Surrogate Genetics and Metabolic Pro ling for ... · deﬁning challenge of human genetics for the foreseeable future: interpreting the impact of variations in the sequences of individual

INVESTIGATION

Surrogate Genetics and Metabolic Profilingfor Characterization of Human Disease Alleles

Jacob A. Mayfield,*,1 Meara W. Davies,*,2 Dago Dimster-Denk,* Nick Pleskac,† Sean McCarthy,*

Elizabeth A. Boydston,*,3,4 Logan Fink,*,3 Xin Xin Lin,*,3 Ankur S. Narain,*,3,5

Michael Meighan,‡ and Jasper Rine*,6

*Department of Molecular and Cell Biology, California Institute of Quantitative Biosciences, University of California, Berkeley,California 94720, †Berkeley High School, Berkeley, California 94704, and ‡University of California, Berkeley, Department of

Molecular and Cell Biology, Berkeley, California 94720

ABSTRACT Cystathionine-b-synthase (CBS) deficiency is a human genetic disease causing homocystinuria, thrombosis, mental re-tardation, and a suite of other devastating manifestations. Early detection coupled with dietary modification greatly reduces pathology,but the response to treatment differs with the allele of CBS. A better understanding of the relationship between allelic variants andprotein function will improve both diagnosis and treatment. To this end, we tested the function of 84 CBS alleles previously sequencedfrom patients with homocystinuria by ortholog replacement in Saccharomyces cerevisiae. Within this clinically associated set, 15% ofvariant alleles were indistinguishable from the predominant CBS allele in function, suggesting enzymatic activity was retained. Anadditional 37% of the alleles were partially functional or could be rescued by cofactor supplementation in the growth medium. Thislarge class included alleles rescued by elevated levels of the cofactor vitamin B6, but also alleles rescued by elevated heme, a secondCBS cofactor. Measurement of the metabolite levels in CBS-substituted yeast grown with different B6 levels using LC–MS revealedchanges in metabolism that propagated beyond the substrate and product of CBS. Production of the critical antioxidant glutathionethrough the CBS pathway was greatly decreased when CBS function was restricted through genetic, cofactor, or substrate restriction,a metabolic consequence with implications for treatment.

THE first complete human genome sequence seeded thedefining challenge of human genetics for the foreseeable

future: interpreting the impact of variations in the sequences

of individual human genomes. Comparative genome se-quencing reveals an average of one single-nucleotide changeper 1200 bp between any two individuals. In the absence ofstrong Mendelian inheritance and linkage, confirming thatany human genotype actually caused a phenotype is a signif-icant challenge given the approximately 3 million geneticvariants per person. Indeed, 4000 traits of medical interestshow evidence for inheritance but lack a clear determinant(Online Mendelian Inheritance in Man 2012). Next-generationsequencing within small pedigrees (Ng et al. 2010a,b; Fanet al. 2011), or a more narrowly defined clinical phenotype(Schubert et al. 1997), can sometimes disentangle the un-derlying contribution of a gene to disease. In this work wehave taken an approach that complements both increasedsequencing capacity and expanded phenotypic description.We used surrogate genetics to assay directly the function ofallelic variants and then evaluate their potential contributionto phenotypes of clinical importance.

Homocystinuria, elevated levels of the sulfur-containingmetabolite homocystine in the urine, illustrates several

Copyright © 2012 by the Genetics Society of Americadoi: 10.1534/genetics.111.137471Manuscript received December 5, 2011; accepted for publication January 9, 2012Supporting information is available online at http://www.genetics.org/content/suppl/2012/01/20/genetics.111.137471.DC1.Reference numbers for publicly available data; GenBank: L14577.1 (CBS); dbSNP:rs17849313 (A69P), rs2229413 (P70L), rs11700812 (R369P); SGD: YGR155W(CYS4) and YDR232W (HEM1).1Present address: Department of Veterinary and Animal Science, 470 IntegratedSciences Bldg., University of Massachusetts, Amherst, MA 01002.

2Present address: Department of Genome Sciences, Foege Bldg. S-250, Box355065, 3720 15th Ave NE, University of Washington, Seattle, WA 98195-5065.

3These authors contributed equally to this work.4Present address: Department of Cellular and Molecular Pharmacology, 403B ByersHall, Howard Hughes Medical Institute, California Institute of QuantitativeBiosciences, University of California, San Francisco, CA 94158.

5Present address: National Institute of Child Health and Human Development, Bldg.6, Rm. 2A01, 6 Center Dr. 2753, Laboratory of Molecular Growth Regulation,National Institutes of Health, Bethesda MD 20892-275.

6Corresponding author: Department of Molecular and Cell Biology, 374A Stanley Hall,California Institute of Quantitative Biosciences, University of California, Berkeley, CA94720. E-mail: [email protected]

Genetics, Vol. 190, 1309–1323 April 2012 1309

http://www.genetics.org/content/suppl/2012/01/20/genetics.111.137471.DC1


mailto:[email protected]

challenges inherent to elucidating the molecular bases ofhuman genetic diseases. Worldwide, 1 in 335,000 individ-uals are affected (Mudd et al. 1995), but the frequencyapproaches 1 in 1800 in certain populations (Gan-Schreieret al. 2010). A few well-characterized alleles of the geneencoding cystathionine b-synthase (CBS) correlate with di-sease symptoms, providing an appealing molecular mechanism.The enzyme CBS converts homocysteine to cystathionine inthe cysteine biosynthesis pathway (Supporting Information,Figure S1). In people with homocystinuria, free homocys-teine accumulates and can covalently bind to proteins or ox-idize to the dimer homocystine. Disease indicators includehomocystinuria or hyperhomocysteinemia, an abnormallyhigh concentration of serum total homocysteine, the sum offree, oxidized, and protein-bound forms.

CBS catalyzes a committed step in the pathway that producescysteine and ultimately glutathione, the major endogenous in-tracellular antioxidant. Upstream of CBS, homocysteine is an in-termediate in the pathway that recycles S-adenosylmethionine(AdoMet), the major methyl donor in the cell, back to methi-onine. The wide range of symptoms may reflect the fact thatCBS and its variants have the potential to alter regulatory me-thylation of DNA and histones, as well as the redox state ofthe cell. Yet, elevated homocysteine levels occur in manypeople, including heterozygotes for some CBS alleles, withoutany clinical symptoms (Motulsky 1996; Guttormsen et al.2001). Additionally, defects in several different genes tangen-tial to cysteine biosynthesis, such as MTHFR, can lead tohomocysteinemia and similar symptoms (Frosst et al. 1995;Gaughan et al. 2001; Pare et al. 2009). Hence, elevatedhomocysteine level is a convenient marker for a metabolicimbalance, but the cause and consequences may be elusive.

The genetic contributions are complex, but because earlymedical intervention, including a diet low in protein andmethionine, successfully alleviates many homocystinuriasymptoms, neonatal screening is widespread (Mudd et al.2001). Vitamin supplementation can replace dietary restric-tion as a therapy in a highly allele-dependent manner. CBSuses a vitamin B6 cofactor to form cystathionine by thecondensation of serine and homocysteine. Hence, elevatedB6 is thought to partially compensate for vitamin-responsivealleles with a lower affinity for the B6 cofactor (Chen et al.2006). Human CBS also forms multimers, coordinates hemewith a bound iron, and contains a regulatory domain thatbinds the metabolite AdoMet as a possible regulatory mech-anism (Shan and Kruger 1998; Meier et al. 2001; Christopheret al. 2002; Scott et al. 2004; Chen et al. 2006; Sen andBanerjee 2007). These features suggest control points forenzyme regulation and function, or targets for nutritionaland pharmaceutical therapies, that CBS alleles may affectdifferently.

Directed sequencing efforts of patients afflicted withhomocystinuria have produced a large catalog of alleles(Kraus et al. 2012), with both common and rare alleles(Mudd et al. 1985; Kraus 1994; Gallagher et al. 1995,1998). However, clinical association does not guarantee cau-

sality. In many cases, the sequenced alleles are further ana-lyzed by genetic or biochemical means, providing most ofour knowledge of CBS deficiency. Despite these heroicefforts, the piecemeal identification of alleles, variations inassessment strategies, diploid nature of the human ge-nome, and increasing numbers of rare alleles all lead touncharacterized alleles that may cause subtle, but impor-tant, differences in phenotype. As ever more CBS alleles arefound, the need for reliable measures of allele impact willincrease. CYS4 is the Saccharomyces cerevisiae ortholog ofCBS and has the same function in yeast as in humans (Onoet al. 1988). Although yeast Cys4p lacks a heme bindingdomain and may differ in details of its biochemical regula-tion, human CBS complements cys4 yeast for cysteine andglutathione production (Kruger and Cox 1994, 1995). Fur-thermore, nonfunctional or B6-remedial CBS alleles reca-pitulate their human phenotypes in yeast cys4 mutants (Kimet al. 1997; Shan and Kruger 1998). We took advantage ofthe foundation built by previous, elegant cross-species com-plementation experiments (Kruger and Cox 1994, 1995) todevelop a quantitative, comprehensive, and direct test of howvariation in a single human disease gene correlated with dis-ease and treatment via nutritional supplementation.

Materials and Methods

Plasmids

The plasmid pHUCBS was the kind gift of Warren Krugerand served as the template for generating alternative CBSalleles using the QuikChange II Kit (Agilent). We selectedsingle-base pair missense mutations from the CBS MutationDatabase (Kraus et al. 1999, 2012), from published litera-ture, and from the RefSeq database for A69P (rs17849313),P70L (rs2229413), and R369P (rs11700812). We verifiedthe sequence of the entire open reading frame of each allele(Table 1). The pHUCBS plasmid and all subsequent clonescontain a single, silent base pair change (909C . T) relativeto the RefSeq sequence for CBS (L14577.1). A BstEII/FseIfragment containing CBS variants was subcloned betweenthe S. cerevisiae TEF1 promoter and CYC1 terminator inpJR2983, a CEN–ARS URA3 shuttle plasmid.

Strains

All S. cerevisiae strains serving as a host for a human CBSallele contained a complete deletion of CYS4 (MATa cys4D::KanMX his3D1 leu2D0 lys2D0 ura3D0, JRY9292) derivedfrom the yeast knockout collection (Winzeler et al. 1999).A hem1 cys4 strain was created by disruption of HEM1 withLEU2 (hem1Δ::LEU2). CBS transformants were selected byuracil prototrophy.

Growth assays

Strains containing CBS plasmids were maintained on com-plete synthetic medium lacking uracil (CSM-Ura) and sup-plemented with glutathione, a stable source of cysteine. cys4complementation was assayed by growth on solid CSM-Ura

1310 J. A. Mayfield et al.

http://www.genetics.org/cgi/data/genetics.111.137471/DC1/1


http://www.yeastgenome.org/cgi-bin/locus.fpl?dbid=S000003387




















medium without glutathione and with 400 ng/ml vitaminB6 (pyridoxine–HCl). CBS alleles that complemented cys4were further characterized in a quantitative growth assayusing a minimal liquid medium made with yeast nitrogenbase lacking vitamin B6 or other vitamins and amino acids(MP Biomedicals). Vitamins (biotin, pantothenate, inositol,niacin, p-aminobenzoic acid, riboflavin, and thiamin) wereincluded at standard concentrations, and vitamin B6 wassupplemented at six different concentrations: 0, 0.5, 1, 2,4, and 400 ng/ml B6. Histidine, leucine, and lysine wereadded to relieve auxotrophies in the parent strain, and me-thionine was included in all minimal media. Growth rateassays used 250 ml volumes and started with cells at OD600 =0.002, inoculated from cells pregrown in minimal mediumthat lacked B6 and contained glutathione. The pregrowthmedium in the hem1 experiments contained 50 mg/mld-aminolevulinc acid (d-ALA) and the cells were washed twicewith minimal medium to prevent carry-over. For the sake ofclarity, we refer to supplementation with the soluble hemeprecursor d-ALA as “heme supplementation” in the text. Hemeis sparingly soluble and d-ALA supplementation was moreefficient. The optical density (A600 nm) was measured every30 min for 96 hr at 28� in a stationary microplate reader(Molecular Devices VersaMax). We accounted for settling ofthe cells over time by resuspending the cells after the finalkinetic read and measuring the OD600 value. Data were thennormalized using the time-weighted ratio of the endpointkinetic OD600 value to the resuspended OD600 value, accordingto the formula,

�Endpoint kinetic ODResuspended OD

21�

time pointfinal time point

þ 1;

and were log10 transformed. Due to the stationary-phaseceiling, growth rate better described the growth of allelesthan end-point measurement. Growth rate was calculated asthe slope of the regression line for data values betweenOD600 = 0.05 and OD600 = 0.1. Data were compared tothe major allele grown in the same medium on the sameplate. Outliers were detected using Grubb’s test and re-moved from growth rate calculations. The raw growth ratedata are available as File S1.

Immunoblots for CBS protein quantification

The total protein concentration of boiled, NaOH-extractedyeast pellets was measured using the Pierce BCA Protein AssayKit (Thermo Scientific) to normalize sample concentrations. Pro-teins were visualized on an Odyssey Infrared Imager (Li-CORBioscience) after separation on a denaturing gel. Mouseanti-CBS polyclonal antibody (Abnova H00000875-A01), arabbit anti-3-phosphoglycerate kinase (PGK) antibody (a giftfrom Jeremy Thorner, University of California Berkeley) andan anti-proliferating cell nuclear antigen (PCNA) antibody(Abcam ab70472) were used to detect target proteins.

Metabolite measurements

Cells were cultured in liquid minimal medium that con-tained glutathione before washing with, and inoculationinto, minimal medium lacking glutathione. Equal numbersof cells from log-phase cultures were harvested 12 hr afterinoculation. Metabolite extraction combined previously de-scribed methods (Canelas et al. 2008; Boer et al. 2010;Godat et al. 2010) as follows: 8.0 · 108 (G307S data set)or 1.9 · 109 (V320A data set) cells were pelleted by centri-fugation at 3200 · g. The cell pellets were resuspended with9.5 ml of their spent medium supernatant, then quenchedwith 20 ml 280� methanol. The cells were pelleted at 4000· g at 210� in a rotor (Sorvall SS-34), prechilled to 280�,and then resuspended with 1.0 ml of 4� extraction solvent[0.1% perchloric acid with 400 mM glycine-1-13C,15N(Sigma 299340) and 20 mM isotopically labeled methioni-ne-13C5,15N (Sigma 608106)]. The samples were boiled for5 min, and cell debris and precipitated proteins were re-moved by centrifugation for 2 min at 4000 · g in a 4� micro-fuge. The supernatants were diluted 1:4 in 0.1% perchloricacid and 0.1% formic acid. Liquid chromatography–massspectrometry (LC–MS) analysis used 20-ml injection vol-umes. Chromatographic separation (2.1 · 250 mm, 5 mmDiscovery HS-F5 column; Supelco) used a water-to-acetoni-trile gradient (Godat et al. 2010) and was followed by de-tection on an LTQ-Orbitrap XL hybrid mass spectrometerequipped with an IonMax electrospray ionization source(Thermo Fisher Scientific, Waltham, MA). For the G307Sdata set, a fourfold dilution series of a mixture of 17 metab-olite standards was added to a pooled cell extract that con-tained equal volumes from each experimental sample, andwas then used for metabolite identification and calibration.A full calibration panel was included in the V320A experi-ment, but was not added to a pooled standard. LC–MS datawere converted to centroids and the mzXML file format us-ing ReAdW 4.3.1 (Deutsch et al. 2010) with an Xcaliburlibrary (ThermoFisher Scientifics, v. 2.0.7). Peak processingused the BioConductor package XCMS (Smith et al. 2006;Tautenhahn et al. 2008); processed data are available as FileS2. Metabolites were identified using the pooled calibrationstandards and the Human Metabolome Database (Wishartet al. 2009) for the G307S study and by exact mass only forthe V320A analysis. Zeros in the data were imputed usinglocal minima, data were normalized using upper quartiles,and intensities were log transformed for analysis using R(scripts and centroided data files are provided as File S3).

Results

The surrogate assessment of clinically associated CBSalleles in Saccharomyces cerevisiae

We selected all alleles of CBS documented prior to 2011 thatcould be generated by a single base-pair change and thataffected an amino acid (Table 1). Each human CBS allelewas synthesized, inserted into a yeast plasmid, and individually

Surrogate Genetics of Human CBS Mutations 1311



http://www.genetics.org/content/vol0/issue2012/images/data/genetics.111.137471/DC1/FileS1.zip





Table 1 CBS alleles tested for function in yeast

pJR

Mutation

Protein cDNAInitial citation and clinical characterization,

or RefSeq number

pJR3044 H65R 194A . G Janosik et al. (2001b)pJR3045 A69P 205G . C rs17849313pJR3046 P70L 209C . T rs2229413pJR3047 P78R 233C . G de Franchis et al. (1994)pJR3048 G85R 253G . C Maclean et al. (2002)pJR3049 T87N 260C . A Kraus et al. (2012)pJR3050 P88S 262C . T Sebastio et al. (1995)pJR3051 L101P 302T . C Gallagher et al. (1998)pJR3052 K102Q 304A . C Kozich et al. (1997)pJR3053 K102N 306G . C de Franchis et al. (1994)pJR3054 C109R 325T . C Gaustadnes et al. (2002)pJR3055 A114V 341C . T Kozich et al. (1993)pJR3056 G116R 346G . A Sperandeo et al. (1996)pJR3057 R121C 361C . T Katsushima et al. (2006)pJR3058 R121H 362G . A Bermudez et al. (2006); Katsushima et al. (2006)pJR3059 R121L 362G . T Kraus et al. (2012)pJR3060 M126V 376A . G de Franchis et al. (1999)pJR3061 E128D 384G . T Coude et al. (1998)pJR3062 E131D 393G . C Marble et al. (1994)pJR3063 G139R 415G . A Shih et al. (1995)pJR3064 I143M 429C . G Orendae et al. (2004)pJR3065 E144K 430G . A Shih et al. (1995)pJR3066 P145L 434C . T Kozich et al. (1993)pJR3067 G148R 442G . C Orendae et al. (2004)pJR3068 G151R 451G . A Kraus et al. (2012)pJR3069 I152M 456C . G Kraus et al. (2012)pJR3070 G153R 457G . C Kraus et al. (2012)pJR3071 L154Q 461T . A Lee et al. (2005)pJR3072 A155T 463G . A Janosik et al. (2001b)pJR3073 A155V 464C . T Lee et al. (2005)pJR3074 A158V 473C . T Shan and Kruger (1998)pJR3075 C165Y 494G . A Kluijtmans et al. (1995)pJR3076 V168M 502G . A Kraus et al. (2012)pJR3077 E176K 526G . A Kozich et al. (1997)pJR3078 V180A 539T . C Kluijtmans et al. (1999)pJR3079 T191M 572C . T Urreizti et al. (2003)pJR3080 D198V 593A . T Kraus et al. (2012)pJR3081 R224H 671G . A Kruger and Cox (1995)pJR3082 A226T 676G . A Kruger et al. (2003)pJR3083 N228S 683A . G Kruger et al. (2003)pJR3084 N228K 684C . A Gallagher et al. (1998)pJR3085 D234N 700G . A De Lucca and Casique (2004)pJR3086 E239K 715G . A de Franchis et al. (1994)pJR3087 T257M 770C . T Sebastio et al. (1995)pJR3088 G259S 775G . A Kraus et al. (2012)pJR3089 T262M 785C . T Kim et al. (1997)pJR3090 R266G 796A . G Katsushima et al. (2006)pJR3091 R266K 797G . A Kim et al. (1997)pJR3092 C275Y 824G . A Urreizti et al. (2003)pJR3093 I278T 833T . C Kozich and Kraus (1992)pJR3094 A288T 862G . A Lee et al. (2005)pJR3095 A288P 862G . C Linnebank et al. (2004)pJR3096 P290L 869C . T De Lucca and Casique (2004)pJR3097 E302K 904G . A Sperandeo et al. (1996)pJR3098 G307S 919G . A Hu et al. (1993)pJR3099 V320A 959T . C Kim et al. (1997)pJR3100 A331E 992C . A Dawson et al. (1997)pJR3101 A331V 992C . T Kruger and Cox (1995)

(continued)


transformed into a cys4 yeast strain lacking the CBS orthologCYS4. Centromere-based vectors were used to reduce copy-number variation. Eighty-one alleles derived from patientswith homocystinuria plus three additional variants found inpublic databases were assayed. This collection of 84 mis-sense mutations included alterations in the heme-binding,catalytic, and AdoMet-binding regulatory domains of theCBS protein. This strain could not grow on minimal media,but the defect in cysteine biosynthesis was bypassed by theaddition of cysteine or the more stable downstream metab-olite glutathione. Critically, the endogenous CYS4 gene sup-ports more robust growth than any CBS allele, suggestingthat CBS function was rate-limiting in yeast. Therefore, allassays necessarily compared CBS alleles to the major alleleof human CBS (major allele, MA), not to CYS4 (Figure 1 andFigure S2).

We discriminated functional from nonfunctional CBSalleles by plating cells onto media containing or lackingglutathione: only CBS alleles that restored CYS4 functionsupported growth on either medium. Of the 84 alleles, 46required glutathione supplementation to support growth,indicating severe loss of function (listed as “nonfunctional”in Table 2). Disease alleles often encode misfolded proteins(Yue et al. 2005), and there is precedence for lower proteinlevels among some nonfunctional CBS alleles due to aggre-gation (Katsushima et al. 2006) or degradation (de Franchiset al. 1994; Urreizti et al. 2006; Singh et al. 2007, 2010).Nonetheless, while we inferred misfolding or aggregation of

some CBS proteins, degradation may differ in yeast andhuman cells, perhaps because an appropriate E3 ligase ismissing. We observed ample steady-state levels of the CBSprotein encoded by 17/17 different human CBS alleles, rep-resentative of different growth classes, regardless of B6availability, as determined by immunoblot (Figure 2, FigureS3, and Table 2). Hence, our data measured the effect ofmutations on the intrinsic functions of the enzyme withoutcomplication from protein turnover.

Although many of the alleles tested were identified inindividuals with clinically significant homocysteinemia, 38CBS alleles were capable of supporting growth on mediumlacking glutathione and hence retained substantial function(alleles that are not listed as nonfunctional in Table 2). These

Table 1, continued

pJR

Mutation

Protein cDNAInitial citation and clinical characterization,

or RefSeq number

pJR3102 R336H 1007G . A Coude et al. (1998)pJR3103 G347S 1039G . A Gaustadnes et al. (2002)pJR3104 S349N 1046G . A Urreizti et al. (2003)pJR3105 S352N 1055G . A Dawson et al. (1997)pJR3106 T353M 1058C . T Dawson et al. (1997)pJR3107 V354M 1060G . A Coude et al. (1998)pJR3108 A355P 1063G . C Gallagher et al. (1998)pJR3109 A361T 1081G . A Castro et al. (1999)pJR3110 R369C 1105C . T Kim et al. (1997)pJR3111 R369H 1106G . A Kraus et al. (2012)pJR3112 R369P 1106G . C rs11700812pJR3113 C370Y 1109G . A Tsai et al. (1997)pJR3114 V371M 1111G . A Kluijtmans et al. (1999)pJR3115 D376N 1126G . A Kruger et al. (2003)pJR3116 R379W 1135C . T Linnebank et al. (2004)pJR3117 K384E 1150A . G Aral et al. (1997)pJR3118 K384N 1152G . T Kraus et al. (2012)pJR3119 M391I 1173G . A Kraus et al. (2012)pJR3120 P422L 1265C . T Maclean et al. (2002)pJR3121 T434N 1301C . A Kraus et al. (2012)pJR3122 I435T 1304T . C Maclean et al. (2002)pJR3123 R439Q 1316G . A Dawson et al. (1997); Tsai et al. (1997)pJR3124 D444N 1330G . A Kluijtmans et al. (1996)pJR3125 L456P 1367T . C Urreizti et al. (2003)pJR3126 S466L 1397C . T Janosik et al. (2001a)pJR3127 Q526K 1572C . A Kruger et al. (2003)

Figure 1 Growth of CBS-complemented yeast on solid media. Culturesgrown to saturation in liquid minimal medium containing glutathione andlacking B6 were plated in a fivefold dilution series onto solid medium 6glutathione. Growth was imaged after 3 days at 30�. The growth of themajor allele and representative alleles of the nonfunctional and B6-responsiveclasses are shown.










Table 2 Summary of clinical and yeast phenotypes

Mutation Clinical Response to B6 Enzyme activity Yeast phenotype and remediation

H65R Nonvariable Lowa Nonfunctionalo

A69P NA Similar to major alleleP70L NA Intermediate growth, B6

remedial, hem1 rescueP78R Variable Meda,b Similar to major alleleG85R Partial response Lowc High growth, B6 and heme remedialT87N ND Low growtho

P88S ND Low growtho

L101P Conflicting Lowd NonfunctionalK102N Variable Meda,b Similar to major alleleo

K102Q ND Similar to major alleleC109R Conflicting Lowd Nonfunctionalo

A114V Conflicting Mede/higha Low growthG116R Variable NonfunctionalR121C ND NonfunctionalR121L ND NonfunctionalR121H Nonvariable NonfunctionalM126V Nonvariable Nonfunctionalo

E128D Nonvariable Similar to major alleleE131D Nonvariable Lowf Low growtho

G139R Variable Intermediate growtho

I143M Nonvariable Lowg NonfunctionalE144K Conflicting Lowa,h Nonfunctionalo

P145L Nonvariable Lowi Nonfunctionalo

G148R Nonvariable Lowa NonfunctionalG151R ND Nonfunctionalo

I152M Conflicting Lowj hem1 rescueG153R NA NonfunctionalL154Q NA Lowk NonfunctionalA155T Conflicting Low growth, B6 and heme remedialA155V NA Lowk hem1 rescueA158V NA Low growthC165Y Conflicting Lowa,j NonfunctionalV168M ND Nonfunctionalo

E176K Nonvariable Lowa hem1 rescueV180A Variable Meda,j Intermediate growth, B6 remedialT191M Nonvariable Lowa,j NonfunctionalD198V Nonvariable Low growtho

R224H ND Low growthA226T Variable Medl Low growtho

N228S Nonvariable NonfunctionalN228K Nonvariable Lowa,c NonfunctionalD234N Nonvariable Intermediate growth, heme remedialE239K Variable NonfunctionalT257M Nonvariable NonfunctionalG259S ND NonfunctionalT262M Nonvariable Low growth, B6 remedialR266G Nonvariable NonfunctionalR266K Variable Meda High growth, B6 and heme remedialC275Y Nonvariable Lowl NonfunctionalI278T Conflicting Lowa,m Low growthA288P NA NonfunctionalA288T NA Lowk NonfunctionalP290L Variable hem1 rescueE302K Conflicting Lowe/higha NonfunctionalG307S Conflicting Lowa,n Nonfunctionalo

V320A Conflicting Intermediate growth, B6 andheme remedial

A331E ND Lowh NonfunctionalA331V ND Low growthR336H Variable Lowl hem1 rescue; not tested in

liquid media

(continued)


alleles were further assayed in liquid medium at varying con-centrations of vitamin B6 to expand the qualitative phenotypeto a quantitative assessment of function and B6 responsive-ness. All CBS alleles grew poorly without B6 supplementation(Table S1). Although S. cerevisiae is a B6 prototroph, theendogenous B6 level was insufficient to support the B6 re-quirement of human CBS. However, cells with the major CBSallele grew relatively well in medium supplemented with aslittle as 1 ng/ml of B6 (Figure S2 and Figure 3A).

When compared to cells with the major allele, the growthphenotypes of cells with other CBS alleles varied greatly(Figure 3B and Table S1). Hierarchical clustering by growthrate under all conditions was used to describe allele behav-ior. This nonbiased method separated alleles into roughlythree bins, in addition to the nonfunctional bin defined

above (Figure 4A). Cells with 14 different CBS alleles hadevidence of some function but grew poorly even at highlevels (400 ng/ml) of B6 (listed as “low growth” in Table2). Cells with 7 alleles showed an intermediate phenotypewith growth rates between that of cells with the major alleleand cells with poorly functioning CBS (listed as “intermedi-ate growth” in Table 2B). Cells with each of the remaining17 alleles grew at rates similar to those of cells with thepredominant allele (listed as “high growth” or “similar tomajor allele” in Table 2). Ten alleles, spanning all growthclasses, shifted from a lower growth-rate class to a highergrowth-rate class at 400 ng/ml B6 (listed as “B6 remedial”in Table 2). All functional alleles benefitted from increasedB6 concentrations; however, cells with these 10 alleles wereespecially sensitive.

Table 2, continued

Mutation Clinical Response to B6 Enzyme activity Yeast phenotype and remediation

G347S Variable Lowd NonfunctionalS349N Nonvariable Lowl NonfunctionalS352N Nonvariable Low growthT353M Conflicting Lowh Low growth, B6 and heme remedialV354M Nonvariable High growth, B6 remedial

A355P ND NonfunctionalA361T Nonresponsive NonfunctionalR369P NA NonfunctionalR369H ND Similar to major alleleR369C Responsive Lowj/meda Similar to major alleleC370Y Responsive NonfunctionalV371M Partial response Lowj Similar to major alleleD376N ND NonfunctionalR379W ND hem1 rescueK384E Responsive Nonfunctionalo

K384N Nonresponsive hem1 rescueM391I Nonresponsive NonfunctionalP422L B6 nonresponsive Higha,c Similar to major alleleT434N B6 responsive High growth, B6 remedialI435T ND Higha,c Similar to major alleleR439Q Conflicting Meda Similar to major alleleD444N Conflicting Higha/medl Similar to major alleleL456P B6 nonresponsive Lowl Intermediate growthS466L ND Higha,c Similar to major alleleQ526K B6 nonresponsive Intermediate growth, hem1 rescueo

Alleles with yeast growth phenotypes inconsistent with clinical (underlined) or in vitro activity (italics) are indicated. Alleles identified in the clinic but without informationabout B6 response (ND, no data) and alleles identified from available sequence only (NA, not applicable) are indicated. Enzyme activity summarizes several different assaysincluding expression in Escherichia coli or in human fibroblast cell culture. Low indicates activity at or below the level of detection, med indicates an intermediate activity andhigh indicates activity indistinguishable from the major allele.a Kozich et al. (2010).b de Franchis et al. (1994).c Maclean et al. (2002).d Gaustadnes et al. (2002).e de Franchis et al. (1999).f Marble et al. (1994).g Orendae et al. (2004).h Dawson et al. (1997).i Kozich et al. (1993).j Kraus et al. (1999).k Lee et al. (2005).l Urreizti et al. (2006).m Kozich and Kraus (1992).n Hu et al. (1993).o The alleles we analyzed by immunoblot. The original publication and a recent reanalysis were cited.


http://www.genetics.org/content/vol0/issue2012/images/data/genetics.111.137471/DC1/TableS1.xls



The importance of cofactor concentration to CBSfunction extended to heme

The CBS enzyme coordinates a second cofactor, heme,through a heme-binding domain. Certain mutations in theheme-binding domain disrupt CBS function in human cells,indicating that heme is critical to protein activity (Janosik et al.2001b). Furthermore, heme increases the activity and dyna-mics of some CBS alleles (Kopecka et al. 2011). Indeed, one ofthe mutations in our set, H65R, alters a heme-coordinating re-sidue and was not functional in yeast. In contrast, S. cerevisiaeCys4p lacks a heme-binding domain and does not requireheme. Yeast produce heme for other purposes, and the mediain our previous experiments lacked additional heme. There-fore, endogenous heme production was sufficient for humanCBS function. We hypothesized that some alleles of CBSmightbe heme responsive under sufficiently challenging conditions.We tested this hypothesis using a yeast hem1 strain that wasunable to synthesize d-aminolevulinic acid (d-ALA), the firstcommitted step in heme biosynthesis, and was thereforea heme auxotroph. We varied in vivo heme levels by amendingthe medium with d-ALA and determined the affect on CBSfunction.

A two-cofactor titration of all alleles in the hem1 back-ground revealed intriguing information about the heme co-factor and about allele function (Figure 4B and Table S1 andFigure S4). hem1 yeast with the predominant CBS allelewere incapable of growth without heme supplementationand showed growth dose dependence on both B6 and hemelevels. Likewise, strains with each of the 38 alleles withmeasurable growth in the B6-only titration grew better inmedia with higher heme concentrations, suggesting hemewas required for CBS function and was limiting for growth

in hem1 yeast. Cells with 6 alleles grew worse than the pre-dominant allele at 2.5 ng/ml heme, but had growth ratesapproaching that of cells with the predominant CBS allele at50 ng/ml heme, indicating that some defective CBS variantswere especially sensitive to heme (listed as “heme remedial”in Table 2). Five of these heme-responsive alleles were alsoremedial with B6, apparently identifying proteins whose de-ficiency could benefit from increased concentration of eithercofactor. The D234N allele alone benefited more from in-creased heme than from increased B6.

The remaining 32 alleles were no more sensitive to hemethan the predominant CBS allele, with two interestingexceptions. Cells with the P70L and Q526K alleles clusteredwith the low-growth alleles in the HEM1 background butwith the predominant CBS allele in the hem1 background.Similarly, cells with 7 of the 46 alleles that appeared non-functional in the HEM1 strain grew in medium containinghigh B6 and high heme, albeit poorly, revealing partial func-tion of these alleles (Figure 4B; listed as “hem1 rescue” inTable 2). Although counterintuitive, rescue of allele functionin the hem1 strain may occur because the hem1 mutationinduced heme uptake (Protchenko et al. 2008) or increasedsubstrate availability. The dynamic range of CBS-dependentgrowth in the hem1 background was larger than in a HEM1

Figure 2 CBS protein levels in yeast whole cell extracts. (A) Immunoblot-ting of yeast cells with the CBS major allele (MA), a B6-responsive allele(A226T), an AdoMet-domain mutation (Q526K), or an empty expressionvector (EV) were grown in minimal medium with 400 ng/ml B6 alone, withglutathione alone, or with glutathione and 400 ng/ml B6. (B) Yeast cellswith the CBS MA and five variant alleles were grown in minimal mediumwith glutathione alone or with glutathione and 400 ng/ml B6. Representa-tive alleles from the nonfunctional (T87N and P88S) and sick (P145L,V168M, and M126V) phenotypic classes were processed for immunoblot-ting. 3-Phosphoglycerate kinase (PGK) was detected as a loading control.

Figure 3 CBS yeast exhibited B6-dependent growth. (A) Representativegrowth curves of yeast with the major allele of human CBS cultures sup-plemented with six different levels of B6 (colored lines). Average growthrate (6SD) is shown for each B6 level (n = 84–90). (B) The growth of eachmutant (n $ 4) was expressed as the percentage of average growth rate ofyeast with the major allele of human CBS at each B6 level (6SD).






http://www.genetics.org/content/suppl/2012/01/20/genetics.111.137471.DC1/FigureS4.pdf











strain, manifested as both saturation at higher cell densityand better growth at lower B6 concentration (Figure S4).

CBS alleles with clinical associationbut no apparent defect

The majority of alleles tested supported less growth than themajor allele, as might be expected for disease-causingalleles, yet 13 appeared indistinguishable from the majorallele (listed as “similar to major allele” in Table 2). Sinceyeast are typically grown at 30�, we considered the possibil-ity that these 13 alleles encoded temperature-sensitive mu-tant proteins whose defects were not apparent at lowertemperature. We tested the growth of 10 nominally benignsubstitutions and found that none had growth defects at 37�(Figure S5), nor on medium containing the denaturant for-mamide, which can reveal partial loss of function (Aguilera1994). One allele, A69P, even appeared less sensitive to de-naturing stress than the predominant allele. Therefore, thesealleles encoded fully functional enzymes within the limits ofthis assay.

Intracellular metabolic imbalances causedby CBS variants

Our previous assays for CBS function relied on growth asa proxy for enzymatic function. As an independent assessmentof CBS function, we used LC–MS to measure directly the me-tabolite profiles of cells with different CBS alleles grown inmedium with different levels of B6. Metabolite levels mirroredthe trends observed in the growth data: cells with the majorCBS allele grown under B6 limitation and cells with a nonfunc-tional CBS allele induced similar metabolic profiles that differedfrom profiles of cells with the major CBS allele grown undernonlimiting B6 conditions (Figure 5, Table S1, and File S2).Yeast cells carrying the G307S allele, a nonfunctional allele inclinical and yeast growth assays, failed to produce glutathioneand instead accumulated homocystine, sharing the namesakediagnostic phenotype of homocystinuria.

Analysis of a second allele class, represented by the B6remedial V320A allele, further defined the correlation be-tween growth rate and metabolite flux (Table S2). Cells re-lying on the V320A allele accumulated significantly more

Figure 4 CBS yeast growth responses to B6and heme grouped alleles into distinct classes.Heat maps of growth rates normalized to thegrowth of the major allele after titration of (A)B6 in HEM1 yeast or (B) B6 and heme in hem1yeast. The column Z-score indicates the meangrowth rate (Z-score of 0) and standard devia-tion (Z-score of 61) of all alleles per column,with positive Z-scores indicating higher than av-erage growth. Arrowheads indicate alleles thatrespond to cofactor titration more strongly thanother alleles in their cluster. Asterisks (*) denotealleles that failed to grow in HEM1 yeast butwere capable of growth in hem1 yeast.




http://www.genetics.org/content/suppl/2012/01/20/genetics.111.137471.DC1/TableS1.xls



homocystine and produced less glutathione than the majorallele regardless of B6 level. However, in contrast to the non-functional G307S allele, glutathione production increased andhomocystine accumulation decreased when cells with theV320A allele were grown with a high dose of B6. These datarevealed perfect concordance with the relative growth rates ofthese alleles at these doses of B6 (Figure 3 and Table S1).

The massively parallel nature of LC–MS allowed us tomeasure metabolites upstream and downstream of CBS, aswell as those in shunt pathways. The accumulation of up-stream metabolites was not restricted to homocystine in cellswith the G307S allele or under B6 limitation of cells with themajor CBS allele. Elevated levels of AdoMet, SAH, and methi-onine, the substrates involved in homocysteine recycling byone-carbon metabolism, were detected (Figure 5 and Figure6, A–D). Overall, our data suggested that the metabolic foot-print of CBS deficiency extended far beyond the immediatesubstrate and product of the enzyme, homocysteine, and cys-tathionine, respectively. For example, the block at CBS,through mutation or B6 limitation, caused a detectable dropin 59-methylthioadenosine, a metabolite in the methioninesalvage pathway with the potential to reduce homocysteinelevels in favor of increased methionine. Instead, flux throughthis pathway was also reduced (Figure 5).

The growth rate of cells with functional CBS alleles in B6-supplemented medium was significantly greater than that inmedium lacking B6 or in cells with a loss-of-function allele.To distinguish the metabolic signature of loss of CBS functionfrom the signature of lack of growth per se, we profiled cellswith the major allele under methionine starvation. Althoughthe csy4 yeast strain used in these assays synthesizes methi-onine, additional methionine supplementation was necessaryfor growth of all CBS-substituted strains. The growth defectwithout exogenous methionine is as severe as without B6;however, the metabolic profile was strikingly different. Spe-cifically, cells limited for methionine produced low levels ofglutathione, but without homocystine accumulation, regard-less of B6 concentration (Figure 5 and Figure 6, C and D).These data confirmed that CBS-deficiency generated a uniquemetabolic profile not due simply to poor growth.

Discussion

Building on Garrod’s Inborn Errors of Metabolism (Garrod1909), technological innovations have shaped our under-standing of how an individual’s genetics cause disease.The Human Genome Project facilitated rapid progress inlinking genes and diseases, but also exposed a gap between

Figure 5 Metabolite profiles of CBS yeast grown undernutrient replete or limiting conditions. Heat map of aminoacid or derivative metabolite levels in cell extracts fromyeast grown with either the major CBS allele (MA) or theG307S (nonfunctional) allele, as measured by mass spec-trometry. Each column represents the average of four bi-ological replicates. B6 was supplemented at doses thatproduced robust growth of the major allele (400 ng/ml)or measurable, but compromised, growth (1 ng/ml). Me-tabolite levels were scaled for each row and both metab-olites and experimental conditions were subject tohierarchical clustering. The row Z-score indicates the meanand standard deviations for each metabolite, such that themean metabolite level has as a Z-score of 0. Duplicatecolumns were independent cell extracts and demonstratedtrial-to-trial variation that was not significant in any of theknown metabolites (t-test P . 0.05). The oxidizing condi-tions used for extraction strongly favored isolation ofhomocystine over homocysteine. Similarly cystathionineand cysteine were not detected because of limitations insample processing or because intracellular pools are small.



an increasing number of minor associations and an actualassessment of causality (Bansal et al. 2010; Cirulli and Gold-stein 2010; McClellan and King 2010). The so-called “miss-ing heritability” lies, in part, in the failure to define diseasewith sufficient phenotypic precision. Here, we developedtechniques that provided a quantitative assessment of clini-cally associated alleles that confirmed some expectationsand led to unexpected insights about one human geneticdisease and presumptive causative alleles.

Using yeast growth, we quantified the relative function of84 alleles of human CBS, binning alleles according to growthrate and ability to be rescued by B6 or heme cofactors. Wealso measured the levels of metabolites in cells with threedifferent human CBS alleles by LC–MS, confirming that yeastgrowth was a relevant proxy for enzyme function and reveal-ing the tight coupling between trans-sulfuration pathway fluxand growth. These quantitative phenotypes confirmed thatmany clinically associated CBS alleles are indeed nonfunc-tional, with a few notable exceptions. Although computa-tional prediction may eventually replace or supplementlaboratory research in the corroboration of genetic associa-tions, the exceptions derived from functional studies offera starting point for future analyses of protein function anddisease (Wei et al. 2010). Similar primary culture-independent,quantitative assays for human alleles in a surrogate orga-nism should be broadly applicable to any gene that fits intoan orthologous pathway (Shan et al. 1999; Zhang et al.

2003; Marini et al. 2008). Methods like this are increasinglyimportant given the expanding sequence landscape: since2010, 38 novel missense alleles of CBS have been identified(NHLBI Exome Sequencing Project 2012).

Eighty-one of the alleles we tested were identified inpeople with homocystinuria. For 33 alleles, there eitherare no clinical data about B6-responsiveness or the evi-dence is conflicting: our data could help to resolve some ofthese cases. For example, the K102Q allele functionedsimilarly to the major allele in our growth assay. Recentexome sequencing revealed that this allele, rare in pre-viously sequenced populations, has an allele frequencyclose to 4% in the African-American population (NHLBIExome Sequencing Project 2012). Therefore, additionalinformation about this allele is critical to assessing diseaserisk. For the remaining alleles, growth in yeast and clinicaldata correlated well, especially for alleles identified inpatients who were B6 nonresponsive (Table 2). In addi-tion to clinical data, the in vitro enzymatic activities ofmany CBS alleles have been assessed. Our growth-ratemeasurements were consistent with published biochemi-cal studies in 36 of the 40 cases of overlap (exceptions areitalicized in Table 2).

Sixteen of the 81 alleles had clinical features that did notmatch our yeast growth data (underlined in Table 2). Somediscrepancies may have resulted from an unrecognized sec-ond mutation in CBS in the patient. Additionally, rare alleles

Figure 6 Levels of metabolites critical in CBSfunction. Scatter plots of the levels of four dif-ferent metabolites measured by mass spectrom-etry. The average of four biological replicates(bars) and their individual measurements (squares)are shown. Duplicated columns show trial-to-trialvariation in independent cell extracts. (A) Methi-onine, (B) AdoMet, (C) homocystine, and (D) glu-tathione. The levels of all four metabolites aresignificantly different (ANOVA P , 0.0001); allsignificant differences between the MA at highB6 and other classes are indicated (Tukey’s hon-est significance test **, P , 0.005; ****, P ,0.0001).


generally occurred in a single individual, heterozygous witha different allele, making it difficult to assess the individualconnection to disease. However, there may be interestingcases in which CBS function in yeast and humans differ. Forexample, the P422L and S466L mutations in the C-terminalregulatory domain encode biochemically active proteins thatare unable to bind AdoMet and cause a distinctive, mild formof homocystinuria (Maclean et al. 2002). We tested both ofthese alleles plus six other AdoMet domain mutations andfound that all supported growth, suggesting that AdoMet re-gulation may not be critical for growth in yeast. However,cells with the L456P and Q526K alleles, both altering theAdoMet domain, had reduced growth, while cells with theT434N allele were B6 responsive, indicating that somemutations in the AdoMet domain diminish CBS function.AdoMet regulatory mutations accounted for some, but notall, discrepancies between yeast growth and clinical data.We emphasize that the power of an allelic series lies in thediversity of phenotypes, which derive from distinct proteinfunctions and reveal allele classes that may respond differentlyto treatment.

The full set of alleles demonstrated that mutation of anyCBS domain could abrogate function, and remediation wasnot specific to cofactor-binding residues (Figure 7). B6 andheme sites are separated in the tertiary structure of CBS(Meier et al. 2001), yet some variants were remedial by eithercofactor. Dual remedial alleles favored a global mechanismfor cofactor rescue over the simpler model that increasing thecofactor concentration overcomes mutations that decrease theKm of cofactor binding (Ames et al. 2002; Wittung-Stafshede2002). Since many characterized disease-causing mutationsalter protein function via folding/stability (Yue et al. 2005;Kozich et al. 2010), alleles encoding unstable proteins maybenefit from the binding energy provided by protein–cofactorinteraction. Rescue of CBS function by biological or chemicalchaperones is consistent with this hypothesis (Singh et al.2007, 2010; Majtan et al. 2010).

Regardless of the biochemical mechanism, cofactor avail-ability regulated enzyme function for all CBS alleles withina narrow and physiologically relevant range of cofactor con-centrations. While fully functional alleles supported growthat lower cofactor concentrations, metabolite levels of cellswith a functional allele grown with a low B6 level and ofcells with a nonfunctional allele were similar. This similarityof profiles may reflect a bone fide regulatory mechanism

coupling pathway flux to nutrient availability. Similarly, sub-strate limitation affected trans-sulfuration flux as strongly ascofactor limitation. Methionine limitation reduced the levelof homocystine in yeast cells regardless of whether CBS wasfunctional or attenuated by limiting B6 (Figure 6). Indeed,since methionine catabolism leads to homocysteine forma-tion, a low methionine diet is part of the treatment strategyfor homocystinuria. Critically, glutathione production wasalso compromised by loss of CBS function or methioninelimitation, with similar consequences to growth but differenteffects on homocystine production. Although patients withhomocystinuria have relatively normal serum glutathionelevels (Hargreaves et al. 2002; Orendac et al. 2003), tissueconcentrations may be significantly lower (Maclean et al.2010). Our data suggest that glutathione deficiency andhomocysteine toxicity should be considered in evaluatingthe pathology of CBS deficiency.

Overall, inability to drive sufficient flux through the trans-sulfuration pathway, regardless of cause, led to growth defects(Figure 5 and Figure 6). Conventional thought about inbornerrors is that metabolites accumulate at the point of the block.However, reversible reactions, circular connections, shunts inor out of a pathway, and feedback regulation, can establishnew ratios among even distant metabolites. Thus, a more thor-ough understanding comes from parsing the symptoms asa function of alleles and related metabolites. For example,our quantitative assays revealed the subset of alleles that weremore sensitive to B6 level and also provided evidence that theproteins encoded by six alleles benefited from increased hemelevel more than the predominant CBS allele. The behavior ofthese alleles suggested that heme deficiencies could compli-cate the diagnosis and treatment of homocystinuria. Conversely,the successful demonstration of heme supplementation couldhave utility in the clinic, either in addition to current treatmentsor as a second treatment formulation for certain alleles.

Acknowledgments

We thank Warren Kruger for plasmid pHUCBS, TonyIvaronie for help with LC–MS, Nicholas Marini for help withyeast assays, Sandrine Dudoit for the impute zeros script,and a reviewer for pointing out the increased K102Q allelefrequency. We thank Georjana Barnes, Susanna Repo, andJonathan Wong for critical evaluation of the manuscript.This work was supported in part by funds from a HowardHughes Medical Institute Professorship in support of under-graduate biology education. Additional support was pro-vided by a grant from the U.S. Department of the Army(W911NF-10-1-0496).

Literature Cited

Aguilera, A., 1994 Formamide sensitivity: a novel conditionalphenotype in yeast. Genetics 136: 87–91.

Ames, B. N., I. Elson-Schwab, and E. A. Silver, 2002 High-dosevitamin therapy stimulates variant enzymes with decreased co-

Figure 7 CBS phenotypes in relation to primary structure. Diagram of thedomain structure of the CBS protein with the location of the 84 allelesused in this analysis represented by colored bars above the diagram. Eachbar represents an allele; colors indicate the affect of the allele on growth.The Robust row reports the position of alleles indistinguishable from thepredominant allele.


enzyme binding affinity (increased K(m)): relevance to geneticdisease and polymorphisms. Am. J. Clin. Nutr. 75: 616–658.

Aral, B., M. Coude, J. London, J. Aupetit, J. F. Chasse et al.,1997 Two novel mutations (K384E and L539S) in the C-terminalmoiety of the cystathionine beta-synthase protein in two Frenchpyridoxine-responsive homocystinuria patients. Hum. Mutat. 9:81–82.

Bansal, V., O. Libiger, A. Torkamani, and N. J. Schork, 2010 Statisticalanalysis strategies for association studies involving rare variants.Nat. Rev. Genet. 11: 773–785.

Bermudez, M., N. Frank, J. Bernal, R. Urreizti, I. Briceno et al.,2006 High prevalence of CBS p.T191M mutation in homocys-tinuric patients from Colombia. Hum. Mutat. 27: 296.

Boer, V. M., C. A. Crutchfield, P. H. Bradley, D. Botstein, and J. D.Rabinowitz, 2010 Growth-limiting intracellular metabolites inyeast growing under diverse nutrient limitations. Mol. Biol. Cell21: 198–211.

Canelas, A. B., C. Ras, A. ten Pierick, J. C. van Dam, J. J. Heijnenet al., 2008 Leakage-free rapid quenching technique for yeastmetabolomics. Metabolomics 4: 226–239.

Castro, R., S. Heil, I. Rivera, M. Segurado, C. Jakobs et al., 1999 Twonovel mutations at the Cystathionine-beta-synthase gene of homo-cystinuric portuguese patients. J. Inherit. Metab. Dis. 22: 84.

Chen, X., L. Wang, R. Fazlieva, and W. D. Kruger, 2006 Contrastingbehaviors of mutant cystathionine beta-synthase enzymes asso-ciated with pyridoxine response. Hum. Mutat. 27: 474–482.

Christopher, S. A., S. Melnyk, S. J. James, and W. D. Kruger,2002 S-Adenosylhomocysteine, but not homocysteine, is toxicto yeast lacking cystathionine beta-synthase. Mol. Genet. Metab.75: 335–343.

Cirulli, E. T., and D. B. Goldstein, 2010 Uncovering the roles ofrare variants in common disease through whole-genome se-quencing. Nat. Rev. Genet. 11: 415–425.

Coude, M., J. Aupetit, M. T. Zabot, P. Kamoun, and B. Chadefaux-Vekemans, 1998 Four novel mutations at the cystathioninebeta-synthase locus causing homocystinuria. J. Inherit. Metab.Dis. 21: 823–828.

Dawson, P. A., A. J. Cox, B. T. Emmerson, N. P. Dudman, J. P. Krauset al., 1997 Characterisation of five missense mutations in thecystathionine beta-synthase gene from three patients with B6-nonresponsive homocystinuria. Eur. J. Hum. Genet. 5: 15–21.

de Franchis, R., V. Kozich, R. R. McInnes, and J. P. Kraus,1994 Identical genotypes in siblings with different homocysti-nuric phenotypes: identification of three mutations in cystathio-nine beta-synthase using an improved bacterial expressionsystem. Hum. Mol. Genet. 3: 1103–1108.

de Franchis, R., E. Kraus, V. Kozich, G. Sebastio, and J. P. Kraus,1999 Four novel mutations in the cystathionine beta-synthasegene: effect of a second linked mutation on the severity of thehomocystinuric phenotype. Hum. Mutat. 13: 453–457.

De Lucca, M., and L. Casique, 2004 Characterization of cystathio-nine beta-synthase gene mutations in homocystinuric Venezue-lan patients: identification of one novel mutation in exon 6. Mol.Genet. Metab. 81: 209–215.

Deutsch, E. W., L. Mendoza, D. Shteynberg, T. Farrah, H. Lam et al.,2010 A guided tour of the trans-proteomic pipeline. Proteo-mics 10: 1150–1159.

Fan, H. C., J. Wang, A. Potanina, and S. R. Quake, 2011 Whole-genome molecular haplotyping of single cells. Nat. Biotechnol.29: 51–57.

Frosst, P., H. J. Blom, R. Milos, P. Goyette, C. A. Sheppard et al.,1995 A candidate genetic risk factor for vascular disease:a common mutation in methylenetetrahydrofolate reductase.Nat. Genet. 10: 111–113.

Gallagher, P. M., P. Ward, S. Tan, E. Naughten, J. P. Kraus et al.,1995 High frequency (71%) of cystathionine beta-synthase

mutation G307S in Irish homocystinuria patients. Hum. Mutat.6: 177–180.

Gallagher, P. M., E. Naughten, N. Q. Hanson, K. Schwichtenberg,M. Bignell et al., 1998 Characterization of mutations in thecystathionine beta-synthase gene in Irish patients with homo-cystinuria. Mol. Genet. Metab. 65: 298–302.

Gan-Schreier, H., M. Kebbewar, J. Fang-Hoffmann, J. Wilrich, G.Abdoh et al., 2010 Newborn population screening for classichomocystinuria by determination of total homocysteine fromGuthrie cards. J. Pediatr. 156: 427–432.

Garrod, A., 1909 Inborn Errors of Metabolism. Frowde, Hodder &Stoughton, London.

Gaughan, D. J., L. A. Kluijtmans, S. Barbaux, D. McMaster, I. S.Young et al., 2001 The methionine synthase reductase (MTRR)A66G polymorphism is a novel genetic determinant of plasmahomocysteine concentrations. Atherosclerosis 157: 451–456.

Gaustadnes, M., B. Wilcken, J. Oliveriusova, J. McGill, J. Fletcheret al., 2002 The molecular basis of cystathionine beta-synthasedeficiency in Australian patients: genotype-phenotype correla-tions and response to treatment. Hum. Mutat. 20: 117–126.

Godat, E., G. Madalinski, L. Muller, J. F. Heilier, J. Labarre et al.,2010 Mass spectrometry-based methods for the determinationof sulfur and related metabolite concentrations in cell extracts.Methods Enzymol. 473: 41–76.

Guttormsen, A. B., P. M. Ueland, W. D. Kruger, C. E. Kim, L. Oseet al., 2001 Disposition of homocysteine in subjects heterozy-gous for homocystinuria due to cystathionine beta-synthase de-ficiency: relationship between genotype and phenotype. Am. J.Med. Genet. 100: 204–213.

Hargreaves, I. P., P. J. Lee, and A. Briddon, 2002 Homocysteineand cysteine–albumin binding in homocystinuria: assessment ofcysteine status and implications for glutathione synthesis?Amino Acids 22: 109–118.

Hu, F. L., Z. Gu, V. Kozich, J. P. Kraus, V. Ramesh et al.,1993 Molecular basis of cystathionine beta-synthase deficiencyin pyridoxine responsive and nonresponsive homocystinuria.Hum. Mol. Genet. 2: 1857–1860.

Janosik, M., V. Kery, M. Gaustadnes, K. N. Maclean, and J. P. Kraus,2001a Regulation of human cystathionine beta-synthase byS-adenosyl-L-methionine: evidence for two catalytically activeconformations involving an autoinhibitory domain in the C-terminalregion. Biochemistry 40: 10625–10633.

Janosik, M., J. Oliveriusova, B. Janosikova, J. Sokolova, E. Krauset al., 2001b Impaired heme binding and aggregation of mu-tant cystathionine beta-synthase subunits in homocystinuria.Am. J. Hum. Genet. 68: 1506–1513.

Katsushima, F., J. Oliveriusova, O. Sakamoto, T. Ohura, Y. Kondoet al., 2006 Expression study of mutant cystathionine beta-synthase found in Japanese patients with homocystinuria.Mol. Genet. Metab. 87: 323–328.

Kim, C. E., P. M. Gallagher, A. B. Guttormsen, H. Refsum, P. M.Ueland et al., 1997 Functional modeling of vitamin respon-siveness in yeast: a common pyridoxine-responsive cystathio-nine beta-synthase mutation in homocystinuria. Hum. Mol.Genet. 6: 2213–2221.

Kluijtmans, L. A., H. J. Blom, G. H. Boers, B. A. van Oost, F. J.Trijbels et al., 1995 Two novel missense mutations in the cys-tathionine beta-synthase gene in homocystinuric patients. Hum.Genet. 96: 249–250.

Kluijtmans, L. A., G. H. Boers, E. M. Stevens, W. O. Renier, J. P.Kraus et al., 1996 Defective cystathionine beta-synthase regu-lation by S-adenosylmethionine in a partially pyridoxine respon-sive homocystinuria patient. J. Clin. Invest. 98: 285–289.

Kluijtmans, L. A., G. H. Boers, J. P. Kraus, L. P. van den Heuvel, J. R.Cruysberg et al., 1999 The molecular basis of cystathioninebeta-synthase deficiency in Dutch patients with homocystinuria:


effect of CBS genotype on biochemical and clinical phenotypeand on response to treatment. Am. J. Hum. Genet. 65: 59–67.

Kopecka, J., J. Krijt, K. Rakova, and V. Kozich, 2011 Restoringassembly and activity of cystathionine beta-synthase mutantsby ligands and chemical chaperones. J. Inherit. Metab. Dis.34: 39–48.

Kozich, V., and J. P. Kraus, 1992 Screening for mutations by ex-pressing patient cDNA segments in E. coli: homocystinuria due tocystathionine beta-synthase deficiency. Hum. Mutat. 1: 113–123.

Kozich, V., R. de Franchis, and J. P. Kraus, 1993 Molecular defectin a patient with pyridoxine-responsive homocystinuria. Hum.Mol. Genet. 2: 815–816.

Kozich, V., M. Janosik, J. Sokolova, J. Oliveriusova, M. Orendacet al., 1997 Analysis of CBS alleles in Czech and Slovak pa-tients with homocystinuria: report on three novel mutationsE176K, W409X and 1223 + 37 del99. J. Inherit. Metab. Dis.20: 363–366.

Kozich, V., J. Sokolova, V. Klatovska, J. Krijt, M. Janosik et al.,2010 Cystathionine beta-synthase mutations: effect of muta-tion topology on folding and activity. Hum. Mutat. 31: 809–819.

Kraus, J. P., 1994 Molecular-basis of phenotype expression in ho-mocystinuria. J. Inherit. Metab. Dis. 17: 383–390.

Kraus, J. P., M. Janosik, V. Kozich, R. Mandell, V. Shih et al.,1999 Cystathionine beta-synthase mutations in homocystinu-ria. Hum. Mutat. 13: 362–375.

Kraus, J. P., V. Kozich, and M. Janosik, 2012 CBS Mutation Da-tabase, v. January 3, 2012. http://cbs.lf1.cuni.cz/index.php.

Kruger, W. D., and D. R. Cox, 1994 A yeast system for expressionof human cystathionine beta-synthase: structural and functionalconservation of the human and yeast genes. Proc. Natl. Acad.Sci. USA 91: 6614–6618.

Kruger, W. D., and D. R. Cox, 1995 A yeast assay for functionaldetection of mutations in the human cystathionine beta-synthasegene. Hum. Mol. Genet. 4: 1155–1161.

Kruger, W. D., L. Wang, K. H. Jhee, R. H. Singh, and L. J. Elsas 2nd.2003 Cystathionine beta-synthase deficiency in Georgia(USA): correlation of clinical and biochemical phenotype withgenotype. Hum. Mutat. 22: 434–441.

Lee, S. J., D. H. Lee, H. W. Yoo, S. K. Koo, E. S. Park et al.,2005 Identification and functional analysis of cystathioninebeta-synthase gene mutations in patients with homocystinuria.J. Hum. Genet. 50: 648–654.

Linnebank, M., M. Janosik, V. Kozich, E. Pronicka, J. Kubalska et al.,2004 The cystathionine beta-synthase (CBS) mutationc.1224–2A.C in Central Europe: Vitamin B6 nonresponsivenessand a common ancestral haplotype. Hum. Mutat. 24: 352–353.

Maclean, K. N., M. Gaustadnes, J. Oliveriusova, M. Janosik, E.Kraus et al., 2002 High homocysteine and thrombosis withoutconnective tissue disorders are associated with a novel class ofcystathionine beta-synthase (CBS) mutations. Hum. Mutat. 19:641–655.

Maclean, K. N., J. Sikora, V. Kozich, H. Jiang, L. S. Greiner et al.,2010 Cystathionine beta-synthase null homocystinuric micefail to exhibit altered hemostasis or lowering of plasma homo-cysteine in response to betaine treatment. Mol. Genet. Metab.101: 163–171.

Majtan, T., L. Liu, J. F. Carpenter, and J. P. Kraus, 2010 Rescue ofcystathionine beta-synthase (CBS) mutants with chemical chap-erones: purification and characterization of eight CBS mutantenzymes. J. Biol. Chem. 285: 15866–15873.

Marble, M., M. T. Geraghty, R. de Franchis, J. P. Kraus, and D.Valle, 1994 Characterization of a cystathionine beta-synthaseallele with three mutations in cis in a patient with B6 nonre-sponsive homocystinuria. Hum. Mol. Genet. 3: 1883–1886.

Marini, N. J., J. Gin, J. Ziegle, K. H. Keho, D. Ginzinger et al.,2008 The prevalence of folate-remedial MTHFR enzyme var-iants in humans. Proc. Natl. Acad. Sci. USA 105: 8055–8060.

McClellan, J., and M. C. King, 2010 Genetic heterogeneity in hu-man disease. Cell 141: 210–217.

Meier, M., M. Janosik, V. Kery, J. P. Kraus, and P. Burkhard,2001 Structure of human cystathionine beta-synthase: a uniquepyridoxal 59-phosphate-dependent heme protein. EMBO J. 20:3910–3916.

Motulsky, A. G., 1996 Nutritional ecogenetics: homocysteine-related arteriosclerotic vascular disease, neural tube defects, andfolic acid. Am. J. Hum. Genet. 58: 17–20.

Mudd, S. H., F. Skovby, H. L. Levy, K. D. Pettigrew, B. Wilcken et al.,1985 The natural history of homocystinuria due to cystathio-nine beta-synthase deficiency. Am. J. Hum. Genet. 37: 1–31.

Mudd, S. H., H. L. Levy, and F. Skovby, 1995 Disorders of trans-sulfuration, pp. 1279–1327 in The Metabolic and Molecular Basesof Inherited Disease, Ed. 7, edited by C. R. Scriver, A. L. Beaudet,W. S. Sly, and D. Valle. McGraw-Hill, New York.

Mudd, S. H., H. L. Levy, and J. P. Kraus, 2001 Disorders of trans-sulfuration, pp. 2007–2056 in The Metabolic and Molecular Basesof Inherited Disease, Ed. 8, edited by C. R. Scriver, A. L. Beaudet,D. Valle, W. S. Sly, B. Childs et al. McGraw-Hill, New York.

Ng, S. B., A. W. Bigham, K. J. Buckingham, M. C. Hannibal, M. J.McMillin et al., 2010a Exome sequencing identifies MLL2 mu-tations as a cause of Kabuki syndrome. Nat. Genet. 42: 790–793.

Ng, S. B., K. J. Buckingham, C. Lee, A. W. Bigham, H. K. Tabor et al.,2010b Exome sequencing identifies the cause of a mendeliandisorder. Nat. Genet. 42: 30–35.

NHLBI Exome Sequencing Project, 2012 Exome Variant Server,v. January 3, 2012. http://evs.gs.washington.edu/EVS.

Online Mendelian Inheritance in Man, 2012 McKusick–NathansInstitute of Genetic Medicine, Johns Hopkins University (Balti-more, MD) and National Center for Biotechnology Information,National Library of Medicine, Bethesda, MD. http://omim.org/.

Ono, B., Y. Shirahige, A. Nanjoh, N. Andou, H. Ohue et al.,1988 Cysteine biosynthesis in Saccharomyces cerevisiae: muta-tion that confers cystathionine beta-synthase deficiency. J. Bac-teriol. 170: 5883–5889.

Orendac, M., J. Zeman, S. P. Stabler, R. H. Allen, J. P. Kraus et al.,2003 Homocystinuria due to cystathionine beta-synthase de-ficiency: novel biochemical findings and treatment efficacy. J.Inherit. Metab. Dis. 26: 761–773.

Orendae, M., E. Pronicka, J. Kubalska, M. Janosik, J. Sokolovaet al., 2004 Identification and functional analysis of two novelmutations in the CBS gene in Polish patients with homocystinu-ria. Hum. Mutat. 23: 631.

Pare, G., D. I. Chasman, A. N. Parker, R. R. Zee, A. Malarstig et al.,2009 Novel associations of CPS1, MUT, NOX4, and DPEP1with plasma homocysteine in a healthy population: a genome-wide evaluation of 13 974 participants in the Women’s GenomeHealth Study. Circ. Cardiovasc. Genet. 2: 142–150.

Protchenko, O., M. Shakoury-Elizeh, P. Keane, J. Storey, R. Androphyet al., 2008 Role of PUG1 in inducible porphyrin and hemetransport in Saccharomyces cerevisiae. Eukaryot. Cell 7: 859–871.

Schubert, E. L., M. K. Lee, H. C. Mefford, R. H. Argonza, J. E.Morrow et al., 1997 BRCA2 in American families with fouror more cases of breast or ovarian cancer: recurrent and novelmutations, variable expression, penetrance, and the possibilityof families whose cancer is not attributable to BRCA1 or BRCA2.Am. J. Hum. Genet. 60: 1031–1040.

Scott, J. W., S. A. Hawley, K. A. Green, M. Anis, G. Stewart et al.,2004 CBS domains form energy-sensing modules whose bind-ing of adenosine ligands is disrupted by disease mutations. J.Clin. Invest. 113: 274–284.

Sebastio, G., M. P. Sperandeo, M. Panico, R. de Franchis, J. P. Krauset al., 1995 The molecular basis of homocystinuria due to cys-tathionine beta-synthase deficiency in Italian families, and re-port of four novel mutations. Am. J. Hum. Genet. 56: 1324–1333.


http://cbs.lf1.cuni.cz/index.php

http://evs.gs.washington.edu/EVS

http://omim.org/

Sen, S., and R. Banerjee, 2007 A pathogenic linked mutation inthe catalytic core of human cystathionine beta-synthase disruptsallosteric regulation and allows kinetic characterization of a full-length dimer. Biochemistry 46: 4110–4116.

Shan, X., and W. D. Kruger, 1998 Correction of disease-causingCBS mutations in yeast. Nat. Genet. 19: 91–93.

Shan, X., L. Wang, R. Hoffmaster, and W. D. Kruger, 1999 Functionalcharacterization of human methylenetetrahydrofolate reductase inSaccharomyces cerevisiae. J. Biol. Chem. 274: 32613–32618.

Shih, V. E., J. M. Fringer, R. Mandell, J. P. Kraus, G. T. Berry et al.,1995 A missense mutation (I278T) in the cystathionine beta-synthase gene prevalent in pyridoxine-responsive homocystinu-ria and associated with mild clinical phenotype. Am. J. Hum.Genet. 57: 34–39.

Singh, L. R., X. Chen, V. Kozich, and W. D. Kruger, 2007 Chemicalchaperone rescue of mutant human cystathionine beta-synthase.Mol. Genet. Metab. 91: 335–342.

Singh, L. R., S. Gupta, N. H. Honig, J. P. Kraus, and W. D. Kruger,2010 Activation of mutant enzyme function in vivo by protea-some inhibitors and treatments that induce Hsp70. PLoS Genet.6: e1000807.

Smith, C. A., E. J. Want, G. O’Maille, R. Abagyan, and G. Siuzdak,2006 XCMS: processing mass spectrometry data for metaboliteprofiling using nonlinear peak alignment, matching, and identi-fication. Anal. Chem. 78: 779–787.

Sperandeo, M. P., M. Candito, G. Sebastio, M. O. Rolland, C. Turc-Carelet al., 1996 Homocysteine response to methionine challenge infour obligate heterozygotes for homocystinuria and relationshipwith cystathionine beta-synthase mutations. J. Inherit. Metab. Dis.19: 351–356.

Tautenhahn, R., C. Bottcher, and S. Neumann, 2008 Highly sen-sitive feature detection for high resolution LC/MS. BMC Bioin-format. 9: 504.

Tsai, M. Y., P. W. Wong, U. Garg, N. Q. Hanson, and K. Schwich-tenberg, 1997 Two novel mutations in the cystathioninebeta-synthase gene of homocystinuric patients. Mol. Diagn. 2:129–133.

Urreizti, R., S. Balcells, M. Rodes, L. Vilarinho, A. Baldellou et al.,2003 Spectrum of CBS mutations in 16 homocystinuric pa-tients from the Iberian Peninsula: high prevalence of T191Mand absence of I278T or G307S. Hum. Mutat. 22: 103.

Urreizti, R., C. Asteggiano, M. Cozar, N. Frank, M. A. Vilaseca et al.,2006 Functional assays testing pathogenicity of 14 cystathio-nine-beta synthase mutations. Hum. Mutat. 27: 211.

Wei, Q., L. Wang, Q. Wang, W. D. Kruger, and R. L. Dunbrack Jr.,2010 Testing computational prediction of missense mutationphenotypes: functional characterization of 204 mutations of hu-man cystathionine beta synthase. Proteins 78: 2058–2074.

Winzeler, E. A., D. D. Shoemaker, A. Astromoff, H. Liang, K. Andersonet al., 1999 Functional characterization of the S. cerevisiae ge-nome by gene deletion and parallel analysis. Science 285: 901–906.

Wishart, D. S., C. Knox, A. C. Guo, R. Eisner, N. Young et al.,2009 HMDB: a knowledgebase for the human metabolome.Nucleic Acids Res. 37: D603–D610.

Wittung-Stafshede, P., 2002 Role of cofactors in protein folding.Acc. Chem. Res. 35: 201–208.

Yue, P., Z. Li, and J. Moult, 2005 Loss of protein structure stabilityas a major causative factor in monogenic disease. J. Mol. Biol.353: 459–473.

Zhang, N., M. Osborn, P. Gitsham, K. Yen, J. R. Miller et al.,2003 Using yeast to place human genes in functional catego-ries. Gene 303: 121–129.

Communicating editor: S. Fields


GENETICSSupporting Information


Surrogate Genetics and Metabolic Profilingfor Characterization of Human Disease Alleles

Jacob A. Mayfield, Meara W. Davies, Dago Dimster-Denk, Nick Pleskac, Sean McCarthy,Elizabeth A. Boydston, Logan Fink, Xin Xin Lin, Ankur S. Narain,

Michael Meighan, and Jasper Rine

Copyright © 2012 by the Genetics Society of AmericaDOI: 10.1534/genetics.111.137471

J. A. Mayfield et al. 2 SI

Figure S1 Biochemical pathway of relevant metabolites. Arrowheads represent the direction of the reaction in human cells. Double arrows indicate that intermediate metabolites are not shown. The reaction performed by CBS is circled; shunts in or out of the pathway are summarized in boxes.

Methionine

AdoMet

SAH

Homocysteine

FolateBiosynthesis

Methionine Salvage

Methylthioadenosine

Cystathionine

Cysteine

Glutathione

Homocystine

Methylation

CBS

FIGURE S1


Figure S2 Growth of nonfunctional and low-‐growth alleles on solid medium lacking glutathione. A 5-‐fold dilution series of cultures grown in minimal medium containing glutathione and lacking B6 was replica plated on solid medium +/-‐ glutathione or with defined concentrations of B6. Three independent transformations of representative nonfunctional (V168M, R302K, R266G), low growth (E131D, R224H), and B6 responsive (R266K) alleles are shown, with the major allele (MA) and empty vector (EV) controls included on each plate for reference.

FIGURE S2

V168M

E131D

R266K

R266G

R224H

R302K

4-fold dilution of cells

Glutathione 400 ng/ml B6 1 ng/ml B6 0.1 ng/ml B6

E131DE131D

V168MV168M

R266KR266K

R266GR266G

R224H

R302K

R224H

R302K

EV

MA

EV

MA

EVMA


Figure S3 CBS protein levels of 11 alleles in yeast cell extracts. Immunoblot of yeast cell extracts from cells containing the major allele of CBS (MA) or one of 11 other alleles after growth in minimal medium containing 400 ng/ml B6.

FIGURE S3

CBS

PCNAMA G1

39R

D198

V

E144

K

E131

D

G151

R

H65R

K384

E

C109

R

K102

N

G307

S

A226

T


Figure S4 hem1 CBS yeast responses to B6 and heme. Growth rates of hem1 yeast with the major human CBS allele were measured under 6 conditions of B6 and heme supplementation. The average (± SD) is shown (n=4).

FIGURE S4

0

0.04

0.08

0.12

ng/ml B6

ng/ml B6

ng/ml B6

ng/ml B6

ng/ml B6

ng/ml B6

HEM1 hem1

grow

th ra

te (l

ogOD

/hr)

0550

Heme (ng/ml)

1 2 400 1 2 400


FIGURE S5

MAvector only

V320A (remedial)K102N V354M V371M P422L P78R

MAvector only

V320A (remedial)K102N V354M V371M P422L P78R

MAvector only

V320A (remedial)E128DK102QS466L A69P I435T

MAvector only

V320A (remedial)E128D K102Q S466LA69P

I435T

Day:Glutathione:

Temperature:Formamide:

3+30-

3-

30-

3-

30+

Day:Glutathione:


5-

37-

5-

37+

5-

30+

Day:Glutathione:


3+30-

3-

30-

3-

30+

Day:Glutathione:


5-

37-

5-

37+

5-

30+


Figure S5 CBS function during denaturing stress. A 5-‐fold dilution series of cultures grown in minimal medium containing glutathione and lacking B6 was replica plated on solid medium +/-‐ glutathione or +/-‐ 1% formamide. Plates were grown at 30° or 37° and imaged at 3 or 5 days after plating to allow equivalent growth. The major allele (MA) is shown for reference.


File S1

Raw growth rate data and normalized averages of growth rates in the HEM1 and hem1 strains, as used in heat

maps.

File S1 is available for download at http://www.genetics.org/content/suppl/2012/01/20/genetics.111.137471.DC1 as

a compressed folder. This folder contains Raw growth rate data.xls, which is an Excel spreadsheet containing yeast

growth data. All raw growth data are included on sheet "Growth Rate Data". The sheet "Summary HEM1" includes

mean growth rate at each B6 concentration for each allele normalized to the major allele control grown on the same

plate after removal of outliers using Grubbs test. "Summary hem1-‐" includes mean growth rates after two-‐way

titration of B6 and heme in the hem1 strain. Each allele is normalized to the major allele controls grown on the same

plate and outliers were removed after using Grubbs test.


File S2

Summary of metabolite data and analysis methods.


a compressed folder. This folder contains metabolite data.xls, which is an Excel spreadsheet containing the measured

metabolite levels from LC/MS analyses. These data can be regenerated using the raw data files and R scripts included

as File S3. For the G307S dataset, all samples were generated in a single Orbitrap run and are therefore directly

comparable. The metabolite extraction was performed in two batches on separate days, indicated as experiment 1 or

2. There were slight, but not significant, differences in metabolite levels of positively identified compounds between

the two days. The sheet “G307S peaks” contains an abbreviated XCMS output. Missing peaks were replaced either

through the fillPeaks function in XCMS or through imputation using local minima, the intensities were normalized

using upper quartiles and the data were log transformed. Note that zero values occurred, but were replaced to

facilitate normalization and transformation. Columns 1 and 2 were appended according to matches between the

XCMS output and the HMDB database and give the names of the top 3 matches for each given peak and the number

of potential hits. The Metlin column gives the URL matches to the Metlin database. The Anova column indicates peaks

that were significantly different between sample classes.

The sheet “G307S Metabolites” contains the subset of data used to generate Figure 5 and includes amino acids and

other metabolites. In addition to database matching using the mass/charge ratio, seventeen target compounds were

included in Calibration Standards (Cal) in a 4-‐fold dilution series added to a pooled experimental sample. Hence,

peaks that corresponded to a chemical included in the calibration standard decreased in intensity as the amount

added decreases, while the retention time and mass/charge ratio remain accurate. Peaks with the expected decrease

in intensity were identified by two statistical methods. The “p value exp” columns use a linear model to match the

measured intensity of a peak in the calibration standard dilution series to the theoretical value, such that a low p

value indicates a good fit. The “Pearsons R exp” columns measure the correlation coefficient between the measured

level of an identified metabolite in the calibration standards and the theoretical level, such that an R value

approaching one indicates better correlation. The undiluted calibration standard contained 100 μg/ml glycine, 25

μg/ml serine, 5 μg/ml proline, 25 μg/ml threonine, 15 μg/ml leucine, 50 μg/ml aspartic acid, 15 μg/ml lysine, 50 μg/ml

glutamic acid, 25 μg/ml methionine, 5 μg/ml histidine, 20 μg/ml cystathionine, 25 μg/ml glutathione, 50 μg/ml


cystine, 15 μg/ml homocystine, 10 μg/ml methylthioadenosine, 30 μg/ml s-‐adenosyl homocysteine (SAH), and 50

μg/ml s-‐adenosyl methionine (AdoMet). Isotopically labeled glycine and isotopically labeled methionine were spiked

into all samples at 50 μM and 2.5 μM, respectively. Glycine was not detected; isotopically labeled methionine did not

vary significantly in any sample or sample class (ANOVA p > 0.05).

The V320A dataset differed in several significant ways, reflected in the data. First, ~2X as many cells were used in the

extraction, increasing both metabolite concentrations and noise. Second, the calibration panel was analyzed in

solvent, not in a pooled standard; hence, metabolite identification was by exact mass and more stringent criteria

were employed in metabolite identification. Finally, the 400 and 1 ng/ml B6 experiments were performed on different

days and although they can be grouped together, are not directly comparable by ANOVA. The sheet “V320A” peaks

contains the XCMS output of peak groups, appended and treated for G307S. The sheet “V320A Metabolites” lists only

the peaks with a single, unambiguous exact mass match. When multiple group peaks shared the same identity, the

more abundant peak was used.


File S3

Data files and R script for processing metabolite data.


a compressed folder. Centroided mass spectrum provided in the mzXML format, organized according to class, are

included as Mayfield_LC_MS_Files.zip. A full executable R script, CBS_Analysis_Scripts.R.zip, is provided to align peaks

using XCMS, normalize and transform the output, match peaks to metabolite databases and target compounds, and

output data tables or figures.


Table S1 Cofactor responses of 44 functional alleles. Mutant growth rates are expressed as a percent of the major

allele at the same dose of B6 or heme; averages represent data from 4-‐24 biological replicates and error bars show

standard deviation.

Table S2 Critical metabolites measured in the major allele and a B6 remedial allele under high and low

concentration of B6. Metabolite levels from cell extracts are expressed as the log2 of the mean intensity ± standard

deviation (SD) measured by LC/MS for the cells containing the major allele (MA) or the B6 remedial allele V320A,

grown at 400 or 1 ng/ml B6. Significant differences are indicated (T test, NS = not significant). Homocystine was not

detected in two 400 ng/ml B6 replicates. High and low B6 experiments were performed on separate dates.

Tables S1 and S2 are available for download at

http://www.genetics.org/content/suppl/2012/01/20/genetics.111.137471.DC1 as Excel files.

Documents

Surrogate Genetics and Metabolic Pro ling for ... · deﬁning challenge of human genetics for the foreseeable future: interpreting the impact of variations in the sequences of individual