secretion yields of Ile, Lys, Met, Phe, Asp, Cys, and Glu. These results are confirmed by Ardalan et al. (2020), who found that abomasal infusions of GAA linearly increased the plasma concentrations of His, Thr, Arg, Tyr, and Asp, and quadratically increased the plasma concentrations of total AA, Glu, Asp, and Ser, because of the Arg-sparing effect of GAA. The above results were supported by Ding et al. (2019), who found that jugular infusion of Arg resulted in a greater supply of Arg, Glu, His, Met, and Thr to the mammary. The increase in milk protein synthesis due to GAA can be attributed to various effects. The action of Arg, spared by GAA, on the mTOR and JAK2-STAT5 pathways (Kong et al., 2014; Wang et al., 2014) cannot be ignored, as confirmed by the expression of proteins concerned with milk protein synthesis in the mammary tissue.

Increased milk protein synthesis resulted from enhanced AA availability and activation of key pathways. The linear increase in the daily secretion yields of His, Thr, Trp, Arg, Cys, and Ser, and the quadratic increase in Ile, Lys, Met, Phe, Asp, Cys, and Glu, indicate that RPGAA supplementation improved the supply of these essential AA to the mammary gland. This finding is consistent with previous studies showing that GAA has an Arg-sparing effect, which allows more Arg to be available for milk protein synthesis (Ardalan et al., 2020). The spared Arg can then activate the mTOR and JAK2-STAT5 signaling pathways, which are crucial for promoting protein synthesis in the mammary gland (Kong et al., 2014; Wang et al., 2014). Future research should focus on elucidating the specific mechanisms by which GAA affects AA metabolism and signaling pathways in the mammary gland.

4.3. Nutrient digestibility and blood metabolites

The observed improvements in nutrient digestibility, particularly for DM, OM, CP, EE, ADF, and NDF, suggested that RPGAA supplementation enhances the efficiency of ruminal degradation. This finding is supported by previous studies showing increased total VFA concentration, enzymatic activity, and bacterial populations in the rumen following GAA supplementation (Liu et al., 2021b, 2023; Li et al., 2020a). Although only 19.3% of RPGAA was degraded in the rumen in this study, the released GAA likely stimulated the activity of rumen cellulolytic bacteria, thereby improving the digestibility of fibrous components such as NDF and ADF. This hypothesis (i.e., that released GAA stimulates rumen cellulolytic bacterial activity) requires further validation through targeted experiments. Additionally, the role of creatine in promoting intestinal health and function in monogastric animals suggests that RPGAA may also enhance post-ruminal nutrient absorption by improving gut morphology and function (Sistermans et al., 1995; Ahmadipour et al., 2018; Ren et al., 2018). Future studies should investigate the specific effects of RPGAA on rumen microbiota and intestinal health in dairy cows.

The observed increase in serum glucose levels following RPGAA supplementation can be attributed to the sparing effect of Arg, which has been shown to enhance glucose metabolism in previous studies (Green et al., 2017; Souza Simões et al., 2024). The elevated levels of serum total protein and albumin indicate improved nitrogen utilization efficiency, suggesting that RPGAA promotes better protein metabolism and utilization in dairy cows (Nousiainen et al., 2004). Additionally, the increases in serum IGF-1, prolactin, and E2 levels observed in this study are significant, as these hormones play crucial roles in mammary gland development and epithelial cell proliferation (Rosen, 2012; Mallepell et al., 2006). The activation of the PI3K/Akt signaling pathway by IGF-1 further supports the notion that RPGAA stimulates mammary gland development (Zhou et al., 2017). The linear increase in serum creatine and Arg levels following RPGAA supplementation is consistent with previous findings in dairy cows and growing steers (Liu et al., 2023; Speer et al., 2022). These increases likely contribute to the enhanced lactation performance by providing more energy and essential nutrients to the mammary gland. Future studies should explore the specific mechanisms by which RPGAA affects hormone levels and signaling pathways related to mammary gland development.

4.4. Expression of proteins implicated in cell proliferation

Mammary gland development depends primarily on the proliferation of MEC, and stimulating MEC proliferation is key to sustaining milk production. Markers that reflect MEC proliferation include proliferation markers (such as cyclin and PCNA) and apoptosis markers (such as BCL2, Bax, and Caspase). The cell cycle from the G1 to the S phase is regulated by cyclin D, whereas cyclin A is necessary to initiate and complete DNA replication during the S phase (Lents and Piszczatowski, 2023). Proliferating cell nuclear antigen is also integral for DNA replication and repair (Wang et al., 2023). Furthermore, cyclins and PCNA are involved in regulating MEC proliferation (Zhang et al., 2023a,b). Adding GAA increased the expression levels of PCNA and cyclin A1 in the study, suggesting that GAA stimulates MEC proliferation. The BCL2 family contains anti-apoptotic (such as BCL2) and pro-apoptotic (such as BAX) proteins, which are involved in apoptosis regulation (Van Delft and Dewson, 2023). The BCL2/BAX ratio reflects the status of cellular apoptosis in BMEC owing to the prevention of apoptosis by BCL2 via the inhibition of BAX activity. Caspase-9 is the initiator involved in the internal apoptosis pathway and finally activates Caspase-3, the final and key enzyme in the execution stage of apoptosis (Khodajou-Masouleh et al., 2022). Reduction in their expression levels inhibits apoptosis. In the study, BCL2 protein expression and BCL2/BAX ratio were increased, whereas apoptosis-related proteins, such as BAX, Caspase-3, and Caspase-9, were upregulated, suggesting that RPGAA prevented apoptosis in the mammary gland. These findings collectively imply that RPGAA promotes mammary gland development by enhancing cell proliferation and preventing apoptosis. Future studies should investigate the specific mechanisms by which RPGAA affects cell cycle regulation and apoptosis signaling pathways in MEC.

The Akt signaling pathway is intricately involved in regulating cell proliferation, differentiation, apoptosis, and other cellular functions (Fujiwara et al., 2014; Li et al., 2020). Estradiol, progesterone, and various growth factors have been shown to stimulate MEC proliferation through the Akt signaling pathway (Diep et al., 2016; Huang et al., 2021). Phosphorylated Akt activates key components of the mTOR signaling pathway, thereby mediating cell proliferation and biosynthesis activities (Bilanges et al., 2019; Qiao et al., 2022). Consistent with previous findings in lamb muscle (Liu et al., 2021a; Zhang et al., 2022; Li et al., 2023), the current study observed increased phosphorylation of Akt and mTOR in the mammary gland following GAA administration. These results suggest that RPGAA promotes mammary gland development by activating the Akt/mTOR signaling pathway, which is crucial for cell proliferation and biosynthesis. Future research should explore the specific interactions between RPGAA and the Akt/mTOR pathway in the context of mammary gland development.

4.5. Expression of proteins associated with milk fat synthesis

In the mammary organs, both PPARγ and SREBP1 are activators of lipogenesis (Abdelatty et al., 2017; Kadegowda et al., 2009) and regulate ACACA, FASN, and SCD expression (Abdelatty et al., 2017; Faulconnier et al., 2019). Both FASN and ACACA are involved in de novo FA biosynthesis from acetic acid and β-hydroxybutyric acid in the mammary gland (Faulconnier et al., 2019) and concerned with the secretion of de novo-synthesized FA (Abdelatty et al., 2017; Tian et al., 2022). Stearoyl-CoA desaturase promotes the biosynthesis of monounsaturated FA from saturated FA (SFA; Conte et al., 2010). The increased expression levels of PPARγ, SREBP1, p-ACACA/ACACA, FASN, and SCD1 following RPGAA addition indicated that the de novo biosynthesis of FA and the desaturation of SFA were enhanced in the mammary tissues. The increased expression of these proteins suggests that RPGAA promotes the efficient synthesis and secretion of FA in the mammary gland. This finding is consistent with previous studies showing that arginine infusion upregulates PPARγ, ACACA, and SCD, thereby enhancing de novo FA biosynthesis (Ding et al., 2022). Future research should investigate the specific mechanisms by which RPGAA affects the expression of these lipogenic proteins and their role in milk fat synthesis.

Adenosine monophosphate activated protein kinase serves as a key energy sensor in cells, and its expression and activity are influenced by the cellular energy status (Ding et al., 2022; Inoki et al., 2012). In this study, the observed inhibition of AMPK phosphorylation following RPGAA supplementation suggests that the cells in the mammary gland were in a positive energy balance, likely due to the increased availability of glucose and other energy substrates. This finding is consistent with previous studies showing downregulation of AMPK during arginine infusion (Ding et al., 2022). The inhibition of AMPK may further promote lipogenesis by reducing its inhibitory effects on key lipogenic enzymes. Future studies should explore the specific interactions between RPGAA and AMPK signaling in the context of mammary gland metabolism.

4.6. Expression of proteins involved in milk protein synthesis

The observed upregulation of casein proteins (αs1-, β-, and κ-casein) and the activation of the JAK2/STAT5 signaling pathway following RPGAA supplementation suggest that GAA promotes milk protein synthesis through multiple mechanisms. The Arg-sparing effect of GAA ensures that sufficient Arg is available for transport to the mammary glands, where it plays a crucial role in milk protein synthesis (Ding et al., 2019; Zhang et al., 2020). Arginine not only regulates cell proliferation and protein turnover but also activates key signaling pathways such as JAK2/STAT5 and mTOR, which are essential for casein gene expression and milk protein synthesis (Ma et al., 2018; Wang et al., 2014). The findings of this study are consistent with previous reports showing that Arg supplementation enhances αs1-casein synthesis (Sun et al., 2023) and increases milk protein percentage and casein production in dairy cows (Zhang et al., 2020). Future research should further investigate the specific mechanisms by which GAA affects milk protein synthesis, particularly the role of Arg in regulating gene expression and signaling pathways in the mammary gland.

Signal transduction and transcriptional activator 5 is a key transcription factor that regulates casein gene expression and is activated by phosphorylation via JAK2 (Levy and Darnell, 2002). The activity of STAT5 is particularly important during lactation, as it responds to lactogenic factors and promotes milk protein synthesis (Yang et al., 2000). The observed increases in JAK2 and STAT5 phosphorylation levels following RPGAA supplementation in this study further support the role of these signaling molecules in enhancing milk protein synthesis. Future research should explore the specific interactions between RPGAA, JAK2, and STAT5, and their downstream effects on milk protein synthesis.

4.7. Practical implications and future research

The findings of this study have significant practical implications for dairy farming. The use of RPGAA as a dietary supplement could potentially enhance milk production and improve the nutritional quality of milk by increasing the content of beneficial FA and AA. This could lead to higher economic returns for dairy farmers and improved milk quality for consumers. Further research is needed to optimize the dosage and application conditions of RPGAA in different breeds and production systems, as well as to explore its long-term effects on cow health and sustainability of dairy production.

5. Conclusion

Adding rumen-protected guanidinoacetic acid to the dairy cow diet can positively influence milk yield and milk FA and AA composition in a dose-dependent manner. The provision of medium levels of rumen-protected guanidinoacetic acid can stimulate the biosynthesis of milk FA and proteins in the mammary glands. Furthermore, medium levels of rumen-protected guanidinoacetic acid activated the Akt/mTOR, AMPK, and JAK2/STAT5 signaling pathways and promoted the expression of proteins involved in cell proliferation, milk FA synthesis, and protein synthesis. Future studies should focus on identifying the mechanisms by which milk FA and protein synthesis can be enhanced.

Credit Author Statement

Jing Zhang: Writing – review & editing, Writing – original draft, Project administration, Conceptualization. Yanchu Tang: Resources, Investigation, Formal analysis. Changjian Xue: Resources, Investigation, Formal analysis. Jiaojiao Lang: Software, Investigation, Formal analysis. Wenjie Huo: Validation, Methodology, Data curation. Caixia Pei: Validation, Software, Methodology, Data curation. Qiang Liu: Writing – review & editing, Supervision, Funding acquisition, Conceptualization.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.