Compensatory Regulation of Two Closely Related Hemoglobin Loci in Chzronomus T E N T a N S 2

Regulatory mutants in the hemoglobin system of Chironomus larvae can be detected by shifts in the relative quantity of specific components among the 10-12 mixed monomers. One such mutant, at or near a hemoglobin structural locus in the right tip of chromosome 3, greatly increases the quantity of that minor hemoglobin and greatly diminishes the quantity of the hemoglobin normally most abundant. Heterozygotes for the mutant are quantitatively intermediate, suggesting a transcriptional basis.-Structural similarities of the two hemoglobins indicate a close evolutionary relationship, and their interdependent but non-coordinate regulation is interpreted as competition for a common factor which functions in transcription. If evolutionary duplication of both structural genes and promotors is assumed, this factor may be a sigma subunit of RNA polymerase which recognizes similar but non-identical promotor regions. Parallels in the compensatory control of human hemoglobins are described. HE larval hemoglobins of Chironomus tentans provide an unusually favorTable situation for quantitative studies of synthesis within a family of proteins which, by their evolution from a single ancestral globin form, are related in structure and function and are encoded by multiple loci in various stages of evolutionary divergence. To summarize the salient features: ( 1 ) Individual larvae have as many as 10-12 different hemoglobins, combined in solution in the hemolymph of the open circulatory system (THOMPSON and ENGLISH 1966; TICHY 1968). (2) In Chironomus tentans, these hemoglobins are all monomers with a molecular weight of about 15,900 and are apparently encoded by an equivalent number of structural genes (THOMPSON, BLEECKER and ENGLISH 1968; TICHY 1968). Quantitative analysis is presumably not complicated by regulatory mechanisms involving dimer or tetramer assembly. ( 3 ) The composite hemoglobins comprise an estimated 40% of the total dissolved proteins of the hemolymph (BRAUN 1964; BRAUN, FORMANEK and BRAUNITZER 1968), and are so much more abundant than any other specific protein that extraordinarily small samples, with no prior purification, can be used effectively in routine electrophoretic analysis. Dedicated to Professor CURT STERN on the occasion of his retirement. Supported by Grant GB129Ol from the National Science Foundation. Genetics 5 0 : 275-290 February, 1972. 276 P. THOMPSON A N D G. PATEL (4) Although phenotypically visible mutants are not extensively known in Chironomus, many of the hemoglobin factors can be mapped in racial or species crosses (Ch. tentans x Ch. pallidiuittatus) by their linkage to cytologically visible inversions. This correlation of hemoglobin types with salivary chromosome analysis has shown that at least 7 hemoglobin factors are linked on chromosome 3 (TICHY 1968, 1970; THOMPSON 1968; THOMPSON, ENGLISH and BLEECKER 1969), and that they are distributed in at least 3 well-spaced regions (TICHY 1970). (5) Differences within species usually involve relative quantity of the various hemoglobins, rather than numbers and kinds of different components (THOMPSON. unpublished). Genetic heterozygosity rarely contributes to the total number of hemoglobins (THOMPSON and ENGLISH 1966). (6) Structural relationships among the 10 or so larval hemoglobins of a related species, Ch. thummi, have been studied in the greatest detail by BRAUTITZER and his associates. These studies have included the determination of Nterminal and C-terminal amino acids, amino acid compositions and fingerprinting (BRAUN 1964; BRAUNITZER and BRAUN 1965; BRAUN, CRICHTON and BRAUNITZER 1968) and to some extent amino acid sequences (BR.~L--NITZER and NEUWIRTH 1968; BUSE, BRAIG and BRAUNITZER 1969). The various chain types show a wide range of similarity and difference, but are obviously evolved from a common ancestral form. A similar situation is assumed in Ch. tentans on the basis of comparable multiplicity and identical molecular weight. These multiple hemoglobins offer another important advantage over the enzymatic proteins in studies of rate of synthesis. The assay for a particular component is independent of activity and can be based on a straightforward physical parameter, usually absorptioll at 412 mp as an index of quantity of heme in that electrophoretic component. Furthermore, the pattern of relative quantity of the various hemoglobins is only slightly variable among individuals from any of the strains reported here, so that total hemoglobin or other peak heights provide an internal standard of quantity for any single hemoglobin. Finally, differences in relative quantity from one developmental stage to another are consistent among individuals, so that strain variations are easily separated from developmmtal changes in rate of synthesis. An exception to the observation that most intraspecific variations are minor shifts in relative quantities of hemoglobins is found in a quantitative “polymorphism” among larvae from a locality in western Iowa. This variation. apparently due to a single regulatory mutation, greatly decreases the relative quantity (and synthesis) of what normally would constitute the most abundant hemoglobin. At the same time, this mutation greatly increases the synthesis of an erstwhile minor hemoglobin, so that in mutant larvae it represents the major component. Although the loci of the structural genes for these hemoglobins appear to be closely linked and although the hemoglobins themselves are quite similar in primary structure. the pattern of quantitative change from normal to mutant is essentially the reverse of coordinate control as found in bacterial COMPENSATORY REGULATION 277 systems, and is mare similar to the compensatory changes in quantity associated with ,&thalassemia in man. MATERIALS A N D METHODS Chironomus stocks: The larvae for this study came from laboratory populations deriving from collections at two localities only five miles apart, in order to minimize differences of genetic background. Larvae showing the pattern normally found in North American collections were from a collection in the west bay of Lake Okoboji, Dickinson Co., Iowa. Larvae showing the mutant pattern of hemoglobin quantities were from a collection at Jemmerson Slough, Dickinson Co., Iowa. The mutant, which was initially found in a number of heterozygous individuals, was established in a homozygous stock by analyzing a small quantity of hemolymph (about 0.025 ml) to identify mutant individuals without killing them, then mating adults raised from these larvae. The designation “Jemmerson” refers to this homozygous mutant strain. Analyiical polyacrylamide gel electrophoresis: Polyacrylamide disc gels for scanning and quantification were prepared by a modification of the method of ORNSTEIN (1964) and DAVIS (1964), as described by BLEECKER (1969). In this continuous system, both gel and chamber buffers consisted of 0.0023 M Tris and 0.0413 M glycine at pH 8.3. Gels were prerun in the Buchler electrophoresis apparatus for 2 h r in order to remove the catalysts. Larvae were punctured anally in order to obtain adequate hemoglobin samples with very low mortality. The unrefined hemolymph was combined with a small quantity of reference dye (brom phenol blue) in 10% sucrose prior to sample application. Electrophoresis of samples was carried out at 0.75 mA/tube and was complete when the dye had traveled 6 cm, usually in 15-20 min. The system was cooled to 5°C with a Savant recirculating water cooler. The standard procedure for analyzing hemoglobin separations was to scan the unstained gel in its glass tube at 412 rqp, using the Gilford Model 240 spectrophotometer with Model 2410 linear transport. Another scan was carried out at 605 m p to place the dye peak, and R, value for each hemoglobin component was based on the latter. Preparaiion of indiuidual hemoglobins: Large numbers of larvae (300-1000) were washed, cut open and bled into Tris-glycine buffer identical to that used in electrophoresis. The larval debris was removed by centrifugation at 40,000 x g for 30 min. The supernatant was reduced in volume with dry Sephadex G-25 in Seprafuge tubes (Occomy Associates) and a 5 ml solution was filtered by passage through a column of Sephadex G-100 equilibrated with Tris-glycine buffer. At this point, about 0.5 mg total hemoglobin per larva is normally recovered. The single hemoglobin peak from Sephadex filtration was collected, reduced in volume, and applied to a polyacrylamide gel column in the Buchler Poly-Prep apparatus for preparative electrophoresis. A continuous system identical to that described for analytical electrophoresis was used, except that the concentration of Tris and glycine in the lower buffer chamber and membrane holder was increased three fold. Separations very similar to those from analytical runs were obtained. The eluted peaks from preparative electrophoresis were then electrophoresed once more with the continuous system in an E-C vertical slab apparatus. Clearly separated bands were cut from the gel, the hemoglobin eluted with Tris-glycine buffer, and its purity tested by analytical electrophoresis of a small aliquot. Further evidence for purity was found in the identification of single N-terminal amino acids. Determination of N-ierminal residues: A 5 mg sample of each individual hemoglobin, following isolation by repeated electrophoresis, was precipitated with cold trichloroacetic acid and washed with distilled water. The protein was redissolved in a small volume of 10% sodium bicarbonate and reacted with dinitrofluorobenzene to produce DNP labelling of free amino groups, in order to identify the amino-terminal residue. The hydrolysis of DNP-protein was carried out with 6 N hydrochloric acid, refluxed for 6 hr. Aqueous and ether-soluble fractions were washed and evaporated to near dryness. The dried amino acids were redissolved, ether-soluble in acetone and aqueous in 0.5 N hydrochloric acid, and were chromatographed with standard DNP amino acids (Sigma Chemical CO.) using solvent systems described by BRENNER, NIEDERWIESER and PATAKI (1965). 278 P. THOMPSON A N D G . PATEL Tr>.ptic digestion and fingerprinting: A 20 mg sample of an isolated hemoglobin was precipitated with 20% trichloroacetic acid at 5°C for 12 hr. The denatured protein was washed with distilled water and suspended in 3 ml of 2% ammonium bicarbonate. A 0.5 mg quantity of trypsin (Worthington TPCK; chymotryptic activity inhibited with L (1 -Tosylamido-%phenyl) ethyl chloromethyl ketone) was added and the suspension was mixed for 8 hr at 38°C. The reaction was terminated by lowering the pH to 5.8 with 2 N acetic acid. Trypsin and possible undigested protein were removed by heating to 90°C for 5 min and centrifugation at 15,000 x g for 30 min. The supernatant was lyophilized and the residue re-dissolved in a 1% pyridine solution. Fingerprinting was carried out essentially as described by INGRAM (1958). A sample was spotted on Whatman 3 MM paper and electrophoresis was applied at 3600 volts for 90 min in a Savant high-voltage tank. The paper was dried for 2 hr and turned 90" for chromatography in a 3 n-butanol: 1 glacial acetic acid: 1 water solvent. The paper was first placed in the chromatography cabinet with fumes of the solvent for 2 hr, then made contact with the solvent for 16 hr. Peptides were localized by dipping the paper in ninhydrin solution and heating to 80°C for 20 min.