In silico pathway reconstruction: Iron-sulfur cluster biogenesis in Saccharomyces cerevisiae

Background Current advances in genomics, proteomics and other areas of molecular biology make the identification and reconstruction of novel pathways an emerging area of great interest. One such class of pathways is involved in the biogenesis of Iron-Sulfur Clusters (ISC). Results Our goal is the development of a new approach based on the use and combination of mathematical, theoretical and computational methods to identify the topology of a target network. In this approach, mathematical models play a central role for the evaluation of the alternative network structures that arise from literature data-mining, phylogenetic profiling, structural methods, and human curation. As a test case, we reconstruct the topology of the reaction and regulatory network for the mitochondrial ISC biogenesis pathway in S. cerevisiae. Predictions regarding how proteins act in ISC biogenesis are validated by comparison with published experimental results. For example, the predicted role of Arh1 and Yah1 and some of the interactions we predict for Grx5 both matches experimental evidence. A putative role for frataxin in directly regulating mitochondrial iron import is discarded from our analysis, which agrees with also published experimental results. Additionally, we propose a number of experiments for testing other predictions and further improve the identification of the network structure. Conclusion We propose and apply an iterative in silico procedure for predictive reconstruction of the network topology of metabolic pathways. The procedure combines structural bioinformatics tools and mathematical modeling techniques that allow the reconstruction of biochemical networks. Using the Iron Sulfur cluster biogenesis in S. cerevisiae as a test case we indicate how this procedure can be used to analyze and validate the network model against experimental results. Critical evaluation of the obtained results through this procedure allows devising new wet lab experiments to confirm its predictions or provide alternative explanations for further improving the models.


Supplementary Appendix
Generating a curated pathway model using expert knowledge After an analysis of the relevant literature, the following detailed information was considered to build the networks shown in Figure 2. Bold letters shown parenthetically identify which part of Figure 2 is described: Iron is imported into the mitochondrial matrix. Once at the mitochondrial matrix, iron is loaded directly into the enzyme ferrochelatase [1,2] and is directed for heme synthesis and ISC biogenesis. Some results show that, in vitro, Fe can also be stored by frataxin (Yfh1) for posterior usage [3,4]. The in vivo relevance of this is disputed. To a first approximation one can simplify the reaction network by assuming that a mitochondrial iron pool exists and that the iron from this pool is used for both heme A synthesis and for ISC synthesis. We also consider a production flux that accounts for iron import into that mitochondrial pool and a sink flux that accounts for any other usage or export of iron from the mitochondria. According to the available information, possible alternative roles for Yfh1 in Fe processing are: a) Regulation of Fe import and usage (I) and b) Regulation of ISC synthesis (S), transfer (T) and repair (R) through the regulation of Fe supply [5,6].
The affinity of the clusters for the scaffolds (and thus the transfer of the ISC) is modulated by the reduction state of the cluster [22]. Herein lays another possible role for Arh1-Yah1 in ISC biogenesis. These proteins could provide electrons to regulate the process of ISC transfer (T) to Apo-proteins.

iv)
If the ISC remains on the scaffold proteins, it can be transformed into a 4Fe-4S ISC [8,23]. Roles for Yfh1, Nfs1, and Arh1-Yah1 in this additional ISC synthesis (S) step are similar to those described before.

v)
Although we know of no direct evidence for this, one can not rule out the possibility that the 4Fe-4S ISC can be transferred directly to 4Fe-4S apoproteins, represented in Figure 2 by Apo P2. Arh1-Yah1 could have a role in providing electrons to regulate cluster affinity and transfer (T).

vi)
There is a natural turnover of ISC, both in scaffold proteins and in the ISC proteins, for example due to oxidative stress [e. g. [24,25]]. Nfs1 homologues are able to repair (R) damaged ISC directly in situ [23,26]. A role for Arh1-Yah1 in providing electrons to facilitate this repair (R) is possible.

vii)
Finally, once assembled in the scaffold proteins, the ISC can be transferred to the cytoplasm [27,28], most likely through Atm1.
viii) Grx5 is a monothyolic reductase that catalyzes the reaction GSSG PSH GSH SSG P + ↔ + − [29]. Because ISC are coordinated between cysteine residues, glutathionylation of such residues would prevent formation of ISC and thus disturb normal ISC biogenesis and ISC dependent protein activity.
Thus, Grx5 could be active in regulating the glutathionylation state of cysteine residues in Arh1, Yah1, scaffold or Nfs1 proteins. Grx5 could also be involved in regulating the formation/destruction of disulfide bridges in these proteins [29]. These two modes of action are lumped into one by defining for each ISC assembly protein an inactivated pool that can be reactivated by Grx5.

ix)
There is a possibility that Grx5 protein-protein disulfide bridge reducing activity could act upon such bridges formed between different ISC proteins.
There have been reports that, in the absence of iron on the scaffold dimers, such a bridge forms between Isu and Nfs1 homologues, leading to a dead end complex between the two proteins [30][31][32]. Grx5 could be active in reducing these bridges and returning both proteins for active duty in ISC assembly.
x) The HSP70-type protein chaperone Ssq1 is important for proper folding of the proteins involved in ISC biogenesis pathway and for the proper functioning of the pathway. Mge1 is used by Ssq1 as a nucleotides exchanging factor. Jac1 is a HSP40-type co-chaperone homologue that is important for the appropriate functioning of Ssq1 in the ISC biogenesis pathway. Ssq1, Mge1 and Jac1 work together and are activated by Isu 4 homologues [7,[33][34][35][36][37][38][39], which suggests that these proteins may be involved in: a) stabilizing (St) ISC assembly in the scaffolds until a productive transfer occurs [7,40], b) in initial folding (F) of scaffold proteins or other ISC proteins or c) in both processes.

Derivation of the GMA Model a) The power-law approximation:
If the flux of a given process is regulated by species X 1 , X 2 , …, X n but the functional form of the rate expression is unknown one can write the following equation to describe its rate.
( ) In general, the exact form of F is unknown. If one assumes that this function is a rational function [7,[40][41][42], one can write the same equation in logarithmic space and approximate that function using a Taylor series. Then, by truncating the series in the first order term and returning to a Cartesian space, the unknown form of the function can be approximated by a power-law function [43,44]: Where γ j is an apparent rate constant and f j,k an apparent kinetic order: In the case of the ISC biogenesis, the reproduction of experimental results requires that the concentration of several dependent variables decreases steadily until it ultimately is zero, for example when reproducing a Grx5 knock-out cell line. Many of the rate expressions approach zero as the concentrations of different species decrease, which suggest that the Generalized Mass Action (GMA) representation is more adequate for our models. By using such a representation, we ensure that as one protein is knocked out of the model it will affect only the process it is supposedly involved in.

b) Reactions that constitute a base model for ISC biogenesis
The reactions used in the model, together with rate expressions are shown in Supplementary Table 1. The assemble model and differential equations are given as supplementary material in an SBML file. This model is obtained in the following way.
First, the mass balance equations for each internal variable are written. For instance, in the case of the pool of scaffold proteins (ISS in Figure 2), we have the following processes:  Another simplifying assumption that we made was that while scanning for the role of one protein, the kinetic orders for the remaining proteins remained constant. Thus, for example when studying the role of Arh1-Yah1, the kinetic orders that regard the role of   15 15  36 36 1 Heme_Fe -Heme molecules with iron.
b Species in parenthesis and brackets in the equations are modifiers and are not represented in the flux diagram because they contribute to the catalysis of the reaction but are neither produced nor consumed in the reaction Supplementary