Abstract
Genes and organisms often evolve in response to a specific driving force, as new functions are fixed in response to selective pressures. The presence of self-resistance genes in antibiotic producers creates a conundrum- how can antibiotic resistance evolve in the absence of the antibiotic, and how can an antibiotic producer evolve in the absence of resistance? This project focuses on the selective pressure that mycophenolic acid (MPA), an uncompetitive NAD+ analog inhibitor of inosine monophosphate dehydrogenase (IMPDH), applies to the Penicillium fungi, some of which are MPA producers. IMPDH catalyzes the rate-limiting step of de novo guanine nucleotide biosynthesis, the conversion of inosine monophosphate (IMP) to xanthosine monophosphate through the co-reduction of NAD+ to NADH. Many Penicillium species contain two copies, “A” and “B”, of IMPDH whereas most Eurotiale family fungi have only one copy. Previous work characterized “A” copies from both an MPA producer and a non-producer. These enzymes are highly active and sensitive to MPA, as is typical for eukaryotic IMPDHs. The B copy gene exists within the MPA biosynthetic gene cluster (BGC) in the MPA producer and is the source of MPA self-resistance. The producer B copy was active with MPA resistance 1000-fold greater than A copies. However, the non-producer B copy was inactive. With significant evolutionary distance between known producers, as well as protein phylogenetic distance between MPA producer B copies, this research seeks to answer the following questions: How has MPA resistance and production evolved, are all non-producer B copies inactive, and why do non-producers retain B copies? As the number of sequenced fungal genomes has grown extensively since the initial report, an updated amino acid phylogeny of fungal IMPDHs was warranted to aid in studying the evolution of MPA self-resistance. In Chapter 2, genomic analysis of the self-resistance conferring B copy IMPDHs was conducted towards this end. Genomic searches of B copy possessing non-producers found that almost all had at least partial matches to MPA BGC genes near B copy genes, suggesting modern non-producers retained B copies from cluster-containing ancestors. After generating amino acid phylogenetic trees of available fungal IMPDHs, the B copy clade was found have a branching pattern distinct from the consensus pattern seen in the A copy clade and species tree from literature, indicating that the B copy has a different evolutionary history and suggesting it evolved under a different selective pressure. Two fungi from the Sordariomycetes class with the MPA BGC were identified. Closely related phylogenetics, synteny, sequence identity provide evidence that the MPA BGC of one of these species with an Aspergillus producer provided evidence of a horizontal gene transfer between the two. Horizontal gene transfers are thus a possible explanation for the divergent branching of the B copy IMPDH clade and the evolutionary history of the MPA BGC.
The evolution of drug self-resistance as drug production falls in and out of related species is a phenomenon that lacks study. In Chapter 3, I characterized activity and MPA resistance of modern and ancestor MPA producer and non-producer B copy IMPDHs. Modern MPA producer B copies were moderately active and MPA resistant, while non-producer B copies were weakly active and resistant. Ancestor B copies were active and resistant to MPA, supporting a hypothesis of direct descent of the MPA BGC from ancestors to modern producers. Modern and ancestor B copies were also found to be resistant to the synthetic competitive IMP analog inhibitor ribavirin monophosphate. Greater drug resistance was found to correlate with lower enzymatic efficiency toward substrates. B copy IMPDH fixation in MPA non-producing species may derive from the selective pressure of other natural product IMP analog inhibitors such as mizoribine and oxanosine that are produced by species in overlapping ecological niches with MPA producers.
A few non-producer B copies were very poor enzymes, questioning the fitness advantage of their existence. IMPDHs are square planar tetramers with active sites at the subunit interfaces. In Chapter 4, A-B copy heterotetramer formation was explored as a possible mechanism to activate the B copy. Dual bacterial expression of A and B copy IMPDHs in bacteria resulted in heterotetramers. The activity of the B copy increased ~10-fold in an A-B heterotetramer, which afforded significantly greater activity in the presence of MPA than either A or B copy alone, justifying the continued presence of the B copy. The results show the first example of activation of an inactive homolog with an active, drug-sensitive homolog for significant gains in effective drug resistance.
The structural determinants of B copy IMPDH drug resistance and heterocomplexing effects are not easily understood due to the conserved nature of the active site. In Chapter 5, the amino acid differences between A and B copy IMPDHs were examined in relation to their corresponding locations on existing IMPDH structures. Several positions of fixed differences between A and B copies involved in interaction networks at interfaces and in the active site were identified. These changes, along with differences in the active site flap, were also identified as possible determinants of activity and drug resistant differences. Additionally, the heterotetrameric activation of the non-producer B copy was speculated to come from compensation of a two-residue deletion in the active site flap.