Proteomic analysis of the signaling pathway mediated by the heterotrimeric G? protein Pga1 of Penicillium chrysogenum
Pga1 signaling and penicillin production
Decreases of penicillin yields in strains lacking Pga1 activity and increases in strains expressing a constitutively active Pga1 have been previously reported [10]. Transketolase was less abundant in strain ?pga1 than in strains Wis54-1255 and PgaG42Rpyr-T (Figs. 3, 4). This enzyme is part of the pentose phosphate pathway, the main source of the reduced form of the NADP+ coenzyme (NADPH). Previous studies described that penicillin production requires high NADPH concentrations [27], and that transketolase and other enzymes of the pentose phosphate pathway are overexpressed in the high yield producer strain AS-P-78 [28]. The high demand for NADPH is due to its requirement for the biosynthesis of penicillin precursors, such as the amino acids valine and cysteine [29]. The low abundance of transketolase in strain ?pga1 would reduce the pentose phosphate pathway flux rate, affecting NADPH formation, which in turn would contribute to the low penicillin production in this strain. Transketolase abundance is not recovered by restoring normal cAMP levels in strain ?pga1 (i.e., transketolase regulation by Pga1 signaling is cAMP-independent), which agrees with the previously reported lack of effect of induced high cAMP levels on penicillin production [10].
Penicillin production also demands high levels of ATP [30], which is mainly due to the high ATP requirement of the enzyme ?-(l-?-aminoadipyl)-l-cysteinyl-d-valine synthetase, which catalyzes the first step of penicillin biosynthesis [31]. The ?pga1 strain is probably inefficient in ATP production in comparison to strains with a functional Pga1 due to the lower abundance of the F1-ATPase alpha-subunit Atp1 (see above), which would affect penicillin production in this strain. Interestingly, the inducer of penicillin biosynthesis 1,3-diaminopropane also causes an increase in the abundance of the F1-ATPase alpha-subunit Atp1 when added to cultures of the Wis54-1255 strain, as observed in a comparative proteomics study [32].
Amino acid metabolism is also affected by Pga1 signaling. Of particular relevance is the abundance of the sulfate adenylyltransferase (Pc20g07710), which is lower in strain ?pga1 (?1.7-fold), and even more so in condition ?pga1 + cAMP (?3.8-fold), than in the strains with a functional Pga1 (Figs. 3, 4). This enzyme is involved in the biosynthesis of cysteine, one of the three amino acids precursors of penicillin biosynthesis, which is a substrate of the ?-(l-?-aminoadipyl)-l-cysteinyl-d-valine synthetase and required in large quantities for efficient penicillin production [30]. In strain DS17690, a very high yield penicillin producer, the genes encoding enzymes of the cysteine biosynthetic pathway were upregulated [33], among them the sulfate adenylyltransferase gene. Similarly, the AS-P-78 strain, a high yield penicillin producer, showed higher abundance of proteins related to cysteine biosynthesis than the wild type NRRL 1951 and the Wis54-1255 strains [28]. The decreased expression of the sulfate adenylyltransferase in strain ?pga1 likely contributes to its lower penicillin production [10]. The induced increase of cAMP decreases the abundance of sulfate adenylyltransferase (Fig. 3), and interestingly cAMP concentration in penicillin non-producing conditions was found to be higher than in producing conditions in a controlled steady-state flux chemostat culture [27]. All evidence suggests that penicillin production is not regulated or is even negatively regulated by cAMP.
According to a metabolome study of a high yield industrial strain in penicillin producing and non-producing conditions [30] and to a proteomics study of low, intermediate and high yield penicillin producing strains [28], the main features of primary metabolism that have a major impact on penicillin production are: (1) high cysteine, but not valine or ?-aminoadipate, availability, (2) high NADPH supply, and (3) high ATP supply. In this study, we have observed that Pga1 has a role in all these processes, regulating the expression of proteins related to the biosynthesis of cysteine, NADPH and ATP. Therefore, we conclude that Pga1 signaling is an important regulator of the primary metabolism processes that lead to penicillin biosynthesis.
A dnaK-type molecular chaperone (Pc22g10220), similar to the S. cerevisiae Ssb2, is less abundant in strain ?pga1 (independently of cAMP) than in strains Wis54-1255 and PgaG42Rpyr-T. Interestingly, this protein has been shown to increase its abundance in the presence of 1,3-diaminopropane [32]. Therefore, this chaperone is more abundant in conditions that stimulate penicillin production, i.e. when there is a functional Pga1 and in the presence of 1,3-diaminopropane. The Ssb2 protein participates in glucose sensing in S. cerevisiae. Ssb2 is required to maintain the Snf1 protein kinase in a dephosphorylated, and thus inactive, form [34]. The Snf1 protein kinase allows expression of genes involved in the utilization of alternative carbon sources in the absence of glucose by inhibiting the action of the repressing complex Mig1/Ssn6/Tup1 [35], which is responsible for carbon catabolite repression. In P. chrysogenum the penicillin genes are subject to carbon catabolite repression which is exerted by the transcription factor CreA [36]. It will therefore be of great interest to study the possible involvement of Pc22g10220 in glucose sensing and carbon catabolite repression, and analyze if the regulation of the penicillin biosynthetic genes by Pga1 [10] is mediated by this chaperone and CreA.
A putative dienelactone hydrolase (Pc22g24530) is more abundant in strain PgaG42Rpyr-T than in strain Wis54-1255 (1.63-fold). This protein was also reported to be more abundant in the high yield penicillin producer AS-P-78 in comparison with lower producing strains [28]. Strains expressing a constitutively active Pga1 G? subunit (Pga1G42R) produce higher amounts of penicillin than strains with a wild type Pga1 [10]. The significance of these findings and the possible relation of dienelactone hydrolase to penicillin production are still unclear.