Multiple origins of the i heterokaryon gene in Rockefeller-Lindegren strains

J. F. Wilson and M. L. Holden, Biology Department, University of North Carolina at Greensboro, Greensboro, NC 27402

The Fungal Genetics Stock Center has been able to revive four old lyophilized cultures of the mutants used by Garnjobst in her pioneering 1953 study of vegetative incompatibility in Neurospora crassa. One of them, inositolless (37401), tests as het-c,* het-d, het-E, as described by Garnjobst. This strain cannot be tested completely for het-I/i, but it is not het-I/i compatible with the FGSC riboflavinless het-cdE I tester strain. The other early cultures were the standard Rockefeller-Lindegren het-CDEI. A lineage chart of the het-e strains in the Wilson-Garnjobst heterokaryon testers shows that many of the inl testers are het-i, which probably accounts for the erratic performance observed in growth tubes with these testers. A list of het-I/i genotypes of 31 FGSC strains is included.

______________________________________________________________________________

The recently reported Hi/hi genes in Rockefeller-Lindegren strains are similar, if not identical, to the I/i genes in Oak Ridge-St. Lawrence strains

(J.F. Wilson, et al.,1999. Fungal Genet. Newsl. 46:25-30). Assuming identity, (I/i ) will be used in the rest of the article in place of (HI/hi).

Het-I/i is unlike the other heterokaryon incompatibility genes that prevent the initial formation of a heterokaryon by killing the fused hyphal segments. In I/i combinations, forced heterokaryons form vigorous heterokaryons initially, but these heterokaryons are unstable, and become homokaryotic on transfer, due to elimination of the het-i component. Most, but not all OR-SL nutritional mutants examined are het-i genotype, and most RL mutants tested are het-I genotype, suggesting that the OR-SL wild types are het-i; the RL wild types; het-I , although they cannot be tested directly. However, in a cross between a RL het-I nutritional mutant and a St. Lawrence wild type carrying a presumed het-i gene, only one incompatibility gene segregated (Wilson, et al., 1999).

Since the het-I genotype seemed to be the norm in RL strains, it was surprising to find that five inositolless f3 het-CDE strains sent to us by Dr. Garnjobst in 1960 were het-i instead of het-I. These isolates were only three generations removed from the original inositolless mutant produced from a Lindegren wild type by Beadle and Tatum (G.W. Beadle and E.L. Tatum, 1945. Am. J. Bot. 32:678-686). We know that the present RL wild types were developed directly from the Lindegren wild types, and the two wild types are compatible in all other het genes. (Beadle and Tatum, 1945; D. Newmeyer, et al., 1987, Fungal Genet. Newsl. 34: 46-51).

We asked FGSC to send us the earliest representatives they had of the five mutants Dr. Garnjobst used in her initial experiments (Laura Garnjobst, 1953. Am. J. Bot. 40:607-614; Beadle and Tatum, 1945). Dr. McCluskey was able to revive four of the mutants from early lyopilized cultures (Kevin McCluskey, 2000. Fungal Genet. Newsl. 47:110)). We used our own silica gel stock as a substitute for the missing fifth mutant, al-2 a(15300).

These cultures, with isolation numbers and lyopil dates were:

inl A (37401)1958

pan-1 A (5531)1946

nic-3 A (Y31881) 1956

rib-2 a (Y30539) 1962

al-2 a(15300) 1961 (silica gel)

We first checked nutritional genotype and mating type,then determined the het-CDE genotype with known het-CDE testers (which are not necessarily the same as FGSC testers), and finally, tested for het-I/i genotype, using the protocol outlined in J.F.Wilson, et al., 1999.

The five mutants were tested against all combinations of het-C/c and D/d, but only in het-E at first, on the assumption that het-E was probably unchanged. If het-E had changed we would know soon enough, because all the combinations would be incompatible.The results of all tests except those of inl (37401) were positive with het-CDE testers only (FGSC 2000. Catalog of Strains, Part VII.D.1). The five mutants were then tested against all het-e combinations, and all tests were negative.

The mutant, inl (37401), gave a positive heterokaryon test solely with

het-cdE. As an additional test, hyphal fusions between inl and all het-E combinations were examined. Interstrain incompatibility reactions were observed with het-CDE, cDE, and CdE, but fusion and continuing flow were seen through interstrain fusions with het-cdE.

The results of the I/i tests for all mutants except inl were also unambiguous. All were I, suggesting strongly again that the original Lindegren wild types were het-C,D,E, I. The inl culture could not be tested thoroughly because we had no testers of genotype het-cdE with known het-I/i genotypes. However, we tested the inl culture against rib-2 (y30539) het-cdE A, FGSC 475, and it was het-I/i incompatible. Limited heterokaryon tests across a single gene het-E/e difference, while not definitive, suggest that FGSC 475 is het-I. If so, the 1958 inl mutant is het-i. More crosses will be needed to provide additional proof.

If the 1958 inl is indeed het-i, it could account for the presence of the allele in Rockefeller-Lindegren mutant stocks prior to the het E/e work (Wilson, J.F.and L.Garnjobst 1966. Genetics 53:(No. 3) 621-631). The het-E/e research and consequent production of the het-testers introduced still more het-i genes, since the OR-SL wild-type strain we used in the initial cross was het-i. What we did not know at the time was that the inl (37401) f3 het-CDE mutant we picked as the other parent was also het-i! The progeny of this cross was the source of all inl het-e combinations placed in FGSC. Chart 1 shows the lineage of the inl het-e FGSC tester strains. The lineage of the pan-1; al-2 (5531;15300) FGSC testers is not available for publication yet. Our records show that het-I/i genotypes of the critical early pan-1;al-2 parents were never checked, but we have them in silica gel. However, since all the inl het-e FGSC testers are probably het-i, two or three serial transfers of heterokaryons of inl and the companion pan-1;al-2 strains should establish their het-I/i. genotype. As noted below, some combinations may give ambivalent responses.

Tables 1 and 2 list all the FGSC strains we have tested for het-I/i. Table 2 illustrates the problems encountered in determining the het-I/i genotypes of some het-Cde strains. In het-Cde and other combinations we have found a number of strains of uncertain het-I/i genotype. The apparent ambivalence in OR strains may be due to the operation of the "Pittenger effect," in which het-i nuclei are not eliminated when they reach a threshold of 70% of the nuclei in the heterokaryon. The results were unambiguous in all the het-CDE cultures (over a hundred) we have tested. The "Pittenger effect" does not seem to be operable in a het-CDE environment (Wilson, et al.1999).

It should be emphasized again that differences in het-I/i in the Wilson-Garnjobst het testers do not interfere with het-CDE determinations as long as growth tube tests are avoided. Jacobson, et al. (1995 Fungal Genet. Newsl. 42: 34-40) encountered just such a problem with their growth tube tests of the Wilson-Garnjobst testers. We ran the same tests on minimal slants and found that all except one of the pairs they found to be erratic were het I/i incompatible. The one exception was a combination including a mat mutant. This combination was perfectly compatible in our serial transfer tests (Wilson, et al.1999).

If growth tubes are used, het-I/i differences will become apparent in 36-48 hr as the supply of het-i nuclei runs out. In contrast, if one grows the components at 30°C for 24 hr, and inoculates minimal slants with superimposed aerial hyphae of the two strains, the resulting heterokaryon, if fully compatible, will cover the slant in 24 hr or less at 30°C. Mixed aerial hyphal-conidial inocula should be avoided, because het-C/c and het-E/e differences can be blurred. Heterozygosity in het-I/i shows up with 1-3 serial transfers of the heterokaryon (Wilson, et al.1999). In het-I/i tests one can use mixed mycelial-conidial inocula with no problem.

There seems to be no way of proving that this lyophilized sample of inl(37401) is indeed a subculture of the original Beadle-Tatum inl(37401). However, it has the same het-cd genotype that Garnjobst attributed to the original mutant. In addition, all the cultures Dr. Garnjobst ever sent to our laboratory were clearly and extensively labelled with mutant name, isolation number, cross number, ascus and ascospore number, and filial or backcross number. This culture was simply marked inos 37401 A, and it was deposited by Dr. Garnjobst about the time she left Stanford to go to Rockefeller University.

We do not know whether the information is available, but it would be of interest to know whether subsequent early inl mutants were tested against this one for complementation. If they were, we may have an explanation for the curious lack of new inl mutations. A two-gene difference, involving het-C/c,D/d

would certainly have been enough to prevent the formation of the heterokaryon.

* We are using the older terminology for het-C OR/het-c PA because combining these superscripts with het genes D/d and E/e makes the text difficult to read.

 

 

Chart 1. Lineage of het-e Tester Strains in FGSC

 

1) Incompatible b9 St. L het-CDe i a X compatible b9 St. L het-Cde i A

|

ascus 23

1, 2 het-Cde i A

5, 6het-Cde i a

3, 4het-CDe i A

7, 8het-CDe i a

2) 23-5 het-Cde i a X inl 37401-f3-1 (4-5) CDE i A *

|

inl 37401 het-CDe i

inl
37401 het-Cde i

3)inl 37401 het-Cde i X inl 37401 het-cdE i

|

inl 37401 het-cde i

 

4)inl 37401 het-Cde i X inl 37401 het-cDE i

|

inl 37401 het-cDe i

*I/i genotype not known at the time of the cross

Analysis of the heterokaryon genotype of the Oak Ridge-St. Lawrence wild types started with twenty isolates sent to us by Dr. St. Lawrence (Wilson, J.F.and L. Garnjobst 1966. Genetics 53:(No. 3) 621-631). These isolates were from a sib cross of b9 isolates arising from a series of back crosses to ST74A. Only half the mat-A isolates were compatible with ST74A, as determined by the presence or absence of incompatibility reactions at hyphal fusions. Similar results were obtained in tests of the mat-a isolates against 74-OR21a. Above is a reconstruction of the lineage of what is now known about the presence of het-I/i in the Wilson-Garnjobst heterokaryon testers in FGSC. Although all the inl het e testers cannot be checked, the inadvertent selection of inl het-i A for the first cross suggests that they are het-i .

Table 1. Het I/i Genotypes of Some FGSC Cultures

het

type
mating

type
het-

I/i

marker

FGSC #
CDE A I rib-2 478
CDE A i inl 479*
CDE A I pan-1;al-2 1423
CDE a I pan-1;al-2 1427
Cde A I inl 1453
Cde a i arg-12 1527
Cde a i inl 1438
Cde A i pan-1;al-2 2658
Cde a i ad-2 ;al-2; 956
Cde a I pan-1;al-2 2657
Cde A i alcoy 997
Cde a i alcoy 998
Cde a i ad-3B 4564
Cde a i cot-1 4066
Cde a i arg-12 1527
Cde A I arg-5;ylo 6828
CDe A I pan-1;al-2 2656
CDe a i inl 1439
CDe a I pan-1;al-2 2661
cDE A i pan-1;al-2 1425
cDE A I inl;al-2 476
cDE a I pan-1;al-2 142

* No longer listed in FGSC collection, but JFW strain is healthy.

Table 2. Ambivalent Strains

het

type

mating

type

 

het-I/i

 

marker

 

FGSC #

Cde

a

I/i <

his-1

680

Cde

A

I/i <

his-1

681

Cde

A

I/i <

his-3

455

Cde

A

I/i <

his-5

456

Cde

A

I/i <

his-6

457

Cde

A

I/i <

ad-6

664

Cde

A

I/i <

ad-1

672

Cde

A

I/i <

ad-8

451

Cde

a

I/i <

arg-5;ylo

6829

< Inconsistent results with Cde testers of known I/i genotype.


Return to the FGN 48 Table of contents