Development 101. 805-813 (1987)Printed in Great Britain © The Company of Biologists Limited 19S7

805

The minor haemoglobins of primitive and definitive erythrocytes of the

chicken embryo. Evidence for haemoglobin L

C. CIROTTO, F. PANARA and I. ARANGI

Istitwo di Biologia Cellulare, University, Via Eke di sono. 06100 Pemgia, Italv

Summary

A new minor haemoglobin, L, was isolated from thehaemolysates of chicken embryos more than 7 daysold. Electrophoresis in denaturing conditions andtryptic peptide maps of the globins show that the fi-tikeglobin of HbL is identical to that of the minorhaemoglobin H(fiH) while the a-like globin is verysimilar to that of the adult haemoglobin D (a"). HbLcompletes the description of the map of the minorchicken haemoglobins during embryonic develop-ment. In early embryos two minor haemoglobins, Mand E, are produced which have the same /J-likc globin(e) and differ in their o--like globins (o^ and cA,respectively). The same two <*-like globins will makeup the minor haemoglobins of the late embryo, L andH, which differ from HbM and HbE on account oftheir 0-like globin (j8H).

The native tetramers L and M are hard to dis-tinguish from each other. However the constituent

e globin can be easily separated from /S11 by electro-phoresis on poly aery lamide gel in formic acid. Withthis method we found that the switch of the minorhaemoglobins in the blood of chicken embryos starts atthe 7th incubation day.

The two red cell populations, primitive and defini-tive, present in the blood of 7-day-old embryos wereseparated on an albumin gradient and their minorhaemoglobins analysed. The haemoglobin couple M/Ewas found in the primitive erythroid cells whereas theL/H couple was found in the definitive ones. Thedisappearance of the early haemoglobin couple and itssubstitution by the late one during embryonic develop-ment correlates with the replacement of erythroidlines in the blood.

Key words: haemoglobin, chick embryo, erythrocytes.

Introduction

Haemolysates of chicken embryos show differenthaemoglobin patterns during development dependingon the various differentiation stages and the differentrespiratory needs. There is widespread agreementamong researchers about the structural character-istics and the production time of the major haemo-globin fractions. The haemolysates of early embryoscontain two major haemoglobins named P and P'whose globin composition is JI2P2 a nd n'2P2 (Bruns &Ingram, 1973a; Brown & Ingram, 1974; Cirotto,Scotto di Telia & Geraci, 1975; Schalekamp & VanGoor, 1984). Their primary structure is well known(Chapman, Tobin & Hood, 1980; Chapman et al.1981).

At about 6 days of development, the two haemo-globins D and A, with globin composition a-J?/̂ anda£f$2, begin to appear in the embryo haemolysates(Bruns & Ingram, 1973a; Cirotto et al. 1975). Theyconstitute the two major haemoglobin fractions of theembryos from the 7-8th day of development andpersist for the whole adult life (Moss & Hamilton,1974). The primary structure of these two haemo-globins is also well known (Takei et al. 1975; Matsuda,Takei, Wu & Shiozawa, 1971; Matsuda, Maita,Mizuno & Ota, 1973; Vandecasserie, Paul, Schnek &Leonis, 1975; Knochel et al. 1982).

Although the description of the major haemo-globins is complete the same cannot be said for theminor ones. Different authors report different re-sults. Only recently has isolation and characterizationof some of the minor fractions contributed to the

806 C. Cirotto, F. Panara and I. Arangi

partial solution of this problem. Two minor haemo-globins, M and E, have been isolated from the bloodof embryos of up to 6 days development (Brown &Ingram, 1974; Schalekamp, Schalekamp, Van Goor& Slingerland, 1972; Cirotto et al. 1975). The subunitcomposition of HbM has been investigated by variousmethods (Keane, Abbott, Brown & Ingram, 1974)and more recently by determining the primary struc-ture of its globin chains (Chapman, Hood & Tobin,1982a). All the data indicate that the globin compo-sition of HbM is c^e2 where a° has the samestructure as the aMike globin of adult HbD and £ is a/Mike globin different from both the adult fi and allthe other /Mike embryo globins.

Similar investigations were carried out on the otherminor haemoglobin E. The primary structure of itsconstituent globins was determined (Chapman, Hood& Tobin, 19826). The amino acid sequence of the a-like globin is the same as that reported by Knochel etal. (1982) for the adult a*. The /S-like subunit has aprimary structure identical to the e globin of HbM.

There is no such clear description in the literatureon the minor haemoglobins of late embryo. Someauthors describe a haemoglobin (H) typical of the lateembryos and of the newly hatched chicks (Godet,1974; Bruns & Ingram, 1973a,b; Moss & Hamilton,1974). Its globin composition is of a^fi™ type, the a-like globin being the same as adult HbA and the /3-like globin differing from both the adult /? and theother /Mike globins of the early embryos (Moss &Hamilton, 1974). Other authors, using different ex-perimental approaches, have described two minorhaemoglobin fractions of late embryo which havebeen characterized only in part (Cirotto et al. 1975).

Many investigations have been carried out todefine the localization of the major haemoglobins P,P', D and A in the primitive and definitive erythro-cyte populations of chicken embryo. All the authorsagree that HbP and HbP' are present in the primitivered cells while HbD and HbA are absent (Shimizu,1976; Chapman & Tobin, 1979; Schalekamp & VanGoor, 1984) and that HbD and HbA are present inthe definitive red cells while HbP and HbP' areabsent (Cirotto etal. 1975; Beaupain, 1985) or presentonly in small quantities (Chapman & Tobin, 1979;Schalekamp & Van Goor, 1984). As regards theerythrocyte localization of the minor haemoglobins,it has been shown that HbM and HbE are containedin a subpopulation of the primitive erythrocyte line(Cirotto, Panara & Geraci, 1977).

Our aim was to examine in more detail the chickenminor haemoglobins at various stages of embryonicdevelopment, paying special attention to the lateembryos, giving as complete a description as possibleof their distinctive properties and their relations withthe erythrocyte populations.

Materials and methods

Preparation and isolation of haemoglobinsFertilized eggs of different chick strains were purchasedfrom CI.C. ZOO (Perugia). Erythroid cells were obtainedfrom the embryos at various stages of development aspreviously described (Cirotto et al. 1975), washed severaltimes in 7mM-phosphate buffer pH7-4 containing 0-9%NaCl (PBS) and lysed in H2O-CC14 (1-5:0-5 V/V) (Bruns& Ingram, 1973a). The ghosts were centrifuged off and theclear supernatant used directly for electrofocusing analysisor dialysed against lOmM-phosphate buffer pH6-2 forchromatographic separation. Isoelectric focusing on 4%polyacrylamide gels in the pH gradient 7-9 was carried outas described by Drysdale, Righetti & Bunn (1971). Proteinbands were stained with Coomassie Brilliant Blue by theMalik & Berrie (1972) method. Their densitometricanalyses were performed on a Varian 634 gel scanner atA = 650nm. In some experiments, as soon as focusing wascompleted, the gels were frozen and sections of 2 mm werecut with a gel cutter. To elute the haemoglobin fractions theslices were collected separately, vortexed and allowed tostand overnight in 0-5 ml 50mM-phosphate buffer pH7-0.

Chromatographic isolation of the haemoglobins wasperformed on Whatman CM 52 cellulose columns(1-8x20cm). Elution was done with a pH and salt concen-tration linear gradient obtained from a gradient mixercontaining 0-51 of lOniM-potassium phosphate buffer,pH 6-2 in the mixing chamber and 0-51 of 20 mM-potassiumphosphate buffer, pH8-0 in the reservoir (Cirotto et al.1975).

Haemoglobin concentration was determined spectro-scopically using e= 11-lxlO3 at 540nm for the cyan-metderivative (Antonini, 1965). Quantitative evaluation of thechromatographic fractions was obtained by cutting andweighing each peak of the tracings.

Separation of globin chainsHaemoglobin molecules were depleted of the haem groupsby the Rossi Fanelli, Antonini & Caputo (1958) method.The analytical separation of the globins was obtained byelectrophoresis on polyacrylamide gels and the quantitativeseparation was performed by chromatography of CM cellu-lose columns.

Electrophoresis on polyacrylamide gel in formic acid wascarried out as reported in a previous paper (Cirotto, Arangi& Panara, 1980). In a typical procedure, 100ml of thepolymerization mixture contained 15 g acrylamide, 0-1 gbisacrylamide, 6ml of 99 % formic acid, 0-25 g ammoniumpersulfate and 7-8 mg silver nitrate. Both pre-electrophor-esis and electrophoresis were carried out at 4 mA per tube,using methyl green as tracing dye. Protein bands werestained with Coomassie brilliant blue as described by Malik& Berrie (1972).

For the quantitative separation of the globin chains, twodifferent chromatographic techniques were used. The glo-bins constituting HbD, HbM and HbL were separated bythe Gander, Luppis, Stewart & Scherrer (1972) method.About 80 mg of globins were loaded on a CM cellulosecolumn (2-5xl0cm) equilibrated with 15mM-acetate buf-

Minor chick Hbs 807

fer pH5-4. 8M-urea, 50mM-2-mercaptoethanol and wereeluted by means of a linear gradient of acetate concen-tration ranging from 15 mM to 0-3 M. pH5-4 in 8 M urea,50 mM-2-mercaptoethanol.

The aA and /3H globins of HbH were separated on CMcellulose columns (2-5x 10cm) using a linear gradient offormic acid (01 -M to 1-5 M). In both cases, the chromato-graphic fractions were lyophilized after dialysis against 1 %formic acid and their purity tested by gel electrophoresis.

Fingerprint analysisSulphydryl groups of isolated globins were blocked byreaction with a 10-fold molar excess of iodoacetamide in8M-urea, 50mM-potassium phosphate pH7-5.

Tryptic digestion of the carboxymethylated globins wasaccording to Hunt, Hunter & Munro (1969). Fingerprintanalysis was carried out by the Ingram method for theelectrophoresis (Ingram, 1958) and by the Waley & Watson(1953) method for the chromatography. Tryptic maps werefirst strained with ninhydrin and then with the specificstainings for histidine, tyrosine, arginine and tryptophan asdescribed by Lehmann & Huntsman (1974).

Separation of erythrocyte populationsPrimitive and definitive erythroid cells from 7-day-oldembryos were separated on 12 ml gradients of bovine serumalbumin (BSA) 2-5% in PBS. The cells (2xlO6) weresuspended in 0-2 ml of 2 % BSA, layered on the gradientand left standing for 14 h at unit gravity at 4°C. The lowercell band contains the primitive erythrocytes, the upperband contains the definitive erythrocytes. The homogeneityof cell bands was judged microscopically. Only pure frac-tions were pooled (Fucci, Vitale, Cirotto & Geraci, 1987).

Results

In chicken embryos of less than 6 days of develop-ment, the chromatographic profile of haemoglobinsobtained by CM cellulose columns does not differqualitatively from that shown in Fig. 1A which refersto the haemolysate of 5-day-old embryos. The elutionpattern is quite similar to that previously reported byBrown & Ingram (1974) and Cirotto et al. (1975). Themajor haemoglobins P and P' are eluted first, fol-lowed by the two minor haemoglobins M and E.Under the chromatographic profile, a gel of isoelec-tric focusing is shown (Fig. IB) on which the 5-dayembryo haemoglobins were separated. The twomajor haemoglobins P and P' and the two minor onesM and E are evident.

In Fig. 2, the chromatography and isoelectric fo-cusing patterns of 15-day-old embryos are shown.Two major and two minor haemoglobins are evident.The two major ones differ in their chromatographicproperties and in their isoelectric points from theearly major haemoglobins P and P'. They are HbDand HbA, characteristic of the mature embryo andadult chick. The two minor haemoglobin fractions do

not differ to any great extent from HbM and HbE ofearly embryos. From literature, it is known that atthis developmental stage the last haemoglobin to beeluted by cation-exchange chromatography is HbH(Moss & Hamilton, 1974) which can be distinguishedfrom HbE only by electrophoresis in a definite pH(Bruns & Ingram, 19736). For these reasons andbecause of the data reported further on in this paper,the last chromatographic fraction has been identifiedas HbH. The other minor fraction, eluted from thecolumn between HbD and HbA has been named

P'

20 40 60 80 100Fraction number

120 140 160

I P'MP

B

Fig. 1. (A) Elution profile of 5-day-old chicken embryohaemoglobins from a CM cellulose column. (B) Thehaemoglobins of 5-day-old chicken embryos afterisoelectric focusing separation. Top of the polyacrylamidegel column: pH9, bottom: pH7.

808 C. Cirotto, F. Panara and I. Arangi

HbL (Late haemoglobin) to distinguish it from HbMof the early embryo. The quantitative evaluation ofminor haemoglobins from chromatographic patternsgives the same values as those obtained from isoelec-tric focusing profiles: HbM 10 %, HbE 5 %, HbL 4 %andHbH5%.

Electrophoresis on polyacrylamide gel in formicacid proved useful for separating the constituentglobin chains of the minor haemoglobins of chickenembryo. Fig. 3 shows the electrophoretic patterns ofthe globins from HbE, HbH, HbM and HbL. Asalready demonstrated (Cirotto et al. 1980), this elec-trophoretic technique gives good separation of the a-type globins from the /Mype ones. The latter have a

" A

0-6

0-4

Q

b0-2

20 40 60 80 100 120 140 160Fraction number

more cathodic migration than the former. Comparingthe bands of the two gels on which the globins of HbEand HbH were loaded confirmed the results obtainedby other investigations. The two haemoglobins differfrom each other on account of the /Mike subunit.Electrophoretic migration of the y8-]ike subunit ofHbH is more cathodic than the corresponding globinof HbE. The electrophoretic mobility of the two a-type globins is instead identical in both cases. Theglobins of HbM and HbL behave in the same electro-phoretic manner. The migration of the HbL /Mikeglobin is more cathodic than that of the HbM and thetwo a-globins cannot be distinguished. Comparing ongel the positions of all the /Mike globins it clearlyappears that the /3 globins of HbE and HbM are equalto those of HbH and HbL respectively. Instead, themobility of the <*-like globins of HbE and HbH aredifferent from the corresponding a--like globins ofHbM and HbL.

These electrophoretic results agree perfectly withdata in literature which show that the o--like globin ofHbE and HbH is very similar to the a^ globin of theadult (Chapman etal. 1982b; Moss & Hamilton, 1974)and the a-like globin of HbM is identical to the <xu ofthe adult (Chapman etal. 1982a).

HbL is the only one for which the composition ofthe subunit is not adequately described. From ourelectrophoretic analysis, HbL appears to be com-posed of an a-type globin which is exactly the same asaD and by a jS-type globin which is exactly the same asfiH. For more detailed knowledge about the subunit

— H

aD--of

*-£

D

B

Fig. 2. (A) Elution profile of 15-day-old chicken embryohaemoglobins from a CM cellulose column. (B) Thehaemoglobins of 15-day-old chicken embryos afterisoelectric focusing separation. Top of the polyacrylamidegel column: pH9, bottom: pH7.

L M H E

Fig. 3. Polyacrylamide disc gel electrophoresis in formicacid of the globin chains constituting the haemoglobins L,M, H, and E, obtained in a pure form by ion exchangechromatography, as shown in Figs 1, 2.

Minor chick Hbs 809

composition of HbL, we did fingerprint analysis ofthe tryptic peptides of each globin chain.

Fig. 4 shows the fingerprinting of the /3-like globinof HbL and, as reference, those of /3H obtained fromHbH and of E obtained from HbM. The /3L map, asregards number of peptides, their position on thepaper sheet and distribution of peptides positive forspecific amino acid stainings, is very similar to /3H butis clearly different from e. Comparison of the finger-prints of the o'-like globins obtained from HbD and

HbL shows that the two maps are very similar(Fig. 5).

These data, together with those previouslyreported in literature, show that during developmentthe early haemoglobins M and E, with a globincomposition O^E2 and o^e2, are replaced by thehaemoglobin couple L/H with a globin compositiona?/3" and a^/J". These differ from the minor haemo-globins M and E of the early embryo on account ofthe /3-type subunit.

o+ -

tO

u

+ -

ts1 ^o

a ot t ;

Electrophoresis

00

r

0

*• Electrophorcsis

Fig. 4. Fingerprint photographs of/3L (A), /3H (B) and e (C) globins. Beside each photograph a chart shows thosepeptides positive for histidine (IB), tyrosine ( • ) , arginine {M) and tryptophan (ID). The dotted peptide of/3L was absentin some experiments.

810 C. Cirotto, F. Panara and I. Arangi

o

o+ —

o00

. Electrophoresis

o

B

o

» Electrophoresis

Fig. 5. Tracings of the tryptic maps of o^ (A) and oP (B)globins showing those peptides positive for histidine (W),tyrosine (•) and arginine (11). Tryptophan residues areabsent in both samples. Uncertainties in the specificstaining are indicated by asterisks.

With formic acid electrophoresis, the time coursefor the substitution of the haemoglobin couple M/Eby L/H during embryonic development can be fol-lowed. Fig. 6 shows the electrophoretic patterns ofthe globin chains present in M/L and E/H chromato-graphic peaks obtained from embryos at variousstages of development. The band patterns show thatreplacement of the e globin by the /3H one begins at 7days of embryonic development. From 7 days toabout 14 days the minor chromatographic peak elutedafter HbD contains both HbM and HbL and the lastpeak contains both HbE and HbH. After the 14th dayof incubation the two peaks are homogeneous andcomposed solely of the two molecular forms HbL andHbH respectively. The substitution time in the hae-molysate of HbM and HbE by HbL and HbH is thesame as the substitution time of primitive erythro-cytes by the definitive ones in the circulating blood(Romanoff, 1960). This suggests that the two haemo-globin couples are localized in the two different cellpopulations. This hypothesis was confirmed by theresults of the experiments below.

On the albumin gradient prepared as described in'Materials and methods', the primitive red cells of7-day embryos were easily separated from the defini-tive erythrocytes. The haemoglobins of the twoerythrocyte types were isolated by electrofocusing onpolyacrylamide gel. The two minor haemoglobinswere then collected separately and their constituentglobin chains analysed by electrophoresis on poly-acrylamide gel in formic acid. The results are shownin Fig. 7. Fig. 7A shows three gels; in the control gel,the globin components of the M/L fraction isolatedfrom the haemolysate of 7-day embryos are analysed.One a-like globin and two different /3-like globins are

H

6 7 8 17B

5 6 7 810 12 1417

Fig. 6. Polyacrylamide disc gel electrophoresis informic acid of the globins present in (A) the M/Lchromatographic fraction obtained from embryosat 6, 7, 8 and 17 days of development and (B) theE/H chromatographic fraction obtained fromembryos at 5, 6. 7. 8, 10. 12, 14 and 17 days ofdevelopment.

Minor chick Hbs 811

BP T D P T D

Fig. 7. (A) Electrophoretic analysis in formic acid of theglobins constituting the M/L electrofocalized fractionobtained from the haemolysates of the total red cells (T),the isolated primitive (P) and definitive (D) erythrocytesof 7-day-old embryos. (B) Electrophoretic analysis informic acid of the globins constituting the E/Helectrofocalized fraction obtained from the haemolysatesof the total red cells (T), the isolated primitive (P) anddefinitive (D) erythrocytes of 7-day-old embryos.

clearly distinguishable: the e forming an intenselycoloured band and the fiH forming a weak one. Theother two gels show results obtained by the same typeof analysis on each of the cell fractions separated onthe albumin gradient. The primitive red cells haveonly the e globin while the definitive ones have onlythe fiH globin. Identical results were obtained analys-ing the E/H fraction. The globin patterns obtained onpolyacrylamide gel in formic acid are shown inFig. 7B. In this case, the eand/SH globins also seem tobe compartmentalized. The former is contained in theprimitive red cells, the latter in the definitive redcells.

The minimum amount of /3-like globin that can beevidenced by this electrophoretic technique is about1 % of total proteins. This value was determinedadding known quantities of e or /3H to the analysedsamples. Therefore, the measure of compartmental-ization of the globins e and fiH in the two erythrocytepopulations cannot be less than 99%.

Discussion

The data reported in this paper fully confirm theprevious observations about the globin compositionof the minor haemoglobins M, E and H of chickenembryo (Chapman et al. 1982a,b; Moss & Hamilton,

1974). They also describe a new minor haemoglobin,named L, which is typical of the late embryo. Theelectrophoretic analysis of its globins and. more so,the analysis of the tryptic peptides of each globinshow that HbL is made up of an cr-like globin verysimilar to a-D and a /3-like globin identical to /3H.

HbL completes the minor heamoglobin map of thelate embryo, showing it is very similar to that of theearly embryo. In both cases, in fact, the haemo-globins are present in couples constituted by the twoadult a--like subunits, a° and a*, and by a character-izing /3-like globin. It has been demonstrated that oPand o^ are synthesized from the very beginning ofembryonic development (Fucci, Cirotto. Tomei &Geraci, 1983). Their production continues for thewhole life and, combining with different /Mike glo-bins, they constitute the various haemoglobincouples. Therefore, in the early embryo, aa and cr*joined to the e globin chain form the haemoglobincouple M/E. In the late embryo, the same o--likeglobins, joined to the /3, form the couple of adulthaemoglobins D/A. If, besides this couple of majorhaemoglobins, there was only one minor haemo-globin produced, HbH (till now the only one de-scribed) it would mean that either /3H is incapable ofbinding aD or that HbH is confined to red cells wherethe oP globin is not produced. Neither of thesehypotheses seems tenable. The existence of HbL,therefore, introduces a rational order in the chickenembryo haemoglobin map. The two globins a° anda*, being synthesized at all ages, combine withdifferent /3-like globins and lead to the formation ofdifferent haemoglobin couples.

The period of embryonic life during which alter-nation of the haemoglobin couples is most evident isabout 6-7 days. At this age, the minor coupleHbM/HbE begins to be substituted by the HbL/HbHcouple and the major couple HbP/HbP' by theHbD/HbA couple. From the data presented hereabout the cytological localization of the minorhaemoglobins, it seems that HbM/HbE is confined tothe primitive erythrocyte population and HbL/HbHto the definitive one. It cannot be said, on the basis ofour electrophoretic data alone, that there is completecompartmentalization of the two haemoglobincouples in the two erythrocyte populations. In fact,formic acid electrophoresis is not able to evidencebands of /3-like globin quantitatively less than 1 % ofthe total proteins so any eventual reciprocal contami-nation at less than this value would not be evidenced.

The debate about cytologic compartmentalizationof the major haemoglobin couples P/P' and D/A ofchicken embryo as an index of a special switchmechanism, is very topical in literature. The data ofChapman & Tobin (1979) on the presence of small

812 C. Cirotlo, F. Panara and I. Arangi

n

n'

aP

a*

Q

Primitive

C

erythrocytes

<tfw

Definitive erythrocytes

Fig. 8. The haemoglobins of developing chicken embryosare composed of four a--like and four /5-like globins. oPand (x* globins are synthesized in both primitive anddefinitive erythrocytes. n, K', p and e are found in theprimitive red cells, while /3H and fi are found in thedefinitive ones. The existence of couples of a--like globinsleads to the constitution of couples of tetramers sharingthe y3-like subunit.

quantities of HbP and HbP' (about 5 %) in erythro-cytes of the definitive line, indicating a 'molecular'mechanism of haemoglobin switch, have not beenconfirmed either by Beaupain (1985) who has evenused the same immunological method, or by Fucci etal. (1987) who have used different techniques. Theresults reported here about the minor haemoglobincouples rule out reciprocal contamination of morethan 1 % therefore suggesting a compartmentaliz-ation of the two haemoglobin couples in the twodifferent erythrocyte populations.

If the parallel between the major and minorhaemoglobin couples is valid for the cytologicallocalization, it is not so for the switch time. The firstappearance of HbA and HbD in the haemolysate, infact, is observed about 24 h before the appearance ofHbL and HbH (Schalekamp et al. 1972; Cirotto et al.1975). This fact could be interpreted in two differentways: (1) in the red cells of the definitive line thesynthesis of the HbL/HbH couple takes place afterthe synthesis of HbD/HbA, (2) the HbD/HbA andHbL/HbH couples are contained in different defini-tive erythrocyte subpopulations which enter the cir-culation at different times. The first of these twoexplanations is the most difficult to accept since thedefinitive red cells, when released into the bloodstream are already rich in haemoglobin and at anadvanced stage of maturation (Romanoff, 1960). It isdifficult to think that the synthesis of HbL/HbH takesplace only when the red cells are completely mature.The second hypothesis seems more probable and isalso supported by what is known about the cytological

localization of the minor couple HbM/HbE of theearly embryo. It has in fact been demonstrated thatHbM and HbE are contained in a subpopulation ofthe primitive line which has been clearly evidencedimmunologically (Cirotto et al. 1977). Therefore, it isplausible that, in the mature embryo also, HbL andHbH are localized in a definitive subpopulation of redcells, released into circulation at 7 incubation days.

Keeping in mind all the data given in this investi-gation about the globin composition of minor haemo-globins, the times of their appearance and disappear-ance and the results reported by other authors aboutmajor haemoglobins, the scheme of Fig. 8 is pro-posed as a summing up.

This scheme is fundamentally that given by Brown& Ingram (1974) with the addition of HbL and theindications about the cytological localization of thevarious major and minor haemoglobins.

This work was financially supported by grants from theItalian Ministry of Education. The authors thank Mr L.Barberini for his skilful technical assistance.

References

ANTONINI, E. (1965). Interrelationship between structureand function in hemoglobin and myoglobin. Physiol.Rev. 45, 123-170.

BEAUPAIN, D. (1985). Line-restricted hemoglobinsynthesis in chick embryonic erythrocytes. CellDifferentiation 16, 101-107.

BROWN, J. L. & INGRAM, V. M. (1974). Structural studieson chick embryonic hemoglobins. J. biol. Chem. 249,3960-3972.

BRUNS, G. A. P. & INGRAM, V. M. (1973a). The

erythroid cells and haemoglobins of the chick embryo.Phil. Trans. R. Soc. Lond. B 226, 225-305.

BRUNS, G. A. P. & INGRAM, V. M. (1973/?).

Erythropoiesis in the developing chick embryo. DeviBiol. 30, 455^159.

CHAPMAN, B. S., HOOD, L. E. & TOBIN, A. J. (1982a).Minor early embryonic chick hemoglobin M. Aminoacid sequences of the e and aP chains. J. biol. Chem.257, 651-658.

CHAPMAN, B. S., HOOD, L. E. & TOBIN, A. J. (1982/?).Amino acid sequences of the e and aP globins of HbE,a minor early embryonic hemoglobin of the chicken. J.biol. Chem. 257, 643-650.

CHAPMAN, B. S. & TOBIN, A. J. (1979). Distribution ofdevelopmentally regulated hemoglobins in embryonicerythroid populations. Devi Biol. 69, 375-387.

CHAPMAN, B. S., TOBIN, A. J. & HOOD, L. E. (1980).Complete amino acid sequences of the major earlyembryonic o'-like globins of the chickens. /. biol. Chem.255, 9051-9059.

CHAPMAN, B. S., TOBIN, A. J. & HOOD, L. E. (1981).Complete amino acid sequence of the major earlyembryonic /3-like globin in chickens. J. biol. Chem. 256,5524-5531.

Minor chick Hbs 813

CIROTTO, C , ARANGI, A. & PANARA, F. (1980). Thehaemoglobins of developing duck embryos. J. Embryol.exp. Morph. 60, 389-404.

CIROTTO, C , PANARA, F. & GERACI, G. (1977). Twodifferent populations of primitive erythroid cells in thechick embryo. Devi Biol. 61, 384-387.

CIROTTO, C , SCOTTO DI TELLA, A. & GERACI, G. (1975).The hemoglobins of the developing chicken embryos.Fractionation and globin composition of the individualcomponent of total erythrocytes and of a singleerythrocyte type. Cell Differentiation 4, 87-99.

DRYSDALE, J. W., RIGHETTI, P. & BUNN, H. F. (1971).The separation of human and animal hemoglobins byisoeletric focusing in polyacrylamide gel. Biochim.biophys. Ada 229, 42-50.

Fucci, L., CIROTTO, C , TOMEI, L. & GERACI, G. (1983).Synthesis of globin chains in the erythropoietic sites ofthe early chick embryo. J. Embryol. exp. Morph. 77,153-165.

Fucci, L., VITALE, E., CIROTTO, C. & GERACI, G. (1987).Evidences that hemoglobin switch in the chick embryodepends on erythroid cell line substitution. CellDifferentiation 20, 55-63.

GANDER, E. S., LUPPIS, B., STEWART, A. & SCHERRER, K.

(1972). Dissociation and reassociation of globin-synthesizing polyribosomes from immature avian redcells. Eur. J. Biochem. 29, 369-376.

GODET, J. (1974). HbF synthesis in chicken embryonicand postnatal development: studies in variousexplanted erythropoietic tissues. Devi Biol. 40,199-207.

HUNT, T., HUNTER, T. & MUNRO, A. (1969). Control ofhaemoglobin synthesis: rate of translation of themessenger RNA for the a- and jS chains. J. molec. Biol.43, 123-133.

INGRAM, V. M. (1958). Abnormal human haemoglobins.I. The comparison of normal human and sickle-cellhaemoglobins by fingerprinting. Biochim. biophvs. Ada28, 539-544.

KEANE, R. W., ABBOTT, U. K., BROWN, J. L. & INGRAM,

M. V. (1974). Ontogeny on hemoglobins: evidence forhemoglobin M. Devi Biol. 38, 229-236.

KNOCHEL, W., WITTIG, B., WITTIG, S., JOHN, M. E.,GRUNDMANN, V., OBERTHUR, W., GODOVAC, J. &BRAUNITZER, G. (1982). No evidence for "stress" a-globin genes in chicken. Nature, Land. 295, 710-712.

LEHMANN, H. & HUNTSMAN, R. G. (1974). Man'shaemoglobins, pp. 431-448. Amsterdam: NorthHolland.

MALIK, N. & BERRIE, A. (1972). New stain fixative forproteins separated by gel isoelectric focusing based onCoomassie Brilliant Blue. Analyt. Biochem. 49,173-176.

MATSUDA, G., MAITA, T., MIZUNO, K. & OTA, H. (1973).Amino acid sequence of a /3 chain of An component ofadult chicken haemoglobin. Nature New Biol. 244, 244.

MATSUDA, G., TAKEI, H., WU, K. C. & SHIOZAWA, T.

(1971). The primary structure of the a polypeptidechain of An component of adult chicken hemoglobin.Int. J. Protein Research III, 173-174.

Moss, B. A. & HAMILTON, E. A. (1974). Chickendefinitive erythrocyte haemoglobins. Biochim. biophys.Ada 371, 379-391.

ROMANOFF, A. L. (1960). The Avian Embryo, pp.571-678. New York: The Macmillan Co.

Rossi FANELLI, A., ANTONINI, E. & CAPUTO, A. (1958).Studies on the structure of haemoglobin.I. Physicochemical properties of human globin.Biochim. biophys. Ada 30, 608-615.

SCHALEKAMP, M., SCHALEKAMP, M., VAN GOOR, D. &SLINGERLAND, R. (1972). Re-evaluation of the presenceof multiple haemoglobins during the ontogenesis of thechicken. /. Embryol. exp. Morph. 28, 681-713.

SCHALEKAMP, M. & VAN GOOR, D. (1984). Haemoglobinand globin synthesis in the isolated primitive anddefinitive erythroid cells of chicken embryos. Evidencefor a non-clonal mechanism at the haemoglobin switch.J. Embryol. exp. Morph. 84, 125-148.

SHIMIZU, K. (1976). Identification of hemoglobin typescontained in single chicken erythrocytes by fluorescentantibody technique. Devi Biol. 48, 317-326.

TAKEI, H., OTA, Y., WU, K. C, KIYOHARA, T. &MATSUDA, G. (1975). Amino acid sequence of the a-chain of chicken AT hemoglobin. J. Biochem. 77,1345-1347.

VANDECASSERIE, C , PAUL, C , SCHNEK, A. G. & LEONIS,J. (1975). Probable identity of the fl chains from thetwo chicken hemoglobin components. Biochimie 57,843-844.

WALEY, S. G. & WATSON, J. (1953). The action of trypsinon polylysine. Biochem. J. 55, 328-335.

{Accepted 12 August 1987)