enzimas alostericas

Instituto Tecnológico

de Veracruz

ING. BIOQUÍMICA

PROFESORA: ZAIDA ORTA FLORES

BIOQUÍMICA 1

UNIDAD 2

TRADUCCIÓN DE ARTÍCULO DE INVESTIGACIÓN:

“El estado bionergético y antioxidante de las neuronas es controlado por la continua degradación de una llave enzima

glicolítica por APC/C-Cdh1”

ESTUDIANTE: LUZ MARÍA VIÑAS CERÓN

El estado bionergético y antioxidante de las neuronas es controlado por la continua degradación de una llave enzima glicolítica por APC/C-Cdh1 Angel Herrero Mendez, Angeles Almeida, Emilio Fernández, Carolina Maestre, Salvador Moncada y Juan P. Bolaños.

Las neuronas son conocidas por tener una tasa glicolítica menor que los astrocitos, y cuando son estresados son incapaces de sobreregular la glicolisis por una baja actividad de Pfkfb3 (6- fosfofructo- 2 -cinasa/fructuosa-2, 6 bifosfatasa-3). Esta enzima genera fructuosa-2, 6-bifosfatasa (F2,6P2), el más potente activador de 6-fosfofructo-1cinasa (Pfk1; ref.4), un regulador maestro de glicosis. Aquí mostramos que el Pfkfb3 está ausente en las neuronas de la corteza cerebral y ese Pfkfb3 es constantemente objeto de una degradación proteasomal por la acción de la E3 ubiquitina ligasa, promovedor de la anafase compleja ciclosoma (APC/C)-Cdh1. En contraste, los astrocitos tienen baja actividad APC/C-Cdh1 y por tanto Pfkfb3 está presente en esas células. La regulación a la alza de Pfkfb3 ya sea por la inhibición de Cdh1 o por la sobreexpresión de Pfkfb3 en las neuronas resultó en la activación de glicolisis. Esto, sin embargo, fue acompañado por un marcado descenso de la oxidación de la glucosa a través de la vía pentasa fosfato (una ruta metabólica involucrada en la regeneración de glutatión reducido) resultando en un estrés oxidativo y muerte apoptótica. Así, regulando activamente a la baja la glicólisis por APC/C-Cdh1, las neuronas usan glucosa para mantener su estado antioxidante a expensas de su utilización para propósitos bioenergéticos.

Usando una reacción de cadena de polimerasa transcriptasa inversa (RT-PCR) en RNA extraído de neuronas corticales de ratas terminales y astrocitos, establecimos que Pfkfb m RNA es expresado en neuronas y ese isoforma 3 (Pfkfb3) mRNA es el más abundante en ambos tipos de células. (Información adicional en figura S1a). Esto fue confirmado por el secante norte (Fig 1ª panel superior). Teniendo demostrado por RT-PCR

que K6 fue el más abundante variante de empalme de pfkfb3 mRNA, alcanzamos un anticuerpo en contra de su dominio carboxi terminal (Informacion adicional en fig s1b,c) y lo utilizaron para evaluar la expresión de la proteína Pfkfb3 en neuronas y atrocitos. Pfkfb3 fue indetectable para el secante oeste en las neuronas, mientras estuvo presente en astrocitos (fig 1ª, paneles más bajos; info adicional fig 1Sd y e). La hinmunohistoquimica en secciones coronales de la corteza cerebral de la rata mostró que el Pfkfb3 no colocalizó con el marcador neuronal nuclear NeuN pero si colocalizó con el marcador astrocito GFAP (Fig 1b). Así, aunque Pfkfb3 mRNA está presente en las neuronas, la proteína Pfkfb3 está ausente, sugiriendo que la enzima es regulada a la baja post transcripcionalmente en estas células.La incubación de neuronas con inhibidores de proteosoma MG132 o lactasistina por una hora resultó en la acumulación de proteína Pfkfb3 (Fig 1c. panel superior). Además, la inmunoprecipitación de Pfkfb3 en cualquiera de ambas MG132 o lactasistina, seguido por la secante oeste usando un anticuerpo anti ubiquitina, reveló un incremento en la ubiquitylation de Pfkfb3 (fig 1c panel más bajo). Este resultado indica que en neuronas Pfkfb3 es degradada a través de la via ubiquitina-proteosoma, la cual ha sido descrita en las células miogenicas durante la diferenciación. Investigamos entonces posibles motivos orientando Pfkfb3 para ubiquitylation y hallamos ese Pfkfb3, pero no los isómeros 1, 2 ó 4, contiene una caja de KEN en la posición 142 (Info adicional fig S1f).Una caja KEN escoge proteínas para ubiquitylation por la APC/C. La activación de la APC/C requiere la formación de un complejo de Cdc20 o Cdh1 (ref 10); sin embargo, Cdh1 es el único activador posible de APC/C en las neuronas terminalmente diferenciadas usadas en este estudio. Cdh1 fue derribadaen

NATURE CELL BIOLOGY ADVANCE ONLINE PUBLICATION 1

L E T T E R S

The bioenergetic and antioxidant status of neurons is controlled by continuous degradation of a key glycolytic enzyme by APC/C–Cdh1Angel Herrero-Mendez1, Angeles Almeida1,2, Emilio Fernández1, Carolina Maestre1,2, Salvador Moncada3,4 and Juan P. Bolaños1,4

Neurons are known to have a lower glycolytic rate than astrocytes and when stressed they are unable to upregulate glycolysis1 because of low Pfkfb3 (6-phosphofructo-2-kinase/fructose-2, 6-bisphosphatase-3) activity2. This enzyme generates fructose-2,6-bisphosphate (F2,6P2)

3, the most potent activator of 6-phosphofructo-1-kinase (Pfk1; ref. 4), a master regulator of glycolysis5. Here, we show that Pfkfb3 is absent from neurons in the brain cortex and that Pfkfb3 in neurons is constantly subject to proteasomal degradation by the action of the E3 ubiquitin ligase6, anaphase-promoting complex/cyclosome (APC/C)–Cdh1. By contrast, astrocytes have low APC/C–Cdh1 activity and therefore Pfkfb3 is present in these cells. Upregulation of Pfkfb3 by either inhibition of Cdh1 or overexpression of Pfkfb3 in neurons resulted in the activation of glycolysis. This, however, was accompanied by a marked decrease in the oxidation of glucose through the pentose phosphate pathway (a metabolic route involved in the regeneration of reduced glutathione7) resulting in oxidative stress and apoptotic death. Thus, by actively downregulating glycolysis by APC/C–Cdh1, neurons use glucose to maintain their antioxidant status at the expense of its utilization for bioenergetic purposes.

Using reverse-transcriptase polymerase-chain reaction (RT–PCR) in RNA extracts from terminally differentiated rat cortical neurons and astrocytes, we established that Pfkfb mRNA is expressed in neurons and that isoform 3 (Pfkfb3) mRNA is the most abundant in both cell types (Supplementary Information, Fig. S1a). This was confirmed by northern blotting (Fig. 1a, upper panel). Having demonstrated by RT–PCR that K6 was the most abundant splice variant of Pfkfb3 mRNA, we raised an anti-body against its carboxy-terminal domain (Supplementary Information, Fig. S1b, c) and used it to assess the expression of Pfkfb3 protein in

neurons and astrocytes. Pfkfb3 was undetectable by western blotting in neurons, whereas it was present in astrocytes (Fig. 1a, lower panels; Supplementary Information, Fig. S1d, e). Immunohistochemistry in coronal sections of rat brain cortex showed that Pfkfb3 did not colocal-ize with the neuronal nuclear marker NeuN but it did colocalize with the astrocyte marker GFAP (Fig. 1b). Thus, although Pfkfb3 mRNA is present in neurons, Pfkfb3 protein is absent, suggesting that the enzyme is downregulated post-transcriptionally in these cells.

Incubation of neurons with the proteosome inhibitors MG132 or lactacystine for 1 h resulted in accumulation of Pfkfb3 protein (Fig. 1c, upper panel). Furthermore, immunoprecipitation of Pfkfb3 in either MG132- or lactacystine-treated neurons, followed by western blotting using an anti-ubiquitin antibody, revealed an increase in Pfkfb3 ubiq-uitylation (Fig. 1c, lower panel). These results indicate that in neurons Pfkfb3 is degraded through the ubiquitin–proteosome pathway, which has been described in myogenic cells during differentiation8. We then investigated possible motifs targeting Pfkfb3 for ubiquitylation and found that Pfkfb3, but not the 1, 2 or 4 isomers, contains a KEN box at position 142 (Supplementary Information, Fig. S1f). A KEN box targets proteins for ubiquitylation by the APC/C9. Activation of APC/C requires the formation of a complex with Cdc20 or Cdh1 (ref. 10); however, Cdh1 is the only possible activator of APC/C in the terminally differentiated neurons used in this study11. Cdh1 was knocked down in neurons using short interfering RNA (siRNA) and Pfkfb3 protein was found to accu-mulate (Fig. 1d). Cdh1 protein abundance (Fig. 1d) and APC/C activity (Fig. 1e) were lower in astrocytes than in neurons, and overexpression of Cdh1 in astrocytes decreased Pfkfb3 protein (Fig. 1f). Transfection of neurons with wild-type Pfkfb3 or a mutant form in which the KEN box was altered to AAA by site-directed mutagenesis, followed by flow cytometry sorting (using green fluorescent protein (GFP)+ labelling) revealed greater accumulation of the mutant Pfkfb3 form than of the wild-type (Fig. 1g). These results demonstrate that although Pfkfb3

1Departamento de Bioquimica y Biologia Molecular, Universidad de Salamanca, Instituto de Neurociencias de Castilla y Leon, 37007 Salamanca, Spain. 2Unidad de Investigacion, Hospital Universitario de Salamanca, Instituto de Estudios de Ciencias de la Salud de Castilla y Leon, 37007 Salamanca, Spain. 3Wolfson Institute for Biomedical Research, University College London, Gower Street, London WC1E 6BT, UK.4Correspondence should be addressed either to J.P.B. or S.M. (e-mail: [email protected]; [email protected])

Received 12 December 2008; accepted 17 February 2009; published online 17 May 2009; DOI: 10.1038/ncb1881


© 2009 Macmillan Publishers Limited. All rights reserved.

L E T T E R S

transfected with Pfkfb3 is consequent to the diversion of G6P from the PPP to glycolysis.

Treatment of neurons with the cytochrome c oxidase inhibitor nitric oxide (NO) is known to cause a marked drop in mitochondrial mem-brane potential (∆ψm), which is associated with an increase in apoptosis1. This does not occur in astrocytes because they are able to upregulate glycolysis and can use glycolytically generated ATP to maintain their ∆ψm. We therefore investigated the effect of Pfkfb3 expression (at a con-centration that did not itself cause apoptosis) on the response of neurons to inhibition of mitochondrial respiration. Treatment with NO, admin-istered as DETA–NONOate, resulted in a marked drop in ∆ψm (Fig. 4a, left panel) and an increase in apoptosis (Fig. 4b, left panel). Transfection with Pfkfb3 transiently delayed, but did not prevent the onset of the NO-induced fall in ∆ψm (Fig. 4a, right panel) and apoptosis (Fig. 4b, right panel), an effect that could be abolished by silencing PGI (Fig. 4c). Similar results were obtained when antimycin A was used to inhibit the electron transport chain (Supplementary Information, Fig. S3f–h). Thus, following cellular stress induced by inhibition of mitochondrial respira-tion, transfection of neurons with Pfkfb3 activates glycolysis; however

this produces only limited protection, as glycolysis diverts glucose away from the PPP, resulting in oxidative stress and death.

Thus, Pfkfb3 is degraded by an active mechanism that seems to be physiological as the enzyme is absent in neurons in the normal rat brain. This explains the lower rate of glycolytic metabolism in neurons than in astrocytes1. Furthermore, under stress conditions, this mecha-nism prevents upregulation of glycolysis, which is normally observed in astrocytes as part of their defence response2. Indeed, we show that enhancement of glycolysis in neurons leads to their apoptotic death from oxidative stress, consequent to a decrease in the regeneration of reduced glutathione.

Our results also suggest that neuronal consumption of glucose by the PPP to maintain their antioxidant status may take priority over the use of glucose to fulfill their bioenergetic requirements, which can be met by other sources. Increasing evidence indicates that neurons can use lactate generated by astrocytes to produce energy17 and that this is not a uniform process but varies as a result of glutamatergic activation18. The fact that Pfkfb3 is subject to proteasomal degradation suggests that this mechanism is amenable to modulation, under conditions that now

a

b

c

*

*

*

*

*

*

*

*

*

*

**

*

Time (h) Time (h)

Time (h) Time (h)

0 2 4 6 80 2 4 6 8

0 2 4 6 80 2 4 6 8

Apo

ptot

ic G

FP+

neur

ons

(per

cent

age)

m in

GFP

+ ne

uron

s (p

erce

ntag

e)

Apo

ptot

ic G

FP+

neur

ons

(per

cent

age)

m in

GFP

+ ne

uron

s (p

erce

ntag

e)

120

90

60

0

20

24

18

12

0

6

NO-treated (2 h) neurons

ControlControl + NO

ControlControl + NO

Pfkfb3Pfkfb3 + NO

Pfkfb3Pfkfb3 + NO

120

100

80

0

20

60

40

120

100

80

0

20

60

40

60

50

40

0

10

30

20

60

50

40

0

10

30

20

Pfkf

b3Pf

kfb3

+PG

I siR

NA

Con

trol

Pfkf

b3Pf

kfb3

+PG

I siR

NA

Con

trol

Figure 4 Expression of Pfkfb3 transiently protects neurons from loss of mitochondrial membrane potential (∆ψm) and apoptotic death triggered by nitric oxide (NO). (a) Primary neurons were transfected with control plasmid vector (left panel) or Pfkfb3 (right panel) and then exposed to NO (released from the NO donor DETA–NONOate, 0.5 mM). NO triggered a rapid loss of ∆ψm, which was initially prevented in neurons transfected with Pfkfb3. (b) Apoptotic

death was assessed in neurons treated as in a. NO triggered apoptotic death of neurons transfected with control plasmid vector (left panel). However, in neurons transfected with Pfkfb3, the effect of NO was delayed for 4 h (right panel). (c) The protective effect of Pfkfb3 against NO-mediated loss of ∆ψm (left panel) and apoptotic death (right panel) was prevented by PGI siRNA. Results are means ± s.e.m. (n = 3). *P < 0.05 versus the corresponding control.


neuronas usando interferencias cortas RNA (siRNA) y la proteína Pfkfb3 fue encontrada para acumular. La abundancia de proteína Cdh1 (fig 1d) y la actividad de APC/C (Fig 1e) fue mas baja en astrocitos que en las neuronas, y una sobre expresión de Cdh1 en astrocitos decrementa la proteína Pfkfb3 (Fig 1f). La transfección de neuronas con Pfkfb3 tipo salvaje o una forma mutante en la que la caja de KEN fue alterada a AAA por mutagenesis directa en sitio seguida de flujocitometría de clasificación (usando

proteína verde fluorescente etiquetada como GFP) reveló mayores acumulaciones de forma Pfkfb3 mutante que el tipo salvaje (fig 1g). Estos resultados demuestran que aunque Pfkfb3 esta sujeta a degradación por la acción de APC/C-Cdh1 en neuronas, Pfkfb3 proteína es estable en astrocitos como un resultado de su baja actividad de APC/C-Cdh1. Los resultados también indican que represión post-transcripcional de Pfkfb1,2 y 4 en neuronas, si esto ocurre, es a travéss de un mecanismo que es independiente de APC/C-Cdh1.


L E T T E R S

is subject to degradation by the action of APC/C–Cdh1 in neurons, Pfkfb3 protein is stable in astrocytes as a result of their low APC/C–Cdh1 activity. The results also indicate that post-transcriptional repression of Pfkfb1, 2 and 4 in neurons, if it occurs, is through a mechanism that is independent of APC/C–Cdh1.

We then investigated whether Cdh1 regulates glycolysis in neurons. Silencing Cdh1 increased both the glycolytic flux (Fig. 2a; Supplementary Information, Fig. S2a) and the concentration of intracellular lactate (Fig. 2b; Supplementary Information, Fig. S2b). These effects were abol-ished by co-silencing phosphoglucose isomerase (PGI), the glycolytic enzyme responsible for the formation of fructose-6-phosphate (F6P), which is the substrate of Pfk1 (Fig. 2a; Supplementary Information, Fig. S2c,d). The increase in the glycolytic flux induced by silencing Cdh1 was also abolished by co-silencing Pfkfb3 (Fig. 2a). Thus the low glyco-lytic rate in neurons can be accounted for by APC/C–Cdh1-mediated degradation of Pfkfb3.

Next we investigated whether an increase in the basal glycolytic rate would render neurons more resistant to stress. Glycolysis was increased initially by silencing Cdh1; however, this treatment enhanced neuronal apoptotic death, as assessed by flow cytometry of annexin V+/7-AAD– cells (Fig. 2c; Supplementary Information, Fig. S2e). The apoptosis caused by silencing Cdh1 was partially prevented by silencing PGI (Fig. 2c) or Pfkfb3 (Fig. 2d). Cyclin B1 also mediates, in part, apopto-sis induced by silencing Cdh1 in neurons11 (Fig. 2d), but co-silencing cyclin B1 and Pfkfb3 fully abolished apoptosis in Cdh1 siRNA neurons (Fig. 2d). Thus, silencing Cdh1 in these cells triggers apoptosis through both cyclin B1 (ref. 11) and Pfkfb3. Neurons were then transfected with the full-length Pfkfb3 cDNA and sorted by flow cytometry using GFP (Supplementary Information, Fig. S2f, g). This resulted in an increase in glycolytic flux (Fig. 2a) and an accumulation of lactate, which was pre-vented by silencing PGI (Fig. 2e). Expression of Pfkfb3 increased apopto-sis in neurons in a concentration- (Fig. 2f) and time- (Fig. 2g) dependent

a b

Pfkfb3

Cyclophilin

Pfkfb3

Map2

GFAP

GAPDH

Pro

tein

mR

NA

Mr(K)

Mr(K)

Mr(K) Mr(K)

63

280

50

37

NeuN Pfkfb3 DAPI

GFAP Pfkfb3 DAPI Merge

Merge

c

Pfkfb3

GAPDH

IP: anti-Pfkfb3

GAPDHWB: anti-Ub

63

180

115

82

37

55

63

d e f g

Pfkfb3

Cdh1

Cdh1 siRNA

Neurons

63

37

54

+

Astrocytes Neurons

35S

-Cyc

lin B

1

Long

Short

min0 20 60 0 20 60NeuronsAstrocytes

Cdh1

Pfkfb3

GAPDH GAPDH

Pfkfb3

Cdh1+

Pfk

fb3

Pfk

fb3 m

ut

Neu

rons

Astro

cyte

sLa

ct

Non

eM

G

–– Ast

r

Figure 1 Pfkfb3 protein is degraded through the ubiquitin–proteasome pathway mediated by APC/C–Cdh1 in rat cortical neurons but not in astrocytes. (a) Upper panel: northern blotting of total RNA extracts from terminally differentiated rat cortical neurons and astrocytes showed that similar amounts of Pfkfb3 mRNA were expressed in each cell type. Cyclophilin was used as an mRNA loading marker. Lower panel: western blotting using an anti-Pfkfb3 antibody revealed the absence of Pfkfb3 protein expression in neurons, whereas Pfkfb3 was expressed abundantly in astrocytes. Map2 was used as a neuronal marker, GFAP as a glial marker and GADPH as a loading control. (b) Fluorescence microphotographs of coronal sections of rat brain cortex show immunohistochemically the absence of Pfkfb3 in neurons and its presence in astrocytes. Neurons and glial cells were identified using the specific markers NeuN and GFAP, respectively. Nuclei were identified using DAPI. Scale bars, 20 μm. (c) Incubation of neurons with the proteasome inhibitors lactacystine or MG132 for 1 h resulted in

Pfkfb3 accumulation, shown by western blotting. Immunoprecipitation of protein extracts with an anti-Pfkfb3 antibody, followed by western blotting against ubiquitin (Ub) showed increased Pfkfb3 protein ubiquitylation in the lactacystine- and MG132-treated neurons. (d) Silencing Cdh1 in neurons at 3 days in vitro resulted in significant knockdown of Cdh1 protein and accumulation of Pfkfb3 (after 3 days) to approximately 50% of the level found in astrocytes, as assessed by densitometry; astrocytes express very little Cdh1 protein. (e) APC/C activity, measured as the ability to ubiquitylate 35S-cyclin B1, is lower in astrocytes than in neurons (the arrow indicates the most abundant ubiquitylated form of cyclin B1; ‘long’ and ‘short’ indicate the exposure time of the film: short exposure allows visualization of the corresponding decrease in 35S-cyclin B1). (f) Overexpression of Cdh1 in astrocytes decreases Pfkfb3 protein. (g) Wild-type Pfkfb3, but not a site-directed mutant form (142KEN to 142AAA, Pfkfb3mut), is degraded when expressed in neurons using low amounts of cDNA.

2 NATURE CELL BIOLOGY ADVANCE ONLINE PUBLICATION


L E T T E R S






a

b

c

*

*

*

*

*

*

*

*

*

*

**

*

Time (h) Time (h)

Time (h) Time (h)

0 2 4 6 80 2 4 6 8

0 2 4 6 80 2 4 6 8

Apo

ptot

ic G

FP+

neur

ons

(per

cent

age)

m in

GFP

+ ne

uron

s (p

erce

ntag

e)

Apo

ptot

ic G

FP+

neur

ons

(per

cent

age)

m in

GFP

+ ne

uron

s (p

erce

ntag

e)

120

90

60

0

20

24

18

12

0

6


ControlControl + NO

ControlControl + NO

Pfkfb3Pfkfb3 + NO

Pfkfb3Pfkfb3 + NO

120

100

80

0

20

60

40

120

100

80

0

20

60

40

60

50

40

0

10

30

20

60

50

40

0

10

30

20

Pfkf

b3Pf

kfb3

+PG

I siR

NA

Con

trol

Pfkf

b3Pf

kfb3

+PG

I siR

NA

Con

trol




Entonces nosotros investigamos si Cdh1 regu lada la g lucó l i s i s en neuronas . Silenciando Cdh1 incremento tanto el flujo g l u c o l í t i c o ( F i g . 2 a ; I n f o r m a c i ó n S u p l e m e n t a r i a , F i g . S 2 a ) c o m o l a concentración del lacato intracelular (Fig. 2b; Información Suplementaria, Fig. S2b). Estos efectos son abolidos por co-reducida fosfoglucosa isomerasa (PGI), la enzima glucolítica responsable para la formación de fructosa-6-fosfato (F6P), que es el sustrato de Pfk1 (Fig. 2a; Información Suplementaria, Fig. S2c,d). El incremento en el flujo glucolítico inducido por el disminuido (o reducido) Cdh1 fue también abolido por la co-reducida Pfkfb3 (Fig. 2a). Así el bajo valor glucolitico en n e u r o n a s p u e d e e x p l i c a r s e p o r l a degradación de APC/C-Cdh1-mediada a Pfkfb3.

Después, nosotros investigamos si un incremento en la proporción basal glucolítca haría a las neuronas más resistentes al estrés. La glucólisis fue incrementada inicialmente por Cdha1 disminuido ; sin embargo, este tratamiento mejoró la muerte apoptótica

neuronal, según lo evaluado por la cotometría de flujo de la anexina V+/7-AAD- celular (Fig. 2c; Información Suplementaria, Fig. S2e). La apoptosis causada por Cdh1 disminuido fue parcialmente impedida por PGI disminuido (Fig. 2c) o Pfkfb3 (Fig. 2d). Ciclina B1 también mediada, en parte , indujo apoptosis por Cdh1 disminuido en neuronas11 (Fig. 2d), pero co-disminuyendo ciclina B1 y Pfkfb3 fue completamente abolida la apoptosis en neuronas Cdh1 siRNA (Fig. 2d). Así, el Cdh1 disminuido en estas células desencadena apoptosis a través de ambos, ciclina B1y Pfkfb3. Las neuronas fueron transfectadas con Pfkfb3 cDNA completa y ordenadas por citometría de flujo usando GFP (Información Suplementaria, Fig. S2f, g). Esto resulto en un incremento en el flujo glucolítico (Fig. 2a) y una acumulación de lactato, que fue impedida por el PGI disminuido (Fig. 2e). Expresiones de Pfkfb3 aumentó la apoptosis en neuronas en una concentración -(Fig. 2f) y tiempo- (Fig. 2g) de manera dependiente, pero también tuvo efecto sobre apoptosis en astrocitos (Información Suplementaria, Fig. S3a).



L E T T E R S






a

b

c

*

*

*

*

*

*

*

*

*

*

**

*

Time (h) Time (h)

Time (h) Time (h)

0 2 4 6 80 2 4 6 8

0 2 4 6 80 2 4 6 8

Apo

ptot

ic G

FP+

neur

ons

(per

cent

age)

m in

GFP

+ ne

uron

s (p

erce

ntag

e)

Apo

ptot

ic G

FP+

neur

ons

(per

cent

age)

m in

GFP

+ ne

uron

s (p

erce

ntag

e)

120

90

60

0

20

24

18

12

0

6


ControlControl + NO

ControlControl + NO

Pfkfb3Pfkfb3 + NO

Pfkfb3Pfkfb3 + NO

120

100

80

0

20

60

40

120

100

80

0

20

60

40

60

50

40

0

10

30

20

60

50

40

0

10

30

20

Pfkf

b3Pf

kfb3

+PG

I siR

NA

Con

trol

Pfkf

b3Pf

kfb3

+PG

I siR

NA

Con

trol




Expresión de la cinasa, pero no la bisfosfatasa domino de Pfkfb3 imitado este efecto sobre la apoptosis en neuronas. (Fig. 2h). Silenciando PGI abolimos la apoptosis disparada por la expresión de cualquiera de las completas o de la cinasa dominante de Pfkfb3 (Fig. 2h) La apoptosis en neuronas resultando desde la expresión de Pfkfb3 es causada por la activación de la vía intrínseca apoptótica, como demostró por el uso de inhibidores se lect ivos de casposas ( In formación Suplementaria, Fig. S3b). Así, aunque la sobreexpresión de Pfkfb3 en neuronas activa la glucólisis, disparando concomitantemente la muerte apoptótica. Para entender la razón o las razones por muerte apoptótica de las neuronas siguiendo la Pfkfb3- mediada en la activación de la

glucólisis, nosotros decidimos investigar si tal activación afecta el valor de glucosa utilizado por la vía pentosa-fosfato (PPP), que es metabólicamente ligada a la glucólisis a través del intermediario común glucosa-6-fosfato (G6P)12 . Para establecer si la modulación de la glucólisis en neuronas affecta el valor de la oxidacion de G6P a través del PPP, las neuronas fueron transferidas con Pkfkfb3 y células GFP+ son ordenadas por citometría de flujo para accesar la actividad PPP13,14. Dentro del control de neuronas el valor del consumo de glucosa a través de PPP (Fig. 3a) fue doblada a través de glucólisis (Información Suplementaria, Fig. S2a). La transfección de Pfkfb3 inhibió el consumo de glucosa través de PPP en neuronas por aproximadamente el 50% (Fig. 3a); este efecto


L E T T E R S

manner, but had no effect on apoptosis in astrocytes (Supplementary Information, Fig. S3a). Expression of the kinase, but not the bisphos-phatase domain of Pfkfb3 mimicked its effect on apoptosis in neurons (Fig. 2h). Silencing PGI abolished the apoptosis triggered by expression of either the full-length or the kinase domain of Pfkfb3 (Fig. 2h). The apoptosis in neurons resulting from expression of Pfkfb3 was caused by activation of the intrinsic apoptotic pathway, as demonstrated by the use of selective inhibitors of caspases (Supplementary Information, Fig. S3b). Thus, although overexpression of Pfkfb3 in neurons activates glycolysis, it concomitantly triggers apoptotic death.

To understand the reason(s) for apoptotic death of neurons follow-ing Pfkfb3-mediated activation of glycolysis, we decided to investigate whether such activation affected the rate of glucose utilization by the

pentose-phosphate pathway (PPP), which is metabolically linked to glycolysis through the common intermediate glucose-6-phosphate (G6P)12. To establish whether modulation of glycolysis in neurons affects the rate of G6P oxidation through the PPP, neurons were transfected with Pfkfb3 and GFP+ cells were sorted by flow cytom-etry to assess PPP activity13,14. In control neurons the rate of glucose consumption through the PPP (Fig. 3a) was double that through gly-colysis (Supplementary Information, Fig. S2a). Pfkfb3 transfection inhibited glucose consumption through the PPP in neurons by about 50% (Fig. 3a); this effect was abolished by silencing PGI, indicating that such inhibition was a consequence of the activation of glycoly-sis (Fig. 3a). Overexpression of the full-length cDNA coding for glu-cose-6-phosphate dehydrogenase (G6PD; the rate-limiting step of the

a6-

14C

-Glu

cose

con

vers

ion

to 14

C-la

ctat

e(n

mol

/ h

x 10

6 neu

rons

)

Cdh1 siRNAPGI siRNA

Pfkfb3 shRNAPfkfb3

b c d

e f g h

Control siRNA

PGI siRNA

– + + + –– – + – –– – – + –

– – – – +

0.9

0.6

0.3

0

**

* Intr

acel

lula

r lac

tate

(µm

ol/1

06 neu

rons

)

Intr

acel

lula

r lac

tate

(µm

ol /

106 n

euro

ns)

Apo

ptot

ic n

euro

ns (p

erce

ntag

e)

Apo

ptot

ic G

FP+

neur

ons

(per

cent

age)

1.5

1.0

0.5

0

40

30

20

0

10

*

*1.5

1.0

0.5

0

Apo

ptot

ic n

euro

ns (p

erce

ntag

e)

40

30

20

0

10

40

30

20

0

10

40

30

20

0

10

40

30

20

0

10

Apo

ptot

ic G

FP+

neur

ons

(per

cent

age)

Apo

ptot

ic G

FP+

neur

ons

(per

cent

age)

*

*

*

*

*

*

Control siRNA

PGI siRNAControlPfkfb3

GFP–

GFP+

Pfkfb3 (µg mL–1) Days post-transfection

0 0.8 1.6 0 1 2 3

*

*

*

*

*

# #

#

#

Cdh1 siRNACyclin B1 shRNA

Pfkfb3 shRNA

– + + + +– – + – +– – – + +C

ontro

l

Con

trol

Cdh

1 si

RNA

Cdh

1 si

RNA

Cdh

1 si

RNA

+ PG

I siR

NA

Con

trol

Pfkf

b3Pf

kfb3

+PG

I siR

NA

Con

trol

Pfkf

b3Ki

nase

BPas

e

Figure 2 Cdh1 downregulates glycolysis and protects against apoptotic death through Pfkfb3 degradation in neurons. (a) Transfection with Cdh1 siRNA in neurons significantly increased the rate of glycolysis, measured as the rate of 6-14C-glucose conversion to 14C-lactate; this effect was prevented by co-silencing PGI or Pfkfb3, and was mimicked by overexpression of Pfkfb3. Left histogram bar corresponds to control (luciferase) siRNA/shRNA. (b) Silencing Cdh1 increased intracellular lactate concentrations, an effect that was prevented by co-silencing PGI. (c) Cdh1 siRNA increased the proportion of apoptotic (annexin V+/7-AAD–) neurons, an effect that was partially prevented by co-silencing PGI. (d) Apoptotic death in Cdh1-silenced neurons was partially prevented by co-silencing either cyclin B1 or Pfkfb3 and was abolished by silencing both. Left histogram bar corresponds to control (luciferase) siRNA/shRNA.

(e) Transfection of full-length Pfkfb3 cDNA in neurons significantly increased intracellular lactate concentrations, as assessed in GFP+ neurons after sorting by flow cytometry; this effect was prevented by PGI siRNA. (f, g) Transfection of neurons with Pfkfb3 induced apoptotic death in a concentration- (f) and time- (g) dependent manner. Apoptosis was assessed by the presence of annexin V+/7-AAD– cells measured by flow cytometry within the GFP+ population. (h) Apoptotic death was also observed following transfection of Pfkfb3 and of a truncated form of Pfkfb3 containing the kinase domain. Transfection of the bisphosphatase Pfkfb3 domain did not affect apoptosis. The increase in apoptotic death in neurons caused by Pfkfb3 or the kinase domain was prevented by silencing PGI. Results are mean ± s.e.m. (n = 3). *P < 0.05 versus the corresponding control; #P < 0.05 versus Cdh1 siRNA (d).



L E T T E R S






a

b

c

*

*

*

*

*

*

*

*

*

*

**

*

Time (h) Time (h)

Time (h) Time (h)

0 2 4 6 80 2 4 6 8

0 2 4 6 80 2 4 6 8

Apo

ptot

ic G

FP+

neur

ons

(per

cent

age)

m in

GFP

+ ne

uron

s (p

erce

ntag

e)

Apo

ptot

ic G

FP+

neur

ons

(per

cent

age)

m in

GFP

+ ne

uron

s (p

erce

ntag

e)

120

90

60

0

20

24

18

12

0

6


ControlControl + NO

ControlControl + NO

Pfkfb3Pfkfb3 + NO

Pfkfb3Pfkfb3 + NO

120

100

80

0

20

60

40

120

100

80

0

20

60

40

60

50

40

0

10

30

20

60

50

40

0

10

30

20

Pfkf

b3Pf

kfb3

+PG

I siR

NA

Con

trol

Pfkf

b3Pf

kfb3

+PG

I siR

NA

Con

trol




fue abolido por el PGI silenciado, indicando que tal inhibición fue una consecuencia de la activación de la glucólisis (Fig. 3a). La s o b r e e x p r e s i ó n c o m p l e t a d e l c D N A c o d i fi c a n d o p o r g l u c o s a - 6 - f o s f a t o deshidrogenasa (G6PD; el valor-limitante al paso de PPP) incremento significativamente la actividad PPP, como como lo hizo la sobreexpresión de la glucosa transportadora GLUT1 (Fig. 3a.).

Más ade lan te , e l i nc remen to en l a acumulación de lactato y la apoptosis inducida por la transfección de Pfkfb3 puede ser prevenida por la co-expresión de G6PD (Fig.

3b, c). Estos resultados sugieren que dentro de las neuronas una porción significativa de G6P es dirigida hacia el PPP y entonces la regulación por incremento de Pfkfb3 es desviada hacia la glucólisis. En astrocitos, sin embargo, a pesar de la considerable actividad del PPP, hay un sustancial metabolismo de g lucosa po r g l ucó l i s i s ( I n fo rmac ión Suplementaria Fig. S3c), consistente con la mayor expresión observada de ambos G9PD (Información Suplementaria, Fig. S3d) y Pfkfb3

(Fig. 1a), cuando se comparó con las neuronas.


L E T T E R S

PPP) significantly increased PPP activity, as did overexpression of the glucose transporter GLUT1 (Fig. 3a). Furthermore, the increase in lactate accumulation and the apoptosis induced by Pfkfb3 transfection could be prevented by co-expression of G6PD (Fig. 3b, c). These results suggest that in neurons a significant proportion of G6P is directed towards the PPP and that Pfkfb3 upregulation diverts it towards glyco-lysis. In astrocytes, however, despite considerable PPP activity14, there is substantial metabolism of glucose by glycolysis (Supplementary Information, Fig. S3c), consistent with the observed higher expres-sion of both G6PD (Supplementary Information, Fig. S3d) and Pfkfb3 (Fig. 1a) when compared with neurons.

A major function of the PPP is the regeneration of reduced glu-tathione at the expense of NADPH(H+)7,13 to provide neuroprotec-tion14–16. We found that Pfkfb3 transfection increased the oxidation of glutathione (Fig. 3d), indicating oxidative stress. We therefore meas-ured the formation of reactive oxygen species (ROS). Pfkfb3 expres-sion enhanced the formation of ROS in neurons, an effect that was prevented either by silencing PGI or by co-expressing G6PD (Figs 3e, f; Supplementary Information, Fig. S3e). Moreover, incubation of neu-rons with a plasma-membrane-permeable form of glutathione (glu-tathione ethyl ester) prevented apoptosis caused by Pfkfb3 expression (Fig. 3g). These results indicate that production of ROS in neurons

a b c d

e f g

nmol

/ h

x 10

6 neu

rons

1.2

0.8

0.4

0

40

30

20

0

10

PPP activityGSSG GSH

1.8

1.2

0.6

0

nmol

/ 10

6 neu

rons

Apo

ptot

ic n

euro

ns (p

erce

ntag

e)

Intracellularlactate

nmol

mg–1

pro

tein

nmol

mg–1

pro

tein

2.0

1.5

1.0

0

0.5

20

15

10

0

5

GFP-

GFP+

G6PD (µg mL–1)

0 0.4 0.8 1.2 1.6

*

*

*

*

*

**

None+GSH-EE

Apo

ptot

ic G

FP+

neur

ons

(per

cent

age)

30

20

10

0

30

20

10

0

Mito

Sox

Red

fluo

resc

ence

inG

FP+

neur

ons

(arb

itrar

y un

its)

GFP HEt Merge

Pfkfb3

Pfkfb3+ PGI siRNA

Pfkfb3+ G6PD

***

*

Con

trol

Con

trol

Pfkf

b3

Pfkf

b3

Pfkf

b3+

PGI s

iRN

AG

6PD

Pfkf

b3 +

G6P

D

GLU

T1

Con

trol

Pfkf

b3

Con

trol

Pfkf

b3

Con

trol

Con

trol

Pfkf

b3

Pfkf

b3

Pfkf

b3+

PGI s

iRN

APf

kfb3

+G

6PD

Figure 3 Pfkfb3 expression in neurons triggers a decrease in glucose oxidation through the PPP, causing oxidative stress. (a) The rate of PPP activity, measured as the difference in the rates of oxidation of 1-14C- and 6-14C-glucose, was reduced in neurons transfected with Pfkfb3; this effect was prevented by PGI siRNA. Expression of a plasmid vector encoding the full-length cDNA of G6PD, or the glucose transporter GLUT1 increased the rate of 1-14C-glucose oxidation. These measurements were performed in GFP+ cells sorted by flow cytometry. (b) Transfection of Pfkfb3 in neurons resulted in increased intracellular lactate concentrations, an effect that was prevented by co-expressing G6PD. These measurements were performed in GFP+ cells sorted by flow cytometry. (c) Neuronal apoptotic death (annexin V+/7-AAD–) caused by Pfkfb3 expression was prevented by co-expression of G6PD. GFP+ neurons indicate those efficiently transfected with the Pfkfb3 cDNA construct, whereas GFP– neurons represent

those that were not transfected. (d) Pfkfb3 expression in neurons followed by flow cytometric sorting of the GFP+ neurons revealed an increase of about 4-fold in the intracellular concentration of GSSG and a corresponding decrease in reduced glutathione (GSH). (e) Pfkfb3 expression in neurons increased the intensity of hydroethidine (HEt) fluorescence in the GFP+ population, but not in the GFP– cells. Increased HEt fluorescence was abolished either by silencing PGI (PGI siRNA) or by G6PD expression. Scale bars, 20 μm. (f) Pfkfb3 expression in neurons increased the intensity of MitoSox fluorescence in the GFP+ population, but not in the GFP– cells; this effect was abolished either by silencing PGI (PGI siRNA) or by G6PD expression. (g) Incubation of neurons with a plasma-permeable form of glutathione (glutathione ethyl ester, GSH-EE) prevented apoptotic death caused by Pfkfb3 expression. Results are means ± s.e.m. (n = 3). *P < 0.05 versus the corresponding control.

4 NATURE CELL BIOLOGY ADVANCE ONLINE PUBLICATION


L E T T E R S






a

b

c

*

*

*

*

*

*

*

*

*

*

**

*

Time (h) Time (h)

Time (h) Time (h)

0 2 4 6 80 2 4 6 8

0 2 4 6 80 2 4 6 8

Apo

ptot

ic G

FP+

neur

ons

(per

cent

age)

m in

GFP

+ ne

uron

s (p

erce

ntag

e)

Apo

ptot

ic G

FP+

neur

ons

(per

cent

age)

m in

GFP

+ ne

uron

s (p

erce

ntag

e)

120

90

60

0

20

24

18

12

0

6


ControlControl + NO

ControlControl + NO

Pfkfb3Pfkfb3 + NO

Pfkfb3Pfkfb3 + NO

120

100

80

0

20

60

40

120

100

80

0

20

60

40

60

50

40

0

10

30

20

60

50

40

0

10

30

20

Pfkf

b3Pf

kfb3

+PG

I siR

NA

Con

trol

Pfkf

b3Pf

kfb3

+PG

I siR

NA

Con

trol




Una mayor función de el PPP es la regeneración reducida de glutatión a expensas d e N A D P H ( H + ) 7 , 1 3 p a r a p r o v e e r neuroprotección14-16. Encontramos que la transfección de Pfkfb3 incrementó la oxidación de glutatión (Fig. 3D), indicando estrés oxidativo. Por lo tanto, medimos la formación de de especies reactivas al oxigeno (ROS). La expresión de Pfkfb3 mejoró la formación de ROS en las neuronas, un efecto que también

fue impedido por el silenciado de PGI o por co-expresado G6PD (Figs 3e,f; Información Suplementaria, Fig. S3e). Mas allá, la incubación de neuronas con una membrana de plasma permeable a partir de Glutatión (glutaniónetilester) impidió la apoptosis causada por la expresión de Pfkfb3 (Fig. 3g). Estos resultados indican que la producción de ROS en neuronas transfectadas con Pfkfb3 son consecuencia de la desviación de G6P desde el PPP hasta la glucólisis.

El tratamiento de neuronas con el citocromo c oxidasa inhibidor oxido nitrico (NO) es conocido por causar una marcado descenso en la potencial de la membrana mitocondrial (∆ψm), que es asociado con un incremente en apoptosis. Esto no ocurre en astrocitos porque el lo son capaces de regular positivamente la glucólisis y pueden usar ATP glucolíticamente generado para mantener su ∆ψm. Por tanto, investigamos el efecto de la expresión de Pfkfb3 (a una concentración que

no causara en si mismo la apoptosis) en la repuesta de neuronas para la inhibición de la respiración mitocondrial. El tratamiento con NO, administrado como DETA-NONOate, resultó dentro del descenso marcado en ∆ψm (Fig. 4a, panel izquierdo), un efecto que podría ser abolido por el PGI silenciado (Fig. 4c). Resultados similares fueron obtenidos cuando antimicina A (antimycin A) fue usado para inhibir la cadena transportadora de electrones (Información Suplementaria, Fig. S3f-h).


L E T T E R S






a

b

c

*

*

*

*

*

*

*

*

*

*

**

*

Time (h) Time (h)

Time (h) Time (h)

0 2 4 6 80 2 4 6 8

0 2 4 6 80 2 4 6 8

Apo

ptot

ic G

FP+

neur

ons

(per

cent

age)

m in

GFP

+ ne

uron

s (p

erce

ntag

e)

Apo

ptot

ic G

FP+

neur

ons

(per

cent

age)

m in

GFP

+ ne

uron

s (p

erce

ntag

e)

120

90

60

0

20

24

18

12

0

6


ControlControl + NO

ControlControl + NO

Pfkfb3Pfkfb3 + NO

Pfkfb3Pfkfb3 + NO

120

100

80

0

20

60

40

120

100

80

0

20

60

40

60

50

40

0

10

30

20

60

50

40

0

10

30

20

Pfkf

b3Pf

kfb3

+PG

I siR

NA

Con

trol

Pfkf

b3Pf

kfb3

+PG

I siR

NA

Con

trol





L E T T E R S






a

b

c

*

*

*

*

*

*

*

*

*

*

**

*

Time (h) Time (h)

Time (h) Time (h)

0 2 4 6 80 2 4 6 8

0 2 4 6 80 2 4 6 8

Apo

ptot

ic G

FP+

neur

ons

(per

cent

age)

m in

GFP

+ ne

uron

s (p

erce

ntag

e)

Apo

ptot

ic G

FP+

neur

ons

(per

cent

age)

m in

GFP

+ ne

uron

s (p

erce

ntag

e)

120

90

60

0

20

24

18

12

0

6


ControlControl + NO

ControlControl + NO

Pfkfb3Pfkfb3 + NO

Pfkfb3Pfkfb3 + NO

120

100

80

0

20

60

40

120

100

80

0

20

60

40

60

50

40

0

10

30

20

60

50

40

0

10

30

20

Pfkf

b3Pf

kfb3

+PG

I siR

NA

Con

trol

Pfkf

b3Pf

kfb3

+PG

I siR

NA

Con

trol




Así, siguiendo el estrés celular inducido por la inhibición de la respiración mitocondrial, la transfección de neuronas con Pfkfb3 activa la glucól isis; sin embargo esto produce solamente una protección limitada, como la glucólisis desvía glucosa lejos de la PPP, resultando en estrés oxidativo y muerte.

En consecuencia, Pfkfb3 es degradada por un mecanismo activo que parece ser fisiológico donde la enzima esta ausente en neuronas dentro de los valores normales cerebrales. Esto explica el bajo valor del metabolismo glucolitico en neuronas que están en astrocitos1. Además, bajo condiciones de estrés, este mecanismo impide la regulación por incremento ( o regulación positiva) de glucólisis, que es normalmente observada en astrocitos como parte de su respuesta de defensa2. En efecto, mostramos que el realce de glucólisis en neuronas lleva a su muerte apoptótica a partir de estrés oxidativo, c o n s e c u e n t e a u n d e s c e n s o e n l a regeneración del glutanión reducido.

Nuestros resultados también sugieren que el consumo neuronal de glucosa por PPP para mantener sus estado antioxidante puede adoptar prioridad sobre el uso de glucosa para cumplir sus requerimientos de bioenergéticos, que pueden ser tomados por otras fuentes. Creciente evidencia indica que las neuronas pueden usar lactato generado por astrocitos para producir energía17 y que esto no es un proceso uniforme pero varia como resultado de la activación glutamatérgica18. El hecho que Pfkfb3 este sujeto a degradación proteasomal sugiere que el mecanismo es susceptible a modulación, bajo condiciones que por ahora requieren investigación. Ciertamente, es probable que el aumento del uso de lactato por neuronas es acoplado a un incremento en su regeneración para reducir glutatión a partir de glucosa. La regulación de la estabilidad de Pfkfb3 por la APC/C-Cdh1, vía que nosotros ahora describimos, podría subrayar este proceso.



L E T T E R S






a

b

c

*

*

*

*

*

*

*

*

*

*

**

*

Time (h) Time (h)

Time (h) Time (h)

0 2 4 6 80 2 4 6 8

0 2 4 6 80 2 4 6 8

Apo

ptot

ic G

FP+

neur

ons

(per

cent

age)

m in

GFP

+ ne

uron

s (p

erce

ntag

e)

Apo

ptot

ic G

FP+

neur

ons

(per

cent

age)

m in

GFP

+ ne

uron

s (p

erce

ntag

e)

120

90

60

0

20

24

18

12

0

6


ControlControl + NO

ControlControl + NO

Pfkfb3Pfkfb3 + NO

Pfkfb3Pfkfb3 + NO

120

100

80

0

20

60

40

120

100

80

0

20

60

40

60

50

40

0

10

30

20

60

50

40

0

10

30

20

Pfkf

b3Pf

kfb3

+PG

I siR

NA

Con

trol

Pfkf

b3Pf

kfb3

+PG

I siR

NA

Con

trol




REFERENCIAS


5

each step, the sections were carefully rinsed three times each for 10 minutes in PB.

After rinsing, sections were mounted with an antifading medium. Sections were

examined using both a microscope (Provis AX70; Olympus, Tokyo, Japan) equipped

with epifluorescence and appropriated filters sets, and a confocal microscope (TCS SP2;

Leica, Mannheim, Germany).

Accession numbers. Pfkfb1 (NM_012621), Pfkfb2 (NM_080477), Pfkfb3

(NM_057135), Pfkfb4 (NM_019333), PGI (NM_207192), Cdh1 (NM_016263), and

cyclin B1 (AY338491).

References

1. Watanabe, F., Sakai, A. & Furuya, E. Novel isoforms of rat brain fructose 6-

phosphate 2-kinase/fructose 2,6-bisphosphatase are generated by tissue-specific

alternative splicing. J Neurochem 69, 1-9 (1997).

2. Hammerstedt, R. H. The use of Dowex-1-borate to separate 3HOH from 2-3H-

glucose. Anal Biochem 56, 292-3 (1973).

3. Huang, M. & Veech, R. L. The quantitative determination of the in vivo

dephosphorylation of glucose 6-phosphate in rat brain. J Biol Chem 257, 11358-

63 (1982).

4. Thauer, R. K., Rupprecht, E. & Jungermann, K. Separation of 14C-formate from

CO2 fixation metabolites by isoionic-exchange chromatography. Anal Biochem

38, 461-8 (1970).

5. Katz, J., Rognstad, R. & Kemp, R. G. Isotope Discrimination Effects in the

Metabolism of Tritiated Glucose. J Biol Chem 240, PC1484-6 (1965).

6. Hughes, S. D., Quaade, C., Johnson, J. H., Ferber, S. & Newgard, C. B.

Transfection of AtT-20ins cells with GLUT-2 but not GLUT-1 confers glucose-

stimulated insulin secretion. Relationship to glucose metabolism. J Biol Chem

268, 15205-12 (1993).

7. Bergmeyer, H. U., Bernt, E., Schmidt, F. & Stork, H. in Methods of Enzymatic

Analysis (ed. Bergmeyer, H. U.) 1196-1201 (Verlag Chemie GmbH, Weinheim,

1974).

8. Michal, G. in Methods of Enzymatic Analysis (eds. Bergmeyer, J. & Graßl, M.)

191-198 (Verlag Chemie, Weinheim, Deerfield Beach (FL), Basel, 1985).

9. Michal, G. in Methods of Enzymatic Analysis (eds. Bergmeyer, J. & Graßl, M.)

342-350 (Verlag Chemie, Weinheim, Deerfield Beach (FL), Basel, 1985).

10. Gutmann, I. & Wahlefeld, A. W. in Methods of Enzymatic Analysis (ed.

Bergmeyer, H. U.) 1464-1468 (Verlag Chemie GmbH, Weinheim, 1974).

11. Van Schaftingen, E., Lederer, B., Bartrons, R. & Hers, H. G. A kinetic study of

pyrophosphate: fructose-6-phosphate phosphotransferase from potato tubers.

Application to a microassay of fructose 2,6-bisphosphate. Eur J Biochem 129,

191-195 (1982).

12. Tietze, F. Enzyme method for quantitative determination of nanogram amounts

of total and oxidized glutathione: application to mammalian blood and other

tissues. Anal. Biochem. 27, 502-522 (1969).

13. Kimata, Y. et al. A mutual inhibition between APC/C and its substrate Mes1

required for meiotic progression in fission yeast. Dev Cell 14, 446-54 (2008).


L E T T E R S






a

b

c

*

*

*

*

*

*

*

*

*

*

**

*

Time (h) Time (h)

Time (h) Time (h)

0 2 4 6 80 2 4 6 8

0 2 4 6 80 2 4 6 8

Apo

ptot

ic G

FP+

neur

ons

(per

cent

age)

m in

GFP

+ ne

uron

s (p

erce

ntag

e)

Apo

ptot

ic G

FP+

neur

ons

(per

cent

age)

m in

GFP

+ ne

uron

s (p

erce

ntag

e)

120

90

60

0

20

24

18

12

0

6


ControlControl + NO

ControlControl + NO

Pfkfb3Pfkfb3 + NO

Pfkfb3Pfkfb3 + NO

120

100

80

0

20

60

40

120

100

80

0

20

60

40

60

50

40

0

10

30

20

60

50

40

0

10

30

20

Pfkf

b3Pf

kfb3

+PG

I siR

NA

Con

trol

Pfkf

b3Pf

kfb3

+PG

I siR

NA

Con

trol




enzimas alostericas

Documents