j . , . i JL

Barry S. Wilson*, Sem H. Phan and Ric;

,artment of Pathology, University of Michigan Medical School, An~

[eceived 21 August 1985; revised manuscript received 15 October

hmary

method is presented for the purification of human c Lds obtained at autopsy. The procedure involved hom~ queous buffer, salt precipitation, affinity chromatogr~ loclonal antibody (LK2H10) and reverse-phase high ~hy. Chromogranin purified from autopsy adrenal gl~ ~olypeptide heterogeneity when analyzed by silver-st

~ r ~ t e r t h a n Q{'l°/~ n f the~ n r n t e i n w ~ r e ~ n r e ~ n t e d ]"

chromogranin from adrenal homogenization of whole glands

,graphy using a highly specific high-pressure liquid chromato-

lands revealed a high degree silver-stained SDS polyacrylamide

the protein was represented by a cluster of polypeptides :hromogranin A), while the remaining protein was highly ~t. That these various polypeptides were in fact chromo- ,~rn blotting using monoclonal antibody LK2H10. About vere obtained from 97 g of starting adrenals which was d and a 250-fold enrichment from adrenal homogenates. Lable yields of this protein was the need for particular low a of chromogranin after solvent removal steps. Chromo- n adrenal glands was similar in amino acid composition, Linal amino acid sequence (24 residues) to bovine chrom- :luence representing 25% of the total protein and missing the N-terminus suggested the possibility of N-terminal n in situ. The conservation of the N-terminal amino acid

eprint requests: Barry S. Wilson, Department of Cell Biology, Hybritech )iego, CA 92121, U.S.A. Tel.: 619-455-6700.

.,vier Science Publishers B.V. (Biomedical Division)

A glands o in a~ monoclonal graphy. of polv gels. Greater than 90% of with an Mr -- 70 000 (i.e. chromo disperse in molecular weight. granin was shown by Western 6 nmol of chromogranin were estimated to be a 25% yield Critical to achieving reasonable pH buffers for resuspension granin purified from human and identical in the N-terminal ogranin A. A secondary sequence the first three residues of processing of chromogranin

* Present address and address for reprint Inc., 11095 Torreyana Rd., San Die

0167-0115/86/$03.50 © 1986 Elsevier Scie

1 normal human adre onoclonal antibody d partial N-terminal

sequence

:ardo V.

Ann Arbor, M

1985; accel 21 October 1985)

nds: r

acid

U.S.A.

ation

l o w

~m- Lg

:rminal acid

;ech

roduction

2atecholamine containing storage granules of adrenal s soluble proteins which are released with catecholarn on [1-4]. The released granule proteins were original • anins based on their origin from adrenal chromaffin tinct proteins such as dopamine beta-hydroxylase an, ze been identified in the soluble granule fraction (for • omogranin most appropriately describes a complex enting the majority of proteins of the soluble granule t re a unique amino acid composition, being rich in ac t they also have an acidic pI (for review, see Ref. 5). h ent of glycosylation [6]. The most abundant of the chl ;anin A, has a molecular size (Mr) of about 70 000 an all the soluble granule protein (for review see Ref. 5). ] numerous smaller size (between Mr 70 000 and 20 00, • omogranin polypeptides (Mr 85 000 and > 100 000) Lnological cross-reactivity [7-10] and biochemical sil exact relationship between these various size chrom~

ly unclear, recent data implicates both gene duplic~ ~teolvtic r~rocessin~ [9.12.13] in the ~eneration of noi

J t l l a t ~ u l U V l l , ~ l t k , v V O . l f f 1 1 1 L I I ~ t

abundant of the chromogranins, termed chrom and represents more than 40%

In addition to chromograniI 000) as well as some larger siz~

I) have been identified by im similarity [6,11,12]. Althougt

chromogranin polypeptides is pres 91ication [12] and intragranula

2,13] in the generation of polypeptide heterogeneity. granin in hormone storage granules is presently unknown ing bovine adrenal chromaffin vesicles have indicated tha :o stabilize granule osmotic pressure by interacting witl tides [14-18]. However, this hypothesis has not been gen irect relationship between amine storage and chromogranil I. Recent work by our laboratory and others has showl : detected in a wide variety of polypeptide hormone pro q and in neural tissue [11,19]. Thus, any function proposer count for its widespread distribution as a neuroendocrin

t involved in the immunohistochemical detection of humal noclonal antibody, LK2H10, produced in mice [8,10]. Wq ct human chromogranin [8,10,24,25] within secretory gran neuroendocrine cells [26] and their associated neoplasms the use of LK2H10 as part of a purification procedure fo

ogranln of all A, chrom munole the entl~ proteolytic processing [9,12,

The function of chromo A number of studies utilizin chromogranin may help to catecholamine and nucleotides erally accepted because a direct r~ presence is lacking [11,19]. that chromogranin can be ducing cells [8,10-13,20-23 for chromogranin must account marker.

Our laboratory has been chromogranin using a monoclona have used LK2H10 to detect ules of most APUD type In this paper, we describe

e chromogranin contrasts with tl a cross-reactivity and suggests thai for its biological activity.

monoclonal antibody affinity chro~ tion

chromaff ,lamines folk

iginally referr¢ granules

and the opi review s~ family o

fraction. ' acidic resid In additioJ

9nal

:quence linus of

; amino

ain var- I stimu- chrom- ',r, since ~eptides he term tes rep- granins proline, in their chrom-

40% ranm

; lze

lm- h

~res- ranular

known. that with

e n -

Lnln shown

) r o -

9osed docrine

'human We

ran- )lasms.

for

he mouse IgG1 monoclonal antibody LK2H10 is secrc "usion of NS-1 myeloma cells with splenocytes from mn pheochromocytoma tumor. Details on the pro& pecificity for human chromogranins have been descri ~nal antibody containing culture fluid was harvested f t for Western blotting studies. Ascites fluids containi ~ined by injection of 5 x 106 hybridoma cells i.p. i lsly with 0.5 ml pristane. Purified LK2H10 was obt~ ,o step procedure employing salt precipitation and ion. ,'fly, ascitic fluid was precipitated with 1 volume of sa lsted to pH 7 as described by Fahey [27]. The pre inst 0.005 M phosphate buffer, pH 7.4, and applied to ilibrated with the same buffer. The column was elw )5 to 0.2 M phosphate buffer, pH 7.4, essentially as dc • A small aliquot from each fraction was tested for ity by electrophoresis in a discontinuous polyacrylal ;sie Blue R250 (details below). Only those fractions

electrophoresis and silver staining ' r ~ t ~ i n e ,txzoro ~ l a r t r ~ n h n r a ~ d i n l fl N (~7~ e r

~1, 1 . / I r ' , / - ~ F _ . ~ - ~ , 3 C p l l ~ l U S C t u U l l l l l l l l

eluted by a salt gradient from a s described by Peters and Coon

IgG heavy and light chain lacrylamide gel stained with Com-

ctions containing highly purified

)resed in 10 0.075 cm slab gels containing 10°A e (37:1, monomer to linker) and the discontinuous sodiur~ er system of Laemmli [29]. Samples were heated to 100°C tining 2% 2-mercaptoethanol and electrophoresis was per- lecular weight standards were phosphorylase B (92 000) ' 000), ovalbumin (43 000), carbonic anhydrase (29 00ff

were treated at room temperature in crystallizing dishe~ lows: (1) fixation in 50% methanol/10% acetic acid for 3( acetic acid for 30 min, (3) 10% glutaraldehyde for 1 h, (4 ['or 1 h (four changes), (5) dithiothreitol (5 #g/ml) in HzC e (Sigma) 0.1% w/v for 30 min, (7) rinse with 50 ml H20 50/tl 37% formaldehyde in 100 ml 3% sodium carbonate) h 2 M citric acid.

0.005 [28]. purity massle antibody were pooled and used.

Gel Proteins were electrophoresed

cross-linked polyacrylamide dodecyl sulfate (SDS) buffer in SDS-sample buffer containin formed at 14 mA/gel. Molecular bovine serum albumin (67 and trypsin inhibitor (21 000).

For silver staining, gels with gentle agitation as follows: min, (2) 5% methanol/7% wash with deionized H20 for for 30 min, (6) silver nitrate, then twice with developer (50 (8) reaction is stopped with

of human chromogranin of high I opsy adrenal glands)• Using such ce of chromogranin out to 28 resid minal processing•

) to human chromogranin LK2H10 is secreted by a]

a mouse ~roduction of

described previ from late

containing antib~ in Balb/c

obtained fro: nd ion-exchang~

saturated precipitated

a DEAE.

x 8 x

a gen- e have e iden-

~ormed with a dy and . M o n -

~lls and 0 were .~d pre- lids by ;raphy. ~ulfate, ialyzed ;olumn

from Coon chain Com-

,urified

10% lodium IO0°C

3 e r -

000), 000)

dishes 3O (4) O

2 0 ,

:),

) washes in PBS for 5 min each, (5) peroxidase conju 00 dilution in fetal calf serum in PBS) for 60 min, (6) l, (7) color development in 5',5'-diaminobenzidine (1 )2 (0.02%).

rein dot blotting ~etection of chromogranin was assessed by dot blottin dfold 96-well apparatus (Keene, N J). Nitrocellulose lc he transfer buffer of Towbin et al. [30] was clampe( )wing solutions were added for 1.5 min and remove( )mogranin containing solution, (2) wash twice in PBS ine serum albumin in PBS (previously heated to 1000( 'BS (100 #1), (5) 75/A LK2H10 antibody (spent cur t ~g/ml antibody), (6) wash 3 times with PBS (100 #1). T apparatus and then placed in a crystallizing dish a )wing reagents at room temperature: (1) 10 min with ( )°C pretreated), (2) wash twice in PBS, (3) 30 min in I -mouse IgG (1/400 in 10% fetal calf serum, PBS), (4) :lop color with 5',5'-diaminobenzidine (1 mg/ml PBSI latter series of steps was performed in the dish rath

ttus. since lower back~,round stainin~ was achieved 1

culture media containing about The paper was removed from and gently rotated with the

t min with 0.1% bovine serum albumin peroxidase conjugated goat wash 3 times with PBS, (5)

'ml PBS) containing H202 (0.02%). rather than in the dot blot ap-

Lg by this modification.

in patients without disease affecting the adrenal medulla ostmortem and kept at -20°C until used. Glands were aced and extracted in 10 volumes of ice-cold PBS in a z inhibitors added to the extraction buffer included phen- L mM), benzamidine (10 mM), N-ethylmaleimide (5 mM) :etic acid (5 mM). The extract was cleared by centrifuga- Congealed fat which floated to the top of the tube was

ernatant through gauze. The extract was stored at - 20°C; soluble material was removed by centrifugation as above. was subjected to (NH4)ESO4 precipitation at 55% satu- :. The precipitate was spun at 15°C for 10 min at a speed :t suspended in 1/10 original volume in PBS. After several the semi-crude chromogranin preparation was applied to ity column (see below).

10 pgj the followin (lO0°C anti-mouse develo The paratus, :g~

Tissue processing Autopsy adrenal glands

were obtained at 8-24 h pos trimmed of excess fat, minced Waring blender. Proteolytic ylmethylsulfonyl fluoride (1 and ethylenediaminetetraacetic tion at 27 000 × g at 4°C. removed by passing the supernatan after thawing, additional insoluble The crude adrenal extract ration at room temperature. of 27 000 × g and the pellet hours dialysis against PBS, an LK2H10 antibody affinity

insferred to 0.45/~m nitrocellulose a Bio-Rad Transblotting Cell (Bi(

ectrophoretic conditions (60 V fo fter electrophoresis the paper wa, allizing dishes with the following Vl phosphate pH 7.4 and 0.85% 1' ing), (2) 3 washes in PBS for 5 rr zlonal antibodies (about 10 #g/ml

Ljugated go~ 3 washe~ mg/ml il

g in a Sct paper (0.4

3ed in the removed by vacu

(1 O0 td), lO0*C for 5 m

tant

[ipore) mond, as de- tted at 1) 2% for 30 spent

o r l h , se IgG 5 min

:aining

Schfill ~oaked nd the /~1 of a f0.1% a twice

column was utilized and stored at 4°C. The semi-cr tined after (NH4)2SO4 precipitation (see previous s Lmn, and the column washed with PBS until the abso column was then flushed with deionized n20 (cond

;, and the bound chromogranin was eluted with 0.5 a the resulting 2-ml fractions were analyzed for cht ting. Samples containing chromogranin were pooled rant Instruments, Westbury, NY).

h-pressure liquid chromatography ( HPLC) [PLC separation was undertaken using a Varian mot with a Vista 402 data system. Separation at 30oc was

H8) 30 nm diameter pore C18 column. After the all )mogranin was dried and then resuspended with 2 N

injected into the HPLC column. The flow rate was )mplished by a 60-min linear gradient of water to ing 0.1% trifluoroacetic acid. 0.8-ml fractions were brmed at 280 nm. Samples were dried using a Speec ruments) and resuspended in appropriate buffer for

ino acid seauences and comoosition

affinity chromatography step, N acetic acid. 500 #1 samples

0.8 ml/min, and elution was 40% n-propanol, both con- collected and detection was

~eed Vac concentrator (Savant further assays.

:omposition lbout 3 nmol) was sequenced on an Applied Biosystems

as originally described [32]. The phenyl-thiohydantoin and identified by reverse-phase HPLC on an Ultrasphere

gradient containing 88 mM aughout. The column was eluted at 50°C with a flow rate

composition were dried and treated overnight with HCI atein. The amino acid analysis utilized the same HPLC cribed above.

~rom autopsy adrenal glands ands as a source of chromogranin would depend on the

chromo: were accom tainin performed Instruments

Amino sequences and corn Human chromogranin (about

470A gas phase sequencer amino acids were separated and ODS column using a step NH4CH3COO, pH 4.9 thn of 1 ml/min.

Samples for amino acid vapor to hydrolyze the protein. column and conditions described

Experimental Results

Stability of chromogranin from The value of autopsy glands

~y

as coupled to Sepharose CL4B be~ /ml of gel using the cyanogen bro ,~d by gently rotating overnight at lpling at pH 6.5 as described by C no more effective for affinity chr{ d CNBr active sites were blocked )H 8.4. 5 ml of LK2H10 gel was l ) and all buffers were run at a flo

semi-crude chro section) v

absorbance 2~ conductivity :

N NH4{ chromogran

~ooled and dri,

model 5560t was by a reve

of 25-50% CH3CN

tcia) at ,d [31]. (2H10 31] for y than [ng the a l x ) ml/h. faction to the

-~ 0.01. "emove amples by dot ~d Vac

~njunc- upelco

Savant

rstems lantoin

~here

)w rate

21 HPLC

the

;tern blotting of these homogenates also showed id~ abundant chromogranin species for all homogenates results were very similar among autopsy adrenal glal

e as those from surgically excised tumor specimens sion. The data indicate that chromogranin polypeptic not routinely subjected to postmortem proteolysis ntenance at 4°C.

ification of chromogranin hromogranin was purified from adrenal glands by a fc rogenization in neutral pH aqueous buffer, precipitati aturation, ant ibody LK2H 10 affinity chromatograph ' t4)2SO4 precipitation at saturation percentages of 4( ally tested and 50% was the lowest value that precipit rain. Although the salt precipitation step only results

2 4 8 4 0 0 4 8 2 4

40, 50, 60, 70 and 80% were 9itated the majority of chrom-

results in a minimal enrichment of it helps more by eliminating material that would get

P

t ing of c h r o m o g r a n i n f rom an au topsy adrena l g l and s tored at 4°C for

gland was minced and separated into four equal amounts. At the times d and the amount of chromogranin was assessed by dot blotting (left :ight hand panel shows that no breakdown occurred of the large main ,~d Chg) during the 4°C incubation. Hours of incubation (0, 4, 8, 24) are

initialb ogranm. chromogranin (1-2-fold),

8 •

° i O • r. 1. 6 ,m t.-, 32

(,.)

o 1 2 8

O a) 2 5 6 r v -

5 1 2

Fig. 1. D o t b lo t and Wes te rn b lo t t in

various hours. A 4-h post-mortem indicated the sample was extracted hand panel). Western blot in the ri chromogranin polypeptides (labeled listed on top of each figure.

able in situ for reasonable periods ility under storage at -20°C . To 1 gland was obtained and frozen /as minced thoroughly with a razor rocessing into soluble homogena aogenized immediately and the ot[ nd processed 4, 8 and 24 h later. y dot blotting showed similar end f chromogranin were present in all

identical cc (Fig. 1).

lands and v frozen w

)tides in situ during

four step p tion with (i

,graphy and rev

Chg

iii !~i i~i~i iiii!!!i!i

, death 'ss this or one livided ails in )red in Fig. 1,

11/128) mates.

large n blot- dly the n after glands ing or

rolving ~t 55% ~PLC.

of

i level of non-specific protein binding which could n ~ent HPLC step. .fter antibody affinity chromatography, an additional : LC was utilized in order to provide highly purified c lies such as amino acid sequencing. From preliminar :opanol, CH3CN and both C8 and C18 columns, the ~mogranin with the best yield was achieved using a ( ~-40% n-propanol. Chromogranin eluted as a single ol (Fig. 2). Contaminant material absorbing at 280 a before and after the chromogranin peak (Fig. 2). "I ~mogranin HPLC peak were analyzed by silver stain: er to assess their polypeptide composition and their 1 ,dy. As seen in Fig. 3, the main chromogranin fract epeptides of about Mr 70 000 with some Mr > tOO q Ole between Mr 43 000 and 28 000 in size. The Westd ;e same fractions clearly shows that all these vario nbers of the chromogranin family since essentially a ant detected by LK2H10 antibody. The HPLC fracti~ irate runs and were then used for amino acid corer ncing. The HPLC fractions 51-55 were not used bec~ inat in~r n n i v n e n t i d e ~ n n r t i e n l n r l v in f r n e t i n n g ~;~ a n d

l g ; C l . l b l . l V l l . V ¢ / I L I I l a l ' k . f . l l l V O _ l l =

fractions (43-50) contain major 000 and many polypeptides

Western blot (Fig. 3, bottom) ot various polypeptides are in facl

all carry the antigenic deter- fractions 43-50 were pooled from

~osition and N-terminal se- because of the presence of con-

~rticularly in fractions 52 and 53, which were not detected K2H 10 antibody (Fig. 3). About 6 nmol of chromogranin whole human adrenal. A summary profile exhibiting the

~wing the location (arrow) of the chromogranin peak eluting at about tjected was previously purified on the monoclonal antibody LK2H10

polypet visible these members minant separate quencmg. taminating polypeptides, particula by Western blotting with LK2HP were isolated from 97 g of

! E

8

TIME Iminl

Fig. 2. HPLC chromatogram showin 30% n-propanol. The material in affinity column.

/ column used in the next purific d, the trapped material gets elute¢ hromogranin preparation. the purification process was antit ~t preparations from two sets of a~ proximately 20 ml and this was r Virtually all available chromogra It was also critical to keep the col ltilizing a column size of about 5

not be rer

an additional step invo] chromogr

~reliminary studies most effi

C 18 colur peak at nm is u

The main staining and

reactivity

alarl

In this ',olumn

r chro- Is were over a und to e small ~d in a le sub-

~-phase ~lytical ~panol, tion of radient n-pro- eluting om the Lting in I10 an-

Ior

~tides of

fact °_

d from se-

c o n °

etected ranm

the

,ii

:hromogranin detected after C18 HPLC separation. For the purpose 20 rnin) were used for the silver-stained gel and 100/A of each HPLC fractioJ amples were dried in a Speed Vac and resuspended in 20 #1 SDS-sampk oresis. Fraction 45 in the silver-stained gel is estimated to contain abou :periment with other samples.

lexity at each step in the purification process from crude : HPLC step is shown in Fig. 4. Lane 4 of Fig. 5 shows the HPLC purified chromogranin. As expected, a heterogen file is seen and all bands visible are reactive with antibody, tsitometric scanning of lane 4 showed that the Mr 67 00( : (chromogranin A) represented about 85% of the tota ,~s isolated. As can be seen, the greatest purification oc Ffinity chromatography step with additional purificatioi

Fig. 3. Analysis of the purity of chromo pl of each HPLC fraction (43-55 min) were used for the Western blot. Sam buffer prior to SDS-gel electrophoresis. Fr 60/~g of protein based on our ex

protein polypeptide complexity adrenal homogenate to the polypeptide complexity of eous molecular weight profile LK2H10 (see Fig. 3). Densitomet chromogranin polypeptide chromogranin polypeptides curred at the antibody affinity

t¢ .G c

ii ~ 1

• F r a c t i o n

9ose 200 ', fraction

~le ~out

crude the ~n-

ty

000 total

OC-

fication

4. SDS-gel electrophoresis and silver staining of proteins present ication of human chromogranin. Lane 1 contains 20 ttg of whole ad: ; of the 50% (NH4)2SO, precipitation, lane 3 contains 5/~g of cl ty chromatography, lane 4 contains 1 pg of HPLC purified chron ane 5 contains molecular weight standards.

eved by HPLC. From parallel experiments, the yield :d by assessing end point titers in dot blotting and by ion with the Bio-Rad protein assay. After all purl: )mogranin was estimated at 25% and a purification

of chromogranin was mon- measuring protein concen-

~urlfication steps, the yield of of 250-fold.

mogranin dried at pH 2.3 and solubilized with buffers ranging in pH ranin was first treated with 2 N CH3COOH or 0.5 N CH3COOH (pH s listed at the left column. The procedure involved vortexing for 1 min ~ere then rinsed and any remaining insoluble chromogranin was solu- Lcetic acid). Samples either undiluted (N) or diluted as indicated at the for detection of chromogranin by monoclonal antibody LK2HI0.

achieved itored tration chrome

First wash

pH N 2 4 8 16 32 7

6

5

3 0 ~

2.3 • 0 ~

Fig. 5. Dot-blot analysis of chromo from 2.3 to 7. The dried chromog 3) and acetate buffers whose pH is listed at room temperature. The tubes were bilized as above by pH 2.3 (2 N acetic top of the figure were dot blotted

5

9 2 K

6 7 K

4 3 K

2 9 K

,'-- 2 1 K

during the of whole adrenal gland

chromograni tariffed chromogranin (p

Second wash (pH 2.3)

N 2 4 8 1 6 3 2

~s in the contains LK2H10 s 43-50)

of

at the

dysis in Fig. 5 shows that chromogranin was not effici~ ~roaching pH 2 were used. These data, together with nmarized in Table I. The solubility of chromograni rkedly reduced if drying was performed at low pH bt pt. A, Table I). Chromogranin dried at low pH, altt ltral pH, was readily solubilized with either low or h hie I). This phenomenon was observed whether dr) ypropylene tubes. This requirement of pH extremes 1 dified chromogranin was also observed using chrom se-phase HPLC column which contains n-propanol a tout pH 1.8). However, the pH is not the only con ubilization of chromogranin since 0.1% trifluoroaceti J < 1) were not as effective as were 2 N acetic acid (

cts of pH on the solubility of human adrenal chromogranin

(pH 2.3) or 80% formic ack

Solubility b Solubility b

+ + + +

+ + +

+

+ + +

+ + +

+ + + +

+ + + + + + + + + + +

+ + + + + + +

vaporation. The pH 2.3 and 3.0 were achieved using 2 N and 0.5 N acid

of buffers pH 4-7 were 0.1 M acetate, pH 8 and 9 were 0.1 M Tris, an~

valent to 100% solubility assayed by performing end-point titer in a do

1 antibody LK2HI0. Conditions for solubilization involved vortexing fo

(pH < 1).

TABLE I

Effects

Expt. A

Drying a Suspending pH pH

2.3 7 -

3 7 + 4 7 + +

5 7 + + +

6 7 +

7 7 + 8 7 +

9 7 + 10 7 + 11 7 +

a All samples were dried under eva

respectively. The composition pH 10 and 11 were 0.1 M carbonate.

b A + + + + reaction was equivalent to

blotting assay with monoclonal antibo, 1 min at room temperature.

l~ed chromogranin fication process that in several ster romogranin could not be readily fithout neutral pH buffer. Because f chromogranin at each step of pur lity were investigated. Fig. 5 illus! n LK2H10 affinity column at pH were first suspended in various p

pended with 2 N acetic acid (pH 2.1 not efficiently solu

addition ranln in neul

but not at ] although rel

high pH c 'ying occu for efficie

chromogranin ~ and 0.1%

condition re trifluoroacetic acid (pl

Expt. B

Drying Suspending

pH pH

2.3 2.3

2.3 3

2.3 4

2.3 5

2.3 6

2.3 7

2.3 8

2.3 9

2.3 10

2.3 11

aid was in sol-

ortance effects

lubility n dried Lnd any 9lotting buffers

nts, are fer was ~ditions luble at .~xpt. B, ;s or in ltion of the re-

tic acid efficient N HCI

acid

md

a dot f o r

:heroical analysis of human chromogranin [uman autopsy adrenal chromogranin, purified by lity chromatography and reverse-phase HPLC (fracti, lysis by amino acid composition and N-terminal am "able II, the amino acid composition of human chrc Lein, being rich in acidic type residues (Asx and Glx), i

low levels of Cys.

LE II

ao acid composition of bovine chromogranin A and human chro~

mol/100 residues

H u m a n H u m a n chromogranin b chromogranin b

47 149

65 18 6

11 45

67 50 28 53 11 10 7

N D 35 12

36

tromatography.

Bovine chromogranin A a

Asx 8.0 7.2 Glx 22.5 22.9

Ser 7.7 10.0 Thr 2.6 2.8

Cys 0.2 0.9 Met 1.4 1.7 Pro 9.2 6.9

Gly 8. I 10.3 Ala 8.5 7.6 Val 3.9 4.3 Leu 7.3 8.1 lie 1.4 1.8 Phe 1.7 1.5

Tyr 1.0 1.1 Trp 1.4 N D e

Lys 8.4 5.3 His 1.9 1.8

Arg 6.0 5.6

" Cohn et al. [35]. b Purified by LK2H10 affinity ch romato

Data from references 34 and 35. d Based on molecular weight of 70 000. e N D = not done.

to determine if the above describ ,ical of other adrenal homogenate ~nate diluted in 2 N acetic acid (I proteins not soluble at neutral pH The experiment shows that essen!

tions were not solubilized at neut ae to chromogranin. The above de erved for human serum IgG but ca).

salt preci tctions 43-5( amino acid

chromograni particulal

h u m a n chromogranin

Residues/mol d

Bovine chromogranin A c

44 144

47 13

3 10 54

49 55 22 46

6 9 6

7 54 10 41

, prop- or this ]eutral zed by ~ctable icating effects serum

ttibody cted to ks seen of this having

i!i i!i I I !III !ii i iiiiii i ! ii ! i i ii! il ~

?5 i!i .... 21 i ¸

6. SDS-PAGE and silver staining of adrenal gland extract prot¢ ,,hal extract was dried under acid conditions (pH 2.3) and those pr n and at 250C using a neutral pH buffer were discarded. The rem~

ng in SDS and were compared with an original untreated sample. d dilutions) shows that the acid treatment affected the solubility hal extract.

he N-terminal amino acid sequence was defined for t ~se residues given in parentheses are presumptive and :Is or on comparison with the known bovine chrome restingly, a secondary sequence missing the first 3 re,, detected. The 1 °-3 sequence represented about 25%

ttical for all residues which could be detected. The 28 1 ~I1 31"1 lrn~,~x~n roeirhl~,e f n r h . m a n ohrc , m n a r r o n ; n

the first 28 residues (Fig. 7). are based on best expected

chromogranin sequence (i.e. Cys). residues of the main sequence

of the total protein and was residue sequence is identical

s for human chromogranm A [33] and the 24 residues ~,ranin A [11,34,35].

sn -Ser -Pro -Met -Asn - L y s - G l y - A s p - T h r -

~e t -Asn-Lys-Gly -Asp-Thr -G Iu - V a I - M e t -

ys-( C ys) - I le -VaI -GIu - V a l - I l e - Ser -Asp-

uman adrenal chromogranin. The main sequence (1 °) is shown for 28 ;nee missing the N-terminal first three residues (1°-3) is below. All Cys are inferred from the lack of any other residues and are consistent with r residues in brackets are the best fit based on yields, x indicates that

The N- Tho yields Interestin w a s

identical for all with all 20 known residues known for bovine chromogranm

1 5 1 ° (75%) H 2 N - L e u - P t o - V a | - A a

l ° - 3 (25%) H2N-(Asn)-Ser-(Pro)-Met-Asn-

15 G Iu -Va I -Met -L

L y s - ( C y s ) - I l e - V a l -

25 28 Thr -Leu-Ser -Lys

L y s - X - X - X

Fig. 7. N-Terminal sequence of human residues while the secondary sequence residues are in brackets since they previous sequence data [32]. Other no residue could be determined.

7

d • stds

2 9

proteins with di 3roteins solul

The remaining prote The diluti(

of a portio

lO

2o

X - V a l - I l e - S e r - A s p - T h r - X - X -

ubilities. :xing for ilized by

epresent as in the

sed normal adrenals and tumors. The stability of chro: wed us to generate about 6 nmol of chromogranin l real medulla represents about 10% of the total adre )maffin cells that produce chromogranin. n interesting and unexpected finding was the difficult) )mogranin in aqueous buffer following vacuum dryil m drying was performed at low pH, then chromogra~ ~ended using either acidic or basic conditions• This p 11 polypeptides in the adrenal homogenate and affecte lmin. Under our experimental conditions (see Table ] "ts when drying was performed under basic conditions "hose NH4OH to elute chromogranin from the antil~ purification process. However, in practice, we often ~IH4OH under vacuum, low pH conditions (2 N ace nsure that all the chromogram was resuspended. U~ e not been previously described for chromogranin, 1 aes, and complete solvent removal, to our knowledgq l previous chromogranin purification schemes• unique characteristic of chromogranin which should

cpeptide pattern that is revealed when total chromog~ lar weight in SDS ~els A~ seen in Fio ~ silver stni

a l l L I U k / I d ) ' a l l l l l l t f f k, U I U l I I I I I d U l l l l ~

observed that after removal acetic acid) were also required

Unusual solubility properties however, the use of pH ex-

ge, have not been associated

be discussed is the complex ranins are analyzed for mo-

As seen m Fig. 3, silver staining of the fractions of the a high degree of molecular weight heterogeneity from Mr fact that these polypeptides are chromogranins and not ealed by the Western blot which shows that all polypep- terminant detected by our highly characterized LK2H10 • The immunological identity of these polypeptides cannot ross-reactivity since this complex polypeptide pattern for 9served previously by other investigators using convert- '] as well as monoclonal antibodies [12]. What is important arious smaller size chromogranin polypeptides are only purified chromogranin preparation, with the majority ot :omogranin polypeptide(s) exhibiting an approximate M, 9ased on the observation that the polypeptide complexity trge amounts of protein are loaded onto the gels (such as ons) and highly sensitive detection systems such as West- ining are used for protein detection. In fact, Coomassie

of NH to ensure have tremes with

A polype] lecular weight in SDS gels. HPLC protein peak reveals 50 000 to Mr 20 000. The contaminant protein is revealed tides carry the antigenic determina monoclonal antibody [8,10] be explained by spurious cross-rea chromogranin has been observed tional polyclonal antisera [7] to appreciate is that the various minor components of our t material represented by chromogr~ 70 000. This conclusion is based is only readily seen when lar from the HPLC peak fractions' ern blotting and silver stainm

fly simple method to purify signif chromogranin, from a readily a

hal glands). The procedure involve noclonal antibody affinity chroml , adrenal glands taken from 4-24 1 e of chromogranin which was i~ plexity from chromogranin isolat,

of chromogranir from 97

adrenal giant

y encount ymg under ranin could

3henomen affected serum i

I), we did (NH4OI-

antibody affin

inant

reactivit'

granin

ities of lrce of ization nd re- :m au- ible in gically glands Is. The ins the

ending litions. ciently operty serum

e these "eason, during

cannot for

onven- ~ortant

Y of

plexity a s

~st- ~massie

more recently from a human pheochromocytoma [5 role isolation and subsequent chromatographic sepa ',ies (i.e., chromogranin A). Whether these procedure: er size chromogranin polypeptides as does our met ',ss sensitive protein detection methods along with hea te have done. n intriguing issue that remains unsettled pertains to t ous size polypeptides of the chromogranin family. ( rmediate sized chromogranins are known to be glycol, )ohydrate composition [6]. Whether this plays a signil rpeptide heterogeneity is presently unknown. Recent ace analysis of different bovine chromogranin pol,. ~00, 85 000) showed very similar yet distinct sequence olypeptide heterogeneity through gene duplication eve ~oly(A) mRNA isolated from endocrine tissues has )mogranin is synthesized as a large precursor (for p :hesized from various sized mRNA's. Two such stud ~slation product(s) which is appropriate in size as a 1: L13], while another study identified a variety of dil Mation products [37]. Thus, it is unclear whether ch y ~ n o i t v n e e n r ~ he ~ fn re cw a f t e r n r n t ~ i n t r n n ~ l n t i n n

p u ~ y p ~ p u u c ~ Ll.w., lWr u a u u u ,

uences, suggesting the possibility ~lication events [12]. In vitro translation

been used to determine if proteolytic processing) or is

studies have identified a major ~recursor for chromogranin

different sized chromogranin chromogranin structural het-

" after protein translation. pf chromogranin A in bovine chromaffin granule lysates r the possibility of chromogranin post-translational pro- ion, N-terminal processing enzymes for opioid precursors affin vesicles have been suggested as potential processing [19]. The fact that we found 25% of our chromogranin

Gterminal 3 residues would appear to be consistent with )lytic processing. However, we cannot eliminate the pos-

sequence resulted from the use of autopsy tissue as a the other hand, it should be noted that the ] /chain of ropin has similarly been reported to have a secondary epresenting 30% of the material and missing the N-ter-

;ranin is presently unclear. The abundance of this protein ales led to experiments indicating a role in catecholamine y [14-18]. However, the present understanding of chrom-

ofpolypel: of pol: chromo synthesized translation A [9,13], translation erogenelty occurs before or

Proteolytic breakdown of provided direct evidence for cessing in situ [12]. In addition which are present in chromaffin enzymes for chromogranin sequence was missing the N-termir the notion of in situ proteolytic sibility that our secondary chromogranin source. On human chorionic gonadotro amino terminal sequence represen minal 3 amino acid residues [38].

The function of chromogranm in adrenal chromaffin granules storage and granule stabilit,

S only the main Mr 70 000 polype tion, silver staining of the HPLC lgle major cluster of polypeptides the smaller polypeptides were visl n were analyzed. Thus, our purif 4r 70 000 polypeptide(s) which by ce) represents chromogranin A, an chromogranin polypeptides heteJ

)een purified from bovine adrena] [36] by a 1

)aration of )rocedures also pu

method cant heavily load,

the relatic Chromogl

glycosylated ar nificant roh

N-termil polypeptides

minal

~entlng

tot the ~n 4 3 -

at the when

granin chem- n con- n size. ,12,35] loying )rotein Lller or mined ttilized

~'en the I some n their granin cid se- i5 000,

lences present in the data bank of the National Biome ~hington, DC. However, no significant sequence hom he recent human chromogranin N-terminal 20 residut together with our data extending to residue 28 shows J bovine and human sequence at all residues which ca: This was unexpected since conventional antisera to

lin A are not usually cross-reactive [11]. In addition 2H10 to human chromogranin cross-reacts only wit le but not with bovine or any other species [8]. The~ xgence in the sequence has occurred at sites distal tc :rminal sequence of chromogranin appears to be hig ~n and as such may contain the residues most critJ ecule. Monoclonal antibodies generated against a s'. N-terminal sequence should provide a reagent use!

he authors are grateful to Ms. Rachel Thayer for her e t n l ~ r l n h n 1 .nrnh~rt f n r hi~ n ~ i ~ t n n ~ in n h t n i n i n o n

. l t .~ ,a l LLI L I I . ~ I B I I I ~ , L I I d l l U I L I I I 3

ynthetic peptide coding for useful to detect chromogranin

:xcellent technical assistance :~r his assistance in obtaining post-mortem adrenal glands. knowledge the skilled assistance of Ms. Sheila Norten of a Amino Acid Sequence Facility. B.S. Wilson was sup-

Career Development award from the National Cancer alth Service Grant P30AM34933. Part of this work was in Established Investigator award to S.H. Phan from the n. This work was supported by N I H grants HL28737 and d from the American Heart Association with funds from

release of protein from the stimulated adrenal medulla, Biochem. J., 97

ith, W.J. and Kirshner, A.G., Release of catecholamines and specific Science, 154 (1966) 529-531.

the from all animal species.

Acknowledgments

Th and to Mr. John Lambert for In addition, the authors acknowlet the University of Michigan ported by a Public Health Institute and by Public Health done during the tenure of an American Heart Association. This 31963 and by a grant-in-aid its Michigan Affiliate.

References

1 Banks, P. and Helle, K., The release of (1965) 40--41C.

2 Kirshner, N., Sage, H.J., Smith, protein from adrenal glands, Science,

stent with that of amine distributJ on of this molecule might come is found to be homologous to oth, ntified in this report for human ~sition with the 20 known residues le chromogranin A [11,34,35]. We ne residue with the sequence of bo~ 12], however, recent studies of bo~ 5,36]. We compared our 28 residue

:dical Res homologies ,a

residue sequenc identical 1

can be com bovine c

, our mo with monke

These data s~ to the N-t¢

;hly conse Jcal to th

vled

A pos- quence oteins. ain are t3] and tat our granin granin 9rotein :lation, trated. :1 et al. etween lues 1- Iromo- Ltibody tdrenal species us, the ;h evo- of this

tce

of

~er

'as

he nd ~m

97

:ific

- - ~ t " . . . . . . . . y - * . , . . . . . J . . . . ~ . . . . . . . . . . J . . . . . . . . . . . . . . . . . . . . . . . 1 ~ -

197-199. Lloyd, R.V. and Wilson, B.S., Specific endocrine tissue marker C Science, 222 (1983) 628-630. Kilpatrick, L., Gavine, F., Apps, D. and Phillips, J., Biosynthel matrix proteins of the adrenal chromaffin granules, FEBS Lett., 1 Wilson, B.S. and Lloyd, R.V., Detection of chromogranin in neurq antibody, Am. J. Pathol., 115 (1984) 458-468. O'Connor, D.T. and Frigon, R.P., Chromogranin A, the major ca protein, J. Biol. Chem., 259 (1984) 3237-3247. Settleman, T., Fonseca, R., Nolan, J. and Angeletti, R.H., Relatiol granin, J. Biol. Chem., 260 (1985) 1645-1651. Falkensammer, A., Fischer-Colbrie, R., Richter, K. and Winkler, of chromogranin A and B of bovine adrenal medulla, Neuroscien Blaschko, H. and Helle, K.B., Interaction of soluble protein fractiq granules with adrenaline and adenosinetriphosphate, J. Physiol. (] Smith, W.J. and Kirshner, N., A specific soluble protein from th~ bovine adrenal medulla. Purification and chemical characterizatior Berneis, K.H., Goetz, U., DePrada, M. and Pletscher, A., Interac and nucleotides with intragranular proteins, Naunyn-Schmiedeb~

Sharp, R.R. and Richards, E.P., Molecular mobilities of soluble ct chromaffin granules, Biochim. Biophys. Acta, 497 (1977) 260-271 Uvn~is, B. and Aborg, C.-H., In vitro studies on a two pool stora~ in granule material from bovine adrenal medulla, Acta Physiol. S

(Lond.), 169 (1963) 120-121P. the catecholamine storage vesicles o

characterization, Mol. Pharmacol., 3 (1967) 52~2 Interaction of aggregated catecholamine

tn-Schmiedeberg's Arch. Pharmacol., 277 (1973

components in the aqueous phase o 260-271.

storage of adrenaline and noradrenalin Scand., 109 (1980) 345-354.

DePotter, R.W., Fischer-Colbrie, R., Schober, M., Winkler, H. an~ immunoreactivity in the central nervous system. Immunochemical char relationship to catecholamine and enkephalin pathways, Brain Res.,

nd Deftos, L.J., Chromogranin A: Immunohistology reveals its universa ptide hormone producing endocrine glands, Life Sci., 33 (1983) 1657-

and Deftos, L.J., Immunoreactive human chromogranin A in divers :ing human tumors and normal endocrine tissues, J. Clin. Endocrinol

zk, M. and Elde, R., Selective localization of the parathyroid secretor. romogranin A protein family in a wide variety of endocrine cells of th 4) 1963-1974. Lnin: widespread immunoreactivity in polypeptide hormone producin~ Peptides, 6 (1983) 263-280. V., Pike, A.M. and Wilson, B.S., Pituitary adenomas: An immunohisto production and chromogranin localization, Am. J. Pathol., 116 (1984

midt, K., Warner, T. and Wilson, B.S., Immunohistochemical detectiol

15 Smith

16 Berneis.

291-296. 17 Shar

chrom 18 Uvn~is

19 Somogyi, P., Hodgson, A.J., Chubb, I.W., Chromogranin acterization, distribution and relations (1984) 193-230.

20 O'Connor, D.T., Burton, D. and occurrence in normal polypep 1663.

21 O'Connor, D.T., Burton, D. polypeptide hormone producin Metab., 57 (1983) 1084-1086.

22 Cohn, D.V., Elting, J.J., Frick, protein I adrenal medulla chromo rat, Endocrinology, 114 (1984)

23 O'Connor, D.T., Chromogranm: tissues and in serum, Regul. Peptides,

24 DeStephano, D.B., Lloyd, R.V., chemical study of hormone !c 464-472.

25 Lloyd, R.V., Mervak, T., Schmidt

d Winkler, H., Secretion from the adren~ rmacol. Chemother., 31 (1967) 94-104. eider, F.H., Silver, M. and Smith, A.D., St om the adrenal gland after splanchnic st

The chromaffin granule. In A.M. Poisner omedical Press, Amsterdam, 1982, pp. 3-" J., Margolis, R.U. and Margolis, R.U., G matrix, J. Biol. Chem., 257 (1982) 1651-1~ er, H., Immunological studies on the acidic 4.2.1) of bovine chromaffin granules, J.

defined by

'nthetic relation,, 164 (1983) _~

neuroendocrine

or catecholamin

Relationship of mt

, H., Cell-fr~ Neuroscience, 14 (198

fractions from bt

e of adren

onshi

~chemical

~romaffin ture, 215

ro (Eds.),

d proteo-

is and on ]2 (1974)

antibody,

he major

~noclonal

le soluble

chromo-

synthesis

nedullary

of 52~2.

Lolamines 1973)

~hase of

drenaline

and char-

8

universal

diverse docrinol.

.y

the

,roducing

anohisto- 1984)

:tlon

owbin, H., Staehelin, T. and Gordon, J., Electrophoretic transfel els to nitrocellulose sheets: Procedures and some applications, Proq 350-4354. uatrecasas, P., Protein purification by affinity chromatography. 1~ cacrylamide beads, J. Biol. Chem., 245 (1970) 3059-3065. [ewick, R.M., A gas-liquid solid phase peptide and protein sequc 990-7997. ~ruggel, W., O'Connor, D.T. and Lewis, R.V., The amino termin~ hromogranin A and secretory protein I are identical, Biochem. Bi 80-383. mith, A.D. and Winkler, H., Purification and properties of an acidiq f the bovine adrenal medulla, Biochem. J., 103 (1967) 483-492. ohn, D.V., Zangerle, R., Fischer-Colbrie, R., Chu, L.L.H., Elting, t., Similarity of secretory protein I from parathyroid gland to chrox 'roc. Natl. Acad. Sci. USA, 79 (1982) 6056~059. ~'Connor, D.T., Frigon, R.P. and Sokoloff, R.L., Human chrom cterization from catecholamine storage vesicles of human phe(

erck-Hanssen, G.S. and O'Connor, D.T., Immunological iden! hromogranin coded by poly(A) mRNA from bovine adrenal medul hromocytoma, J. Biol. Chem., 259 (1984) 11597-11600. ~organ, F.J., Birken, S. and Canfield, R.E., Human chorionic mino acid sequence, Mol. Cell. Biochem., 2 (1973) 97-99.

~ , l l l I J I l l l d ~ l t l l l l l l /'1. t i t / I l l t l l d t l ~ l l t l l t l l l ~ I d U l l t l ,

chromogranin A: Purification and char- ~heochromocytoma, Hypertension, 6

identification and characterization of from bovine adrenal medulla and pituitary and human phaeo-

gonadotropin: A proposal for the

36 O'Connor acterization (1984) 2-12.

37 Serck-Hanssen chromo chromoc'

38 Mol amino acid

tic enolase in pancreatic endocrine neopl

ocrine system and the APUD concept: Reh a, and anuran cutaneous glands, Med. Bi tion of immunoglobulins. In Williams, C d Immunochemistry, Academic Press, NY

antibody or specific cytochemical agents. munology and Immunochemistry, Acader

aral proteins during assembly of the heac

transfer of protein Proc. Natl. Ac~

Derivativati~

uenator, J. B

terminal sequence,, Biophys. Res

acidic protein frt

, J.J., Hamil chromogranin A

~thol., 8

peptides [15-125. .~, M.W.

LA. and 1976, 5,

lage T4,

:ylamide 6 (1979)

and po-

6 (1981)

i human 7 (1985)

granules

Winkler, medulla

of