Chemically Defined Media for Growth of Haemophilus influenzae Strains

Abstract

A chemically defined medium that supports the growth of both encapsulated and nontypeable Haemophilus influenzae strains in broth to densities that are ≥ 10⁹ CFU/ml or on agar plates is described. The mean generation time of a panel of clinical isolates was comparable to that in rich, chemically undefined media (brain-heart infusion broth supplemented with heme and β-NAD).

Haemophilus influenzae is a fastidious, gram-negative coccus that inhabits the upper respiratory system of humans and has an obligate requirement for heme and β-NAD for aerobic growth. Prior investigators studying Haemophilus influenzae sought a chemically defined medium to facilitate genetic and metabolic studies. Multiple defined media were devised for use with Rd derivatives of H. influenzae (2, 3, 4, 5, 6, 8, 13, 15), but these media would not support the growth of many nontypeable clinical isolates or encapsulated strains.

During the course of experiments studying the invasion of human cell lines by pathogenic H. influenzae, we found that RPMI medium-based tissue culture media appeared to support bacterial growth. Further studies led to the development of the chemically defined media described herein.

Table 1 describes the strains used in this study, which were stored at −70°C in 10% skim milk and were subcultured onto chocolate agar. One liter of chocolate agar was prepared by adding 36 g of GC base (catalog no. 228920; Difco, Detroit, Mich.) and 10 g of hemoglobin (catalog no. ZIZ392; BD Biosciences, Sparks, Md.) and autoclaving at 121°C at 15 lb/in² for 20 min; after cooling to 55°C, 5,000 U of bacitracin and 10 ml of GCHI Rehydrating Solution (catalog no. 450411; Remel, Lenexa, Kans.) were added and the plates were poured. Supplemented brain-heart infusion (sBHI) agar was prepared by adding 37 g of BHI media (catalog no. 211059; BD Biosciences) and 15 g of BactoAgar (catalog no. 214530; BD Biosciences) to 1 liter of deionized water, autoclaving, cooling to 55°C, and adding 10 ml of β-NAD and 10 ml of heme and l-histidine stock. The heme-histidine stock was prepared by dissolving 0.2 g of l-histidine (freebase; Sigma catalog no. H 8000) in 200 ml of deionized water and then adding 0.2 g of hemin HCl (catalog no. H 5533, bovine; Sigma) and 4 ml of 1 N NaOH and steaming over a boiling water bath for 5 to 10 min to solubilize the mixture. The solution was then cooled to room temperature, filter sterilized (0.2-μm pore size), and placed in a foil-covered bottle at 4°C. We found that the histidine decreased the rate at which the hemin precipitated from solution. β-NAD⁺ stock was prepared by dissolving 100 mg of β-NAD⁺ (catalog no. N 7004; Sigma) in 100 ml of deionized water, filter sterilizing (0.2-μm pore size), and storing at 4°C. sBHI broth was prepared without the agar.

TABLE 1.

H. influenzae strains used in this study

Strain	Origin	Supplierb (strain no.)	Reference
R538	Reference type b	ATCC (9795)
R539	Reference type a	ATCC (9006)
R540	Reference type c	ATCC (9007)
R541	Reference type d	ATCC (9008)
R542	Reference type e	ATCC (8142)
R543	Reference type f	ATCC (9833)
R652	Rd KW20	H. O. Smith	14
R2846	Middle ear isolate from an otitis media patient	S. Barenkamp	1
R2866	Nontypeable blood isolate	Author's collection	9
R3001	Bronchoalveolar lavage isolate from a pediatric cystic fibrosis patient	Author's collection	NAa
R3157	Middle ear isolate from an otitis media patient	L. Bakaletz	7
R3164	Middle ear isolate from an otitis media patient	S. Barenkamp	1
DB117	recA mutant Rd	J. Setlow	10
E1a	Type b originally isolated from cerebrospinal fluid	Author's collection	11

^aNA, not applicable.

^bATCC, American Type Culture Collection.

Defined liquid medium was made with the following: 191 ml of RPMI 1640 with l-glutamine and 25 mM HEPES, pH 7.26 (catalog no. 61870036; InVitrogen), 2 ml of a 100 mM MEM sodium pyruvate solution (catalog no. 11360070; InVitrogen), 2 ml of β-NAD⁺ stock, 4 ml of heme-l-histidine stock, 10 ml of a 2-mg/ml uracil solution (catalog no. U 0750; Sigma) dissolved in 0.1 N NaOH, and 20 ml of a 20-mg/ml inosine solution (catalog no. I 4125; Sigma) dissolved in deionized water and filter sterilized (0.2-μm pore size). The final pH of the liquid medium was 7.56. Where specified, NaHCO₃ (2 g/liter) was added to the liquid medium.

Defined medium solidified with agar was made as follows: a package of RPMI 1640 powder containing l-glutamine and 25 mM HEPES (pH 7.26) (catalog no. 31800022; InVitrogen), which when reconstituted will make 1.0 liter of medium, was added to 334 ml of deionized H₂O, 8.7 ml of 100 mM MEM sodium pyruvate solution (as described above), 8.7 ml of β-NAD (1 mg/ml), 17.5 ml of heme-histidine stock, 44 ml of uracil (2 mg/ml), 87 ml of inosine (20 mg/ml), and 1 ml of bacitracin (5,000 U/ml). This 2× solution was adjusted to a pH of 7.18 to 7.2, filter sterilized (0.2-μm pore size), and warmed to 55°C. Fifteen grams of BactoAgar was added to 500 ml of deionized H₂O, autoclaved for 20 min, and cooled to 55°C in a water bath. After the agar reached 55°C, the two solutions were mixed in a sterile fashion and were poured into 100-mm-diameter plates. Agar plates were incubated in either a 37°C Lab-Line Imperial II bacterial incubator or a NuAire 5% CO₂, 37°C water-jacketed tissue culture incubator.

Growth curves were performed at 37°C by using 125-ml Erlenmeyer flasks with a starting volume of 15 ml of medium (sBHI broth or liquid defined medium) shaken at 200 rpm in room air or in 5% CO₂. A 1-ml sample was taken from the flask at time zero and then at hourly intervals, and the A₆₀₀ was measured in a Hitachi U2000 spectrophotometer. The flask contents were incubated in a New Brunswick floor shaker or on a platform shaker placed inside the tissue culture incubator. The defined media allowed growth of all 14 strains tested (Table 2). All strains grew to an A₆₀₀ of >1.10 after 24 h of incubation, a turbidity equivalent to 10⁷ to 10⁹ CFU/ml. However, the final density of the Rd derivative R652 in defined media was about one-third that seen with sBHI broth: a mean of 3.5 × 10⁹ versus 9.0 × 10⁹ CFU/ml. Strains E1a and R2866 reached densities of 9.9 × 10⁷ and 7.3 × 10⁸ CFU/ml, respectively.

TABLE 2.

Turbidity of H. influenzae cultures in defined media in air at 37°C

Strain	Inoculuma	A600 at:
		7 h	24 h
R652	4.3	1.23	1.44
DB117	4.6	1.03	1.28
E1a	4.3	1.36	1.45
R2866	5.1	1.02	1.40
R3001	3.6	0.97	1.32
R3157	5.2	0.95	1.28
R2846	4.1	0.63	1.27
R3164	5.1	0.85	1.14
R538	5.2	1.38	1.52
R539	4.6	0.94	1.24
R540	4.3	1.13	NDb
R541	4.5	1.36	ND
R542	4.6	1.27	ND
R543	4.3	1.31	ND

^aValues given are initial inocula in numbers of CFU/milliliter (10⁶).

^bND, not determined.

Representative H. influenzae strains, R652 (a reference strain), E1a (a prototypic type b), and R2866 (an invasive, unencapsulated strain), were chosen for further study. To estimate mean generation time, 0.1 ml was removed at the end of the lag phase and colony counts were repeated in duplicate at a minimum of three time points until the stationary phase was reached. With strains E1a and R2866, the mean generation time was slightly shorter in sBHI broth when the flasks were shaken in an environment of 5% CO₂ than when they were shaken in room air (Table 3). The mean generation time of the prototypic strain R652 grown in defined medium and incubated in room air was 35.66 ± 9.43 min (n = 7), while in the CO₂-enriched environment it was 31.36 ± 1.15 min (n = 3), differences that were not statistically significant.

TABLE 3.

Mean generation time of selected strains grown in sBHI broth or defined media when incubated in air or 5% CO₂^a

Strain	Medium		Environmentb		n	MGT (SD)
	sBHI	Defined	Air	CO2
R652	x		x		11	30.71 (2.27)
R652	x			x	5	31.36 (1.15)
R652		x	x		7	35.66 (9.43)
E1a	x		x		6	30.78 (2.00)
E1a	x			x	6	26.21 (3.73)
E1a		x	x		7	33.84 (5.12)
R2866	x		x		9	40.56 (6.38)
R2866		x	x		8	49.34 (9.95)
R2866	x			x	7	31.89 (2.83)

^an, the number of replicates; MGT, mean generation time in minutes.

^bx, strain was grown in indicated medium or environment.

We observed that, after 7 h of incubation, phenol red in the defined medium changed from pink to yellow and that there was a strain dependence and incubation environment effect on apparent acidification. To more precisely define the apparent acidification, we measured the pH of sBHI broth and defined media after 18 h of incubation in air and in a CO₂-enriched environment. In general, there was acidification of sBHI broth in air and in 5% CO_2. The same trend was seen with defined media, but there was less acidification in 5% CO₂ (Table 4). Since defined media contain 23.8 mM NaHCO₃, we added the same concentration to sBHI broth and measured the overnight growth and pH (Table 4). In general there were less acidification and less growth. Doubling the NaHCO₃ concentration in defined media increased the final pH after incubation in room air but, as expected, had little effect in the 5% CO₂ environment.

TABLE 4.

Medium pH and turbidity after 18 h of incubation in room air or 5% CO₂^a

Environment	Strain	Result for:
		sBHI		sBHI + NaHCO3		Defined medium		Defined medium + NaHCO3
		pH	A600	pH	A600	pH	A600	pH	A600
Air	Blankb	7.36	0	7.9	0	7.35	0	6.77	0
	R652	5.98	1.16	8.29	0.34	5.13	1.26	6.94	1.23
	E1a	5.85	1.40	8.34	0.28	5.49	1.15	7.15	1.13
	R2866	5.90	1.13	7.76	0.38	5.11	1.12	6.97	1.13
5% CO2	Blank	7.22	0	7.8	0	7.32	0	7.30	0
	R652	6.45	0.97	7.51	0.89	6.61	1.37	6.60	1.15
	E1a	6.38	1.12	7.54	0.87	6.57	0.78	6.75	0.01
	R2866	6.96	1.18	7.50	0.91	6.11	0.85	6.11	1.39

^aValues are the means of results obtained with three or more replicates.

^bBlank is medium alone.

Since H. influenzae is a facultative anaerobe, we examined the growth of strains R652, E1a, and R2866 in liquid defined media under anaerobic conditions. Anaerobic incubations were performed by placing a 125-ml Erlenmeyer flask containing 15 ml of medium and a sterile stirring bar inside a Becton Dickinson anaerobic chamber by using the BBL GasPak Plus anaerobic system envelopes (catalog no. 271040; BD Biosciences) and BBL Dry Anaerobic Indicator strips (catalog no. 271051; BD Biosciences). The chamber was then placed on a magnetic stirrer inside a standard 37°C bacteriological incubator. The chamber was insulated from the surface of the stirrer by a 13-mm-thick styrofoam square. The cell density after overnight incubation was comparable to that observed in air; strain R652 reached 10⁹ CFU/ml, E1a reached 9.9 × 10⁷ CFU/ml, and R2866 reached 7.3 × 10⁹ CFU/ml. Defined media solidified with agar supported growth of all the strains listed in Table 1 incubated in air or 5% CO₂.

We conclude that this medium will facilitate metabolic studies of a wide variety of H. influenzae.