Adding bovine seminal plasma prior to freezing improves post-thaw bull sperm kinematics but decreases mitochondrial activity

ABSTRACT

Variations in fertility between bulls with comparable sperm quality could be due to differences in their seminal plasma (SP). The aim of this study was to investigate the effect of adding bovine SP from bulls of known fertility to SP-free sperm samples. After removal of SP by Single Layer Centrifugation, resuspended sperm pellets were treated with SP from high or low fertility bulls at 0% (control), 1%, or 5% before freezing. Sperm quality was evaluated after thawing. Data were analyzed using Proc MIXED, SAS®. Bovine SP at 1% or 5% SP1 and SP5, respectively, decreased average path velocity, curvilinear velocity, and amplitude of lateral head displacement whereas wobble and linearity were increased. In addition, the proportion of spermatozoa with high mitochondrial membrane potential (MMP) was lowest for treatment with SP5 compared to SP1 and control. The proportion of SP did not affect other parameters of sperm quality. Thus, adding 5% bovine SP produced a favorable effect on some sperm velocity parameters but had an unfavorable effect on MMP. There were no differences in effect between SP from high and low fertility bulls.

Abbreviations: AI: artificial insemination; BCF: beat cross frequency; CASA: computer-assisted sperm analysis; IVF: in vitro fertilization; MMP: mitochondrial membrane potential; SLC: single layer centrifugation; SP: seminal plasma

KEYWORDS:

• Bovine SP
• fertility
• sperm quality
• frozen semen

Introduction

Fertility in inseminated dairy cattle is thought to have declined over the last few decades, although differences in sperm handling procedures and sperm numbers in insemination doses (López-Gatius 2013) may have contributed to this decreased fertility. Semen has two major compartments; spermatozoa and seminal plasma (SP). Bovine SP is composed of secretions from the accessory sex glands together with a small volume of fluid from the testis and epididymis (Maxwell et al. 2007). It contains proteins, minerals, electrolytes, hormones, and enzymes (Poiani 2006) and has been shown to activate and support spermatozoa (Graham 1994; Maxwell et al. 1996; Garner et al. 2001). Furthermore, SP is implicated in sperm decapacitation and in fertilization (Maxwell et al. 2007; Rodriguez-Martinez et al. 2011). A male’s fertility is usually considered to be a function of sperm fertility, whereas the contribution of the SP is usually ignored.

Bovine SP composition is related to sperm fertilizing ability (Maxwell et al. 2007; Juyena and Stelletta 2012), especially the proteins (Leahy and de Graaf 2012). The major components of bovine SP include peptidase proteins, cytokines, enzymes, antioxidants, hormones, ions, sugar, and lipid (Juyena and Stelletta 2012). Variations in specific protein content (Killian et al. 1993; Morrell et al. 2015) and protein expression (Peddinti et al. 2008; D’Amours et al. 2010) between high and low fertility bulls have been reported. The proteins involved in energy metabolism, cell communication, spermatogenesis, and cell motility have greater expression in semen from high fertility bulls than from low fertility bulls (Peddinti et al. 2008). In a study with bovine semen, the concentration of heparin-binding proteins had a strong relationship to calving rates (Bellin et al. 1994). In contrast, a low ratio of phosphorylcholine to heparin-binding proteins was related to high fertility in Holstein bulls, with the proportion of heparin-binding proteins being increased in low fertility bulls (Morrell et al. 2015). Heparin-binding proteins are believed to be involved in sperm capacitation and acrosome reaction processes (Chandonnet et al. 1990). Sperm fertility is evaluated in vivo by analyzing pregnancy data after artificial insemination (AI). The sperm quality is evaluated in the laboratory by tests for structural and functional parameters such as morphology, motility, plasma membrane integrity, acrosomal membrane integrity, mitochondrial function, chromatin structure, capacitation, and fertilization (Graham and Moce 2005). This evaluation is done manually or by automated sperm analysis using computer-assisted sperm analysis (CASA) and flow cytometry, and in vitro fertilization (IVF) (Graham and Moce 2005; Morrell and Rodriguez-Martinez 2011). Correlations between sperm quality and field fertility are reported (Graham 2001; Rodriguez-Martinez 2003) although there is no consensus on which parameters of sperm quality are indicative of field fertility. Thus, fertility was found to be correlated with sperm motility (Zhang et al. 1998; Januskauskas et al. 2003), and with viability and chromatin integrity (Januskauskas et al. 2003; Nongbua et al. 2014). The differences in expression of BSP proteins in normospermic and asthenospermic bulls were shown to be related to sperm motility (Divyashree and Roy, 2018).

When preparing bull semen doses for AI, semen extender is added to the raw ejaculate to produce a sperm suspension of the desired sperm concentration, calculated to deliver a set number of spermatozoa to the cow when the 0.25 mL straw is thawed. Thus the SP is not removed from the semen. The proportion of bovine SP included in the insemination dose varies between ejaculates, depending on the sperm concentration in the original ejaculate, the intended number of spermatozoa per straw, and sometimes on the fertility of the bull (Bromfield 2016). Spermatozoa account for approximately 1.0–1.5 mL of the raw semen (depending on sperm concentration) as observed from the pellet obtained after centrifugation. Using this rough estimation, we estimated that commercial semen doses from the bull stud contained approximately 0.8–12% (v/v) SP, which is in agreement with other reports (Hering et al. 2014; Bromfield 2016). The purpose of the present study was to investigate the impact of adding bovine SP from bulls of high fertility and low fertility to SP-free ejaculated sperm samples before freezing on bull sperm quality after thawing. Our hypothesis was that SP from low fertility bulls might have a different impact on spermatozoa than SP from high fertility bulls.

Results and discussion

The results of the preliminary study are shown in Table 1. The addition of SP1 decreased TM compared to SP510 and SP20, and there was a trend toward significance compared with SP5. Addition of SP1 decreased PM compared with SP20. There was a higher proportion of spermatozoa in velocity class D (immotile spermatozoa) for SP1 than for SP10 and SP20, and a tendency to be higher than SP5. Since adding any proportion of SP to the semen would decrease the ratio of spermatozoa to cryoprotectant and might have an adverse effect on survival during cryopreservation, it was deemed appropriate to use the lowest proportions of SP, i.e., SP1 and SP5, for the main experiment, since SP5 had tended to produce an effect on the motility of fresh sperm samples. Since this preliminary experiment was carried out at the bull station rather than in the laboratory at SLU, the motility analysis was done using the portable Qualisperm® motility analyzer rather than the SpermVision® motility analyzer in the laboratory at SLU that was used for the main experiment.

Table 1. Sperm motility and velocity after addition of bovine seminal plasma (n = 7).

Proportion of seminal plasma (%)	Total motility (%)	Progressive motility (%)	Velocity class A	Velocity class B	Velocity class C	Velocity class D
0	97 ± 1^ab	94 ± 2	46 ± 13	48 ± 5	3 ± 1	3 ± 1^ab
1	95 ± 1^a	90 ± 2^a	75 ± 13	46 ± 5	5 ± 1	5 ± 1^a
5	97 ± 1^b	93 ± 2	50 ± 13	43 ± 5	4 ± 1	3 ± 1^b
10	98 ± 1^b	94 ± 2	47 ± 13	47 ± 5	4 ± 1	2 ± 1^b
20	99 ± 1^b	96 ± 2^b	53 ± 13	43 ± 5	3 ± 1	2 ± 1^b

Different superscripts within a column denote significant differences. For total motility and also for velocity class D, SP1 was different to SP10 (p < 0.05) and SP20 (p < 0.01) and there was a trend toward significance for SP5 (p < 0.08). For progressive motility, SP1 was different to SP20 (p < 0.05).

Based on the preliminary results, each single layer centrifugation (SLC)-selected sperm suspension was divided into three 5-mL aliquots; bovine SP from low or high fertility bulls was added at 0%, 1%, and 5% of the final volume (v/v) (control, SP1, and SP5, respectively) in a crossover design (Figure 1). The kinematics of thawed sperm samples are shown in Table 2. The proportion of SP added had a significant effect on velocity average path (VAP) (p < 0.05), velocity curved line (VCL) (p < 0.01), linearity (LIN) (p < 0.05), wobble (WOB) (p < 0.01), and amplitude of lateral head displacement (ALH) (p < 0.05) but there was no interaction with bull fertility. The VAP was lowest for SP5, which was different to control (p < 0.05) but was not significantly different to SP1. The VCL and ALH were lowest for SP5, which were different to SP1 and control (p < 0.05 for each). Furthermore, LIN was highest for SP5, which was different to SP1 (p < 0.05) but was not different to the control. The WOB was highest for SP5, which was different to SP1 and control (p < 0.01 for each). In addition, the fertility of the bull (low or high) had a significant effect on velocity straight line (VSL) (p < 0.05), VAP (p < 0.05), and VCL (p < 0.05). However, there were no differences between proportion of SP and fertility of bull, or their interaction, for the other sperm kinematics (p > 0.05). Therefore, the data in Table 2 are only shown for SP1 and SP5, regardless of bull fertility.

Table 2. Sperm kinematics patterns (SpermVision® of sperm samples treated with different proportions of seminal plasma; control (0%) (n = 14), 1% of bovine SP (SP1; n = 17) and 5% of bovine SP (SP5; n = 14).

Sperm kinematic	Control	1SP	5SP
MOT (%)	51.6 ± 5.5	55.5 ± 5.2	47.5 ± 5.7
PMOT (%)	44.6 ± 5.3	52.1 ± 5.0	44.0 ± 5.5
VSL (μm/s)	60.5 ± 2.7	57.0 ± 2.5	52.9 ± 2.7
VAP (μm/s)	77.6 ± 2.9^a	73.6 ± 2.6^ab	65.5 ± 2.9^b
VCL (μm/s)	140.8 ± 6.0^a	135.9 ± 5.4^a	114.5 ± 6.0^b
LIN (%)	0.4 ± 0.0^ab	0.4 ± 0.0^a	0.5 ± 0.0^b
STR (%)	0.8 ± 0.0	0.8 ± 0.0	0.8 ± 0.0
WOB (%)	0.5 ± 0.0^a	0.5 ± 0.0^a	0.6 ± 0.0^b
ALH (μm)	5.0 ± 0.2^a	5.0 ± 0.3^a	4.0 ± 0.2^b
BCF (Hz)	27.2 ± 0.7	26.5 ± 0.7	26.9 ± 0.7
Hyper (%)	10.5 ± 2.8	13.6 ± 2.6	9.1 ± 3.0

Different superscript letters within a row indicate significant difference (p < 0.05).

MOT: total motility; PMOT: progressive motility; VSL: velocity straight line; VAP: velocity average path; VCL: velocity curved line; LIN: linearity; STR: straightness; WOB: wobble; ALH: amplitude of lateral head displacement; BCF: beat cross frequency; Hyper: hyperactive.

Figure 1. Experimental design.

Sperm kinematics can be an indication of in vivo fertility, although opinions differ as to which kinematics are predictive of fertility (Kathiravan et al. 2011; Michos et al. 2013). In one study, total motility, progressive motility, or beat cross frequency (BCF) were considered to predict fertility, whereas VCL was associated with low fertility (Oliveira et al. 2012). In cattle, increased VCL and ALH are indicative of hyperactivity, which may not be desirable in sperm doses for insemination but could be acceptable for IVF (Suarez and Ho 2003). Therefore, the decreased VAP reported in the present study following exposure to SP5 could be detrimental to fertility, although the decreased ALH and VCL, as well as increased WOB and LIN, might be beneficial in sperm samples for AI. These results differ from other studies, for example Maxwell et al. (1996) who showed that the proportions of live and motile spermatozoa were greater when SP was included in the collection medium. Nonetheless, their studies did not report an impact of fertility. Similarly, SP from bulls of known fertility increased the motility of epididymal sperm samples although the functionality of the treated spermatozoa was not affected in an IVF system (Holden et al., 2017). Previous studies from our own laboratory showed that adding 5% stallion SP increased motility in SP-free stallion sperm samples (Morrell et al., 2010).

The proportion of SP also had a significant effect on the proportion of spermatozoa with high mitochondrial membrane potential (MMP) (p < 0.01), but again there was no interaction with fertility of bulls (Table 3). The proportion of spermatozoa with high MMP was lowest when SP5 was added, compared to SP1 (p < 0.05) and control (p < 0.01). These results differ from those of White et al. (1987) who showed that the addition of SP to the thawed semen increased the oxygen uptake of spermatozoa. Likewise, Holden et al. (2017) were unable to find an effect of SP from bulls of known fertility on MMP. The decreased high MMP with SP5 in the present study could also be indicative of reduced fertilizing ability since high MMP is considered by some researchers to be related to better fertility in AI (Garner and Thomas 1999). However, others have not found such a link (Ericsson et al., 1993).

Table 3. Sperm quality after adding different proportion of seminal plasma; control (0%) (n = 14), SP1 (n = 17), and SP5 (n = 14).

Sperm parameter	Control (0%)	1SP	5SP
Living spermatozoa (%)	54.9 ± 3.3	52.0 ± 3.2	49.3 ± 3.4
Live superoxide negative (%)	35.8 ± 3.7	39.0 ± 3.3	37.6 ± 3.7
Live superoxide positive (%)	19.4 ± 2.0	20.5 ± 1.8	20.3 ± 2.1
Live hydrogen peroxide negative (%)	53.3 ± 4.9	58.5 ± 4.4	56.2 ± 4.9
Live hydrogen peroxide positive (%)	0.6 ± 0.2	0.3 ± 0.2	0.4 ± 0.2
%DFI (%)	2.7 ± 0.4	2.2 ± 0.4	2.5 ± 0.5
High MMP (%)	36.5 ± 5.7^a	33.3 ± 5.6^a	23.0 ± 5.9^b
Phosphorylated cell (%)	27.3 ± 6.4	40.7 ± 6.1	32.5 ± 6.7

%DFI: % DNA fragmentation index; High MMP: high mitochondrial membrane potential. Different superscript letters within a row indicate significant difference (p < 0.05).

The differences in these MMP results may be due to the use of different fluorochromes: Rhodamine 123 was used in the study by Ericsson et al. (1993) whereas JC-1 was used in the present study and also in the study by Garner and Thomas (1999). The study by Oliveira et al. (2012) linked high MMP to the increased pregnancy rates after AI, confirming the association between these factors. The main role of sperm mitochondria is energy production through oxidative phosphorylation (reviewed by Pena et al., 2009).

No significant differences were observed between the proportion of SP and the fertility of bull on membrane integrity (Table 3) and there was no interaction between the proportion of SP and the fertility of bull. Therefore, the results in Table 3 are shown only for SP1 and SP5, i.e., not divided according to the fertility of the donor. Similarly, SP from bulls of known fertility did not affect sperm membrane integrity in the study by Holden et al. (2017). However, in contrast, the proportions of live spermatozoa were greater when SP was included in the collection medium (Maxwell et al., 1996), and a positive effect SP on sperm viability was found when semen was diluted to a low number/dose (Garner et al., 2001).

Chromatin structure was also not affected by the addition of SP in the present study (Table 3), regardless of the fertility of the bull. Zinc-binding proteins, present in SP, provide a source of zinc ions that protect chromatin (Mogielnicka-Brzozowska et al., 2011), which would support the present findings. However, there may be species differences since a significant increase in chromatin damage was seen in stallion spermatozoa treated with SP and then stored for 24 h (Morrell and Johannisson, 2014).

No changes in production of reactive oxygen species (ROS) were seen in the present study following the addition of SP (Table 3). This protective action of SP could be due to antioxidant enzymes or spermadhesins present in SP that protect spermatozoa against oxidative stress and lipid peroxidation (Schoneck et al., 1996). Furthermore, adding SP did not affect the proportion of phosphorylated spermatozoa (Table 3), indicating that capacitation was not induced by SP in our study, whereas Manjunath and Thérien (2002) suggested that SP promoted capacitation of bovine spermatozoa. Similarly, Almadaly et al. (2017) observed that the addition of SP to bull spermatozoa increased PTP-induced intracellular cAMP. The different effects seen in these studies may be due to the differences in sperm preparation techniques used.

Our observations indicate that the effects on sperm quality were similar regardless of the fertility of the bulls from which the SP originated. Therefore, our hypothesis that SP from bulls of low fertility would have a different effect on spermatozoa than SP from bulls of high fertility was not upheld. These findings are in agreement with those observed by Holden et al. (2017) with epididymal spermatozoa, where they observed an effect from adding SP that was independent of the fertility of the bull from which the SP originated. It is possible that the differences in SP between the high and low fertility bulls in the present study were not sufficiently different for an effect to be observed. It would be interesting to expand the study to include SP from bulls with even higher or lower fertility index scores, and to determine whether SP from another bull has the same effect as the bull’s own SP. Other factors, such as breed of bull or components of SP such as enzymes, antioxidants, and hormones, can affect sperm function; their relationship to field fertility should also be investigated. However, adding bovine SP5 to SP-free sperm samples had a beneficial effect on sperm velocity, independent of bull fertility, and an unfavorable effect on high MMP. Other parameters of sperm quality were not adversely affected by adding SP5. No effect was observed from adding SP1. Furthermore, there were no differences in effect between SP from high or low fertility bulls. Since sperm motility may be related to sperm fertility, adding SP to semen doses before insemination may have a favorable impact on sperm fertility in inseminated cows.

Materials and methods

Bovine semen collection, preparation, and addition of bovine SP

Semen was collected at a commercial semen collection station (Viking Genetics, Skara, Sweden) from bulls kept according to standard husbandry procedures. Bulls were categorized as high or low fertility according to a fertility index used routinely by the breeding company, based on the 56-day nonreturn rate from a minimum of 1000 AIs. This index was adjusted to take into account factors, such as time of year of insemination, farm, age and parity of cow, and inseminator (Rodriguez-Martinez, 2003), based on a cohort study. The nonreturn rate data for each bull were then transformed into an index. A bull of average fertility will score 100 on this index; bulls of higher fertility in the cohort will have a score of >100, whereas lower fertility bulls will have a score of <100 (Hans Stålhammar, personal communication).

Ejaculates were available from Swedish Red and Holstein Friesian bulls of known fertility (high fertility n = 6 and low fertility n = 4). Ejaculates deemed to be acceptable for freezing according to the internal quality control standards of the company were used for the study. The semen was divided into two parts. The first part was used for the preparation of bovine SP by centrifuging at 1800 g for 10 min to pellet the spermatozoa (high fertility n = 4 and low fertility n = 3). The supernatant was removed and checked by light microscopy for spermatozoa. The centrifugation was repeated if spermatozoa were found in the SP. Meanwhile, the second part of the ejaculate (semen from 9 bulls) was extended in warm (35°C) egg yolk medium (20 ml egg yolk, Tris 2.422 g, glucose 1 g, citric acid 1.36 g, de-ionized water 80 ml; and chemicals from VWR, Stockholm, Sweden) to a sperm concentration of 50 × 10⁶ spermatozoa/mL. This suspension was used for SLC with a species-specific colloid, Bovicoll (formerly Androcoll-B) (patent applied for, J. M. Morrell), following the protocol for stallion semen (Morrell et al., 2009) with slight modifications: the extended sperm sample was layered over 15 mL of Bovicoll in a 50 mL conical centrifuge tube. The preparation was centrifuged at 300 × g for 20 min at room temperature. After centrifugation, the supernatant was aspirated and the sperm pellet was harvested from beneath the last 2 mL colloid and resuspended in cryopreservation extender (AndroMed®, Minitube GmbH, Germany). Sperm concentration was measured using the NucleoCounter® SP-100™ (ChemoMetic AS, Allerød, Denmark) as directed by the manufacturer, and adjusted to achieve a final concentration of 69 × 10⁶ spermatozoa/mL.

For a preliminary experiment to determine how much SP to add in the main experiment, the SLC sperm suspensions from seven bulls were each divided into five aliquots and SP was added at 0, 1%, 5%,10%, or 20% of the final volume (= treatments SP0, SP1, SP5, SP10, and SP20, respectively). Motility assessment was carried out using the Qualisperm motility analyzer as described under CASA.

Each SLC-selected sperm suspension was divided into three 5-mL aliquots; bovine SP from low or high fertility bulls was added at 0%, 1%, and 5% of the final volume (v/v) (control, SP1, and SP5, respectively) in a crossover design (Figure 1). The choice of 1% and 5% SP in this study was based on the results of the preliminary study. Finally, the sperm samples were cooled to 4°C and frozen by controlled rate freezing according to the company’s routine practice. The straws were stored in liquid nitrogen until required for analysis. One straw of each sample was thawed at 37°C for 12 s and used for the analyses shown in Figure 1.

Computer-assisted sperm analysis

Sperm motility analysis for a preliminary trial was performed on fresh sperm samples at the bull station with the QualiSperm® motility analyzer (Biophos, AG, Switzerland) attached to a microscope (Nikon, Eclipse E200, Tokyo, Japan) with a heated stage (38°C). The following measurements were made: total motility (TMQ, %), progressive motility (PMQ, %), velocity (μm/s), velocity class A (%, rapid progressive >50 μm/s), velocity class B (%, slow progressive >15 < 50 μm/s), velocity class C (%, slow progressive <15 μm/s), and velocity class D (%, nonprogressive 0 μm/s).

For the main experiment using thawed sperm samples, sperm motility was analyzed at the laboratory at SLU using a SpermVision® (Minitüb GmbH, Germany) connected to an Olympus BX 51 microscope (Olympus, Tokyo, Japan) with a heated stage (38°C) as described by Goodla et al. (2014). Aliquots (5 µL) of the sperm samples on a warm glass slide (38°C) were evaluated for total motility (MOT, %), progressive motility (PMOT, %), VSL (μm/s), VAP (μm/s), VCL (μm/s), LIN (VSL/VCL), straightness (STR, VSL/VAP), WOB (VAP/VCL), ALH (μm), BCF (Hz), and hyperactive (Hyper, %).

Plasma membrane integrity

Plasma membrane integrity was evaluated by flow cytometry using SYBR14 and PI staining according to published procedures (Johannisson et al., 2009), with slight adaption for bull samples (Goodla et al., 2014). The integrity of the plasma membrane was classified according to staining as living (%) (SYBR14-positive/PI-negative), dead (%) (SYBR14- negative/PI-positive), and dying (%) (SYBR14-positive/PI-positive). Only the results for membrane intact spermatozoa are reported here.

Reactive oxygen species

The procedure of Guthrie and Welch (2006) was adapted for bull semen (Goodla et al., 2014). In total, 50,000 spermatozoa were evaluated by flow cytometry and classified as follows: live, superoxide-negative; live, superoxide-positive; dead, superoxide-positive; live, H₂O₂-negative; live, H₂O₂-positive; dead, H₂O₂-negative; and dead, H₂O₂-positive. For the purposes of this study, only the results for living spermatozoa are reported.

Sperm chromatin structure

Sperm chromatin integrity was evaluated according to Evenson et al. (2002) with slight modifications (Goodla et al., 2014). In total, 10,000 spermatozoa were evaluated: the DNA fragmentation index (%DFI) was calculated as the ratio of cells with denatured, single-stranded DNA to total cells (both with stable, double-stranded DNA, and denatured single-stranded DNA); all the results were calculated using FCS Express version 2 (De Novo software, Glendale CA, USA).

Mitochondrial membrane potential

MMP was analyzed using the lipophilic cationic probe 5,5ʹ,6,6ʹ-tetrachloro-1,1ʹ,3,3ʹ-tetraethyl-benzimidazolylcarbocyanine iodide (JC-1; Invitrogen, Eugene, OR) according to Cossarizza et al. (1993) with slight modifications (Goodla et al., 2014). A total of 30,000 cells was analyzed and classified as spermatozoa with high respiratory activity (%) (Orange fluorescence) or low respiratory activity (%) (Green fluorescence).

Global protein tyrosine phosphorylation (PTP)

Bull sperm PTP was evaluated using flow cytometry according to Piehler et al. (2006) with modifications (Goodla et al., 2014). The proportion of phosphorylated cells (%) and the mean fluorescence intensity in 30,000 cells were evaluated using Cell Quest 3.3 software (Becton Dickinson).

Statistical analysis

The treatment means were compared for all analyses. In all response variables, the residuals were investigated for normality and homoscedasticity using diagnostic plots. The comparisons between treatments were performed using mixed statistical models (Olsson, 2011) of the SAS® (Proc Mixed, SAS® 9.3, Cary, NC, USA). The proportion of SP (0%, 1%, and 5%), fertility of bull (high or low fertility), and their interaction were used for the fixed part of the model to observe the response variable of the samples. The bull was considered as a random factor. Post hoc comparisons were adjusted for multiplicity using Tukey’s method. All the values are reported as least squares means ± standard error of the mean (LSMEAN ± SEM). A value of p < 0.05 was considered statistically significant.

Acknowledgments

We are grateful to the staff in the barn and laboratory at Viking Genetics, Skara, for supplying the ejaculates used in this study and for freezing all the samples. We are very grateful to Professor Ulf Olsson for his suggestions and comments on the statistical analyses. TN was funded by Mahasarakham University, Thailand. JMM and AJ were funded by the Swedish Research Council for the environment, agricultural sciences, and spatial planning (FORMAS), project number 221-2010-1241.

Disclosure statement

The authors declare that they have no competing interests. JMM is the inventor and patent holder of the colloid and buffer B used in the experiment. However, since the colloid was not the subject of the study, this is not considered to be a potential conflict of interest.

Additional information

Funding

TN was funded by Mahasarakham University, Thailand. JMM and AJ were funded by the Swedish Research Council for the environment, agricultural sciences, and spatial planning (FORMAS) [project number 221-2010-1241].

Notes on contributors

Thanapol Nongbua

Contributed to the conception and design of the experiment: JMM, AJ, TN; Performed the experiments: TN, AJ, JMM, EA-E; Analyzed the data: TN, AJ; Contributed materials and facilities: AE; Wrote the manuscript: TN, JMM. All authors approved revisions and the final paper.