Molybdenum cofactor biosynthesis in humans Identification of two complementation groups of cofactor deficient patients and preliminary characterization of a diffusible molybdopterin precursor

Molybdenum cofactor biosynthesis in humans.

Identification of two complementation groups of

cofactor-deficient patients and preliminary

characterization of a diffusible molybdopterin

precursor.

J L Johnson, … , R Mandell, V E Shih

J Clin Invest.

1989;

83(3)

:897-903.

https://doi.org/10.1172/JCI113974

.

Molybdenum cofactor deficiency is a devastating disease with affected patients displaying

the symptoms of a combined deficiency of sulfite oxidase and xanthine dehydrogenase.

Because of the extreme lability of the isolated, functional molybdenum cofactor, direct

cofactor replacement therapy is not feasible, and a search for stable biosynthetic

intermediates was undertaken. From studies of cocultured fibroblasts from affected

individuals, two complementation groups were identified. Coculture of group A and group B

cells, without heterokaryon formation, led to the appearance of active sulfite oxidase. Use of

conditioned media indicated that a relatively stable, diffusible precursor produced by group

B cells could be used to repair sulfite oxidase in group A recipient cells. Although the

extremely low levels of precursor produced by group B cells preclude its direct

characterization, studies with a heterologous, in vitro reconstitution system suggest that the

precursor that accumulates in group B cells is the same as a molybdopterin precursor

identified in the Neurospora crassa molybdopterin mutant nit-1, and that a converting

enzyme is present in group A cells which catalyzes an activation reaction analogous to that

of a converting enzyme identified in the Escherichia coli molybdopterin mutant ChlA1.

Research Article

Molybdenum Cofactor

Biosynthesis in Humans

Identification of Two

Complementation

Groups

of

Cofactor-deficient

Patients

and

_Preliminary

Characterization of

a

Diffusible

Molybdopterin

Precursor

JeanL.Johnson,* MargotM.Wuebbens,* Roseann Mandell, and VivianE.Shih

*DepartmentofBiochemistry, Duke University Medical Center, Durham, North Carolina 27710; andAminoAcid Disorder Laboratory, Massachusetts General Hospital, and Department

of

Neurology, Harvard Medical

School,

Boston, Massachusetts02114

Abstract

Molybdenum cofactor

deficiency

is a devastating disease with

affected

patients

displaying

thesymptoms of a

combined

defi-ciency of

sulfite oxidase

and xanthine

dehydrogenase.

Because

of

the extreme

lability of

the

isolated, functional molybdenum

cofactor,

direct

cofactor

replacement therapy

is

not feasible, and a search

for

stable

biosynthetic intermediates

was under-taken. From

studies

of cocultured

fibroblasts from affected

individuals, two complementation groups were identified. Co-culture of group A and group B cells, without heterokaryon

formation,

led to theappearance of

active sulfite

oxidase. Use of

conditioned media indicated

that a

relatively

stable, diffus-ible precursor produced by group B cells could be used to repair sulfite oxidase in group A recipient cells. Although the extremely low levels of precursor produced by group B cells preclude its direct characterization, studies with a heterolo-gous, in

vitro

reconstitution system suggest that the precursor that accumulates in group Bcells

is

the same as a

molybdo-pterin

precursor

identified

in the

Neurospora

crassa

molybdo-pterin

mutant

nit-i,

and that a

converting

enzyme is present in group A cells which

catalyzes

anactivation reaction

analogous

to that

of

a

converting

enzyme

identified

in the

Escherichia coli

molybdopterin

mutant

ChiAJ.

Introduction

Molybdenum

cofactor

deficiency

was

first identified in

1978 (2) and has now been

documented

in more than 20

patients

(3-5). The

molybdenum cofactor is

a

complex of

a reduced

pterin

species

termed

molybdopterin

and

molybdenum

(Fig.

1).

The

cofactor is required for

the

function of

atleast three enzymes in

humans:

sulfite

oxidase,

xanthine

dehydrogenase,

and

aldehyde oxidase.

A

deficiency

of

the

cofactor results

in symptoms

of sulfite oxidase

deficiency,

which

include seizures

and other

neurological

abnormalities,

mental

retardation,

and

dislocated

ocular

lenses,

in

combination with

xanthinuria,

and most

often

leads to deathat ayoung age. No symptoms

attrib-utable

specifically

to

the

absence

of

_aldehyde

oxidase

have been

identified.

Molybdenum

cofactor

deficiency

is inherited as an autosomal

recessive

trait

(5)

andcanbe

diagnosed

pre-Aportion of thiswork waspresented earlierin abstractform (1). Addressreprintrequests to Dr.Johnson.

Received

for publication

27 June 1988 and in

revisedform

27

Sep-tember 1988.

natallyby assay of sulfite oxidase activity in cultured amniotic cells (6) or chorionic

villus

biopsy (7). However, at present there isno cure or effective therapy for molybdenum cofactor deficiency. The cofactor itself is exceedingly unstable, such that its structural

characterization

couldbe

carried

out only by study of stable degradation products (8, 9). Thus direct cofac-tor

replacement

therapy is not

feasible,

and anunderstanding of thebiosynthetic pathway and identification of stable pre-cursormolecules

which

may

be

intermediates

in the pathway become extremely important in

development

of a strategy to alleviate the symptoms of the

deficiency

disease.

Inthe results described below, we show that molybdenum cofactor

deficient patients

can be

assigned

to two complemen-tation groups. A

relatively

stable

diffusible

molybdopterin precursor

which is

produced by group B cells and utilized by group A cells to synthesize active molybdenum cofactor is

identified,

and

evidence is

presented whichsuggests that this diffusibleintermediate is identical to a

molybdopterin

precur-sor that accumulates in a

molybdopterin

mutant

of

Neuro-spora crassa

(10). The

presence

of

a

converting

enzyme

activ-ity

in group A

cells that

activates

theprecursor to yield func-tional

molybdopterin

and is

analogous

to a

converting

enzyme

identified

in a

molybdopterin

mutant

of

E.

coli

is also docu-mented.From the results of studies described in this article, it

is

possible

to

define

a

pathway outlining

the late steps in the

biosynthesis of

the molybdenum

cofactor.

That pathway,with

the

sites that

appear to be

defective in the

two classes

of

molyb-denum

cofactor-deficient patients

and in the Neurospora and

Escherichia coli

molybdopterin mutants

indicated,

is shown in

Fig.

2.

Methods

Assayof sulfiteoxidase in cultured fibroblasts. Fibroblasts were ob-tained andculturedasdescribedpreviously (1 1). For assayofsulfite oxidaseactivity, frozen cell pellets(3-10 X 106 cells) werethawed, suspendedin 1 ml0.01MTris-HCl, pH 8.5, with 50ulof1%sodium deoxycholate, frozenandthawedonce, andpassedthrougha27-gauge needle severaltimes.After the addition of10 ulof50 mMNaCN,the mixturewascentrifugedat12,000gfor5 min. The supernatant frac-tionwasappliedto acolumn ofSephadex G-25 (PD-10, Pharmacia FineChemicals, Piscataway, NJ;9.1 mlbed volume)equilibrated with 0.1 MTris-HCl, pH 8.5, with0.1 mM EDTA. The columnwaswashed with2 mlof thesamebuffer,andtheproteinfractionwaselutedwith 1.5 mlof bufferandassayed for sulfite oxidaseactivity (1 1).Aunitof sulfite oxidase activity is definedastheamountofenzymeproducing

an absorbancechange at550 nmof 0.001 absorbanceunitper min

at25'C.

Each fibroblastcell line describedwasculturedindividuallyand showntolack sulfite oxidaseactivity. Patientswereidentifiedas

mo-lybdenum cofactor deficient by demonstrated absence of xanthine de-hydrogenase activityin combination with sulfite oxidase deficiency

and,inmostcases,bytheir failureto excreteurothione,themetabolic J.Clin.Invest.

© TheAmerican Society for Clinical Investigation,Inc. 0021-9738/89/03/0897/07 $2.00

H

N

HN_~I C =C-CHOHCH20PO3

H2N N N

H

MOLYBDOPTERIN

H

N

HN) C--CHOHCH20PO3

H2N N N \/

H Mo

0

MOLYBDENUM COFACTOR

Figure 1.Structures of molybdopterinand themolybdenum

cofac-tor.The dioxostructureshown for themolybdenumcofactor is char-acteristic of themolybdenumcenterin sulfite oxidase and nitrate

re-ductase. In themolybdenumcenterof xanthinedehydrogenase,one

of theoxogroupsisreplaced byaterminalsulfideligand.The oxida-tionstateof thepterinringmayvaryindifferentmolybdoenzymes.

degradation product of the molybdenumcofactor(12). Forclarity, patientsand fibroblast linesarereferredtobyacode whichdesignates thecomplementationgrouptowhichtheyhavebeenassignedbasedon

the studiesreportedhere. Thecorresponding patientinitials and

refer-encestothecasedescriptionsareasfollows:Al,R.M. (11, 13); A2, E.V. (2-4); A3,K.Z.(4,14), A4,K.M.(4); A5,M.V.(4);A6,F.M.(5); A7, C.K.(5); Bl,S.B.(15); B2,Z.B.(15), B3,W.T. Patients Bl andB2

aresibs,andpatientsA2 andA5aredistantlyrelated.Allotherpatients areunrelated.

Complementation analyses. Complementation experiments were

carriedoutasdescribedpreviously(16). Aliquotsoffibroblastswere

mixed ina 1:1 ratio and inoculated into 100-mm culture dishesata densityof 1.2X106 cellsperdish. Cell fusionwasinitiatedbyexposure

to 50%polyethylene glycol 1000. Fibroblastswereplated 24 hafter fusion, andgrownfor 11 d beforeharvesting.Mixtures oftwo fibro-blast lines(1:1)notexposedtopolyethylene glycolweresubjectedto

thesametreatments.

CONVERTING ENZYME

r___.___

MOLYBDOPTERIN _

PRECURSOR V

defectin group Apatients defectin group Bpatients

and inE.coliChll and in Neurosporanit-I

COMPOUND Z

Culture and preparationofcell-freeextractsofNeurosporaand E. coli. Neurospora nit-i mycelia were grown, induced for nitrate reduc-tase,and harvested as previously described (17). Frozen myceliawere thawed, homogenized in 2 vol of 0.1 M potassium phosphate buffer, pH 7.4, with 5 mM EDTA, 1 mM phenylmethylsulfonyl fluoride, and 1mMdithiothreitol usingaDuall glass grinder (Kontes Co., Vineland, NJ) and centrifugedat15,000 g for 20min. The supernatant fraction was usedfor reconstitution studies.Forstudiesrequiring further frac-tionation ofthe nit-i extract, 2.0 ml ofthe supernatant was applied toa column of Sephadex G-25 (PD-bI, Pharmacia Fine Chemicals) equili-brated with homogenization buffer. The column was washed with 1 ml ofbuffer and the excluded, high molecular weight fractionwaseluted with 2.5mlof buffer. After another 2.0-ml wash whichwasdiscarded, the low molecularweight fractionwaseluted in 4 ml.

E.coliChlAJ cells (18) were grown on LB medium (Gibco, Grand Island, NY) and harvestedatlate logor early stationary phase by centrifugationat5,000 g for 25 min. Cellswerewashedoncewithwater and suspended in 5 vol 0.01 M potassium phosphate, pH 7.4. Extracts werepreparedwith a French pressure cell and storedat-20'C.

Assay ofmolybdopterin precursor in conditioned culture media. Reconstitution mixtures containing 40

Al

ofmedium,25glof ChL41 extract, 25

Al

of nit-ihigh molecular weight fraction, and 10 Ml of 0.5 Msodium molybdatewereincubatedat roomtemperature for 15 min. Reconstituted nitrate reductase activitywasassayedby the addition of 0.4mlofasolution of substratestoyield in the 0.5 ml assay mix: 0.35 mMNADPH, 30mMKNO3, 10 MM flavin adeninedinucleotide, 5 mMNa2SO3, and 0.1 M potassium phosphate, pH 7.2 (19). After incubation for 20 min at room temperature, nitrite produced was quantitated by a colorimetric assay (17) and monitoredat540 nm. Blank reactions with NADPH omitted were also run and used to correctfor anynonspecific contributionstothe 540nmabsorbance.

Assay ofconverting enzyme infibroblastextracts. Cells were lysed as described for assay of sulfite oxidase activity. Lysates were used directly withnocentrifugationorG-25 chromatography steps. Recon-stitution mixtures containing 60 Ml of fibroblast extract, 30 Ml of nit-i extract(completeextract orhigh molecular weight fraction), and 10 M1 of0.5M sodiummolybdatewereincubatedat4°Covernight and then assayed for nitratereductaseactivityasdescribed above. To demon-stratethatfibroblast converting enzyme could activatethenit-i mo-lybdopterin precursor, reconstitution mixtures were prepared as above butwith 30 Ml ofthenit-i low molecular weight fraction. After over-night incubationat4°C to allow activation ofthe precursor species, 50

Mlofthe nit-I

high

molecularweightfractionwasaddedas a sourceof apo nitrate reductasetoaccept activemolybdenum cofactor. The mix-turewasincubatedat roomtemperaturefor 15 min to allow reconsti-tution and then assayed for nitrate reductase activity.

APO SULFITE OXIDASE (humans)

or

APO NITRATE REDUCTASE (E.coliorNeurospora)

MOLYBDOPTERIN J MOLYBDENUM

ACTIVE SULFITE OXIDASE (humans)

or

ACTIVE NITRATE REDUCTASE (E. coliorNeurospora)

Assaysforcompound Z. Fibroblast extracts prepared as for assay of sulfite oxidase(uncentrifuged)andconditionedculturemedium sam-ples were acidified to pH 1 with HC1 and a solution of 1% 12 with 2% KI wasadded in sufficient excess to maintain a yellow color. After 30 min oxidation, samples were centrifuged at 12,000 g for 5 min and chro-matographed in 50 mM ammonium acetate, pH 5, on a C-18 HPLC column (4.6X250 mm, 10

,g,

Alltech Associates, Deerfield, IL) using a Hewlett-Packard 1090 liquid chromatograph with an HP 1040 A diode-array detectorand an HP 1046Afluorescence detector (Hew-lett-Packard Co., Palo Alto, CA). Under these conditions authentic compound Z(10) elutes at 6 min. Samples of urine (10 ml) were acidified to pH 3 with HCI, incubated for 30 min withI2JKI, centri-fuged to clarify,andappliedto a0.8mlcolumn of Florisil resin(Fisher Scientific Co., Springfield, NJ) equilibrated with 0.01 N HCl. The column waswashed with 10ml of 0.01 N HCl,and the material of interest was eluted with 5ml50%acetone in water. The eluant was adjusted to pH 8 with NaOH and applied to a 1.0mlcolumn of QAE Sephadex (acetate form) equilibrated with H20. The column was washed with 3 ml ofH20followed by 10 ml0.01Macetic acid and the material ofinterest was eluted with 0.01MHCI. Fractions of 3mlwere collected and analyzed for the presence ofcompoundZby HPLC in 50 mMammonium acetate, as described above, and also with 5% metha-nolacidified to pH2with HC1 as mobile phase. Authentic compound

Z,purified from E. coli ChlM cells (10) was carried through the same

Florisil and QAE chromatography steps; itwaseluted froma i ml column ofQAE Sephadex with 0.01 NHC1 in fractions 10 and 11.

Activemolybdopterin precursorinurine sampleswasassayed as described for assay in conditioned cellculturemedium.

Results

Reconstitution ofsulfite oxidase in

culturedfibroblasts.

Fibro-blastscultured

from patients with

molybdenum

cofactor

defi-ciency contain

no

active sulfite oxidase

assayable byasensitive

spectrophotometric

assay. However,asshown in Table I,

cul-tures

containing

heterokaryons produced by polyethylene gly-col

fusion of

cells

from

patient

B1 or

sib

B2

with

those

from

patient

Al gave

positive

results when

assayed

for sulfite

oxi-dase. Also

positive

werecultures

containing

Al cellsgrownin thepresenceofB IorB2cells but withnoinduction ofhetero-karyon

formation.

Coculture

without

cell

fusion

has been

ex-tended

to

include

combinations of fibroblasts from

a large number

of

molybdenum

cofactor-deficient patients

and the resultsare

summarized

in Table II.

Initially it

was

found

that allcrossesthat

contained

cells

from

B1 orB2 showed

reconsti-tution of

sulfite oxidase

activity

whilecrossesbetweenany

of

the other

patients

gave

negative

results. These results

define

two

complementation

groups: group A

consisting of

seven

pa-tients (A 1.-A7)

andgroup B

consisting

of BI and B2.

Recently,

another

unrelated

molybdenum

cofactor-deficient

patient

has been

assigned

to group B. Asshown in Table II, cells

from

patient B3 complement

A6 and A7 but

yield

no

activity

when cultured

with cells from

B 1.

Thepresence

of sulfite

oxidase activity

in

cocultured

fibro-blasts in the absence

of

heterokaryon formation

suggeststhat

complementation is

occurring

by

means

of

a

diffusible

mole-cule

produced

and

excreted into

theculture

medium

byone

cell class

which

is then taken up and processed to produce

active molybdenum cofactor by

cells

of

theother

classification

group. To

establish

that such is the case, to

determine

which

complementation

group

produces

and

which

usesthe

putative

precursor

species, and

to

gain

some

insight

into the

stability of

suchaprecursor

species,

experiments

were

undertaken

cultur-ingrecipientcellsin medium

which

had been

conditioned

by

Table

L

_Sulfite

Oxidase ActivityinFibroblast

Heterokaryons

Induced

by

Polyethylene Glycol

Fusion

Cross Sulfite oxidase Polyethylene glycol U/mg solubleprotein

AlX

Bi

5.1 +

7.2

-AlXB2 10.1 +

6.7 _

BlXB2 <0.8 +

<0.8

-BI Xcontrol 6.2 +

7.4

-B2Xcontrol 9.3 +

8.8

-A3XA4 <0.8 +

<0.8

-Foreach cross,fibroblastswerecombined ina 1:1 ratio. Cellswere

incubated with(+)orwithout(-)polyethylene glycol for24h, plated and cultured for 11 d, harvested,andassayed for sulfite oxi-daseactivity.

prior

culture

of cells of

adonor

line.

The

results of

one

experi-ment, shown in Table III,

confirm

that cells from patient

B1

produce a

transferable species

and

demonstrate

that

it is

of

sufficient stability

tobe

utilized

by

A3

cells even

after

B I

cells

havebeen removed

from

theculture. These results

indicate

that

group B

cells

arethe precursor donors and group A cells are the acceptor cells

which

convert the precursor to

active

Table II.Sulfite Oxidase Activity in Mixed Fibroblast Cultures from Molybdenum

Cofactor-deficient

Patients

GroupA XgroupA Group AXgroup B Group BXgroup B Sulfite Sulfite Sulfite Celllines oxidase Cell lines oxidase Celllines oxidase

U/mgsoluble protein

AlXA2 -* AIXBI 5.0 BIXB3

-AlXA3 - _A2X

Bl

_9.6

AlXA4 - _A3X

Bl

_3.8

AlXA5 - A4X

Bl

7.2

AlXA6 - _A5X

Bl

_1.7

AlXA7 - _A6X

Bl

_4.0 A2X A3 - A7XBI 4.7

A2X A4 - _{A6XB3 1.5}

A2XA5 - _{A7XB3 1.6} A3XA4

-A3X A5

-A4X A5

-A6XA7

-Foreachcross,fibroblasts fromtwomolybdenum cofactor-deficient patientswerecombinedandcultured forperiods ranging 3-13d,

harvested,andassayed for sulfite oxidase activity.The valuesgiven

forAlXB1and A3XB1 areaveragesofvaluesobtained from four and threereplicate experiments, respectively.

TableIII.Sulfite Oxidase ActivityinFibroblasts Cultured in Conditioned Media

Donor cell line Recipient cell line Sulfite oxidaseactivity

U/mgsoluble protein

A2 A4 -*

A3 Bi

B1 A3 1.8

Donorfibroblastswerecultured for 3 d under standardconditions;

the culture mediumwasthen collected andcentrifuged.Amixture of conditioned medium(70%)andfresh medium(30%)wasused for platingrecipient cells. Disheswerefedevery 2 dwithamixture of conditioned medium and freshmedium,harvestedonday10,and assayed for sulfite oxidase activity.

*Dashindicatesnoactivity detected (<0.5).

molybdopterin.

It would appearaswell that thegenetic defect in groupAlies in an enzyme that precedes the enzyme which is defective in group B in the molybdopterin biosynthetic

pathway.

Characterization

of

the

molybdopterin

precursor

produced

by group B patients. Becauseof theexceedinglylow levelsof

molybdoenzymes,

molybdopterin, and its precursors which arepresent in cultured

fibroblasts,

itwasclear that

purification

of

the

diffusible intermediate from group B cells or

condi-tioned

medium would beadifficult

task,

further

hampered

by

the

lengthy

assay

procedure relying

onreconstitution of sulfite oxidaseactivity in cultured groupAcells. To overcome these

problems

and facilitate

a

partial

characterization of the

pre-cursor

produced by

group B

patients,

weturnedto a heterolo-gous in vitro reconstitution system which is described below:

Themost

widely used

assay

for active molybdenum

cofac-tor

utilizes

the

nit-i

mutant

of Neurospora

crassa, which is

defective

in

molybdopterin biosynthesis.

The nit-i strain

ac-cumulates

a

soluble, inactive

apo nitrate reductase whichcan

be

readily reconstituted

in vitro

by

a sourceof active

molybde-num

cofactor.

The

reconstitution

takes

place rapidly

at room

temperatureor

somewhat

more

slowly

at

4VC requiring only

a

crudeextract

of nit-i

mycelia,

a sourceof

molybdenum

cofac-tor or

molybdopterin

and

inorganic molybdate.

After

recon-stitution,

the

active

nitrate reductase

is incubated with

sub-strates

and

nitrite

produced

is

quantitated by

a

calorimetric

procedure.

One

precaution

mustbe taken when

utilizing

the

nit-i

sys-tem as anassay

for active

molybdopterin.

While the nit-i mu-tant

contains

no

functional

molybdopterin, it does accumulate

alow molecular

weight

precursor

of molybdopterin

(10). In

the presence

of

an

appropriate converting

enzyme,added

ex-ogenously

tothe

nit-i

extracts,

this

precursorcanbe

activated

to

molybdopterin which

in turnwillreconstitute the apo

ni-trate reductase. The

converting

enzyme

which

activates

the

nit-i

precursorwasfirst

identified

(10) in a mutant of E.

coli,

ChlAi,

which also is defective in

molybdopterin

synthesis.

Specifically,

it

was

observed

that

incubation of

an extract

of

ChlAI cells

with

an

unfractionated

extract

of

nit-i

led tothe appearance

of active

nitrate

reductase. Separation of the low

molecular weight fractions of

the

nit-i

by gel exclusion

chro-matography

produced a

preparation of

aponitrate reductase

(high

molecular weight fraction) whichwasactivatedby

ChlAI

only when the nit-i precursor (low molecular weight fraction) was added back. Thus it is possible to differentiate between nit-i nitrate reductase reconstitution whichoccurs as aresult of the addition of active molybdopterinand thatwhichresults

from addition of a source of convertingenzymewhich acti-vatesthe

nit-i

precursorby the

fact

that the former

requires

only the high molecular weight fraction of the nit-i extract

while

the latter

requires both high

and low molecular

weight

fractions. Recently we have shown that the nit-i assay can be appliedas asensitivemeasure of active molybdopterinin

fi-broblast extracts (13); in such studies it was prudent to use

extracts

of nit-i from which

all

low

molecular

weight

precur-sorshad been removed.

Neither

the low

molecular weight molybdopterin

precursor which accumulates in

nit-i

northe

converting

enzyme

which

is

present

in ChlAI has

been

fully characterized, although

stud-ies to

this

end arecurrently in progress.

Nevertheless,

enough

information

is

already

available

to allow us to ask

whether

theselate steps in the

pathway

of cofactor

biosynthesis/activa-tion observed in microorganisms are present in humans as

well. That is,

is

the

diffusible

precursor that accumulates in group B patients related to the molybdopterin precursor in

Neurospora

nit-i? And, is

there an

activity

infibroblasts from groupA

patients

that

is

equivalent

tothe

converting

enzyme

found

in E.

coli

ChI4i?

Early

experiments

were

directed

atthe

first

question.

Con-ditioned

medium from

group B cells was

incubated

with

ChlAI

extractsandthen addedto extractsofnit-I

(high

molec-ular

weight fraction)

and

assayed

for nitrate reductase

recon-stitution.

No

activity

was

observed,

but from the

negative

re-sult

it

was not

possible

toconclude whether the group B

pre-cursor differs from the nit-i precursor and thus was not

recognized

and

activated

by ChlAI

converting factor

or

whether levels

of

precursor weremerely too low to detect.

TableIV. Demonstration ofConverting Enzyme Activity in ExtractsofFibroblasts fromGroup AMolybdenum

Cofactor-deficient

PatientsbyIn Vitro Reconstitution ofNitrateReductase

Activity in Neurospora crassa

nit-i

Patient Complete High molecular High + low molecular cell line nit-iextract weight fraction weight fraction

Al 0.282 0.013 0.061

A2 0.297 0.029 0.097

A3 0.321 0.013 0.123

BI 0.001 0.009 0.014

Fibroblastextracts were incubated at

40C

overnight with complete unfractionatedextracts ofNeurospora

nit-i

to test for the presence of

aconvertingenzymeactivity inthefibroblastswhich could activate a molybdopterinprecursor in the

nit-i

extract andallow reconstitution oftheNeurosporaaponitratereductase. The nitrate reductase that wasreconstitutedwas thenallowed to react with substrates (nitrate andNADPH)asdescribedin Methods and nitrite formed was quan-titated byacalorimetricprocedure. Incubations of fibroblast extracts withtheisolated high molecular weight fraction of a

nit-i

extract

(containingaponitrate reductasebut no precursor) and with recom-bined highandlow molecular weightfractions were carried out to determinewhetherreconstitutionwas dependent on the presence of themolybdopterinprecursorpresent in the low molecular weight fraction of the

nit-i

extract.

*

_Activity

_is

Amore

fruitful experimental

approachwasthen employed

which demonstrated thatgroupAcellscontain a converting

enzyme which can activate the nit-i precursor. Asshown in

Table IV, extractsof fibroblastsfrom three groupApatients wereabletoreconstitute nitrate reductase activitywhen incu-batedwithcrude, unfractionated extractsofNeurosporanit-i while anextractofcellsfroma group Bpatientfailed to

pro-duce anyactive nitrate reductase. The factor in thegroup A cellsresponsibleforthe reconstitutionofnitrate reductasewas identified as aconverting enzyme rather than active

molyb-dopterin (which must be absentin fibroblasts from molybde-numcofactor-deficient patients) by the observations thatvery

little reconstitution occurred in the absence of the nit-i low

molecular weight precursor fraction and that considerable

stimulation ofreconstitutionwasobserved when the nit-i low

molecularweight fraction wasadded back. Thesuccessful

ac-tivation of nitratereductaseusinggroupAconvertingenzyme and nit-iprecursorsand thefailureof theconverseexperiment

usingapotential sourceofgroupBprecursorand

ChlAi

con-verting factormaybeattributabletothecatalyticnatureof the convertingenzyme.Verylow levelsofenzymewhen incubated

with an excess of substrate can produce significant levels of

active molybdopterin while high levels ofenzymeand limiting precursor canat mostmake molybdopterinatalevel equalto

thatoftheprecursoritself.

The presence ofa converting enzyme in fibroblasts from

groupApatients thatcanactivate the molybdopterin

precur-sorin nit-i implied that the intermediate whichaccumulates in fibroblasts is very closely related to that in Neurospora. Confirmation, however, required some way to analyze the

precursors themselves. Published work (10) directed at char-acterizing the molybdopterin precursor in Neurospora nit-i

suggested thatanunusual fluorescent pterin species thatcould beisolatedafter acidic iodine oxidation ofnit-i extractsmight beaninactive oxidized degradationproductofthe

molybdop-terin

precursor.Morerecently, wehave shownthistobetrue

by thedemonstration that thepurifiedactiveprecursor is

oxi-dized to thissame fluorescent species (termed compound Z; Johnson etal., manuscript in preparation). Thus, anattempt was made to show the presence of compound Z in samples

fromgroupBpatients. Neither conditioned mediumnor

fibro-blastsfromgroupBpatients, whentreatedwithacidiciodine, yielded significant levelsof compound Z. However, asshown

in Fig. 3, aurine sample from patient Bl didindeed contain

compound Z. The peakof material elutingatthesame posi-tion asauthenticcompound

Z,

detectedbyitsabsorption at

300 nm, wasfluorescentand, as shown in Fig. 3, hadan

ab-sorptionspectrumessentiallyidentical tothatofthestandard

material.Significantly,nopeakcorrespondingtocompound Z

wasidentified inchromatogramsof urine frompatientAl or fromacontrolindividual. Asecondaliquotofthesameurine

samplefrom patientB1 wasprocessedin thesamemannerbut

with the iodine oxidation step omitted. The yield of

com-pound Z was identical suggesting that conversion ofactive

precursortocompoundZoccurredeitherbefore excretionor

in theaged samplewhich had been storedfor 5 yrand

sub-jectedto many freeze-thaw cycles. A stored sample ofurine

from patient B2 yielded thesameresults:alevelof compound

Zcomparabletothatin the BI samplewhich didnotvarywith

theinclusionoromission of the iodineoxidationstep. Fresh

urine samplesfrompatientsB1 andB2werenotavailablefor

study;however, afresh samplefrompatient B3wasobtained

A

3-2

E 3-o

.2

c0

-0I "0

0

0 2 4 6 8 10

ml

2.0

CD

C _1.0

°0

o5

'f

250 300 350 400 450 nm

Figure 3.(Left)HPLC elution profilesof urinesamplesfrompatients (A)BI, (B) Al, and (C)acontrolindividual.Sampleswereprepared

asdescribedinMethods usingin eachcase10mlof urine which

contained (A) 3.5, (B) 5.4,or(C)5.3mgcreatinine.In eachcase,

chromatogramswereobtained from 500ulof fraction 10 eluted

from QAE Sephadex with 0.01 MHCl. The mobilephasewas5%

methanol acidifiedtopH2withHCl ataflowrateof1ml/min.

Ab-sorbancewasmonitoredat300nm.Thearrowin A indicates the

elution position ofauthenticcompound Z. (Right) Absorption

spec-traof compoundZisolatedfrom urine frompatientBI ( )and of

the standard compound(---).Spectrawereacquired byon-line diode

arraydetectionof HPLC eluantsin5% methanolacidified topH2

with HCl.

andassayedfor compoundZ.LevelsofcompoundZwere_very

low (15% of that seenin B1)inthesample analyzedwithout

iodine

oxidation,

but equivalenttoB1 after iodineoxidation.

The

iodine-dependent

appearance ofcompoundZ in the

sample from patient B3suggestedthat itmightbe possibleto

detect the presence of the parent functional molybdopterin

precursorinasample which hadnotbeen iodine-oxidized. As

shown in TableV,incubation ofanaliquotof B3 urinewithan

extractofChiAIasa sourceofconvertingenzyme and thehigh molecular weight fraction ofa

Neurospora

nit-i extract did

lead to the reconstitution of high levels of nitrate reductase

activity. No reconstitution occurred in the absence of the

ChLAJ

extract. As expected, the control urine sample, which

Table V. Demonstration ofActiveMolybdopterin Precursor in Urine fromaGroupBMolybdenumCofactor-deficientPatient

by In VitroReconstitutionofNitrateReductaseActivityin

Neurosporacrassanit-i

Urine sample Nitratereductaseactivity*

B3 (fresh sample) 1.512

Conrol (fresh sample) 0.035

B1 (stored sample) 0.025

Reaction mixtures containing aliquotsof urine fromgroup B molyb-denumcofactor-deficientpatientsor acontrolindividual,anextract

ofE. coliChLA1toprovideasourceofconvertingenzyme,and the

high molecular weightfraction ofanextract ofnit-l to_provideapo

nitrate reductasewereincubatedat 40C forIh. The nitrate

reduc-tasethatwasreconstitutedwasallowedto react withsubstratesand

the nitrite formedwasquantitated byacolorimetricprocedure.

*_{Activity is expressedasabsorbanceat540nm.}

', I

yielded nocompound Zevenwith iodine oxidation, and the stored sample of urine from patient B 1, which yielded its full complement of compound Z without iodine oxidation, both were incapable of reconstituting significant levels of nitrate reductase activity. A preliminary investigation of the stability of the precursor activity in the urine sample from patient B3 hasrevealed that activity was not greatly diminished by a num-ber of freeze-thaw cycles but was totally abolished by iodine oxidation.

Discussion

The

presence

of

converting enzyme

in

fibroblasts from

group A patients and active precursor and

compound

Z in urine from group B patients is strong evidence that the late steps in the

synthesis

of

molybdopterin

in man and

microorganisms

are identical. The final stepsinthe

pathway

of molybdopterin biosynthesis have been summarizedinFig. 1.

According

to

this

scheme,

themolecular

defect

in group B patients and in the Neurospora nit-i mutant may be assigned to a gene which codes for or regulates the expression of

con-verting

enzyme. As a

consequence

of this

defect,

the molyb-dopterin precursor, which is the substrate for the converting enzyme,

and/or

itsoxidized

product

accumulate andcanbe

identified

in

culture media

from group B

cells,

in

Neurospora

nit-i cells

(10)

and culture medium

(Johnson

et

al.,

manu-script

in

preparation),

and in

urine samples

from

group

B

patients.

The

absence

of detectable levels of compound

Z in control urine

samples

would suggest that under normal cir-cumstances

levels of the

molybdopterin

precursor are

quite

low; appearance

of

precursor in

extracellular

fluids and culture

media

may

result

only when

intracellular levels

are

raised due

to

absence

of

functional

converting

enzyme.

The

processing of the

molybdopterin

precursor to

yield

molybdopterin is carried

out

by

a

converting

enzyme

in

group A

fibroblasts and

ChlAJ cells. These results

suggest

that the

molecular defect in

group A

patients and

in

ChIAJ is earlier

in

the

biosynthetic pathway

(prior

to

the

converting

enzyme), but

do not

establish whether

the

ChLiA

and groupA

mutations

are at

the

samestep or in

fact whether all

groupA

patients

have

the

same

molecular

defect. If

anyor

all

of the

group A

patients

lack an enzyme

which

directly precedes converting

enzyme on

the

biosynthetic pathway, they

may accumulate

the substrate

of

that

defective

enzyme.

Analysis of urine samples from

group A

patients for the

presence

of unusual pterin species

which

may be

related

to

molybdopterin and

the

molybdop-terin precursor

could

prove very

useful in

future studies of

molybdopterin

biosynthesis

in humans.

The

converting

enzymes

in fibroblasts

and

ChLAJ

which

are

assayed by their activation

of the

nit-i

molybdopterin

pre-cursorwould appear to have been

designed

to carry out the same

catalytic reaction.

Further

studies

will be required to

determine whether

the

reaction mechanisms

are

indeed

iden-tical

and to

ascertain

themolecular form and cellular

localiza-tion of

the enzymes in these

widely divergent life forms.

From

the

results presented above,

it

appears very likely that the

diffusible

precursor

produced

bygroup B cells is

iden-tical

to

the molybdopterin

precursor present in Neurospora

nit-l. The

implications

of such

a

conclusion

are

far-reaching.

The

precursor

in

Neurospora

nit-i

andcertain other E.

coli

molybdopterinmutantsispresentin extremely highamounts

andit,aswellasits oxidized product, compound

Z,

is

amena-bleto

purification

and structural characterization (Johnsonet

al., manuscript

in

preparation).

Such information should

prove invaluable in terms ofunderstanding how the labile

molybdopterin

is

synthesized

andits fragile side chain compo-nentsprotected prior to activation and may ultimatelyprove

to be oftherapeutic valuetomolybdenum cofactor-deficient

patients.

If further studies

verify

that the

purified

molybdop-terin

precursorisolated from nit-i is abletoreconstitutesulfite oxidasewhen addedtothe culture medium ofgroupA

cells,

the possibility of employing theprecursor speciesin vivo to correctmolybdenum cofactor deficiency may indeed become quite real.

Acknowledgments

This work was supported by grants DK-35029 and NS-05096 from the National Institutes of Health.

References

1.Shih, V. E., R. Mandell, and J. L. Johnson. 1984. Complemen-tation in variantsof sulfite oxidasedeficiency.Pediatr.Res. 18:226A. (Abstr.)

2. Duran, M., F.A.Beemer, C.vanderHeiden,J.Korteland, P. K. deBree, M.Brink,and S. K. Wadman. 1978. Combineddeficiencyof xanthine oxidase andsulphiteoxidase:adefect of molybdenum me-tabolismortransport?J.Inherited Metab.Dis. 1:175-178.

3. Johnson,J.L., W. R.Waud, K. V.Rajagopalan,M.Duran, F. A. Beemer, and S. K. Wadman. 1980. Inborn errors ofmolybdenum

metabolism: combineddeficiencies of sulfite oxidase and xanthine dehydrogenase inapatient lackingthemolybdenum cofactor. Proc. Natl.Acad. Sci. USA. 77:3715-3719.

4. Wadman, S. K., M. Duran, F. A. Beemer, B. P. Cats, J. L. Johnson,K. V.Rajagopalan,J.M.Saudubray, H.Ogier,C. Charpen-tier,R. Berger, G. P. A.Smit, J.Wilson, and S. Krywawych. 1983. Absence ofhepatic molybdenum cofactor:aninbornerrorof metabo-lismleadingto acombineddeficiencyofsulphiteoxidase and xanthine dehydrogenase. J.InheritedMetab. Dis.6(Suppl. 1):78-83.

5.Johnson,J.L., and S. K. Wadman.Molybdenumcofactor defi-ciency. In Metabolic Basis ofInherited Disease, Sixth Edition. C. R. Scriver, A. L. Beaudet, W. S. Sly, and D. L. Valle, editors. McGraw-Hill Book Co.,New York. Inpress.

6.Ogier,H., S.K.Wadman, J.L.Johnson,J.M.Saudubray,M. Duran, J. Boue,A. Munnich, and C.Charpentier. 1983. Antenatal

diagnosis of combined xanthine and sulphiteoxidase deficiencies. Lancet.2:1363-1364.

7.Desjacques,P.,P.Baltassat, and J. M. Saudubray. Useof

chori-onic villi for the antenataldiagnosisofxanthineoxidase andsulphite

oxidasecombineddeficiency.J.Inherited Metab. Dis.Inpress. 8.Johnson,J. L., B. E.Hainline, K.V. Rajagopalan,and B. H. Arison. 1984. The pterin component of the molybdenum cofactor. Structural characterization oftwo fluorescent derivatives. J. Bio.

Chem. 259:5414-5422.

9. Kramer, S. P.,J. L.Johnson,A.A.Ribeiro, D. S.Millington, and K.V.Rajagopalan. 1987.Thestructureof the molybdenumcofactor. Characterization ofdi-(carboxamidomethyl)molybdopterin from sul-fite oxidase and xanthine oxidase. J. Biol. Chem. 262:16357-16363.

10.Johnson,M.E., andK. V.Rajagopalan. 1987.In vitrosystem formolybdopterin biosynthesis.J.Bacteriol. 169:110-116.

heredi-tary metabolic disorder in sulfur metabolism. N. Engl. J. Med. 297:1022-1028.

12. Johnson, J. L., and K. V.Rajagopalan. 1982. Structural and metabolic relationshipbetween themolybdenumcofactor and urothi-one. Proc.Natl.Acad. Sci. USA. 79:6856-6860.

13. Johnson, J. L., M. M.Wuebbens,R.Mandell,andV.E. Shih. 1988. Molybdenum cofactor deficiencyin a patientpreviously charac-terized as deficient in sulfite oxidase. Biochem. Med. Metab. Biol. 40:86-93.

14.Ogier,H., J. M.Saudubray,C.Charpentier,A. Munnich, J. L. Perignon,A.Kesseler,and J. Frezal. 1982. Doubledeficiten sulfite et xanthine oxydase, cause d'encephalopathie due a une anomalie hereditaire du metabolisme du molybdene. Ann. Med. Interne.

133:594-596.

15.Munnich,A., J. M. Saudubray, C.Charpentier,H.Ogier, F. X. Coude, J. Frezal,L.Yacoub,A.Harbi,and S. Snoussi. 1983. Multiple molybdoenzymedeficienciesdue to an inborn error of molybdenum

cofactormetabolism:twoadditionalcasesina new

_family.

J. Inherited

Metab.Dis.

6(Suppl. 2):95-96.

16. Shih, V. E., R. Mandell, and A. Herzfeld. 1982. Defective ornithine metabolism in cultured skin fibroblasts frompatientswith thesyndrome ofhyperornithinemia,

hyperammonemia

and homoci-trullinuria. Clin. Chim.Acta. 118:149-157.

17. Amy, N.K.,and K. V.Rajagopalan. 1979.Characterization of molybdenum cofactor fromEscherichia coli. J. Bacteriol. 140:114-124.

18. Johnson,M. E.andK.V.Rajagopalan. 1987. Involvement of

ChiA,

E, M, and Nloci inEscherichia coli

molybdopterin biosynthesis.

J.Bacteriol. 169:117-125.

19. Hawkes,T.R.,andR.C.Bray. 1984.Quantitative transfer of themolybdenum cofactor from xanthine oxidase and fromsulphite

oxidaseto the deficientenzyme of the

nit-i

mutant of Neurospora crassa toyieldactive nitrate reductase. Biochem. J. 219:481-493.