C-phycocyanin protects against low fertility by inhibiting reactive oxygen species in aging mice
Part 1 of 2
Yan-Jiao Li1, Zhe Han1, Lei Ge1, Cheng-Jie Zhou1, Yue-Fang Zhao1, Dong-Hui Wang1, Jing Ren1, Xin-Xin Niu1 and Cheng-Guang Liang1
1The Key Laboratory of National Education Ministry for Mammalian Reproductive Biology and Biotechnology, The Research Center for Laboratory Animal Science, College of Life Science, Inner Mongolia University, Hohhot, Inner Mongolia, People’s Republic of China
Correspondence to: Cheng-Guang Liang, email: email@example.com
Keywords: ovarian aging, oocyte, oxidative stress, C-phycocyanin, D-galactose, Gerotarget
Women over 35 have higher rates of infertility, largely due to deterioration of oocyte quality characterized by fragmentation, abnormal meiotic spindle-chromosome complexes, and oxidative stress. C-phycocyanin (PC) is a biliprotein enriched in Spirulina platensis that is known to possess antioxidant, anti-inflammatory, and radical-scavenging properties. D-galactose-induced aging acceleration in mice has been extensively used to study aging mechanisms and for pharmaceutical screening. In this study, adult female B6D2F/1 mice injected with D-galactose were used as a model to test the age-reversing effects of C-phycocyanin (PC) on degenerated reproductive ability. Our results show that PC can prevent oocyte fragmentation and aneuploidy by maintaining cytoskeletal integrity. Moreover, PC can reverse the expression of antioxidant genes, increase superoxide dismutase (SOD) activity and decrease methane dicarboxylic aldehyde (MDA) content, and normalize mitochondria distribution. PC exerts its benefit by inhibiting reactive oxygen species (ROS) production, which decreases apoptosis. Finally, we observe a significant increase in litter size after C-phycocyanin (PC) administration to D-galactose-induced aging mice. Our study demonstrates for the first time that D-galactose-induced impaired female reproductive capability can be partially rescued by the antioxidant effects of C-phycocyanin (PC).
The human female reproductive system ages more rapidly than most other body systems, and reproductive capacity is negatively correlated with age [1, 2]. For a variety of reasons, many women postpone childbearing, and a considerable proportion of aged female become infertile . A reduction in quantity and deterioration in oocyte quality is universal in women over 40. Poor oocyte quality is characterized by meiotic spindle anomalies, chromosome misalignment, oxidative stress, gene expression changes, shortened telomeres, and loss of cohesion [1, 4-6]. Oxidative stress occurs due to gradual accumulation of damage by free radicals that are generated during normal metabolism, and it is considered one of the major mechanisms underlying aging [7, 8]. Germ cells initiate meiosis and arrest at the dictyate stage of prophase I in the fetal ovary; these postnatal germ cells remain arrested for weeks to months in mice and 10-50 years in humans . During this prolonged interval, reactive oxygen species (ROS) accumulate and decrease oocyte quality and quantity . ROS negatively affect processes from oocyte maturation to fertilization, embryo development, and pregnancy, and they are associated with the agerelated decline in reproduction [3, 11, 12]. Previous reports demonstrated that ROS accumulation in cells can lead to cytoskeletal derangement , shortened telomeres , impaired telomerase activity , antioxidant system dysfunction [16, 17], disturbances of ATP levels  and mitochondrial distribution , and cell apoptosis .
C-galactose (D-gal) is a reducing sugar that can form advanced glycation end products (AGEs) in vivo. Aging is accelerated after mice receive oral or subcutaneous D-gal [21-23]. Some studies focused on the effect of D-gal on oxidative stress, and one found that AGEs can cause the accumulation of ROS, especially superoxide radicals and hydrogen peroxide .
Reducing oxidative stress by antioxidant supplementation could potentially reduce ROS-induced damage, thus maintaining oocyte and follicle number and quality [24-27].
C-phycocyanin (C-Pc, PC) is a major biliprotein in Spirulina platensis that possesses antioxidant, neuroprotective, anti-inflammatory, and radical-scavenging properties [28-30], suggesting PC as a potential agent for preventing ROS-induced aging or ROS damage [31, 32]. However, little is known about whether PC can prevent D-gal-induced aging and impaired reproductive ability. In the present study, we treated mice with PC to investigate whether it could preserve reproductive performance in a D-gal-induced aging model.
C-phycocyanin reversed some organ coefficients in D-galinduced aging mice
We first examined macroscopic views of mouse organs in the control, D-gal, and D-gal+
C-phycocyanin groups and found no obvious morphologic changes (Figure 1A). We then compared organ weight and organ coefficients (organ weight/body weight) among the three groups. There were no statistical differences in the weights of the ovary, liver, spleen, or kidney. Similarly, there was no significant difference in the liver organ coefficient (Figure 1B, Table S1-1 and Table S1-2).
Figure 1: C-phycocyanin reversed the organ coefficients of the ovary, spleen, and kidney in D-gal-induced aging mice. A. Organ views. No obvious abnormal morphology was observed after D-gal or PC administration. Scale bar = 1 cm. L: left, R: right. B. Body weight and organ coefficients. No statistical difference was observed for ovary, liver, spleen, or kidney weight. However, the organ coefficients of the ovary, spleen, and kidney from the control and D-gal+C-phycocyanin groups were higher than those in the D-gal group. Data are presented as means ± SEMs and were processed by one-way ANOVA and Newman-Keuls post hoc tests. Significant differences between groups, *P < 0.05. n indicates the number of mice for each treatment.
Surprisingly, the organ coefficients of ovary, spleen, and kidney from control and D-gal+PC groups were higher than those of the D-gal group (P < 0.05, Figure 1B and Table S1-2). These data indicate that D-gal may impair the ovary, spleen, and kidney, and that this damage can be reversed partially by PC. We next examined the age-reversing effect of C-phycocyanin in the reproductive system of D-gal-induced aging mice.
C-phycocyanin rescued oocyte morphology and developmental competence in D-gal-induced aging mice
We assessed oocytes at the germinal vesicle (GV) and metaphase II (MII) stages generated from in vivo or in vitro maturation by examining their morphology with bright-field microscopy. We found that increased percentages of abnormal oocytes from the D-gal group matured both in vivo and in vitro. These oocytes were characterized by enlarged perivitelline spaces, fragmented or dark cytoplasm, or giant polar bodies, all of which are considered morphological abnormalities . Conversely, after PC administration most D-gal-induced oocytes showed normal morphology, similar to the control group (Figure 2A).
Figure 2: Impaired oocyte quality and developmental competence in D-gal-induced aging mice could be rescued by C-phycocyanin.
A. Morphology of oocytes at the GV stage, 14 hours after in vivo maturation and 14 hours after in vitro maturation. Arrows and triangles indicate morphologically abnormal oocytes after in vivo and in vitro maturation, respectively. Scale bar = 100 μm.
B. There were no significant differences in terms of oocyte numbers per mouse before or 14 hours after hCG injection in the control, D-gal, and D-gal+C-phycocyanin groups.
C. The percent of oocyte polar body extrusion was decreased in D-gal-induced aging mice. This inhibition was reversed by C-phycocyanin.
D. D-gal-induced aging increased the percentage of oocyte fragmentation. This was decreased by C-phycocyanin administration after in vivo or in vitro maturation.
E. D-gal-induced aging induced oocyte aneuploidy, which was decreased by PC after in vivo or in vitro maturation. Data are presented as the means ± SEMs and were processed by one-way ANOVA and Newman-Keuls post hoc tests. Significant differences between groups, *P < 0.05; **P < 0.01; ***P < 0.001. n indicates the number of mice for each treatment.
In superovulated mice, the number of GV-stage oocytes retrieved from ovaries in the control, D-gal, and D-gal+PC groups were not statistically different. Similarly, the numbers of MII -stage oocytes collected from oviduct ampullae were comparable among the three groups (Figure 2B and Table S2-1).
In terms of first polar body (PB1) extrusion from in vivo and in vitro matured oocytes, we observed significant differences. During in vivo maturation, D-gal severely decreased PB1 extrusion, resulting in a much lower frequency compared to that of the control (P < 0.001). Interestingly, PC could apparently reverse impaired PB1 extrusion to rates comparable to the control group (control vs D-gal+PC, P > 0.05; D-gal vs D-gal+PC, P < 0.001). Similarly, the percentage of PB1 extrusion from oocytes matured in vitro was decreased by D-gal compared to control (P < 0.001), and administration of PC improved PB1 extrusion (D-gal vs D-gal+PC, P < 0.01). However, treatment did not rescue to the level of the control (control vs D-gal+PC, P < 0.05) (Figure 2C and Table S2-2).
We also calculated the percentages of fragmented oocytes matured in vivo. We calculated that 13.08% of fragmented oocytes matured in the D-gal group, which was much higher than the control group (P < 0.001). The rate of fragmentation was decreased in the D-gal+PC group to the control group (control vs D-gal+PC, P > 0.05; D-gal vs D-gal+PC, P < 0.001). For oocytes matured in vitro, 36.57% were fragmented in the D-gal group, which was much higher than that of the control group (P < 0.001). This impairment was partially reversed by PC (P < 0.01), but not to the level of control (P < 0.001) (Figure 2D and Table S2-3).
As age-related infertility is associated with chromosome aneuploidy , we checked if D-galinduced aging could increase oocyte aneuploidy. For both the in vivo and in vitro maturation models, we found that the aneuploidy rate of sister chromatids was higher in the D-gal group compared to control (P < 0.001). After PC administration, decreased percentages of aneuploidy oocytes were observed both in vivo and in vitro (P < 0.001). However, PC administration could not normalize aneuploidy to the level of the control group (P < 0.001) (Figure 2E and Table S2-4). These results indicate that PC could reverse the deterioration of oocyte maturation in D-gal-induced aging mice.
C-phycocyanin rescued spindle-chromosome complex (SCC) malformation in D-gal-induced aging mice
PB1 extrusion failure and oocyte fragmentation in D-gal-induced aging mice may be associated with disordered spindle assembly and chromosome alignment. We observed defective spindles in the D-gal group. α-Tubulin labeling revealed that some oocytes had no spindle pole, while others had multiple spindle poles. Chromosome alignment in the D-gal group was also defective. The mid-plate, which appeared in control MII oocytes, was absent in the D-gal group, replaced by dispersed chromosomes in the mid-plate area. Notably, PC reversed all these malfunctions (Figure 3A).
Figure 3: C-phycocyanin rescued SCC malformation and cytoskeletal abnormalities in D-gal-induced aging mice.
A. D-gal could induce various abnormal SCCs including no or multipolar spindle poles and chromosome misalignment in MII oocytes. These abnormalities were reversed by PC. Blue, DNA; red, α-tubulin; scale bar = 20 μm.
B. Abnormal distribution of γ-tubulin in MII oocytes in D-gal-induced mice could be reversed by the administration of PC. Blue, DNA; red, γ-tubulin; scale bar = 20 μm.
C. Malformed oocyte SCCs in D-galinduced aging mice were rescued by C-phycocyanin. Data are presented as the means ± SEMs and were processed by one-way ANOVA and NewmanKeuls post hoc tests. Significant differences between groups, ***P < 0.001. n indicates the number of mice for each treatment.
We also examined SCC integrity by analyzing the distribution of γ-tubulin, which should be localized at the spindle pole area. In D-gal-treated mice, most oocytes lacked specific γ-tubulin localization at the spindle pole, but this was normalized by C-phycocyanin (Figure 3B).
We compared the frequency of abnormal SCC formation among the three groups. Almost half (43.32%) of oocytes generated from D-gal-treated mice had abnormal SCC (P < 0.001). This was halved by PC administration (P < 0.001), which was comparable to the control group (Figure 3C and Table S3).
Aging is associated with oocyte fragmentation and impaired PB1 extrusion. Abnormal SCC formation and translocation is the main reason for oocyte maturation failure and aneuploidy. These data suggest that PC may improve oocyte quality by correcting SCC formation and translocation, leading to the production of high-quality mature oocyte with a lower rate of aneuploidy.
D-gal and C-phycocyanin do not influence telomere length or telomerase activity
Since short telomeres are considered a biomarker of chronic oxidative stress and biological aging [35, 36], we measured telomere length and telomerase activity prior to and after D-gal and PC treatment. Telomere length (indicated by T/S ratio) were not significantly different among the three groups (Figure 4A and Table S4). Similarly, no significant difference was observed in terms of telomerase activity in the whole ovary (Figure 4B and Table S4). Telomere length is considered a useful biomarker in determining biological and chronological aging [37, 38]. Our results showed that neither telomere length nor the telomerase activity was altered, indicating that telomeres were not disturbed in D-gal-induced aging mice.
Figure 4: D-gal and C-phycocyanin did not affect telomere length or telomerase activity in mouse ovaries.
A. Relative telomere length shown as the T/S ratio determined by qPCR analysis. Administration of D-gal or C-phycocyanin had no effect.
B. Telomerase activity of ovaries assessed with by ELISA. Administration of D-gal or PC had no effect. Data are presented as the means ± SEMs and were processed by one-way ANOVA and Newman-Keuls post hoc tests. n indicates the number of mice for each treatment.
C-phycocyanin rescued antioxidant gene expression and antioxidant enzyme activity
As aging is highly correlated with the expression and activity of antioxidant genes and enzymes, we used quantitative real-time reverse transcription polymerase chain reaction (RT-PCR) to measure the ovarian mRNA levels of the antioxidant genes Gclm, Gclc, Gpx1, Gpx3, Gsr, Gsta4, Gstm1, Gstm2, Gstt1, Mgst1, Sod1, Sod2, Cat, Glrx1, Glrx2, Prdx3, Txn2, Txnrd1, Txnrd2, and copper chaperone for SOD (Ccs). D-gal significantly changed expression of some genes such as Gclc, Gpx3, and Cat, and PC reversed these alterations. However, D-gal changed the expression of some other genes such as Gclm, Glrx2, and Txn2, but these changes were not normalized by PC. Expression levels of most of the genes (Gpx1, Gsr, Gsta4, Gstm1, Gstm2, Gstt1, Sod1, Glrx1, Prdx3, Txnrd1, Txnrd2, and Ccs) were not affected by D-gal or PC treatments. Interestingly, the expression levels of Mgst1 and Sod2 were not changed after D-gal treatment but increased after PC administration (Figure 5A and Table S5-1).
Figure 5: C-phycocyanin rescued antioxidant genes expression and enzymes activity.
A. Relative expression levels of antioxidant genes in ovaries by qPCR. D-gal significantly increased Gclc and Gpx3 expression and decreased Cat expression, and PC reversed these alterations.
B. SOD activity in ovaries was decreased by D-gal and partially increased after PC administration.
C. GSH-Px activity in ovaries was not changed by D-gal or PC administration.
D. CAT activity in ovaries was increased by D-gal and PC administration.
E. MDA content in ovaries increased after D-gal treatment and decreased after C-phycocyanin administration. Data are presented as the means ± SEMs and were processed by one-way ANOVA and Newman-Keuls post hoc tests. Significant differences between groups, *P < 0.05; **P < 0.01; ***P < 0.001. n indicates the number of mice for each treatment.
We also measured antioxidant enzyme levels and activities. Superoxide dismutase (SOD) content was decreased after D-gal treatment (P < 0.001), and PC partially reversed this decrease (control vs D-gal, P < 0.001; control vs D-gal+PC, P < 0.001) (Figure 5B and Table S5-2). There was no significant difference in glutathione peroxidase (GSH-Px) activity among the three groups (Figure 5C and Table S5-3). Catalase (CAT) content was higher in the D-gal group than in the control group (P < 0.05), and this increase could not be reversed by PC administration (control vs D-gal+PC, P < 0.01) (Figure 5D and Table S5-4). Methane dicarboxylic aldehyde (MDA) content was also measured as a biomarker for oxidative stress and aging. MDA was higher in the D-gal group compared to control (P < 0.01), and this increase was inhibited by PC (P < 0.01) (Figure 5E and Table S5-5).
Consistent with the generally accepted view that antioxidant gene expression and antioxidant enzyme activity are involved in the aging process [39, 40], we found alterations in some genes and enzymes. Moreover, MDA was increased in the D-gal group and could be inhibited by C-phycocyanin. This led us to explore whether mitochondria and ROS are downstream of PC’s effects.
Aggregated mitochondrial distribution in D-galinduced oocytes was normalized by C-phycocyanin
Next, we examined ATP levels and mitochondrial distribution in MII oocytes. Relative ATP levels in cumulus cells were not significantly different in the D-gal or D-gal+PC groups compared to control (Figure 6A and Table S6-1). A similar result was observed in oocytes (Figure 6B and Table S6-1). However, we observed a significant difference when we assessed mitochondrial distribution in MII oocytes. The distribution of mitochondria was classified as “aggregated” or “even.” Only 5.76% of MII oocytes exhibited aggregated distribution in the control group. However, after D-gal administration, this proportion increased to 37.97%, which was significantly higher than control (P < 0.001). After PC administration, 88.95% of oocytes were evenly distributed around mitochondria, which was similar to the oocytes in control (Figure 6C and 6D, Table S6-2). Mitochondrial distribution is a cytoskeleton-dependent intracellular traffic behavior . The distribution of mitochondria in oocytes determines spindle translocation . Although ATP levels in the whole oocytes were not changed, abnormal distribution of mitochondria induced by D-gal and later recovered by PC may be associated with oocyte maturation. Furthermore, ROS accumulation can damage the cytoskeleton and affect mitochondrial distribution [13, 19], which led us to evaluate the ROS level in D-gal- or PC-treated oocytes.
Figure 6: Mitochondrial distribution but not ATP level was altered by D-gal and reversed by C-phycocyanin.
A. Relative ATP levels in cumulus cells and
B. relative ATP levels in MII oocytes were not affected by D-gal or C-phycocyanin administration. Data are presented as means ± SEMs.
C. Percentages of MII oocytes with altered mitochondrial distribution after D-gal injection and C-phycocyaninadministration. Column with large square, aggregated mitochondria distribution; column with small square, even mitochondria distribution. Data are presented as means.
D. Representative images of mitochondria with aggregated or even distribution. Data were processed by one-way ANOVA and NewmanKeuls post hoc tests. Significant differences between groups, ***P < 0.001. Scale bar = 20 μm. n indicates the number of mice for each treatment.
C-phycocyanin reduced high ROS levels induced by D-gal in MII oocytes
ROS levels can indicate oxidative stress. In oocytes, we measured intracellular ROS levels in in vivo matured MII oocytes. Dichlorofluorescein diacetate (DCFH-DA) fluorescence intensity was significantly higher in the MII oocytes of the D-gal-treated group than in the control or D-gal+PC groups, indicating enhanced ROS production after D-gal administration (Figure 7A). We quantified the relative fluorescence intensities and confirmed that ROS levels were much higher in the D-gal group (P < 0.001). Interestingly, when PC was administered, ROS levels decreased to a value comparable to the control group (Figure 7B and Table S7). These results suggest that D-gal significantly increases ROS generation in oocytes. Significantly, PC reduced ROS levels in the MII oocytes of D-gal-induced aging mice (P < 0.01).
Figure 7: C-phycocyanin inhibited ROS level in MII oocytes in D-gal-induced aging mice.
A. Representative images of ROS generation determined by DCFH-DA fluorescence (green). Scale bar = 50 μm.
B. ROS relative fluorescence intensity. ROS generation in oocytes was obviously increased after D-gal treatment; this was suppressed to the control group level after PC administration. Data are presented as the means ± SEMs and were processed by one-way ANOVA and Newman-Keuls post hoc tests. Significant differences between groups, **P < 0.01; ***P < 0.001. n indicates the number of mice for each treatment.
C-phycocyanin inhibited D-gal-induced early apoptosis in MII oocytes
High levels of ROS can induce apoptosis. We therefore performed annexin-V staining to determine whether the frequency of early stage apoptosis in oocytes was altered by D-gal and C-phycocyanin treatment. Oocytes undergoing early apoptosis were characterized by a clear green signal in the membrane and zona pellucida (Figure 8A). We quantified the fluorescence signals and found that 19.89% of oocytes in the D-gal group were apoptotic (P < 0.001). After PC administration, the percentage of apoptotic cells decreased (control vs D-gal+PC, P > 0.05; D-gal vs D-gal+PC, P < 0.001) (Figure 8B and Table S8). Oocyte apoptosis is always accompanied by abnormal morphology changes, which was verified by our previous results. These indicate that D-gal triggers apoptosis in MII oocytes, which is inhibited by PC.
Figure 8: C-phycocyanin inhibited early stage apoptosis in MII oocytes in D-gal-induced aging mice.
A. Representative images of early stage apoptosis in MII oocytes. Oocytes without green fluorescence signals at the zona pellucida and oocyte membrane were non-apoptotic, and oocytes undergoing early apoptosis were characterized by a clear green signal in the zona pellucida and membrane. Scale bar = 20 μm.
B. Percent oocytes undergoing early stage apoptosis. D-gal induced early stage apoptosis in oocytes, and this was inhibited by C-phycocyanin. Data are presented as the means ± SEMs and were processed by one-way ANOVA and Newman-Keuls post hoc tests. Significant differences between groups, ***P < 0.001. n indicates the number of mice for each treatment.
C-phycocyanin rescued litter size in D-gal-treated mice
Finally, the numbers of offspring in the three groups were evaluated. Each normal female in the control group delivered an average of 8.69 pups. The D-gal-induced aging female exhibited low reproductive ability compared with the control group (P < 0.05). Significantly, PC treatment increased the litter size in D-gal-induced aging mice to the level of the controls (control vs D-gal+PC, P > 0.05; D-gal vs D-gal+PC, P < 0.01) (Figure 9A and Table S9-1). Offspring birth weight was not significantly different among the three groups (Figure 9B and Table S92). Considering that aging mice have a high proportion of birth defects correlated with aneuploidy, we assessed the growth state of offspring in both female (Figure 9C and Table S9-3) and male (Figure 9D and Table S9-4) and detected no significant differences among the three groups. These results indicate that PC has no detectable side effects on the offspring of D-gal-induced aging mice.
Figure 9: C-phycocyanin treatment led to increase in the litter size of D-gal-induced aging mice but did not influence offspring birth weight or growth rate.
A. Litter sizes in the control, D-gal, and D-gal+PC groups.
B. Birth weights in the control, D-gal and D-gal+PC groups.
C. Postnatal growth of female from weeks 1 to 8 in the control, D-gal, and D-gal+PC groups.
D. Postnatal growth of male from weeks 1 to 8 in the control, D-gal, and D-gal+PC groups. Data are presented as the means ± SEMs for A. and B. and means for C. and D.. Data were processed by one-way ANOVA and Newman-Keuls post hoc tests. Significant differences between groups, *P < 0.05; **P < 0.01. N indicates the number of female mice with plugs and that gave birth for each treatment; n shows the total number of offspring for each treatment.