Extraction, identication, and quantication of volatile fatty acids (VFA) in rumen uid samples using Reverse Phase High-Performance Liquid Chromatography with Diode Array Detector (RP HPLC-DAD)

In ruminant animals, volatile fatty acids (VFA) or short-chain fatty acids (SCFAs) are derived from the protein and carbohydrate fermentation by rumen microorganism. Hence, the VFA determination in rumen uid allows the evaluation of the nutritional quality of a diet, as well as its potential impact on the chemical composition of ruminant milk and meat. Thus, we developed a protocol to extract, identify, and quantify acetic, propionic, butyric, valeric, and caproic acids in ruminal uid samples using RP-HPLC-DAD. Despite literature ndings had shown that the most suitable column for VFA chromatographic separation under HPLC-DAD is an ion-exchange column, our protocol showed that a C18 column also allows an ecient VFA separation of the aforementioned acid, except for butyric and iso-butyric acids. This condition may constitute a limitation of the Hypersil GOLD C18 column use for VFA determination. However, considering that the concentration of iso-butyric acid is signicantly lower than that of butyric acid, a good estimation of butyric acid can be obtained.


Introduction
Ruminants obtain nutrients from feeds for milk, meat, and wool production. But nutrient utilization from feeds implies the activity of different digestive processes, in which rumen microorganisms play a central role in. When ruminant consumes forage or concentrates, the carbohydrate and protein molecules enter the rumen and these are initially catabolized by the action of enzymes secreted by rumen bacteria, fungus, and protozoa. The nal products of this catabolic process are the volatile fatty acids (VFA) or short-chain fatty acids (SCFAs).
The VFA are short carbon chain compounds that are produced during the fermentative degradation of feeds in the rumen. These are absorbed by the rumen wall and converted by anabolic processes into carbohydrates, proteins, and lipids being these largely deposited in milk and meat.
Generally, formic, acetic, propionic, butyric, iso-butyric, 2-methyl butyric, valeric, isovaleric, caproic, and caprylic acids are considered as VFA. However, acetic, propionic, and butyric acids are the main VFA produced during the feed rumen fermentation. Hence, the objective of this protocol is to describe a method to extract, identify, and quantify VFA in rumen uid samples, using RP HPLC-DAD as a quanti cation technique.

Extraction solution:
Solution of ortho-phosphoric acid 0.85% (v/v) containing adipic acid in the concentration of 24 mM (100 mL) (working solution): accurately and carefully weigh 0.3507 g of adipic acid (internal standard; Sigma Aldrich®) in a beaker and dissolve them in 50 mL of distilled water (this process needs to be done with magnetic stirring and slight heating). When adipic acid is dissolved, carefully add 1 mL of 85% v/v orthophosphoric acid (Sigma Aldrich®). Transfer the resulting mixture to a 100 mL class A volumetric ask and complete to 100 mL with distilled water.
Note: this solution will ensure after dilutions, that the internal standard will be in the ruminal uid sample at a concentration of approximately 6 mM (the same concentration of internal standard included in the calibration curve).

Mobile phase components for HPLC running:
Phase A: sodium phosphate buffer solution 10 mM (pH = 2.6) (1 L): weigh 0.78 g of Sodium dihydrogen phosphate dihydrate (Sigma Aldrich ®) and mix them with 0.34 mL of 85% ortho-phosphoric acid in a beaker. Add 10 mL of Milli-Q water and transfer the mixture to a class A volumetric ask of 1L. Complete the volume to 1L with Milli-Q water.
Phase B: Acetonitrile HPLC grade (Sigma Aldrich ®). 3.3.2 Preparation of the calibration curve from 2 to 8 mM of VFA using a commercial mix VFA mixture (CRM46975, Sigma Aldrich) and containing adipic acid 6 mM (internal standard): Considering that the concentration of each fatty acid in the commercial VFA mixture is equal to 10 mM, prepare 5 chromatographic vials and add to each one, the reagent amounts described in table 1 (see supplementary les) to prepare the calibration curve in the range of concentrations mentioned above.

Equipment
-HPLC equipped with an autosampler, Hypersil GOLD C18 column, Hypersil GOLD C18 guard column, a column oven, and a Diode Array Detector.
-Centrifuge tubes of at least 5 mL of capacity.
-Transfer pipettes from 100 uL to 5 mL plus tips.
-Chromatographic vials with septa. Procedure 5.1 Ruminal uid collection and sample preparation: 1. From a stulated cattle, sheep or goat, collect at least 50 mL of rumen uid ltering it through three cheese-cloth layers. Make the collection in a 39 ºC pre-warmed thermos and transport it to the laboratory as soon as possible.
Notes: 1) to thermos pre-warming, ll it with pre-warmed water at 39 ºC, and empty it at the moment to collect the rumen uid. 2) the rumen uid sample must be composed of rumen uid from three rumen regions: front and half of the ventral sac, and from cranial sac (Zijderveld et al., 2011).
2. Analyze each point of the curve by RP HPLC-DAD.
3. Identify and integrate the resulting chromatographic peaks.

Chromatographic analysis of the ruminal uid samples:
1. Place 500 μL of the nal extract in a chromatographic vial.
3. Analyze by RP HPLC-DAD. 4. Identify and integrate the resulting chromatographic peaks.
Note: According to the literature ndings, the most suitable column for VFA chromatographic separation by HPLC-DAD is an ion-exchange column. However, the Hypersil GOLD C18 column allows an e cient VFA separation, except for butyric and iso-butyric acids. This condition may constitute a limitation of the Hypersil GOLD C18 column use for VFA determination. However, considering that the concentration of iso-butyric acid is signi cantly lower than that of butyric acid, a good estimation of butyric acid can be obtained. Troubleshooting 1. According to Hypersil GOLD C18 column manufacturer, the secure pH range of buffer use is between 2 -8. Therefore, in case of future optimizations of this chromatographic method, the use of solvents or mobile phases with a pH out of this range must be avoided to preserve the activity of the stationary phase.
2. The column used in the proposed method does not allow the use of 100% aqueous mobile phases (example: the use of solutions of inorganic acids). Therefore, in case of future optimizations of this method, the use of a mobile phase 100% aqueous must be avoided.
3. Depending on the rumen uid sample volume used, the quantities of reagents to be used during the VFA extraction can be proportionally reduced without signi cant effects on the chromatographic separation.
4. Verify that the concentration of the internal standard in the nal rumen samples and the calibration curve point tends to be the same, assuming that internal standard remains constant after VFA extraction.
In this protocol, we calculated the quantities of reagents to preserve a similar concentration of the internal standard in both samples and calibration points (6 mM).

Calculations (for each VFA using the internal standard method):
A: for calibration points: B: for samples: 4. Determination 3: compute VFA peak area / internal standard peak area.
5. Using determination 3, calculate VFA peak concentration / internal standard peak concentration of samples from the linear regression equation tted in step 3. After that, multiply this value by the internal standard concentration initially de ned (i.e., 6 mM in our protocol).
Time Taken

Anticipated Results
The representative chromatograms of standards and a rumen uid sample are shown in gures 1 and 2, respectively (see gures section) Figure 1 Chromatogram of 10 mM VFA mixture standard including adipic acid as the internal standard. Note: the term "butyric acid" in the chromatogram may represent a mixture of butyric and iso-butyric acids. Figure 2