Mitochondrial biogenesis implicates the generation of new healthy mitochondria to meet the requirement of biological energy and replenish damaged mitochondria (Popov, 2020). Mitochondrial biogenesis is an extremely intricate process including the synthesis of the mtDNA encoded proteins and the imports of nuclear encoded mitochondrial proteins and replication of mitochondrial DNA (mtDNA) (Li et al., 2017; Uittenbogaard and Chiaramello, 2014). Mitochondrial biogenesis is dependent on coordinated regulation of mitochondrial and nuclear factors (Ploumi et al., 2017). It has been reported that PGC-1α acts as a critical regulator of mitochondrial biogenesis via the transcriptional machinery to increase mitochondrial mass. Stressors (nutrient deprivation, hypoxia, oxidant stress, or exercise) activate PGC-1α activity and enhance its level, then inducing its location from the cytoplasm to the nucleus. Activated PGC-1α results in increase of NRF1 and NRF2 expression. Activation of NRF1 and NRF2 promotes the transcription of many mitochondrial genes involved in subunits mitochondrial respiratory chain complexes (Figure 2) (Wu et al., 1999). Simultaneously, NRF1 and NRF2 also stimulate the synthesis of TFAM, which subsequently mediates mtDNA replication and transcription (Figure 2) (Kelly and Scarpulla, 2004; Gleyzer et al., 2005; Vina et al., 2009). Finally, the PGC-1a—NRF1/2—TFAM pathway contributes to the formation of new mitochondria.

PGC-1α expression and activity are modulated by transcriptional and posttranslational levels (Scarpulla, 2008). Transcriptional regulation is the central approach to increase the total expression and activity of PGC-1α. Increase of cyclic adenosine monophosphate (cAMP) concentration activates protein kinase A (PKA), mediating phosphorylation of cAMP-response element-binding protein (CREB) at Ser 133 (Herzig et al., 2001). Ca²⁺ is interacted with Ca²⁺/calmodulin‐dependent protein kinase (CaMK), which induces phosphorylation of CREB (Mattson, 2012) Phosphorylation of CREB eventually results in enhancement of PGC-1α level (Figure 3) (Uittenbogaard and Chiaramello, 2014). Ca²⁺ also triggers the activation of calcineurin A (CnA). once activated, CnA interacts with myocyte enhancer factors 2C and 2D (MEF2C and MEF2D) and strongly drives PGC-1α expression (Figure 3) (Handschin et al., 2003; Stotland and Gottlieb, 2015). Besides, Ca²⁺ can contribute to the activation of AMPK, which increases expression of PGC-1α (Choi et al., 2016). Mammalian target of rapamycin (mTOR) can modulate yin yang 1(YY1)–PGC-1α interaction, which then mediates increases of PGC-1α promoter activity (Cunningham et al., 2007). On the contrary, TWEAK, an inflammation factor, induces activation of nuclear factor-κB (NF-κB). Then, NF-κB translates to nuclear along with recruitment of histone deacetylase (HDAC), which subsequently reduces histone acetylation, suppressing PGC-1α expression (Ruiz-Andres et al., 2016; Shi et al., 2013). In addition, Hes1 directly binding the PGC-1α promoter region (a downstream target of fibrotic Notch signaling) decreases PGC-1α levels (Figure 3) (Han et al., 2017). Moreover, transforming growth factorβ (TGFβ)—induced phosphorylation of Smad3 directly binding to PGC-1α promoter represses PGC-1α expression (Figure 3) (Yadav et al., 2011).

Posttranslational modifications including methylation, phosphorylation and deacetylation can regulate PGC-1α levels. PGC-1α is methylated by protein arginine methyltransferase1 (PRMT1) at arginine (Arg) 665, 667, and 669 (Figure 4), which induces the enhancement of PGC-1α activity, thus mediating the expression of essential target genes that are involved in mitochondrial biogenesis (Teyssier et al., 2005). Phosphorylated PGC-1α at threonine (Thr) 262, serine (Ser) 265, and Thr 298 by p38 mitogen-activated protein kinase (MAPK) (Figure 4) disrupts the interaction between PGC-1α and its inhibitor p160MBP, which increases its activity (Barger et al., 2001; Tang, 2016). Recent a study shows that inhibition of p38 MAPK markedly repressed the expression of PGC-1α (Ye et al., 2019). Activation of AMP-activated protein kinase (AMPK) induced by the elevated AMP/ATP ratio directly phosphorylates PGC-1α at Thr 177 and Ser 538 Jager et al., 2007. Furthermore, this phosphorylation can increase the occupancy of PGC-1α at the promoters of its target genes. In addition, AMPK increases nicotinamide adenine dinucleotide (NAD⁺) levels, thus enhancing Sirtuin1 (SIRT1) activity, which results in activation of PGC-1α by deacetylation (Figure 4) (Canto et al., 2009; Nemoto et al., 2005). Conversely, when NAD⁺ intracellular concentrations decrease, general control of amino acid synthesis 5 (GCN5) acetylates PGC-1α with a decrease in its transcriptional activation (Dominy et al., 2010; Kelly et al., 2009; Gerhart-Hines et al., 2007). Inhibition of PGC-1α can occur via phosphorylation by AKT at Ser 570 Li et al., 2007, S6 kinase at Ser 568 and Ser 572 Lustig et al., 2011, or glycogen synthase kinase 3β (GSK3β) at Thr 295 (Figure 4) (Anderson et al., 2008).

Mitochondria are highly dynamic organelles that constantly undergo mitochondrial fusion and division. Fusion and fission are both regulated by members of the dynamin-related protein (DRP) family, including MFN1 and 2, optic atrophy 1 (OPA1), and DRP1. These proteins comprise a large self-assembling GTPases (Praefcke and McMahon, 2004). Fusion of outer mitochondrial membrane (OMM) requires MFN1 and MFN2 to promote fusion of adjacent organelles via GTP hydrolysis. In contrast, the fusion of the inner mitochondrial membranes is regulated by the inner membrane protein, OPA1 (Liesa et al., 2009). The mitochondrial fission requires DRP1, which is localized in the cytosol. Upon the recruitment by receptor proteins (mitochondrial fission factor (MFF), fission protein 1(FIS1), mitochondrial dynamics proteins of 49 and 51 kDa (MiD49/51), DRP1 transfers to OMM. DRP1 and receptors form an oligomeric complex that results in constricting to garrote the organelle (Figure 5) (Smirnova et al., 2001).

PGC-1α is involved in the regulation of mitochondrial dynamics through the control of expression of core genes. It has been reported that exercise drives the enhancement of expression of MFN1 andMFN2, as well as PGC-1α and its coactivators of mitochondrial biogenesis, estrogen-related receptor α (Cartoni et al., 2005). Furthermore, the upregulation of MFN1 and MFN2 expression is related to PGC-1α in muscle cells. PGC-1α stimulated the transcriptional activity of the Mfn2 promoter, which was mediated by the endogenous ERRα, thus indicating that PGC-1α and ERRα play a synergic role in increasing Mfn2 mRNA (Soriano et al., 2006). Moreover, MitoQ treatment upregulates MFN2 expression in vitro. However, when PGC-1α was knockdown by siPGC-1α, MFN2 levels did not markedly change treated with MitoQ (Xi et al., 2018). These results demonstrate that PGC-1α leads to the transcriptional upregulation of Mfn2 mediated by MitoQ. Consistent with Vitro results, MFN1, MFN2and DRP1expression are significantly reduced in muscle deletion of PGC-1α ^−/− in mice compared to wild type (WT) (Zechner et al., 2010). Analysis results of mitochondrial morphology by electron microscopy reveals small, fragmented, and thin mitochondria with largely different in sizes and a reduction in mitochondrial density in PGC-1α ^−/- mice compared with WT (Zechner et al., 2010). In accordance, recent studies also reveal that mitochondria are higher fragmented in the muscle of PGC-1α ^−/- young mice compared with young WT (Halling et al., 2017; Halling et al., 2019). Accumulating evidence shows that PGC-1α directly interacts with the promoter of the Drp1 gene to regulate DRP1 levels. DRP1 alteration leads to mitochondrial fission (Ding et al., 2018; Lei et al., 2021). A recent study shows that upregulation of PGC-1α can increase expression of MFN2 and OPA1 and decrease expression of DRP1 and FIS1, which mediates the balance fusion and fission (Sui et al., 2021). Together, PGC-1α serves as an important modulator of mitochondrial fusion and fission via mainly regulating MFN1, MFN2, and DRP1, which keeps mitochondrial network balance.

Mitophagy selectively eliminates superfluous and damaged mitochondria to maintain mitochondrial homeostasis (Pickles et al., 2018). Increasing evidence demonstrates that PINK1/PARKIN pathway is the most critical ubiquitination-dependent mitophagy pathway (Youle and Narendra, 2011). Upon depolarization of mitochondrial membrane potential, PINK1 accumulates at the OMM. Furthermore, it recruits the E3 ubiquitin ligase PARKINand phosphorylates PARKIN at Ser65 Lazarou et al., 2015. Activation of PARKINpolyubiquitinates mitochondrial proteins, which are then recognized by autophagy receptors (optineurin (OPTN), p62 [or SQSTM1), NDP52 and neighbor of BRCA1 gene 1(NBR1)] (Lazarou et al., 2015). Then, these complexes bind to Microtubule-associated proteins 1A/1B light chain 3 (LC3) to form the autophagosome, which fuses with the lysosome, resulting in degradation of the mitochondria (Figure 5) (Lazarou et al., 2015). In addition to classical PINK1/PARKIN—related mitophagy, other mitophagy receptors have been reported to involve in mitophagy, including FUN14 domain-containing protein 1(FUNDC1), and Nip3-like protein X (NIX)/BCL-2/adenovirus 19-kd interacting protein 3(BNIP3), AMBRA1, Bcl-2-like protein 13 (Bcl2-L-13), FKBP8, and prohibitin2 (PHB2) (Bhujabal et al., 2017; Fivenson et al., 2017; Wei et al., 2017; Lampert et al., 2019). These receptors can directly interact with LC3-II and induce mitophagy.

Emerging evidence has shown that PGC-1α is also an essential element in the regulation of mitophagy. The PGC-1α may potentially mediate PINK1 transcriptional activity, which then increases PINK1 levels (Choi et al., 2014; Zhang et al., 2019). PGC-1α can also indirectly elevates activation of the PINK1-PARKIN pathway via the ERRα-SIRT3 pathway, thus mediating the degradation of damaged mitochondria (Ziviani and Whitworth, 2010). Recently, a study shows that PGC-1α activates NRF1 to promote mitochondrial biogenesis. Activation of NRF1 also binds to the classic consensus site (‐186/‐176) in the promoter of FUNDC1 to enhance its expression. FUNDC1 interacted with LC3 induces autophagic flux (Liu et al., 2021). Vivo studies demonstrate overexpression of PGC-1α in mice elevates autophagy flux. Similarly, it has detected thatthe expression of BNIP3, LC3II, and Beclin1 is upregulated and p62 is downregulated in muscle-specific PGC-1α transgenic mice than WT (Lira et al., 2013; Greene et al., 2015). This sign indicates increased basal autophagy flux. Yet another study finds that the downregulation of PGC-1α can upregulate BNIP3 in chondrocytes, ultimately inducing clearance of damaged mitochondrial (Kim et al., 2021). This distinct result might attribute to the different genetic background. Taken together, PINK1-Parkin dependent or independent mitophagy pathway is under control by PGC-1α.

A study found that the protein level of PGC-1α was unchanged in heart failure patients compared to normal donors (Hu et al., 2011). However, different study groups detected downregulation of PGC-1α in heart and in serum (Garnier et al., 2009; Chen et al., 2019; Joseph, 2019). Xu’s group further found that levels of PGC-1α is low along with low left ventricular ejection fraction (LVEF). This indicates that serum PGC-1α is associated with left ventricular ejection fraction (LVEF) in patients with HF (Chen et al., 2019). In accordance with heart failure patients, the levels of PGC-1α in animal models also is diverse from different studies. Multiple reports illustrate that PGC-1α expression is reduced after transverse aortic constriction (TAC), which inhibits the expression of mitochondrial genes and causes important deficiencies in cardiac energy reserves and function (Lehman and Kelly, 2002; Arany et al., 2006; Lu et al., 2010; Piquereau et al., 2017). Yet some studies observed that myocardial PGC-1α level was not decreased in the mice following pressure overload-induced heart failure (Hu et al., 2008; Bhat et al., 2019). Recently, Wang and his coworker observed the expression of PGC-1α at multiple time points after TAC. The result showed that the expression of PGC-1α initially increased 5 days after TAC, but the expression of PGC-1α began to reduce 14 days after TAC (Wang et al., 2018). This finding suggests that the expression of PGC-1α is fluctuant in the development and progression of HF. This also explains the various outcomes of these different studies regarding PGC-1α expression in heart failure. A large body of evidence demonstrates that the mice developed cardiac hypertrophy at 7 days and heart failure at 28 days under overload pressure. PGC-1α, as a primary mitochondrial biogenesis regulator, is almost coincident with this condition, indicating that PGC-1α-mediated MQC play an important role in the pathogenesis of HF.

Genetic ablation mice are used to explore the function of PGC-1α (Table 1). A report from Kelly’s group exhibited normal chamber sizes and ventricular function in deletion of PGC-1α in mice at ages 4∼6 months (Leone et al., 2005). In accordance, Chen and colleagues showed that PGC-1α ^−/− mice did not reveal significant differences in cardiac phenotype and heart ratio under basal conditions compared with WT (Lu et al., 2010). The Spiegelman group displayed that heart structure was normal and mitochondrial biogenesis was not impaired in PGC-1α ^−/− mice. Whereas, the heart contractile function was significantly deficient in PGC-1α lacking mice compared to WT (Arany et al., 2005). Compared with systemic PGC-1α knockout mice, cardiac-specific PGC-1α knockout mice have more severe impairment in cardiac function. PGC-1α reduction represses mitochondrial biogenesis which induces the inhibition of mitophagy further imparing MQC. The mitochondrial content is further alleviated by the accumulation of damaged mitochondria. The Oka group observed that cardiac deletion of PGC-1α in mice resulted in enlargement of left ventricular diameters accompanied by cardiac systolic dysfunction (Bhat et al., 2019). Gene expression analyses unveiled increased expression of HF markers atrial natriuretic peptide (ANP), B-type natriuretic peptide (BNP), which suggests PGC-1α ^−/− mice might develop HF (Bhat et al., 2019). Another study also supports this conclusion and found that ejection fraction (EF%∼29%) significantly depressed and left ventricular (LV) volume increased (Karkkainen et al., 2019). PGC-1α deficiency in cardiomyocytes leads to compromised metabolism as well as reduced mitochondrial function. Furthermore, mice lacking PGC-1α developed cardiomyopathy at 17 weeks and premature death occurred at 25 weeks (Karkkainen et al., 2019).

Together, these studies indicate that PGC-1α deficiency, affects MQC and metabolism, ultimately leads to the development and progression of HF under basal conditions. Moreover, PGC-1α ^−/− mice develop more profound cardiac dysfunction and clinical heart failure under stressful stimuli such as TAC than WT (Arany et al., 2006; Lu et al., 2010; Bhat et al., 2019). After TAC, PGC-1α ^−/− mice showed a higher ratio of heart weight to body weight and increase of LV fibrosis compare with sham, which is a sigh of heart failure. Meanwhile contractile performance was aberrant and mortality rate are high. Taken together, PGC-1α expression is an essential factor in maintaining normal heart function.

PGC-1α ^−/- mice from different group almost show abnormal cardiac baseline phenotype. Its cardiac function was more worsened in response to pressure overload stimulation than sham. Therefore, it suggests that the enhancement of PGC-1α might serve as a therapeutic strategy (Table 1). Cardiomyocyte-specific overexpression of PGC-1α contributed to a significant mitochondrial proliferation (Lehman et al., 2000). Uncontrolled mitochondrial proliferation replaced the sarcomeric assembly, which impaired the sarcomeric structure. These transgenic mice also showed increase of heart size, enlargement of four-chamber consistent with a dilated cardiomyopathy and severe decrease of global contractile function. Finally, all transgenic mice died at 6 weeks (Lehman et al., 2000). Another study found that doxycycline (DOX) - induced PGC-1α expression in adult mouse hearts also elevated mitochondrial biogenesis, but the mitochondrial ultrastructure appeared abnormal such as vacuoles. (Russell et al., 2004). PGC-1α knock in mice occurred cardiac hypertrophy and biventricular dilatation. Echocardiograms revealed repression of ventricular function. These alterations can reverse by removing DOX or cessation of PGC-1α overexpression.

These findings manifest that excessive PGC-1α expression does not exert a therapeutic role but facilitates the development of heart failure. Recently, several study groups generated a transgenic (TG) mouse model of moderate overexpression of PGC-1α(∼3-fold) in the heart, whose cardiac function at baseline was not altered (Karamanlidis et al., 2014; Pereira et al., 2014; Zhu et al., 2019). Moderate PGC-1α expression maintained mitochondrial biogenesis, mitophagy and cardiac homeostasis during aging. Nevertheless, a moderate level of PGC-1α overexpression did not preserve cardiac function during pressure overload. Directly excessive increase in PGC-1α can contribute to various changes, including dramatic enhancement of mitochondrial numbers, enlargement of heart chambers and impairment of cardiac function. Fine-tuning the expression of PGC-1α can maintain cardiac homeostasis, but the degree of increase of PGC-1α is not sufficient to protect the heart from overload pressure. Thus, it is necessary to consider the dose and period of enhancement of PGC-1α so that this strategy can achieve an optimal effect.